From 4352b8aee98005853aa63f57d6377282de17a33f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 23 Aug 2026 13:28:57 -0700 Subject: Merge tag 'i3c/for-7.3' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux Pull I3C updates from Alexandre Belloni: "A new driver this cycle, for the AMD AXI I3C controller. There are also new features such as IBI wakeup support, SETAASA device discovery and ACPI support for the the DAA methods, meaning we can now communicate with the SPD devices on DDR5 modules. New driver: - AMD AXI I3C master controller Subsystem: - I3C Common Command Code (CCC) handling improvements, especially around GET CCCs - SETAASA device discovery support - ACPI support for all existing DAA methods like SETDASA, SETNEWDA as well as I2C devices on I3C bus - IBI-based wakeup support Drivers: - dw: SETAASA support - mipi-i3c-hci: advertise IBI wakeup capability, AMD_PT I3C controller support, PIO queue management support for HCI v1.2 - renesas: improve suspend to RAM support, add runtime PM support" * tag 'i3c/for-7.3' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux: (70 commits) i3c: dw: reduce do_daa time if there's no client i3c: mipi-i3c-hci: Add support for AMD_PT I3C controller i3c: mipi-i3c-hci: Add PIO queue management support for HCI v1.2 i3c: mipi-i3c-hci: Fix missing STAT_IBI_STATUS_THLD in PIO mode i3c: dw: rename "pclk" to "apb" to match dt-binding i3c: dw: make struct dw_i3c_cmd smaller i3c: dw: use COMMAND_PORT_TRANSFER_ARG instead of hardcoding i3c: renesas: Don't register devices when ENTDAA times out i3c: master: dw-i3c-master: fix OD timing for first broadcast i3c: mipi-i3c-hci: Advertise IBI wakeup capability i3c: mipi-i3c-hci: Factor out i3c_hci_sysdev() i3c: mipi-i3c-hci-pci: Propagate I3C wakeup requirements to PCI i3c: master: Reject IBI requests from non-IBI-capable devices i3c: master: Add helper to query bus wakeup requirements i3c: master: Report wakeup events for IBIs i3c: master: Support IBI-based wakeup capability i3c: master: Fix potential UAF in i3c_device_match() i3c: master: Fix potential UAF in i3c_device_uevent() i3c: Make dev->desc locking assumptions explicit i3c: master: Fix use-after-free of master->this ... --- fs/nfs/Kconfig | 217 + fs/nfs/Makefile | 38 + fs/nfs/blocklayout/Makefile | 7 + fs/nfs/blocklayout/blocklayout.c | 1047 +++ fs/nfs/blocklayout/blocklayout.h | 204 + fs/nfs/blocklayout/dev.c | 599 ++ fs/nfs/blocklayout/extent_tree.c | 726 ++ fs/nfs/blocklayout/rpc_pipefs.c | 270 + fs/nfs/cache_lib.c | 158 + fs/nfs/cache_lib.h | 32 + fs/nfs/callback.c | 355 + fs/nfs/callback.h | 204 + fs/nfs/callback_proc.c | 756 ++ fs/nfs/callback_xdr.c | 1111 +++ fs/nfs/client.c | 1542 ++++ fs/nfs/delegation.c | 1619 +++++ fs/nfs/delegation.h | 141 + fs/nfs/dir.c | 3433 +++++++++ fs/nfs/direct.c | 1112 +++ fs/nfs/dns_resolve.c | 480 ++ fs/nfs/dns_resolve.h | 37 + fs/nfs/export.c | 166 + fs/nfs/file.c | 982 +++ fs/nfs/filelayout/Makefile | 6 + fs/nfs/filelayout/filelayout.c | 1155 +++ fs/nfs/filelayout/filelayout.h | 118 + fs/nfs/filelayout/filelayoutdev.c | 305 + fs/nfs/flexfilelayout/Makefile | 6 + fs/nfs/flexfilelayout/flexfilelayout.c | 3069 ++++++++ fs/nfs/flexfilelayout/flexfilelayout.h | 269 + fs/nfs/flexfilelayout/flexfilelayoutdev.c | 644 ++ fs/nfs/fs_context.c | 1793 +++++ fs/nfs/fscache.c | 387 + fs/nfs/fscache.h | 202 + fs/nfs/getroot.c | 133 + fs/nfs/inode.c | 2793 +++++++ fs/nfs/internal.h | 1012 +++ fs/nfs/io.c | 204 + fs/nfs/iostat.h | 61 + fs/nfs/localio.c | 1120 +++ fs/nfs/mount_clnt.c | 471 ++ fs/nfs/namespace.c | 386 + fs/nfs/netns.h | 48 + fs/nfs/nfs.h | 30 + fs/nfs/nfs2super.c | 33 + fs/nfs/nfs2xdr.c | 1088 +++ fs/nfs/nfs3_fs.h | 37 + fs/nfs/nfs3acl.c | 343 + fs/nfs/nfs3client.c | 144 + fs/nfs/nfs3proc.c | 1128 +++ fs/nfs/nfs3super.c | 34 + fs/nfs/nfs3xdr.c | 2516 +++++++ fs/nfs/nfs40.h | 19 + fs/nfs/nfs40client.c | 245 + fs/nfs/nfs40proc.c | 395 + fs/nfs/nfs42.h | 69 + fs/nfs/nfs42proc.c | 1687 +++++ fs/nfs/nfs42xattr.c | 1067 +++ fs/nfs/nfs42xdr.c | 1830 +++++ fs/nfs/nfs4_fs.h | 682 ++ fs/nfs/nfs4client.c | 1203 ++++ fs/nfs/nfs4file.c | 459 ++ fs/nfs/nfs4getroot.c | 39 + fs/nfs/nfs4idmap.c | 793 ++ fs/nfs/nfs4idmap.h | 68 + fs/nfs/nfs4namespace.c | 564 ++ fs/nfs/nfs4proc.c | 10797 ++++++++++++++++++++++++++++ fs/nfs/nfs4renewd.c | 160 + fs/nfs/nfs4session.c | 653 ++ fs/nfs/nfs4session.h | 158 + fs/nfs/nfs4state.c | 2713 +++++++ fs/nfs/nfs4super.c | 317 + fs/nfs/nfs4sysctl.c | 52 + fs/nfs/nfs4trace.c | 40 + fs/nfs/nfs4trace.h | 2859 ++++++++ fs/nfs/nfs4xdr.c | 7786 ++++++++++++++++++++ fs/nfs/nfsroot.c | 316 + fs/nfs/nfstrace.c | 15 + fs/nfs/nfstrace.h | 1986 +++++ fs/nfs/pagelist.c | 1496 ++++ fs/nfs/pnfs.c | 3535 +++++++++ fs/nfs/pnfs.h | 928 +++ fs/nfs/pnfs_dev.c | 381 + fs/nfs/pnfs_nfs.c | 1225 ++++ fs/nfs/proc.c | 781 ++ fs/nfs/read.c | 499 ++ fs/nfs/super.c | 1463 ++++ fs/nfs/symlink.c | 80 + fs/nfs/sysctl.c | 46 + fs/nfs/sysfs.c | 474 ++ fs/nfs/sysfs.h | 33 + fs/nfs/unlink.c | 532 ++ fs/nfs/write.c | 2212 ++++++ 93 files changed, 85428 insertions(+) create mode 100644 fs/nfs/Kconfig create mode 100644 fs/nfs/Makefile create mode 100644 fs/nfs/blocklayout/Makefile create mode 100644 fs/nfs/blocklayout/blocklayout.c create mode 100644 fs/nfs/blocklayout/blocklayout.h create mode 100644 fs/nfs/blocklayout/dev.c create mode 100644 fs/nfs/blocklayout/extent_tree.c create mode 100644 fs/nfs/blocklayout/rpc_pipefs.c create mode 100644 fs/nfs/cache_lib.c create mode 100644 fs/nfs/cache_lib.h create mode 100644 fs/nfs/callback.c create mode 100644 fs/nfs/callback.h create mode 100644 fs/nfs/callback_proc.c create mode 100644 fs/nfs/callback_xdr.c create mode 100644 fs/nfs/client.c create mode 100644 fs/nfs/delegation.c create mode 100644 fs/nfs/delegation.h create mode 100644 fs/nfs/dir.c create mode 100644 fs/nfs/direct.c create mode 100644 fs/nfs/dns_resolve.c create mode 100644 fs/nfs/dns_resolve.h create mode 100644 fs/nfs/export.c create mode 100644 fs/nfs/file.c create mode 100644 fs/nfs/filelayout/Makefile create mode 100644 fs/nfs/filelayout/filelayout.c create mode 100644 fs/nfs/filelayout/filelayout.h create mode 100644 fs/nfs/filelayout/filelayoutdev.c create mode 100644 fs/nfs/flexfilelayout/Makefile create mode 100644 fs/nfs/flexfilelayout/flexfilelayout.c create mode 100644 fs/nfs/flexfilelayout/flexfilelayout.h create mode 100644 fs/nfs/flexfilelayout/flexfilelayoutdev.c create mode 100644 fs/nfs/fs_context.c create mode 100644 fs/nfs/fscache.c create mode 100644 fs/nfs/fscache.h create mode 100644 fs/nfs/getroot.c create mode 100644 fs/nfs/inode.c create mode 100644 fs/nfs/internal.h create mode 100644 fs/nfs/io.c create mode 100644 fs/nfs/iostat.h create mode 100644 fs/nfs/localio.c create mode 100644 fs/nfs/mount_clnt.c create mode 100644 fs/nfs/namespace.c create mode 100644 fs/nfs/netns.h create mode 100644 fs/nfs/nfs.h create mode 100644 fs/nfs/nfs2super.c create mode 100644 fs/nfs/nfs2xdr.c create mode 100644 fs/nfs/nfs3_fs.h create mode 100644 fs/nfs/nfs3acl.c create mode 100644 fs/nfs/nfs3client.c create mode 100644 fs/nfs/nfs3proc.c create mode 100644 fs/nfs/nfs3super.c create mode 100644 fs/nfs/nfs3xdr.c create mode 100644 fs/nfs/nfs40.h create mode 100644 fs/nfs/nfs40client.c create mode 100644 fs/nfs/nfs40proc.c create mode 100644 fs/nfs/nfs42.h create mode 100644 fs/nfs/nfs42proc.c create mode 100644 fs/nfs/nfs42xattr.c create mode 100644 fs/nfs/nfs42xdr.c create mode 100644 fs/nfs/nfs4_fs.h create mode 100644 fs/nfs/nfs4client.c create mode 100644 fs/nfs/nfs4file.c create mode 100644 fs/nfs/nfs4getroot.c create mode 100644 fs/nfs/nfs4idmap.c create mode 100644 fs/nfs/nfs4idmap.h create mode 100644 fs/nfs/nfs4namespace.c create mode 100644 fs/nfs/nfs4proc.c create mode 100644 fs/nfs/nfs4renewd.c create mode 100644 fs/nfs/nfs4session.c create mode 100644 fs/nfs/nfs4session.h create mode 100644 fs/nfs/nfs4state.c create mode 100644 fs/nfs/nfs4super.c create mode 100644 fs/nfs/nfs4sysctl.c create mode 100644 fs/nfs/nfs4trace.c create mode 100644 fs/nfs/nfs4trace.h create mode 100644 fs/nfs/nfs4xdr.c create mode 100644 fs/nfs/nfsroot.c create mode 100644 fs/nfs/nfstrace.c create mode 100644 fs/nfs/nfstrace.h create mode 100644 fs/nfs/pagelist.c create mode 100644 fs/nfs/pnfs.c create mode 100644 fs/nfs/pnfs.h create mode 100644 fs/nfs/pnfs_dev.c create mode 100644 fs/nfs/pnfs_nfs.c create mode 100644 fs/nfs/proc.c create mode 100644 fs/nfs/read.c create mode 100644 fs/nfs/super.c create mode 100644 fs/nfs/symlink.c create mode 100644 fs/nfs/sysctl.c create mode 100644 fs/nfs/sysfs.c create mode 100644 fs/nfs/sysfs.h create mode 100644 fs/nfs/unlink.c create mode 100644 fs/nfs/write.c (limited to 'fs/nfs') diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig new file mode 100644 index 000000000..6bb30543e --- /dev/null +++ b/fs/nfs/Kconfig @@ -0,0 +1,217 @@ +# SPDX-License-Identifier: GPL-2.0-only +config NFS_FS + tristate "NFS client support" + depends on INET && FILE_LOCKING && MULTIUSER + select CRC32 + select LOCKD + select SUNRPC + select NFS_COMMON + select NFS_ACL_SUPPORT if NFS_V3_ACL + help + Choose Y here if you want to access files residing on other + computers using Sun's Network File System protocol. To compile + this file system support as a module, choose M here: the module + will be called nfs. + + To mount file systems exported by NFS servers, you also need to + install the user space mount.nfs command which can be found in + the Linux nfs-utils package, available from http://linux-nfs.org/. + Information about using the mount command is available in the + mount(8) man page. More detail about the Linux NFS client + implementation is available via the nfs(5) man page. + + Below you can choose which versions of the NFS protocol are + available in the kernel to mount NFS servers. Support for NFS + version 2 (RFC 1094) is always available when NFS_FS is selected. + + To configure a system which mounts its root file system via NFS + at boot time, say Y here, select "Kernel level IP + autoconfiguration" in the NETWORK menu, and select "Root file + system on NFS" below. You cannot compile this file system as a + module in this case. + + If unsure, say N. + +config NFS_V2 + tristate "NFS client support for NFS version 2" + depends on NFS_FS + default n + help + This option enables support for version 2 of the NFS protocol + (RFC 1094) in the kernel's NFS client. + + If unsure, say N. + +config NFS_V3 + tristate "NFS client support for NFS version 3" + depends on NFS_FS + default y + help + This option enables support for version 3 of the NFS protocol + (RFC 1813) in the kernel's NFS client. + + If unsure, say Y. + +config NFS_V3_ACL + bool "NFS client support for the NFSv3 ACL protocol extension" + depends on NFS_V3 + help + Some NFS servers support an auxiliary NFSv3 ACL protocol that + Sun added to Solaris but never became an official part of the + NFS version 3 protocol. This protocol extension allows + applications on NFS clients to manipulate POSIX Access Control + Lists on files residing on NFS servers. NFS servers enforce + ACLs on local files whether this protocol is available or not. + + Choose Y here if your NFS server supports the Solaris NFSv3 ACL + protocol extension and you want your NFS client to allow + applications to access and modify ACLs on files on the server. + + Most NFS servers don't support the Solaris NFSv3 ACL protocol + extension. You can choose N here or specify the "noacl" mount + option to prevent your NFS client from trying to use the NFSv3 + ACL protocol. + + If unsure, say N. + +config NFS_V4 + tristate "NFS client support for NFS version 4" + depends on NFS_FS + select KEYS + select SUNRPC_BACKCHANNEL + help + This option enables support for version 4.1 of the NFS protocol + (RFC 5661) in the kernel's NFS client. + + To mount NFS servers using NFSv4, you also need to install user + space programs which can be found in the Linux nfs-utils package, + available from http://linux-nfs.org/. + + If unsure, say N. + +config NFS_SWAP + bool "Provide swap over NFS support" + default n + depends on NFS_FS && SWAP + select SUNRPC_SWAP + help + This option enables swapon to work on files located on NFS mounts. + +config NFS_V4_0 + bool "NFS client support for NFSv4.0" + depends on NFS_V4 + default y + help + This option enables support for minor version 0 of the NFSv4 protocol + (RFC 3530) in the kernel's NFS client. + + If unsure, say N. + +config NFS_V4_2 + bool "NFS client support for NFSv4.2" + depends on NFS_V4 + help + This option enables support for minor version 2 of the NFSv4 protocol + in the kernel's NFS client. + + If unsure, say N. + +config PNFS_FILE_LAYOUT + tristate + depends on NFS_V4 + default NFS_V4 + +config PNFS_BLOCK + tristate + depends on NFS_V4 && BLK_DEV_DM + default NFS_V4 + +config PNFS_FLEXFILE_LAYOUT + tristate + depends on NFS_V4 + default NFS_V4 + +config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN + string "NFSv4.1 Implementation ID Domain" + depends on NFS_V4 + default "kernel.org" + help + This option defines the domain portion of the implementation ID that + may be sent in the NFS exchange_id operation. The value must be in + the format of a DNS domain name and should be set to the DNS domain + name of the distribution. + If the NFS client is unchanged from the upstream kernel, this + option should be set to the default "kernel.org". + +config NFS_V4_1_MIGRATION + bool "NFSv4.1 client support for migration" + depends on NFS_V4 + default n + help + This option makes the NFS client advertise to NFSv4.1 servers that + it can support NFSv4 migration. + + The NFSv4.1 pieces of the Linux NFSv4 migration implementation are + still experimental. If you are not an NFSv4 developer, say N here. + +config NFS_V4_SECURITY_LABEL + bool + depends on NFS_V4_2 && SECURITY + default y + +config ROOT_NFS + bool "Root file system on NFS" + depends on NFS_FS=y && IP_PNP + help + If you want your system to mount its root file system via NFS, + choose Y here. This is common practice for managing systems + without local permanent storage. For details, read + . + + Most people say N here. + +config NFS_FSCACHE + bool "Provide NFS client caching support" + depends on NFS_FS + select NETFS_SUPPORT + select FSCACHE + help + Say Y here if you want NFS data to be cached locally on disc through + the general filesystem cache manager + +config NFS_USE_LEGACY_DNS + bool "Use the legacy NFS DNS resolver" + depends on NFS_V4 + help + The kernel now provides a method for translating a host name into an + IP address. Select Y here if you would rather use your own DNS + resolver script. + + If unsure, say N + +config NFS_USE_KERNEL_DNS + bool + depends on NFS_V4 && !NFS_USE_LEGACY_DNS + select DNS_RESOLVER + default y + +config NFS_DEBUG + bool + depends on NFS_FS && SUNRPC_DEBUG + default y + +config NFS_DISABLE_UDP_SUPPORT + bool "NFS: Disable NFS UDP protocol support" + depends on NFS_FS + default y + help + Choose Y here to disable the use of NFS over UDP. NFS over UDP + on modern networks (1Gb+) can lead to data corruption caused by + fragmentation during high loads. + +config NFS_V4_2_READ_PLUS + bool "NFS: Enable support for the NFSv4.2 READ_PLUS operation" + depends on NFS_V4_2 + default y + help + Choose Y here to enable use of the NFS v4.2 READ_PLUS operation. diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile new file mode 100644 index 000000000..c895521f2 --- /dev/null +++ b/fs/nfs/Makefile @@ -0,0 +1,38 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Makefile for the Linux nfs filesystem routines. +# + +obj-$(CONFIG_NFS_FS) += nfs.o + +CFLAGS_nfstrace.o += -I$(src) +nfs-y := client.o dir.o file.o getroot.o inode.o super.o \ + io.o direct.o pagelist.o read.o symlink.o unlink.o \ + write.o namespace.o mount_clnt.o nfstrace.o \ + export.o sysfs.o fs_context.o +nfs-$(CONFIG_ROOT_NFS) += nfsroot.o +nfs-$(CONFIG_SYSCTL) += sysctl.o +nfs-$(CONFIG_NFS_FSCACHE) += fscache.o +nfs-$(CONFIG_NFS_LOCALIO) += localio.o + +obj-$(CONFIG_NFS_V2) += nfsv2.o +nfsv2-y := nfs2super.o proc.o nfs2xdr.o + +obj-$(CONFIG_NFS_V3) += nfsv3.o +nfsv3-y := nfs3super.o nfs3client.o nfs3proc.o nfs3xdr.o +nfsv3-$(CONFIG_NFS_V3_ACL) += nfs3acl.o + +obj-$(CONFIG_NFS_V4) += nfsv4.o +CFLAGS_nfs4trace.o += -I$(src) +nfsv4-y := nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o nfs4super.o nfs4file.o \ + delegation.o nfs4idmap.o callback.o callback_xdr.o callback_proc.o \ + nfs4namespace.o nfs4getroot.o nfs4client.o nfs4session.o \ + dns_resolve.o nfs4trace.o pnfs.o pnfs_dev.o pnfs_nfs.o +nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o +nfsv4-$(CONFIG_SYSCTL) += nfs4sysctl.o +nfsv4-$(CONFIG_NFS_V4_0) += nfs40client.o nfs40proc.o +nfsv4-$(CONFIG_NFS_V4_2) += nfs42proc.o nfs42xattr.o + +obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/ +obj-$(CONFIG_PNFS_BLOCK) += blocklayout/ +obj-$(CONFIG_PNFS_FLEXFILE_LAYOUT) += flexfilelayout/ diff --git a/fs/nfs/blocklayout/Makefile b/fs/nfs/blocklayout/Makefile new file mode 100644 index 000000000..7668a1bfb --- /dev/null +++ b/fs/nfs/blocklayout/Makefile @@ -0,0 +1,7 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Makefile for the pNFS block layout driver kernel module +# +obj-$(CONFIG_PNFS_BLOCK) += blocklayoutdriver.o + +blocklayoutdriver-y += blocklayout.o dev.o extent_tree.o rpc_pipefs.o diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c new file mode 100644 index 000000000..d54a141a8 --- /dev/null +++ b/fs/nfs/blocklayout/blocklayout.c @@ -0,0 +1,1047 @@ +/* + * linux/fs/nfs/blocklayout/blocklayout.c + * + * Module for the NFSv4.1 pNFS block layout driver. + * + * Copyright (c) 2006 The Regents of the University of Michigan. + * All rights reserved. + * + * Andy Adamson + * Fred Isaman + * + * permission is granted to use, copy, create derivative works and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the university of michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. if + * the above copyright notice or any other identification of the + * university of michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * this software is provided as is, without representation from the + * university of michigan as to its fitness for any purpose, and without + * warranty by the university of michigan of any kind, either express + * or implied, including without limitation the implied warranties of + * merchantability and fitness for a particular purpose. the regents + * of the university of michigan shall not be liable for any damages, + * including special, indirect, incidental, or consequential damages, + * with respect to any claim arising out or in connection with the use + * of the software, even if it has been or is hereafter advised of the + * possibility of such damages. + */ + +#include +#include +#include +#include +#include /* struct bio */ +#include + +#include "../pnfs.h" +#include "../nfs4session.h" +#include "../internal.h" +#include "blocklayout.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Andy Adamson "); +MODULE_DESCRIPTION("The NFSv4.1 pNFS Block layout driver"); + +static bool is_hole(struct pnfs_block_extent *be) +{ + switch (be->be_state) { + case PNFS_BLOCK_NONE_DATA: + return true; + case PNFS_BLOCK_INVALID_DATA: + return be->be_tag ? false : true; + default: + return false; + } +} + +/* The data we are handed might be spread across several bios. We need + * to track when the last one is finished. + */ +struct parallel_io { + struct kref refcnt; + void (*pnfs_callback) (void *data); + void *data; +}; + +static inline struct parallel_io *alloc_parallel(void *data) +{ + struct parallel_io *rv; + + rv = kmalloc_obj(*rv, GFP_NOFS); + if (rv) { + rv->data = data; + kref_init(&rv->refcnt); + } + return rv; +} + +static inline void get_parallel(struct parallel_io *p) +{ + kref_get(&p->refcnt); +} + +static void destroy_parallel(struct kref *kref) +{ + struct parallel_io *p = container_of(kref, struct parallel_io, refcnt); + + dprintk("%s enter\n", __func__); + p->pnfs_callback(p->data); + kfree(p); +} + +static inline void put_parallel(struct parallel_io *p) +{ + kref_put(&p->refcnt, destroy_parallel); +} + +static struct bio * +bl_submit_bio(struct bio *bio) +{ + if (bio) { + get_parallel(bio->bi_private); + dprintk("%s submitting %s bio %u@%llu\n", __func__, + bio_op(bio) == READ ? "read" : "write", + bio->bi_iter.bi_size, + (unsigned long long)bio->bi_iter.bi_sector); + submit_bio(bio); + } + return NULL; +} + +static bool offset_in_map(u64 offset, struct pnfs_block_dev_map *map) +{ + return offset >= map->start && offset < map->start + map->len; +} + +static struct bio * +do_add_page_to_bio(struct bio *bio, int npg, enum req_op op, sector_t isect, + struct page *page, struct pnfs_block_dev_map *map, + struct pnfs_block_extent *be, bio_end_io_t end_io, + struct parallel_io *par, unsigned int offset, int *len) +{ + struct pnfs_block_dev *dev = + container_of(be->be_device, struct pnfs_block_dev, node); + u64 disk_addr, end; + + dprintk("%s: npg %d rw %d isect %llu offset %u len %d\n", __func__, + npg, (__force u32)op, (unsigned long long)isect, offset, *len); + + /* translate to device offset */ + isect += be->be_v_offset; + isect -= be->be_f_offset; + + /* translate to physical disk offset */ + disk_addr = (u64)isect << SECTOR_SHIFT; + if (!offset_in_map(disk_addr, map)) { + if (!dev->map(dev, disk_addr, map) || !offset_in_map(disk_addr, map)) + return ERR_PTR(-EIO); + bio = bl_submit_bio(bio); + } + disk_addr += map->disk_offset; + disk_addr -= map->start; + + /* limit length to what the device mapping allows */ + end = disk_addr + *len; + if (end >= map->disk_offset + map->len) + *len = map->disk_offset + map->len - disk_addr; + +retry: + if (!bio) { + bio = bio_alloc(map->bdev, bio_max_segs(npg), op, GFP_NOIO); + bio->bi_iter.bi_sector = disk_addr >> SECTOR_SHIFT; + bio->bi_end_io = end_io; + bio->bi_private = par; + } + if (bio_add_page(bio, page, *len, offset) < *len) { + bio = bl_submit_bio(bio); + goto retry; + } + return bio; +} + +static void bl_mark_devices_unavailable(struct nfs_pgio_header *header, bool rw) +{ + struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg); + size_t bytes_left = header->args.count; + sector_t isect, extent_length = 0; + struct pnfs_block_extent be; + + isect = header->args.offset >> SECTOR_SHIFT; + bytes_left += header->args.offset - (isect << SECTOR_SHIFT); + + while (bytes_left > 0) { + if (!ext_tree_lookup(bl, isect, &be, rw)) + return; + extent_length = be.be_length - (isect - be.be_f_offset); + nfs4_mark_deviceid_unavailable(be.be_device); + isect += extent_length; + if (bytes_left > extent_length << SECTOR_SHIFT) + bytes_left -= extent_length << SECTOR_SHIFT; + else + bytes_left = 0; + } +} + +static void bl_end_io_read(struct bio *bio) +{ + struct parallel_io *par = bio->bi_private; + + if (bio->bi_status) { + struct nfs_pgio_header *header = par->data; + + if (!header->pnfs_error) + header->pnfs_error = -EIO; + pnfs_set_lo_fail(header->lseg); + bl_mark_devices_unavailable(header, false); + } + + bio_put(bio); + put_parallel(par); +} + +static void bl_read_cleanup(struct work_struct *work) +{ + struct rpc_task *task; + struct nfs_pgio_header *hdr; + dprintk("%s enter\n", __func__); + task = container_of(work, struct rpc_task, u.tk_work); + hdr = container_of(task, struct nfs_pgio_header, task); + pnfs_ld_read_done(hdr); +} + +static void +bl_end_par_io_read(void *data) +{ + struct nfs_pgio_header *hdr = data; + + hdr->task.tk_status = hdr->pnfs_error; + INIT_WORK(&hdr->task.u.tk_work, bl_read_cleanup); + schedule_work(&hdr->task.u.tk_work); +} + +static enum pnfs_try_status +bl_read_pagelist(struct nfs_pgio_header *header) +{ + struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg); + struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 }; + struct bio *bio = NULL; + struct pnfs_block_extent be; + sector_t isect, extent_length = 0; + struct parallel_io *par; + loff_t f_offset = header->args.offset; + size_t bytes_left = header->args.count; + unsigned int pg_offset = header->args.pgbase, pg_len; + struct page **pages = header->args.pages; + int pg_index = header->args.pgbase >> PAGE_SHIFT; + const bool is_dio = (header->dreq != NULL); + struct blk_plug plug; + int i; + + dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__, + header->page_array.npages, f_offset, + (unsigned int)header->args.count); + + par = alloc_parallel(header); + if (!par) + return PNFS_NOT_ATTEMPTED; + par->pnfs_callback = bl_end_par_io_read; + + blk_start_plug(&plug); + + isect = (sector_t) (f_offset >> SECTOR_SHIFT); + /* Code assumes extents are page-aligned */ + for (i = pg_index; i < header->page_array.npages; i++) { + if (extent_length <= 0) { + /* We've used up the previous extent */ + bio = bl_submit_bio(bio); + + /* Get the next one */ + if (!ext_tree_lookup(bl, isect, &be, false)) { + header->pnfs_error = -EIO; + goto out; + } + extent_length = be.be_length - (isect - be.be_f_offset); + } + + if (is_dio) { + if (pg_offset + bytes_left > PAGE_SIZE) + pg_len = PAGE_SIZE - pg_offset; + else + pg_len = bytes_left; + } else { + BUG_ON(pg_offset != 0); + pg_len = PAGE_SIZE; + } + + if (is_hole(&be)) { + bio = bl_submit_bio(bio); + /* Fill hole w/ zeroes w/o accessing device */ + dprintk("%s Zeroing page for hole\n", __func__); + zero_user_segment(pages[i], pg_offset, pg_len); + + /* invalidate map */ + map.start = NFS4_MAX_UINT64; + } else { + bio = do_add_page_to_bio(bio, + header->page_array.npages - i, + REQ_OP_READ, + isect, pages[i], &map, &be, + bl_end_io_read, par, + pg_offset, &pg_len); + if (IS_ERR(bio)) { + header->pnfs_error = PTR_ERR(bio); + bio = NULL; + goto out; + } + } + isect += (pg_len >> SECTOR_SHIFT); + extent_length -= (pg_len >> SECTOR_SHIFT); + f_offset += pg_len; + bytes_left -= pg_len; + pg_offset = 0; + } + if ((isect << SECTOR_SHIFT) >= header->inode->i_size) { + header->res.eof = 1; + header->res.count = header->inode->i_size - header->args.offset; + } else { + header->res.count = (isect << SECTOR_SHIFT) - header->args.offset; + } +out: + bl_submit_bio(bio); + blk_finish_plug(&plug); + put_parallel(par); + return PNFS_ATTEMPTED; +} + +static void bl_end_io_write(struct bio *bio) +{ + struct parallel_io *par = bio->bi_private; + struct nfs_pgio_header *header = par->data; + + if (bio->bi_status) { + if (!header->pnfs_error) + header->pnfs_error = -EIO; + pnfs_set_lo_fail(header->lseg); + bl_mark_devices_unavailable(header, true); + } + bio_put(bio); + put_parallel(par); +} + +/* Function scheduled for call during bl_end_par_io_write, + * it marks sectors as written and extends the commitlist. + */ +static void bl_write_cleanup(struct work_struct *work) +{ + struct rpc_task *task = container_of(work, struct rpc_task, u.tk_work); + struct nfs_pgio_header *hdr = + container_of(task, struct nfs_pgio_header, task); + + dprintk("%s enter\n", __func__); + + if (likely(!hdr->pnfs_error)) { + struct pnfs_block_layout *bl = BLK_LSEG2EXT(hdr->lseg); + u64 start = hdr->args.offset & (loff_t)PAGE_MASK; + u64 end = (hdr->args.offset + hdr->args.count + + PAGE_SIZE - 1) & (loff_t)PAGE_MASK; + u64 lwb = hdr->args.offset + hdr->args.count; + + ext_tree_mark_written(bl, start >> SECTOR_SHIFT, + (end - start) >> SECTOR_SHIFT, lwb); + } + + pnfs_ld_write_done(hdr); +} + +/* Called when last of bios associated with a bl_write_pagelist call finishes */ +static void bl_end_par_io_write(void *data) +{ + struct nfs_pgio_header *hdr = data; + + hdr->task.tk_status = hdr->pnfs_error; + hdr->verf.committed = NFS_FILE_SYNC; + INIT_WORK(&hdr->task.u.tk_work, bl_write_cleanup); + schedule_work(&hdr->task.u.tk_work); +} + +static enum pnfs_try_status +bl_write_pagelist(struct nfs_pgio_header *header, int sync) +{ + struct pnfs_block_layout *bl = BLK_LSEG2EXT(header->lseg); + struct pnfs_block_dev_map map = { .start = NFS4_MAX_UINT64 }; + struct bio *bio = NULL; + struct pnfs_block_extent be; + sector_t isect, extent_length = 0; + struct parallel_io *par = NULL; + loff_t offset = header->args.offset; + struct page **pages = header->args.pages; + int pg_index = header->args.pgbase >> PAGE_SHIFT; + unsigned int pg_len; + struct blk_plug plug; + int i; + + dprintk("%s enter, %u@%lld\n", __func__, header->args.count, offset); + + /* At this point, header->page_aray is a (sequential) list of nfs_pages. + * We want to write each, and if there is an error set pnfs_error + * to have it redone using nfs. + */ + par = alloc_parallel(header); + if (!par) + return PNFS_NOT_ATTEMPTED; + par->pnfs_callback = bl_end_par_io_write; + + blk_start_plug(&plug); + + /* we always write out the whole page */ + offset = offset & (loff_t)PAGE_MASK; + isect = offset >> SECTOR_SHIFT; + + for (i = pg_index; i < header->page_array.npages; i++) { + if (extent_length <= 0) { + /* We've used up the previous extent */ + bio = bl_submit_bio(bio); + /* Get the next one */ + if (!ext_tree_lookup(bl, isect, &be, true)) { + header->pnfs_error = -EINVAL; + goto out; + } + + extent_length = be.be_length - (isect - be.be_f_offset); + } + + pg_len = PAGE_SIZE; + bio = do_add_page_to_bio(bio, header->page_array.npages - i, + REQ_OP_WRITE, isect, pages[i], &map, + &be, bl_end_io_write, par, 0, &pg_len); + if (IS_ERR(bio)) { + header->pnfs_error = PTR_ERR(bio); + bio = NULL; + goto out; + } + + offset += pg_len; + isect += (pg_len >> SECTOR_SHIFT); + extent_length -= (pg_len >> SECTOR_SHIFT); + } + + header->res.count = header->args.count; +out: + bl_submit_bio(bio); + blk_finish_plug(&plug); + put_parallel(par); + return PNFS_ATTEMPTED; +} + +static void bl_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct pnfs_block_layout *bl = BLK_LO2EXT(lo); + int err; + + dprintk("%s enter\n", __func__); + + err = ext_tree_remove(bl, true, 0, LLONG_MAX); + WARN_ON(err); + + kfree_rcu(bl, bl_layout.plh_rcu); +} + +static struct pnfs_layout_hdr *__bl_alloc_layout_hdr(struct inode *inode, + gfp_t gfp_flags, bool is_scsi_layout) +{ + struct pnfs_block_layout *bl; + + dprintk("%s enter\n", __func__); + bl = kzalloc_obj(*bl, gfp_flags); + if (!bl) + return NULL; + + bl->bl_ext_rw = RB_ROOT; + bl->bl_ext_ro = RB_ROOT; + spin_lock_init(&bl->bl_ext_lock); + + bl->bl_scsi_layout = is_scsi_layout; + return &bl->bl_layout; +} + +static struct pnfs_layout_hdr *bl_alloc_layout_hdr(struct inode *inode, + gfp_t gfp_flags) +{ + return __bl_alloc_layout_hdr(inode, gfp_flags, false); +} + +static struct pnfs_layout_hdr *sl_alloc_layout_hdr(struct inode *inode, + gfp_t gfp_flags) +{ + return __bl_alloc_layout_hdr(inode, gfp_flags, true); +} + +static void bl_free_lseg(struct pnfs_layout_segment *lseg) +{ + dprintk("%s enter\n", __func__); + kfree(lseg); +} + +/* Tracks info needed to ensure extents in layout obey constraints of spec */ +struct layout_verification { + u32 mode; /* R or RW */ + u64 start; /* Expected start of next non-COW extent */ + u64 inval; /* Start of INVAL coverage */ + u64 cowread; /* End of COW read coverage */ +}; + +/* Verify the extent meets the layout requirements of the pnfs-block draft, + * section 2.3.1. + */ +static int verify_extent(struct pnfs_block_extent *be, + struct layout_verification *lv) +{ + if (lv->mode == IOMODE_READ) { + if (be->be_state == PNFS_BLOCK_READWRITE_DATA || + be->be_state == PNFS_BLOCK_INVALID_DATA) + return -EIO; + if (be->be_f_offset != lv->start) + return -EIO; + lv->start += be->be_length; + return 0; + } + /* lv->mode == IOMODE_RW */ + if (be->be_state == PNFS_BLOCK_READWRITE_DATA) { + if (be->be_f_offset != lv->start) + return -EIO; + if (lv->cowread > lv->start) + return -EIO; + lv->start += be->be_length; + lv->inval = lv->start; + return 0; + } else if (be->be_state == PNFS_BLOCK_INVALID_DATA) { + if (be->be_f_offset != lv->start) + return -EIO; + lv->start += be->be_length; + return 0; + } else if (be->be_state == PNFS_BLOCK_READ_DATA) { + if (be->be_f_offset > lv->start) + return -EIO; + if (be->be_f_offset < lv->inval) + return -EIO; + if (be->be_f_offset < lv->cowread) + return -EIO; + /* It looks like you might want to min this with lv->start, + * but you really don't. + */ + lv->inval = lv->inval + be->be_length; + lv->cowread = be->be_f_offset + be->be_length; + return 0; + } else + return -EIO; +} + +static int decode_sector_number(__be32 **rp, sector_t *sp) +{ + uint64_t s; + + *rp = xdr_decode_hyper(*rp, &s); + if (s & 0x1ff) { + printk(KERN_WARNING "NFS: %s: sector not aligned\n", __func__); + return -1; + } + *sp = s >> SECTOR_SHIFT; + return 0; +} + +static struct nfs4_deviceid_node * +bl_find_get_deviceid(struct nfs_server *server, + const struct nfs4_deviceid *id, const struct cred *cred, + gfp_t gfp_mask) +{ + struct nfs4_deviceid_node *node; + int err = -ENODEV; + +retry: + node = nfs4_find_get_deviceid(server, id, cred, gfp_mask); + if (!node) + return ERR_PTR(-ENODEV); + + /* + * Devices that are marked unavailable are left in the cache with a + * timeout to avoid sending GETDEVINFO after every LAYOUTGET, or + * constantly attempting to register the device. Once marked as + * unavailable they must be deleted and never reused. + */ + if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) { + unsigned long end = jiffies; + unsigned long start = end - PNFS_DEVICE_RETRY_TIMEOUT; + + if (!time_in_range(node->timestamp_unavailable, start, end)) { + /* Uncork subsequent GETDEVINFO operations for this device */ + nfs4_delete_deviceid(node->ld, node->nfs_client, id); + goto retry; + } + goto out_put; + } + + if (!bl_register_dev(container_of(node, struct pnfs_block_dev, node))) { + /* + * If we cannot register, treat this device as transient: + * Make a negative cache entry for the device + */ + nfs4_mark_deviceid_unavailable(node); + goto out_put; + } + + return node; + +out_put: + nfs4_put_deviceid_node(node); + return ERR_PTR(err); +} + +static int +bl_alloc_extent(struct xdr_stream *xdr, struct pnfs_layout_hdr *lo, + struct layout_verification *lv, struct list_head *extents, + gfp_t gfp_mask) +{ + struct pnfs_block_extent *be; + struct nfs4_deviceid id; + int error; + __be32 *p; + + p = xdr_inline_decode(xdr, 28 + NFS4_DEVICEID4_SIZE); + if (!p) + return -EIO; + + be = kzalloc_obj(*be, GFP_NOFS); + if (!be) + return -ENOMEM; + + memcpy(&id, p, NFS4_DEVICEID4_SIZE); + p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); + + be->be_device = bl_find_get_deviceid(NFS_SERVER(lo->plh_inode), &id, + lo->plh_lc_cred, gfp_mask); + if (IS_ERR(be->be_device)) { + error = PTR_ERR(be->be_device); + goto out_free_be; + } + + /* + * The next three values are read in as bytes, but stored in the + * extent structure in 512-byte granularity. + */ + error = -EIO; + if (decode_sector_number(&p, &be->be_f_offset) < 0) + goto out_put_deviceid; + if (decode_sector_number(&p, &be->be_length) < 0) + goto out_put_deviceid; + if (decode_sector_number(&p, &be->be_v_offset) < 0) + goto out_put_deviceid; + be->be_state = be32_to_cpup(p++); + + error = verify_extent(be, lv); + if (error) { + dprintk("%s: extent verification failed\n", __func__); + goto out_put_deviceid; + } + + list_add_tail(&be->be_list, extents); + return 0; + +out_put_deviceid: + nfs4_put_deviceid_node(be->be_device); +out_free_be: + kfree(be); + return error; +} + +static struct pnfs_layout_segment * +bl_alloc_lseg(struct pnfs_layout_hdr *lo, struct nfs4_layoutget_res *lgr, + gfp_t gfp_mask) +{ + struct layout_verification lv = { + .mode = lgr->range.iomode, + .start = lgr->range.offset >> SECTOR_SHIFT, + .inval = lgr->range.offset >> SECTOR_SHIFT, + .cowread = lgr->range.offset >> SECTOR_SHIFT, + }; + struct pnfs_block_layout *bl = BLK_LO2EXT(lo); + struct pnfs_layout_segment *lseg; + struct xdr_buf buf; + struct xdr_stream xdr; + struct folio *scratch; + int status, i; + uint32_t count; + __be32 *p; + LIST_HEAD(extents); + + dprintk("---> %s\n", __func__); + + lseg = kzalloc_obj(*lseg, gfp_mask); + if (!lseg) + return ERR_PTR(-ENOMEM); + + status = -ENOMEM; + scratch = folio_alloc(gfp_mask, 0); + if (!scratch) + goto out; + + xdr_init_decode_pages(&xdr, &buf, + lgr->layoutp->pages, lgr->layoutp->len); + xdr_set_scratch_folio(&xdr, scratch); + + status = -EIO; + p = xdr_inline_decode(&xdr, 4); + if (unlikely(!p)) + goto out_free_scratch; + + count = be32_to_cpup(p++); + dprintk("%s: number of extents %d\n", __func__, count); + + /* + * Decode individual extents, putting them in temporary staging area + * until whole layout is decoded to make error recovery easier. + */ + for (i = 0; i < count; i++) { + status = bl_alloc_extent(&xdr, lo, &lv, &extents, gfp_mask); + if (status) + goto process_extents; + } + + if (lgr->range.offset + lgr->range.length != + lv.start << SECTOR_SHIFT) { + dprintk("%s Final length mismatch\n", __func__); + status = -EIO; + goto process_extents; + } + + if (lv.start < lv.cowread) { + dprintk("%s Final uncovered COW extent\n", __func__); + status = -EIO; + } + +process_extents: + while (!list_empty(&extents)) { + struct pnfs_block_extent *be = + list_first_entry(&extents, struct pnfs_block_extent, + be_list); + list_del(&be->be_list); + + if (!status) + status = ext_tree_insert(bl, be); + + if (status) { + nfs4_put_deviceid_node(be->be_device); + kfree(be); + } + } + +out_free_scratch: + folio_put(scratch); +out: + dprintk("%s returns %d\n", __func__, status); + switch (status) { + case -ENODEV: + /* Our extent block devices are unavailable */ + set_bit(NFS_LSEG_UNAVAILABLE, &lseg->pls_flags); + fallthrough; + case 0: + return lseg; + default: + kfree(lseg); + return ERR_PTR(status); + } +} + +static void +bl_return_range(struct pnfs_layout_hdr *lo, + struct pnfs_layout_range *range) +{ + struct pnfs_block_layout *bl = BLK_LO2EXT(lo); + sector_t offset = range->offset >> SECTOR_SHIFT, end; + + if (range->offset % 8) { + dprintk("%s: offset %lld not block size aligned\n", + __func__, range->offset); + return; + } + + if (range->length != NFS4_MAX_UINT64) { + if (range->length % 8) { + dprintk("%s: length %lld not block size aligned\n", + __func__, range->length); + return; + } + + end = offset + (range->length >> SECTOR_SHIFT); + } else { + end = round_down(NFS4_MAX_UINT64, PAGE_SIZE); + } + + ext_tree_remove(bl, range->iomode & IOMODE_RW, offset, end); +} + +static int +bl_prepare_layoutcommit(struct nfs4_layoutcommit_args *arg) +{ + return ext_tree_prepare_commit(arg); +} + +static void +bl_cleanup_layoutcommit(struct nfs4_layoutcommit_data *lcdata) +{ + ext_tree_mark_committed(&lcdata->args, lcdata->res.status); +} + +static int +bl_set_layoutdriver(struct nfs_server *server, const struct nfs_fh *fh) +{ + dprintk("%s enter\n", __func__); + + if (server->pnfs_blksize == 0) { + dprintk("%s Server did not return blksize\n", __func__); + return -EINVAL; + } + if (server->pnfs_blksize > PAGE_SIZE) { + printk(KERN_ERR "%s: pNFS blksize %d not supported.\n", + __func__, server->pnfs_blksize); + return -EINVAL; + } + + return 0; +} + +static bool +is_aligned_req(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, unsigned int alignment, bool is_write) +{ + /* + * Always accept buffered writes, higher layers take care of the + * right alignment. + */ + if (pgio->pg_dreq == NULL) + return true; + + if (!IS_ALIGNED(req->wb_offset, alignment)) + return false; + + if (IS_ALIGNED(req->wb_bytes, alignment)) + return true; + + if (is_write && + (req_offset(req) + req->wb_bytes == i_size_read(pgio->pg_inode))) { + /* + * If the write goes up to the inode size, just write + * the full page. Data past the inode size is + * guaranteed to be zeroed by the higher level client + * code, and this behaviour is mandated by RFC 5663 + * section 2.3.2. + */ + return true; + } + + return false; +} + +static void +bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +{ + if (!is_aligned_req(pgio, req, SECTOR_SIZE, false)) { + nfs_pageio_reset_read_mds(pgio); + return; + } + + pnfs_generic_pg_init_read(pgio, req); + + if (pgio->pg_lseg && + test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) { + pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg); + pnfs_set_lo_fail(pgio->pg_lseg); + nfs_pageio_reset_read_mds(pgio); + } +} + +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t +bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + if (!is_aligned_req(pgio, req, SECTOR_SIZE, false)) + return 0; + return pnfs_generic_pg_test(pgio, prev, req); +} + +/* + * Return the number of contiguous bytes for a given inode + * starting at page frame idx. + */ +static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t end; + + /* Optimize common case that writes from 0 to end of file */ + end = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); + if (end != inode->i_mapping->nrpages) { + rcu_read_lock(); + end = page_cache_next_miss(mapping, idx + 1, ULONG_MAX); + rcu_read_unlock(); + } + + if (!end) + return i_size_read(inode) - (idx << PAGE_SHIFT); + else + return (end - idx) << PAGE_SHIFT; +} + +static void +bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +{ + u64 wb_size; + + if (!is_aligned_req(pgio, req, PAGE_SIZE, true)) { + nfs_pageio_reset_write_mds(pgio); + return; + } + + if (pgio->pg_dreq == NULL) + wb_size = pnfs_num_cont_bytes(pgio->pg_inode, req->wb_index); + else + wb_size = nfs_dreq_bytes_left(pgio->pg_dreq, req_offset(req)); + + pnfs_generic_pg_init_write(pgio, req, wb_size); + + if (pgio->pg_lseg && + test_bit(NFS_LSEG_UNAVAILABLE, &pgio->pg_lseg->pls_flags)) { + + pnfs_error_mark_layout_for_return(pgio->pg_inode, pgio->pg_lseg); + pnfs_set_lo_fail(pgio->pg_lseg); + nfs_pageio_reset_write_mds(pgio); + } +} + +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t +bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + if (!is_aligned_req(pgio, req, PAGE_SIZE, true)) + return 0; + return pnfs_generic_pg_test(pgio, prev, req); +} + +static const struct nfs_pageio_ops bl_pg_read_ops = { + .pg_init = bl_pg_init_read, + .pg_test = bl_pg_test_read, + .pg_doio = pnfs_generic_pg_readpages, + .pg_cleanup = pnfs_generic_pg_cleanup, +}; + +static const struct nfs_pageio_ops bl_pg_write_ops = { + .pg_init = bl_pg_init_write, + .pg_test = bl_pg_test_write, + .pg_doio = pnfs_generic_pg_writepages, + .pg_cleanup = pnfs_generic_pg_cleanup, +}; + +static struct pnfs_layoutdriver_type blocklayout_type = { + .id = LAYOUT_BLOCK_VOLUME, + .name = "LAYOUT_BLOCK_VOLUME", + .owner = THIS_MODULE, + .flags = PNFS_LAYOUTRET_ON_SETATTR | + PNFS_LAYOUTRET_ON_ERROR | + PNFS_READ_WHOLE_PAGE, + .read_pagelist = bl_read_pagelist, + .write_pagelist = bl_write_pagelist, + .alloc_layout_hdr = bl_alloc_layout_hdr, + .free_layout_hdr = bl_free_layout_hdr, + .alloc_lseg = bl_alloc_lseg, + .free_lseg = bl_free_lseg, + .return_range = bl_return_range, + .prepare_layoutcommit = bl_prepare_layoutcommit, + .cleanup_layoutcommit = bl_cleanup_layoutcommit, + .set_layoutdriver = bl_set_layoutdriver, + .alloc_deviceid_node = bl_alloc_deviceid_node, + .free_deviceid_node = bl_free_deviceid_node, + .pg_read_ops = &bl_pg_read_ops, + .pg_write_ops = &bl_pg_write_ops, + .sync = pnfs_generic_sync, +}; + +static struct pnfs_layoutdriver_type scsilayout_type = { + .id = LAYOUT_SCSI, + .name = "LAYOUT_SCSI", + .owner = THIS_MODULE, + .flags = PNFS_LAYOUTRET_ON_SETATTR | + PNFS_LAYOUTRET_ON_ERROR | + PNFS_READ_WHOLE_PAGE, + .read_pagelist = bl_read_pagelist, + .write_pagelist = bl_write_pagelist, + .alloc_layout_hdr = sl_alloc_layout_hdr, + .free_layout_hdr = bl_free_layout_hdr, + .alloc_lseg = bl_alloc_lseg, + .free_lseg = bl_free_lseg, + .return_range = bl_return_range, + .prepare_layoutcommit = bl_prepare_layoutcommit, + .cleanup_layoutcommit = bl_cleanup_layoutcommit, + .set_layoutdriver = bl_set_layoutdriver, + .alloc_deviceid_node = bl_alloc_deviceid_node, + .free_deviceid_node = bl_free_deviceid_node, + .pg_read_ops = &bl_pg_read_ops, + .pg_write_ops = &bl_pg_write_ops, + .sync = pnfs_generic_sync, +}; + + +static int __init nfs4blocklayout_init(void) +{ + int ret; + + dprintk("%s: NFSv4 Block Layout Driver Registering...\n", __func__); + + ret = bl_init_pipefs(); + if (ret) + goto out; + + ret = pnfs_register_layoutdriver(&blocklayout_type); + if (ret) + goto out_cleanup_pipe; + + ret = pnfs_register_layoutdriver(&scsilayout_type); + if (ret) + goto out_unregister_block; + return 0; + +out_unregister_block: + pnfs_unregister_layoutdriver(&blocklayout_type); +out_cleanup_pipe: + bl_cleanup_pipefs(); +out: + return ret; +} + +static void __exit nfs4blocklayout_exit(void) +{ + dprintk("%s: NFSv4 Block Layout Driver Unregistering...\n", + __func__); + + pnfs_unregister_layoutdriver(&scsilayout_type); + pnfs_unregister_layoutdriver(&blocklayout_type); + bl_cleanup_pipefs(); +} + +MODULE_ALIAS("nfs-layouttype4-3"); +MODULE_ALIAS("nfs-layouttype4-5"); + +module_init(nfs4blocklayout_init); +module_exit(nfs4blocklayout_exit); diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h new file mode 100644 index 000000000..6da40ca19 --- /dev/null +++ b/fs/nfs/blocklayout/blocklayout.h @@ -0,0 +1,204 @@ +/* + * linux/fs/nfs/blocklayout/blocklayout.h + * + * Module for the NFSv4.1 pNFS block layout driver. + * + * Copyright (c) 2006 The Regents of the University of Michigan. + * All rights reserved. + * + * Andy Adamson + * Fred Isaman + * + * permission is granted to use, copy, create derivative works and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the university of michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. if + * the above copyright notice or any other identification of the + * university of michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * this software is provided as is, without representation from the + * university of michigan as to its fitness for any purpose, and without + * warranty by the university of michigan of any kind, either express + * or implied, including without limitation the implied warranties of + * merchantability and fitness for a particular purpose. the regents + * of the university of michigan shall not be liable for any damages, + * including special, indirect, incidental, or consequential damages, + * with respect to any claim arising out or in connection with the use + * of the software, even if it has been or is hereafter advised of the + * possibility of such damages. + */ +#ifndef FS_NFS_NFS4BLOCKLAYOUT_H +#define FS_NFS_NFS4BLOCKLAYOUT_H + +#include +#include +#include + +#include "../nfs4_fs.h" +#include "../pnfs.h" +#include "../netns.h" + +#define PAGE_CACHE_SECTORS (PAGE_SIZE >> SECTOR_SHIFT) +#define PAGE_CACHE_SECTOR_SHIFT (PAGE_SHIFT - SECTOR_SHIFT) +#define SECTOR_SIZE (1 << SECTOR_SHIFT) + +struct pnfs_block_dev; + +#define PNFS_BLOCK_MAX_UUIDS 4 +#define PNFS_BLOCK_MAX_DEVICES 64 + +/* + * Random upper cap for the uuid length to avoid unbounded allocation. + * Not actually limited by the protocol. + */ +#define PNFS_BLOCK_UUID_LEN 128 + +struct pnfs_block_volume { + enum pnfs_block_volume_type type; + union { + struct { + int len; + int nr_sigs; + struct { + u64 offset; + u32 sig_len; + u8 sig[PNFS_BLOCK_UUID_LEN]; + } sigs[PNFS_BLOCK_MAX_UUIDS]; + } simple; + struct { + u64 start; + u64 len; + u32 volume; + } slice; + struct { + u32 volumes_count; + u32 volumes[PNFS_BLOCK_MAX_DEVICES]; + } concat; + struct { + u64 chunk_size; + u32 volumes_count; + u32 volumes[PNFS_BLOCK_MAX_DEVICES]; + } stripe; + struct { + enum scsi_code_set code_set; + enum scsi_designator_type designator_type; + int designator_len; + u8 designator[256]; + u64 pr_key; + } scsi; + }; +}; + +struct pnfs_block_dev_map { + u64 start; + u64 len; + u64 disk_offset; + struct block_device *bdev; +}; + +struct pnfs_block_dev { + struct nfs4_deviceid_node node; + + u64 start; + u64 len; + + enum pnfs_block_volume_type type; + u32 nr_children; + struct pnfs_block_dev *children; + u64 chunk_size; + + struct file *bdev_file; + u64 disk_offset; + unsigned long flags; + + u64 pr_key; + + bool (*map)(struct pnfs_block_dev *dev, u64 offset, + struct pnfs_block_dev_map *map); +}; + +/* pnfs_block_dev flag bits */ +enum { + PNFS_BDEV_REGISTERED = 0, +}; + +/* sector_t fields are all in 512-byte sectors */ +struct pnfs_block_extent { + union { + struct rb_node be_node; + struct list_head be_list; + }; + struct nfs4_deviceid_node *be_device; + sector_t be_f_offset; /* the starting offset in the file */ + sector_t be_length; /* the size of the extent */ + sector_t be_v_offset; /* the starting offset in the volume */ + enum pnfs_block_extent_state be_state; /* the state of this extent */ +#define EXTENT_WRITTEN 1 +#define EXTENT_COMMITTING 2 + unsigned int be_tag; +}; + +struct pnfs_block_layout { + struct pnfs_layout_hdr bl_layout; + struct rb_root bl_ext_rw; + struct rb_root bl_ext_ro; + spinlock_t bl_ext_lock; /* Protects list manipulation */ + bool bl_scsi_layout; + u64 bl_lwb; +}; + +static inline struct pnfs_block_layout * +BLK_LO2EXT(struct pnfs_layout_hdr *lo) +{ + return container_of(lo, struct pnfs_block_layout, bl_layout); +} + +static inline struct pnfs_block_layout * +BLK_LSEG2EXT(struct pnfs_layout_segment *lseg) +{ + return BLK_LO2EXT(lseg->pls_layout); +} + +struct bl_pipe_msg { + struct rpc_pipe_msg msg; + wait_queue_head_t *bl_wq; +}; + +struct bl_msg_hdr { + u8 type; + u16 totallen; /* length of entire message, including hdr itself */ +}; + +#define BL_DEVICE_UMOUNT 0x0 /* Umount--delete devices */ +#define BL_DEVICE_MOUNT 0x1 /* Mount--create devices*/ +#define BL_DEVICE_REQUEST_INIT 0x0 /* Start request */ +#define BL_DEVICE_REQUEST_PROC 0x1 /* User level process succeeds */ +#define BL_DEVICE_REQUEST_ERR 0x2 /* User level process fails */ + +/* dev.c */ +bool bl_register_dev(struct pnfs_block_dev *d); +struct nfs4_deviceid_node *bl_alloc_deviceid_node(struct nfs_server *server, + struct pnfs_device *pdev, gfp_t gfp_mask); +void bl_free_deviceid_node(struct nfs4_deviceid_node *d); + +/* extent_tree.c */ +int ext_tree_insert(struct pnfs_block_layout *bl, + struct pnfs_block_extent *new); +int ext_tree_remove(struct pnfs_block_layout *bl, bool rw, sector_t start, + sector_t end); +int ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start, + sector_t len, u64 lwb); +bool ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect, + struct pnfs_block_extent *ret, bool rw); +int ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg); +void ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status); + +/* rpc_pipefs.c */ +dev_t bl_resolve_deviceid(struct nfs_server *server, + struct pnfs_block_volume *b, gfp_t gfp_mask); +int __init bl_init_pipefs(void); +void bl_cleanup_pipefs(void); + +#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */ diff --git a/fs/nfs/blocklayout/dev.c b/fs/nfs/blocklayout/dev.c new file mode 100644 index 000000000..368d20daf --- /dev/null +++ b/fs/nfs/blocklayout/dev.c @@ -0,0 +1,599 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2014-2016 Christoph Hellwig. + */ +#include +#include +#include +#include +#include +#include +#include + +#include "blocklayout.h" +#include "../nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +static void bl_unregister_scsi(struct pnfs_block_dev *dev) +{ + struct block_device *bdev = file_bdev(dev->bdev_file); + const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops; + int status; + + status = ops->pr_register(bdev, dev->pr_key, 0, false); + if (status) + trace_bl_pr_key_unreg_err(bdev, dev->pr_key, status); + else + trace_bl_pr_key_unreg(bdev, dev->pr_key); +} + +static bool bl_register_scsi(struct pnfs_block_dev *dev) +{ + struct block_device *bdev = file_bdev(dev->bdev_file); + const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops; + int status; + + if (test_and_set_bit(PNFS_BDEV_REGISTERED, &dev->flags)) + return true; + + status = ops->pr_register(bdev, 0, dev->pr_key, true); + if (status) { + trace_bl_pr_key_reg_err(bdev, dev->pr_key, status); + return false; + } + trace_bl_pr_key_reg(bdev, dev->pr_key); + return true; +} + +static void bl_unregister_dev(struct pnfs_block_dev *dev) +{ + u32 i; + + if (dev->nr_children) { + for (i = 0; i < dev->nr_children; i++) + bl_unregister_dev(&dev->children[i]); + return; + } + + if (dev->type == PNFS_BLOCK_VOLUME_SCSI && + test_and_clear_bit(PNFS_BDEV_REGISTERED, &dev->flags)) + bl_unregister_scsi(dev); +} + +bool bl_register_dev(struct pnfs_block_dev *dev) +{ + u32 i; + + if (dev->nr_children) { + for (i = 0; i < dev->nr_children; i++) { + if (!bl_register_dev(&dev->children[i])) { + while (i > 0) + bl_unregister_dev(&dev->children[--i]); + return false; + } + } + return true; + } + + if (dev->type == PNFS_BLOCK_VOLUME_SCSI) + return bl_register_scsi(dev); + return true; +} + +static void +bl_free_device(struct pnfs_block_dev *dev) +{ + bl_unregister_dev(dev); + + if (dev->nr_children) { + int i; + + for (i = 0; i < dev->nr_children; i++) + bl_free_device(&dev->children[i]); + kfree(dev->children); + } else { + if (dev->bdev_file) + fput(dev->bdev_file); + } +} + +void +bl_free_deviceid_node(struct nfs4_deviceid_node *d) +{ + struct pnfs_block_dev *dev = + container_of(d, struct pnfs_block_dev, node); + + bl_free_device(dev); + kfree_rcu(dev, node.rcu); +} + +static int +nfs4_block_decode_volume(struct xdr_stream *xdr, struct pnfs_block_volume *b) +{ + __be32 *p; + int i; + + p = xdr_inline_decode(xdr, 4); + if (!p) + return -EIO; + b->type = be32_to_cpup(p++); + + switch (b->type) { + case PNFS_BLOCK_VOLUME_SIMPLE: + p = xdr_inline_decode(xdr, 4); + if (!p) + return -EIO; + b->simple.nr_sigs = be32_to_cpup(p++); + if (!b->simple.nr_sigs || b->simple.nr_sigs > PNFS_BLOCK_MAX_UUIDS) { + dprintk("Bad signature count: %d\n", b->simple.nr_sigs); + return -EIO; + } + + b->simple.len = 4 + 4; + for (i = 0; i < b->simple.nr_sigs; i++) { + p = xdr_inline_decode(xdr, 8 + 4); + if (!p) + return -EIO; + p = xdr_decode_hyper(p, &b->simple.sigs[i].offset); + b->simple.sigs[i].sig_len = be32_to_cpup(p++); + if (b->simple.sigs[i].sig_len > PNFS_BLOCK_UUID_LEN) { + pr_info("signature too long: %d\n", + b->simple.sigs[i].sig_len); + return -EIO; + } + + p = xdr_inline_decode(xdr, b->simple.sigs[i].sig_len); + if (!p) + return -EIO; + memcpy(&b->simple.sigs[i].sig, p, + b->simple.sigs[i].sig_len); + + b->simple.len += 8 + 4 + \ + (XDR_QUADLEN(b->simple.sigs[i].sig_len) << 2); + } + break; + case PNFS_BLOCK_VOLUME_SLICE: + p = xdr_inline_decode(xdr, 8 + 8 + 4); + if (!p) + return -EIO; + p = xdr_decode_hyper(p, &b->slice.start); + p = xdr_decode_hyper(p, &b->slice.len); + b->slice.volume = be32_to_cpup(p++); + break; + case PNFS_BLOCK_VOLUME_CONCAT: + p = xdr_inline_decode(xdr, 4); + if (!p) + return -EIO; + + b->concat.volumes_count = be32_to_cpup(p++); + if (b->concat.volumes_count > PNFS_BLOCK_MAX_DEVICES) { + dprintk("Too many volumes: %d\n", b->concat.volumes_count); + return -EIO; + } + + p = xdr_inline_decode(xdr, b->concat.volumes_count * 4); + if (!p) + return -EIO; + for (i = 0; i < b->concat.volumes_count; i++) + b->concat.volumes[i] = be32_to_cpup(p++); + break; + case PNFS_BLOCK_VOLUME_STRIPE: + p = xdr_inline_decode(xdr, 8 + 4); + if (!p) + return -EIO; + + p = xdr_decode_hyper(p, &b->stripe.chunk_size); + b->stripe.volumes_count = be32_to_cpup(p++); + if (b->stripe.volumes_count > PNFS_BLOCK_MAX_DEVICES) { + dprintk("Too many volumes: %d\n", b->stripe.volumes_count); + return -EIO; + } + + p = xdr_inline_decode(xdr, b->stripe.volumes_count * 4); + if (!p) + return -EIO; + for (i = 0; i < b->stripe.volumes_count; i++) + b->stripe.volumes[i] = be32_to_cpup(p++); + break; + case PNFS_BLOCK_VOLUME_SCSI: + p = xdr_inline_decode(xdr, 4 + 4 + 4); + if (!p) + return -EIO; + b->scsi.code_set = be32_to_cpup(p++); + b->scsi.designator_type = be32_to_cpup(p++); + b->scsi.designator_len = be32_to_cpup(p++); + p = xdr_inline_decode(xdr, b->scsi.designator_len); + if (!p) + return -EIO; + if (b->scsi.designator_len > 256) + return -EIO; + memcpy(&b->scsi.designator, p, b->scsi.designator_len); + p = xdr_inline_decode(xdr, 8); + if (!p) + return -EIO; + p = xdr_decode_hyper(p, &b->scsi.pr_key); + break; + default: + dprintk("unknown volume type!\n"); + return -EIO; + } + + return 0; +} + +static bool bl_map_simple(struct pnfs_block_dev *dev, u64 offset, + struct pnfs_block_dev_map *map) +{ + map->start = dev->start; + map->len = dev->len; + map->disk_offset = dev->disk_offset; + map->bdev = file_bdev(dev->bdev_file); + return true; +} + +static bool bl_map_concat(struct pnfs_block_dev *dev, u64 offset, + struct pnfs_block_dev_map *map) +{ + int i; + + for (i = 0; i < dev->nr_children; i++) { + struct pnfs_block_dev *child = &dev->children[i]; + + if (child->start > offset || + child->start + child->len <= offset) + continue; + + child->map(child, offset - child->start, map); + return true; + } + + dprintk("%s: ran off loop!\n", __func__); + return false; +} + +static bool bl_map_stripe(struct pnfs_block_dev *dev, u64 offset, + struct pnfs_block_dev_map *map) +{ + struct pnfs_block_dev *child; + u64 chunk; + u32 chunk_idx; + u64 disk_chunk; + u64 disk_offset; + + chunk = div_u64(offset, dev->chunk_size); + disk_chunk = div_u64_rem(chunk, dev->nr_children, &chunk_idx); + + if (chunk_idx >= dev->nr_children) { + dprintk("%s: invalid chunk idx %d (%lld/%lld)\n", + __func__, chunk_idx, offset, dev->chunk_size); + /* error, should not happen */ + return false; + } + + /* truncate offset to the beginning of the stripe */ + offset = chunk * dev->chunk_size; + + /* disk offset of the stripe */ + disk_offset = disk_chunk * dev->chunk_size; + + child = &dev->children[chunk_idx]; + child->map(child, disk_offset, map); + + map->start += offset; + map->disk_offset += disk_offset; + map->len = dev->chunk_size; + return true; +} + +static int +bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask); + + +static int +bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + struct pnfs_block_volume *v = &volumes[idx]; + struct file *bdev_file; + dev_t dev; + + dev = bl_resolve_deviceid(server, v, gfp_mask); + if (!dev) + return -EIO; + + bdev_file = bdev_file_open_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_WRITE, + NULL, NULL); + if (IS_ERR(bdev_file)) { + printk(KERN_WARNING "pNFS: failed to open device %d:%d (%ld)\n", + MAJOR(dev), MINOR(dev), PTR_ERR(bdev_file)); + return PTR_ERR(bdev_file); + } + d->bdev_file = bdev_file; + d->len = bdev_nr_bytes(file_bdev(bdev_file)); + d->map = bl_map_simple; + + printk(KERN_INFO "pNFS: using block device %s\n", + file_bdev(bdev_file)->bd_disk->disk_name); + return 0; +} + +static bool +bl_validate_designator(struct pnfs_block_volume *v) +{ + switch (v->scsi.designator_type) { + case PS_DESIGNATOR_EUI64: + if (v->scsi.code_set != PS_CODE_SET_BINARY) + return false; + + if (v->scsi.designator_len != 8 && + v->scsi.designator_len != 10 && + v->scsi.designator_len != 16) + return false; + + return true; + case PS_DESIGNATOR_NAA: + if (v->scsi.code_set != PS_CODE_SET_BINARY) + return false; + + if (v->scsi.designator_len != 8 && + v->scsi.designator_len != 16) + return false; + + return true; + case PS_DESIGNATOR_T10: + case PS_DESIGNATOR_NAME: + pr_err("pNFS: unsupported designator " + "(code set %d, type %d, len %d.\n", + v->scsi.code_set, + v->scsi.designator_type, + v->scsi.designator_len); + return false; + default: + pr_err("pNFS: invalid designator " + "(code set %d, type %d, len %d.\n", + v->scsi.code_set, + v->scsi.designator_type, + v->scsi.designator_len); + return false; + } +} + +static struct file * +bl_open_path(struct pnfs_block_volume *v, const char *prefix) +{ + struct file *bdev_file; + const char *devname __free(kfree) = NULL; + + devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/%s%*phN", + prefix, v->scsi.designator_len, v->scsi.designator); + if (!devname) + return ERR_PTR(-ENOMEM); + + if (tsk_is_kthread(current)) { + scoped_with_init_fs() + bdev_file = bdev_file_open_by_path(devname, + BLK_OPEN_READ | BLK_OPEN_WRITE, + NULL, NULL); + } else { + bdev_file = bdev_file_open_by_path(devname, + BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL); + } + if (IS_ERR(bdev_file)) { + dprintk("failed to open device %s (%ld)\n", + devname, PTR_ERR(bdev_file)); + } else { + pr_info("pNFS: using block device %s\n", + file_bdev(bdev_file)->bd_disk->disk_name); + } + + return bdev_file; +} + +static int +bl_parse_scsi(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + struct pnfs_block_volume *v = &volumes[idx]; + struct block_device *bdev; + const struct pr_ops *ops; + struct file *bdev_file; + int error; + + if (!bl_validate_designator(v)) + return -EINVAL; + + /* + * Try to open the RH/Fedora specific dm-mpath udev path first, as the + * wwn- links will only point to the first discovered SCSI device there. + * On other distributions like Debian, the default SCSI by-id path will + * point to the dm-multipath device if one exists. + */ + bdev_file = bl_open_path(v, "dm-uuid-mpath-0x"); + if (IS_ERR(bdev_file)) + bdev_file = bl_open_path(v, "wwn-0x"); + if (IS_ERR(bdev_file)) + bdev_file = bl_open_path(v, "nvme-eui."); + if (IS_ERR(bdev_file)) { + pr_warn("pNFS: no device found for volume %*phN\n", + v->scsi.designator_len, v->scsi.designator); + return PTR_ERR(bdev_file); + } + d->bdev_file = bdev_file; + bdev = file_bdev(bdev_file); + + d->len = bdev_nr_bytes(bdev); + d->map = bl_map_simple; + d->pr_key = v->scsi.pr_key; + + if (d->len == 0) { + error = -ENODEV; + goto out_blkdev_put; + } + + ops = bdev->bd_disk->fops->pr_ops; + if (!ops) { + pr_err("pNFS: block device %s does not support reservations.", + bdev->bd_disk->disk_name); + error = -EINVAL; + goto out_blkdev_put; + } + + return 0; + +out_blkdev_put: + fput(d->bdev_file); + return error; +} + +static int +bl_parse_slice(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + struct pnfs_block_volume *v = &volumes[idx]; + int ret; + + ret = bl_parse_deviceid(server, d, volumes, v->slice.volume, gfp_mask); + if (ret) + return ret; + + d->disk_offset = v->slice.start; + d->len = v->slice.len; + return 0; +} + +static int +bl_parse_concat(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + struct pnfs_block_volume *v = &volumes[idx]; + u64 len = 0; + int ret, i; + + d->children = kzalloc_objs(struct pnfs_block_dev, + v->concat.volumes_count, gfp_mask); + if (!d->children) + return -ENOMEM; + + for (i = 0; i < v->concat.volumes_count; i++) { + ret = bl_parse_deviceid(server, &d->children[i], + volumes, v->concat.volumes[i], gfp_mask); + if (ret) + return ret; + + d->nr_children++; + d->children[i].start += len; + len += d->children[i].len; + } + + d->len = len; + d->map = bl_map_concat; + return 0; +} + +static int +bl_parse_stripe(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + struct pnfs_block_volume *v = &volumes[idx]; + u64 len = 0; + int ret, i; + + d->children = kzalloc_objs(struct pnfs_block_dev, + v->stripe.volumes_count, gfp_mask); + if (!d->children) + return -ENOMEM; + + for (i = 0; i < v->stripe.volumes_count; i++) { + ret = bl_parse_deviceid(server, &d->children[i], + volumes, v->stripe.volumes[i], gfp_mask); + if (ret) + return ret; + + d->nr_children++; + len += d->children[i].len; + } + + d->len = len; + d->chunk_size = v->stripe.chunk_size; + d->map = bl_map_stripe; + return 0; +} + +static int +bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d, + struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask) +{ + d->type = volumes[idx].type; + + switch (d->type) { + case PNFS_BLOCK_VOLUME_SIMPLE: + return bl_parse_simple(server, d, volumes, idx, gfp_mask); + case PNFS_BLOCK_VOLUME_SLICE: + return bl_parse_slice(server, d, volumes, idx, gfp_mask); + case PNFS_BLOCK_VOLUME_CONCAT: + return bl_parse_concat(server, d, volumes, idx, gfp_mask); + case PNFS_BLOCK_VOLUME_STRIPE: + return bl_parse_stripe(server, d, volumes, idx, gfp_mask); + case PNFS_BLOCK_VOLUME_SCSI: + return bl_parse_scsi(server, d, volumes, idx, gfp_mask); + default: + dprintk("unsupported volume type: %d\n", d->type); + return -EIO; + } +} + +struct nfs4_deviceid_node * +bl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, + gfp_t gfp_mask) +{ + struct nfs4_deviceid_node *node = NULL; + struct pnfs_block_volume *volumes; + struct pnfs_block_dev *top; + struct xdr_stream xdr; + struct xdr_buf buf; + struct folio *scratch; + int nr_volumes, ret, i; + __be32 *p; + + scratch = folio_alloc(gfp_mask, 0); + if (!scratch) + goto out; + + xdr_init_decode_pages(&xdr, &buf, pdev->pages, pdev->pglen); + xdr_set_scratch_folio(&xdr, scratch); + + p = xdr_inline_decode(&xdr, sizeof(__be32)); + if (!p) + goto out_free_scratch; + nr_volumes = be32_to_cpup(p++); + + volumes = kzalloc_objs(struct pnfs_block_volume, nr_volumes, gfp_mask); + if (!volumes) + goto out_free_scratch; + + for (i = 0; i < nr_volumes; i++) { + ret = nfs4_block_decode_volume(&xdr, &volumes[i]); + if (ret < 0) + goto out_free_volumes; + } + + top = kzalloc_obj(*top, gfp_mask); + if (!top) + goto out_free_volumes; + + ret = bl_parse_deviceid(server, top, volumes, nr_volumes - 1, gfp_mask); + + node = &top->node; + nfs4_init_deviceid_node(node, server, &pdev->dev_id); + if (ret) + nfs4_mark_deviceid_unavailable(node); + +out_free_volumes: + kfree(volumes); +out_free_scratch: + folio_put(scratch); +out: + return node; +} diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c new file mode 100644 index 000000000..7393118c4 --- /dev/null +++ b/fs/nfs/blocklayout/extent_tree.c @@ -0,0 +1,726 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2014-2016 Christoph Hellwig. + */ + +#include + +#include "blocklayout.h" +#include "../nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +static inline struct pnfs_block_extent * +ext_node(struct rb_node *node) +{ + return rb_entry(node, struct pnfs_block_extent, be_node); +} + +static struct pnfs_block_extent * +ext_tree_first(struct rb_root *root) +{ + struct rb_node *node = rb_first(root); + return node ? ext_node(node) : NULL; +} + +static struct pnfs_block_extent * +ext_tree_prev(struct pnfs_block_extent *be) +{ + struct rb_node *node = rb_prev(&be->be_node); + return node ? ext_node(node) : NULL; +} + +static struct pnfs_block_extent * +ext_tree_next(struct pnfs_block_extent *be) +{ + struct rb_node *node = rb_next(&be->be_node); + return node ? ext_node(node) : NULL; +} + +static inline sector_t +ext_f_end(struct pnfs_block_extent *be) +{ + return be->be_f_offset + be->be_length; +} + +static struct pnfs_block_extent * +__ext_tree_search(struct rb_root *root, sector_t start) +{ + struct rb_node *node = root->rb_node; + struct pnfs_block_extent *be = NULL; + + while (node) { + be = ext_node(node); + if (start < be->be_f_offset) + node = node->rb_left; + else if (start >= ext_f_end(be)) + node = node->rb_right; + else + return be; + } + + if (be) { + if (start < be->be_f_offset) + return be; + + if (start >= ext_f_end(be)) + return ext_tree_next(be); + } + + return NULL; +} + +static bool +ext_can_merge(struct pnfs_block_extent *be1, struct pnfs_block_extent *be2) +{ + if (be1->be_state != be2->be_state) + return false; + if (be1->be_device != be2->be_device) + return false; + + if (be1->be_f_offset + be1->be_length != be2->be_f_offset) + return false; + + if (be1->be_state != PNFS_BLOCK_NONE_DATA && + (be1->be_v_offset + be1->be_length != be2->be_v_offset)) + return false; + + if (be1->be_state == PNFS_BLOCK_INVALID_DATA && + be1->be_tag != be2->be_tag) + return false; + + return true; +} + +static struct pnfs_block_extent * +ext_try_to_merge_left(struct rb_root *root, struct pnfs_block_extent *be) +{ + struct pnfs_block_extent *left = ext_tree_prev(be); + + if (left && ext_can_merge(left, be)) { + left->be_length += be->be_length; + rb_erase(&be->be_node, root); + nfs4_put_deviceid_node(be->be_device); + kfree(be); + return left; + } + + return be; +} + +static struct pnfs_block_extent * +ext_try_to_merge_right(struct rb_root *root, struct pnfs_block_extent *be) +{ + struct pnfs_block_extent *right = ext_tree_next(be); + + if (right && ext_can_merge(be, right)) { + be->be_length += right->be_length; + rb_erase(&right->be_node, root); + nfs4_put_deviceid_node(right->be_device); + kfree(right); + } + + return be; +} + +static void __ext_put_deviceids(struct list_head *head) +{ + struct pnfs_block_extent *be, *tmp; + + list_for_each_entry_safe(be, tmp, head, be_list) { + nfs4_put_deviceid_node(be->be_device); + kfree(be); + } +} + +static void +__ext_tree_insert(struct rb_root *root, + struct pnfs_block_extent *new, bool merge_ok) +{ + struct rb_node **p = &root->rb_node, *parent = NULL; + struct pnfs_block_extent *be; + + while (*p) { + parent = *p; + be = ext_node(parent); + + if (new->be_f_offset < be->be_f_offset) { + if (merge_ok && ext_can_merge(new, be)) { + be->be_f_offset = new->be_f_offset; + if (be->be_state != PNFS_BLOCK_NONE_DATA) + be->be_v_offset = new->be_v_offset; + be->be_length += new->be_length; + be = ext_try_to_merge_left(root, be); + goto free_new; + } + p = &(*p)->rb_left; + } else if (new->be_f_offset >= ext_f_end(be)) { + if (merge_ok && ext_can_merge(be, new)) { + be->be_length += new->be_length; + be = ext_try_to_merge_right(root, be); + goto free_new; + } + p = &(*p)->rb_right; + } else { + BUG(); + } + } + + rb_link_node(&new->be_node, parent, p); + rb_insert_color(&new->be_node, root); + return; +free_new: + nfs4_put_deviceid_node(new->be_device); + kfree(new); +} + +static int +__ext_tree_remove(struct rb_root *root, + sector_t start, sector_t end, struct list_head *tmp) +{ + struct pnfs_block_extent *be; + sector_t len1 = 0, len2 = 0; + sector_t orig_v_offset; + sector_t orig_len; + + be = __ext_tree_search(root, start); + if (!be) + return 0; + if (be->be_f_offset >= end) + return 0; + + orig_v_offset = be->be_v_offset; + orig_len = be->be_length; + + if (start > be->be_f_offset) + len1 = start - be->be_f_offset; + if (ext_f_end(be) > end) + len2 = ext_f_end(be) - end; + + if (len2 > 0) { + if (len1 > 0) { + struct pnfs_block_extent *new; + + new = kzalloc_obj(*new, GFP_ATOMIC); + if (!new) + return -ENOMEM; + + be->be_length = len1; + + new->be_f_offset = end; + if (be->be_state != PNFS_BLOCK_NONE_DATA) { + new->be_v_offset = + orig_v_offset + orig_len - len2; + } + new->be_length = len2; + new->be_state = be->be_state; + new->be_tag = be->be_tag; + new->be_device = nfs4_get_deviceid(be->be_device); + + __ext_tree_insert(root, new, true); + } else { + be->be_f_offset = end; + if (be->be_state != PNFS_BLOCK_NONE_DATA) { + be->be_v_offset = + orig_v_offset + orig_len - len2; + } + be->be_length = len2; + } + } else { + if (len1 > 0) { + be->be_length = len1; + be = ext_tree_next(be); + } + + while (be && ext_f_end(be) <= end) { + struct pnfs_block_extent *next = ext_tree_next(be); + + rb_erase(&be->be_node, root); + list_add_tail(&be->be_list, tmp); + be = next; + } + + if (be && be->be_f_offset < end) { + len1 = ext_f_end(be) - end; + be->be_f_offset = end; + if (be->be_state != PNFS_BLOCK_NONE_DATA) + be->be_v_offset += be->be_length - len1; + be->be_length = len1; + } + } + + return 0; +} + +int +ext_tree_insert(struct pnfs_block_layout *bl, struct pnfs_block_extent *new) +{ + struct pnfs_block_extent *be; + struct rb_root *root; + int err = 0; + + switch (new->be_state) { + case PNFS_BLOCK_READWRITE_DATA: + case PNFS_BLOCK_INVALID_DATA: + root = &bl->bl_ext_rw; + break; + case PNFS_BLOCK_READ_DATA: + case PNFS_BLOCK_NONE_DATA: + root = &bl->bl_ext_ro; + break; + default: + dprintk("invalid extent type\n"); + return -EINVAL; + } + + spin_lock(&bl->bl_ext_lock); +retry: + be = __ext_tree_search(root, new->be_f_offset); + if (!be || be->be_f_offset >= ext_f_end(new)) { + __ext_tree_insert(root, new, true); + } else if (new->be_f_offset >= be->be_f_offset) { + if (ext_f_end(new) <= ext_f_end(be)) { + nfs4_put_deviceid_node(new->be_device); + kfree(new); + } else { + sector_t new_len = ext_f_end(new) - ext_f_end(be); + sector_t diff = new->be_length - new_len; + + new->be_f_offset += diff; + new->be_v_offset += diff; + new->be_length = new_len; + goto retry; + } + } else if (ext_f_end(new) <= ext_f_end(be)) { + new->be_length = be->be_f_offset - new->be_f_offset; + __ext_tree_insert(root, new, true); + } else { + struct pnfs_block_extent *split; + sector_t new_len = ext_f_end(new) - ext_f_end(be); + sector_t diff = new->be_length - new_len; + + split = kmemdup(new, sizeof(*new), GFP_ATOMIC); + if (!split) { + err = -EINVAL; + goto out; + } + + split->be_length = be->be_f_offset - split->be_f_offset; + split->be_device = nfs4_get_deviceid(new->be_device); + __ext_tree_insert(root, split, true); + + new->be_f_offset += diff; + new->be_v_offset += diff; + new->be_length = new_len; + goto retry; + } +out: + spin_unlock(&bl->bl_ext_lock); + return err; +} + +static bool +__ext_tree_lookup(struct rb_root *root, sector_t isect, + struct pnfs_block_extent *ret) +{ + struct rb_node *node; + struct pnfs_block_extent *be; + + node = root->rb_node; + while (node) { + be = ext_node(node); + if (isect < be->be_f_offset) + node = node->rb_left; + else if (isect >= ext_f_end(be)) + node = node->rb_right; + else { + *ret = *be; + return true; + } + } + + return false; +} + +bool +ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect, + struct pnfs_block_extent *ret, bool rw) +{ + bool found = false; + + spin_lock(&bl->bl_ext_lock); + if (!rw) + found = __ext_tree_lookup(&bl->bl_ext_ro, isect, ret); + if (!found) + found = __ext_tree_lookup(&bl->bl_ext_rw, isect, ret); + spin_unlock(&bl->bl_ext_lock); + + return found; +} + +int ext_tree_remove(struct pnfs_block_layout *bl, bool rw, + sector_t start, sector_t end) +{ + int err, err2; + LIST_HEAD(tmp); + + spin_lock(&bl->bl_ext_lock); + err = __ext_tree_remove(&bl->bl_ext_ro, start, end, &tmp); + if (rw) { + err2 = __ext_tree_remove(&bl->bl_ext_rw, start, end, &tmp); + if (!err) + err = err2; + } + spin_unlock(&bl->bl_ext_lock); + + __ext_put_deviceids(&tmp); + return err; +} + +static int +ext_tree_split(struct rb_root *root, struct pnfs_block_extent *be, + sector_t split) +{ + struct pnfs_block_extent *new; + sector_t orig_len = be->be_length; + + new = kzalloc_obj(*new, GFP_ATOMIC); + if (!new) + return -ENOMEM; + + be->be_length = split - be->be_f_offset; + + new->be_f_offset = split; + if (be->be_state != PNFS_BLOCK_NONE_DATA) + new->be_v_offset = be->be_v_offset + be->be_length; + new->be_length = orig_len - be->be_length; + new->be_state = be->be_state; + new->be_tag = be->be_tag; + new->be_device = nfs4_get_deviceid(be->be_device); + + __ext_tree_insert(root, new, false); + return 0; +} + +int +ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start, + sector_t len, u64 lwb) +{ + struct rb_root *root = &bl->bl_ext_rw; + sector_t end = start + len; + struct pnfs_block_extent *be; + int err = 0; + LIST_HEAD(tmp); + + spin_lock(&bl->bl_ext_lock); + /* + * First remove all COW extents or holes from written to range. + */ + err = __ext_tree_remove(&bl->bl_ext_ro, start, end, &tmp); + if (err) + goto out; + + /* + * Then mark all invalid extents in the range as written to. + */ + for (be = __ext_tree_search(root, start); be; be = ext_tree_next(be)) { + if (be->be_f_offset >= end) + break; + + if (be->be_state != PNFS_BLOCK_INVALID_DATA || be->be_tag) + continue; + + if (be->be_f_offset < start) { + struct pnfs_block_extent *left = ext_tree_prev(be); + + if (left && ext_can_merge(left, be)) { + sector_t diff = start - be->be_f_offset; + + left->be_length += diff; + + be->be_f_offset += diff; + be->be_v_offset += diff; + be->be_length -= diff; + } else { + err = ext_tree_split(root, be, start); + if (err) + goto out; + } + } + + if (ext_f_end(be) > end) { + struct pnfs_block_extent *right = ext_tree_next(be); + + if (right && ext_can_merge(be, right)) { + sector_t diff = end - be->be_f_offset; + + be->be_length -= diff; + + right->be_f_offset -= diff; + right->be_v_offset -= diff; + right->be_length += diff; + } else { + err = ext_tree_split(root, be, end); + if (err) + goto out; + } + } + + if (be->be_f_offset >= start && ext_f_end(be) <= end) { + be->be_tag = EXTENT_WRITTEN; + be = ext_try_to_merge_left(root, be); + be = ext_try_to_merge_right(root, be); + } + } +out: + if (bl->bl_lwb < lwb) + bl->bl_lwb = lwb; + spin_unlock(&bl->bl_ext_lock); + + __ext_put_deviceids(&tmp); + return err; +} + +static size_t ext_tree_layoutupdate_size(struct pnfs_block_layout *bl, size_t count) +{ + if (bl->bl_scsi_layout) + return sizeof(__be32) + PNFS_SCSI_RANGE_SIZE * count; + else + return sizeof(__be32) + PNFS_BLOCK_EXTENT_SIZE * count; +} + +static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg, + size_t buffer_size) +{ + if (arg->layoutupdate_pages != &arg->layoutupdate_page) { + int nr_pages = DIV_ROUND_UP(buffer_size, PAGE_SIZE), i; + + for (i = 0; i < nr_pages; i++) + put_page(arg->layoutupdate_pages[i]); + vfree(arg->start_p); + kfree(arg->layoutupdate_pages); + } else { + put_page(arg->layoutupdate_page); + } +} + +static __be32 *encode_block_extent(struct pnfs_block_extent *be, __be32 *p) +{ + p = xdr_encode_opaque_fixed(p, be->be_device->deviceid.data, + NFS4_DEVICEID4_SIZE); + p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT); + p = xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT); + p = xdr_encode_hyper(p, 0LL); + *p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA); + return p; +} + +static __be32 *encode_scsi_range(struct pnfs_block_extent *be, __be32 *p) +{ + p = xdr_encode_hyper(p, be->be_f_offset << SECTOR_SHIFT); + return xdr_encode_hyper(p, be->be_length << SECTOR_SHIFT); +} + +/** + * ext_tree_try_encode_commit - try to encode all extents into the buffer + * @bl: pointer to the layout + * @p: pointer to the output buffer + * @buffer_size: size of the output buffer + * @count: output pointer to the number of encoded extents + * @lastbyte: output pointer to the last written byte + * + * Return values: + * %0: Success, all required extents encoded, outputs are valid + * %-ENOSPC: Buffer too small, nothing encoded, outputs are invalid + */ +static int +ext_tree_try_encode_commit(struct pnfs_block_layout *bl, __be32 *p, + size_t buffer_size, size_t *count, __u64 *lastbyte) +{ + struct pnfs_block_extent *be; + + spin_lock(&bl->bl_ext_lock); + for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) { + if (be->be_state != PNFS_BLOCK_INVALID_DATA || + be->be_tag != EXTENT_WRITTEN) + continue; + + (*count)++; + if (ext_tree_layoutupdate_size(bl, *count) > buffer_size) { + spin_unlock(&bl->bl_ext_lock); + return -ENOSPC; + } + } + for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) { + if (be->be_state != PNFS_BLOCK_INVALID_DATA || + be->be_tag != EXTENT_WRITTEN) + continue; + + if (bl->bl_scsi_layout) + p = encode_scsi_range(be, p); + else + p = encode_block_extent(be, p); + be->be_tag = EXTENT_COMMITTING; + } + *lastbyte = (bl->bl_lwb != 0) ? bl->bl_lwb - 1 : U64_MAX; + bl->bl_lwb = 0; + spin_unlock(&bl->bl_ext_lock); + + return 0; +} + +/** + * ext_tree_encode_commit - encode as much as possible extents into the buffer + * @bl: pointer to the layout + * @p: pointer to the output buffer + * @buffer_size: size of the output buffer + * @count: output pointer to the number of encoded extents + * @lastbyte: output pointer to the last written byte + * + * Return values: + * %0: Success, all required extents encoded, outputs are valid + * %-ENOSPC: Buffer too small, some extents are encoded, outputs are valid + */ +static int +ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p, + size_t buffer_size, size_t *count, __u64 *lastbyte) +{ + struct pnfs_block_extent *be, *be_prev; + int ret = 0; + + spin_lock(&bl->bl_ext_lock); + for (be = ext_tree_first(&bl->bl_ext_rw); be; be = ext_tree_next(be)) { + if (be->be_state != PNFS_BLOCK_INVALID_DATA || + be->be_tag != EXTENT_WRITTEN) + continue; + + (*count)++; + if (ext_tree_layoutupdate_size(bl, *count) > buffer_size) { + (*count)--; + ret = -ENOSPC; + break; + } + + if (bl->bl_scsi_layout) + p = encode_scsi_range(be, p); + else + p = encode_block_extent(be, p); + be->be_tag = EXTENT_COMMITTING; + be_prev = be; + } + if (!ret) { + *lastbyte = (bl->bl_lwb != 0) ? bl->bl_lwb - 1 : U64_MAX; + bl->bl_lwb = 0; + } else { + *lastbyte = be_prev->be_f_offset + be_prev->be_length; + *lastbyte <<= SECTOR_SHIFT; + *lastbyte -= 1; + } + spin_unlock(&bl->bl_ext_lock); + + return ret; +} + +/** + * ext_tree_prepare_commit - encode extents that need to be committed + * @arg: layout commit data + * + * Return values: + * %0: Success, all required extents are encoded + * %-ENOSPC: Some extents are encoded, but not all, due to RPC size limit + * %-ENOMEM: Out of memory, extents not encoded + */ +int +ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg) +{ + struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout); + size_t count = 0, buffer_size = PAGE_SIZE; + __be32 *start_p; + int ret; + + arg->layoutupdate_page = alloc_page(GFP_NOFS); + if (!arg->layoutupdate_page) + return -ENOMEM; + start_p = page_address(arg->layoutupdate_page); + arg->layoutupdate_pages = &arg->layoutupdate_page; + + ret = ext_tree_try_encode_commit(bl, start_p + 1, buffer_size, + &count, &arg->lastbytewritten); + if (unlikely(ret)) { + ext_tree_free_commitdata(arg, buffer_size); + + buffer_size = NFS_SERVER(arg->inode)->wsize; + count = 0; + + arg->layoutupdate_pages = + kzalloc_objs(struct page *, + DIV_ROUND_UP(buffer_size, PAGE_SIZE), + GFP_NOFS); + if (!arg->layoutupdate_pages) + return -ENOMEM; + + start_p = __vmalloc(buffer_size, GFP_NOFS); + if (!start_p) { + kfree(arg->layoutupdate_pages); + return -ENOMEM; + } + + ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size, + &count, &arg->lastbytewritten); + } + + *start_p = cpu_to_be32(count); + arg->layoutupdate_len = ext_tree_layoutupdate_size(bl, count); + + if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) { + void *p = start_p, *end = p + arg->layoutupdate_len; + struct page *page = NULL; + int i = 0; + + arg->start_p = start_p; + for ( ; p < end; p += PAGE_SIZE) { + page = vmalloc_to_page(p); + arg->layoutupdate_pages[i++] = page; + get_page(page); + } + } + + trace_bl_ext_tree_prepare_commit(ret, count, + arg->lastbytewritten, !!ret); + return ret; +} + +void +ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status) +{ + struct pnfs_block_layout *bl = BLK_LO2EXT(NFS_I(arg->inode)->layout); + struct rb_root *root = &bl->bl_ext_rw; + struct pnfs_block_extent *be; + + dprintk("%s status %d\n", __func__, status); + + ext_tree_free_commitdata(arg, arg->layoutupdate_len); + + spin_lock(&bl->bl_ext_lock); + for (be = ext_tree_first(root); be; be = ext_tree_next(be)) { + if (be->be_state != PNFS_BLOCK_INVALID_DATA || + be->be_tag != EXTENT_COMMITTING) + continue; + + if (status) { + /* + * Mark as written and try again. + * + * XXX: some real error handling here wouldn't hurt.. + */ + be->be_tag = EXTENT_WRITTEN; + } else { + be->be_state = PNFS_BLOCK_READWRITE_DATA; + be->be_tag = 0; + } + + be = ext_try_to_merge_left(root, be); + be = ext_try_to_merge_right(root, be); + } + spin_unlock(&bl->bl_ext_lock); +} diff --git a/fs/nfs/blocklayout/rpc_pipefs.c b/fs/nfs/blocklayout/rpc_pipefs.c new file mode 100644 index 000000000..d526f5ba7 --- /dev/null +++ b/fs/nfs/blocklayout/rpc_pipefs.c @@ -0,0 +1,270 @@ +/* + * Copyright (c) 2006,2007 The Regents of the University of Michigan. + * All rights reserved. + * + * Andy Adamson + * Fred Isaman + * + * permission is granted to use, copy, create derivative works and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the university of michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. if + * the above copyright notice or any other identification of the + * university of michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * this software is provided as is, without representation from the + * university of michigan as to its fitness for any purpose, and without + * warranty by the university of michigan of any kind, either express + * or implied, including without limitation the implied warranties of + * merchantability and fitness for a particular purpose. the regents + * of the university of michigan shall not be liable for any damages, + * including special, indirect, incidental, or consequential damages, + * with respect to any claim arising out or in connection with the use + * of the software, even if it has been or is hereafter advised of the + * possibility of such damages. + */ + +#include +#include + +#include "blocklayout.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +static void +nfs4_encode_simple(__be32 *p, struct pnfs_block_volume *b) +{ + int i; + + *p++ = cpu_to_be32(1); + *p++ = cpu_to_be32(b->type); + *p++ = cpu_to_be32(b->simple.nr_sigs); + for (i = 0; i < b->simple.nr_sigs; i++) { + p = xdr_encode_hyper(p, b->simple.sigs[i].offset); + p = xdr_encode_opaque(p, b->simple.sigs[i].sig, + b->simple.sigs[i].sig_len); + } +} + +dev_t +bl_resolve_deviceid(struct nfs_server *server, struct pnfs_block_volume *b, + gfp_t gfp_mask) +{ + struct net *net = server->nfs_client->cl_net; + struct nfs_net *nn = net_generic(net, nfs_net_id); + struct bl_dev_msg *reply = &nn->bl_mount_reply; + struct bl_pipe_msg bl_pipe_msg; + struct rpc_pipe_msg *msg = &bl_pipe_msg.msg; + struct bl_msg_hdr *bl_msg; + DECLARE_WAITQUEUE(wq, current); + dev_t dev = 0; + int rc; + + dprintk("%s CREATING PIPEFS MESSAGE\n", __func__); + + mutex_lock(&nn->bl_mutex); + bl_pipe_msg.bl_wq = &nn->bl_wq; + + b->simple.len += 4; /* single volume */ + if (b->simple.len > PAGE_SIZE) + goto out_unlock; + + memset(msg, 0, sizeof(*msg)); + msg->len = sizeof(*bl_msg) + b->simple.len; + msg->data = kzalloc(msg->len, gfp_mask); + if (!msg->data) + goto out_unlock; + + bl_msg = msg->data; + bl_msg->type = BL_DEVICE_MOUNT; + bl_msg->totallen = b->simple.len; + nfs4_encode_simple(msg->data + sizeof(*bl_msg), b); + + dprintk("%s CALLING USERSPACE DAEMON\n", __func__); + add_wait_queue(&nn->bl_wq, &wq); + rc = rpc_queue_upcall(nn->bl_device_pipe, msg); + if (rc < 0) { + remove_wait_queue(&nn->bl_wq, &wq); + goto out_free_data; + } + + set_current_state(TASK_UNINTERRUPTIBLE); + schedule(); + remove_wait_queue(&nn->bl_wq, &wq); + + if (reply->status != BL_DEVICE_REQUEST_PROC) { + printk(KERN_WARNING "%s failed to decode device: %d\n", + __func__, reply->status); + goto out_free_data; + } + + dev = MKDEV(reply->major, reply->minor); +out_free_data: + kfree(msg->data); +out_unlock: + mutex_unlock(&nn->bl_mutex); + return dev; +} + +static ssize_t bl_pipe_downcall(struct file *filp, const char __user *src, + size_t mlen) +{ + struct nfs_net *nn = net_generic(file_inode(filp)->i_sb->s_fs_info, + nfs_net_id); + + if (mlen != sizeof (struct bl_dev_msg)) + return -EINVAL; + + if (copy_from_user(&nn->bl_mount_reply, src, mlen) != 0) + return -EFAULT; + + wake_up(&nn->bl_wq); + + return mlen; +} + +static void bl_pipe_destroy_msg(struct rpc_pipe_msg *msg) +{ + struct bl_pipe_msg *bl_pipe_msg = + container_of(msg, struct bl_pipe_msg, msg); + + if (msg->errno >= 0) + return; + wake_up(bl_pipe_msg->bl_wq); +} + +static const struct rpc_pipe_ops bl_upcall_ops = { + .upcall = rpc_pipe_generic_upcall, + .downcall = bl_pipe_downcall, + .destroy_msg = bl_pipe_destroy_msg, +}; + +static int nfs4blocklayout_register_sb(struct super_block *sb, + struct rpc_pipe *pipe) +{ + struct dentry *dir; + int err; + + dir = rpc_d_lookup_sb(sb, NFS_PIPE_DIRNAME); + if (dir == NULL) + return -ENOENT; + err = rpc_mkpipe_dentry(dir, "blocklayout", NULL, pipe); + dput(dir); + return err; +} + +static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event, + void *ptr) +{ + struct super_block *sb = ptr; + struct net *net = sb->s_fs_info; + struct nfs_net *nn = net_generic(net, nfs_net_id); + int ret = 0; + + if (!try_module_get(THIS_MODULE)) + return 0; + + if (nn->bl_device_pipe == NULL) { + module_put(THIS_MODULE); + return 0; + } + + switch (event) { + case RPC_PIPEFS_MOUNT: + ret = nfs4blocklayout_register_sb(sb, nn->bl_device_pipe); + break; + case RPC_PIPEFS_UMOUNT: + rpc_unlink(nn->bl_device_pipe); + break; + default: + ret = -ENOTSUPP; + break; + } + module_put(THIS_MODULE); + return ret; +} + +static struct notifier_block nfs4blocklayout_block = { + .notifier_call = rpc_pipefs_event, +}; + +static int nfs4blocklayout_register_net(struct net *net, struct rpc_pipe *pipe) +{ + struct super_block *pipefs_sb; + int ret; + + pipefs_sb = rpc_get_sb_net(net); + if (!pipefs_sb) + return 0; + ret = nfs4blocklayout_register_sb(pipefs_sb, pipe); + rpc_put_sb_net(net); + return ret; +} + +static void nfs4blocklayout_unregister_net(struct net *net, + struct rpc_pipe *pipe) +{ + struct super_block *pipefs_sb; + + pipefs_sb = rpc_get_sb_net(net); + if (pipefs_sb) { + rpc_unlink(pipe); + rpc_put_sb_net(net); + } +} + +static int nfs4blocklayout_net_init(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + int err; + + mutex_init(&nn->bl_mutex); + init_waitqueue_head(&nn->bl_wq); + nn->bl_device_pipe = rpc_mkpipe_data(&bl_upcall_ops, 0); + if (IS_ERR(nn->bl_device_pipe)) + return PTR_ERR(nn->bl_device_pipe); + err = nfs4blocklayout_register_net(net, nn->bl_device_pipe); + if (unlikely(err)) + rpc_destroy_pipe_data(nn->bl_device_pipe); + return err; +} + +static void nfs4blocklayout_net_exit(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + nfs4blocklayout_unregister_net(net, nn->bl_device_pipe); + rpc_destroy_pipe_data(nn->bl_device_pipe); + nn->bl_device_pipe = NULL; +} + +static struct pernet_operations nfs4blocklayout_net_ops = { + .init = nfs4blocklayout_net_init, + .exit = nfs4blocklayout_net_exit, +}; + +int __init bl_init_pipefs(void) +{ + int ret; + + ret = rpc_pipefs_notifier_register(&nfs4blocklayout_block); + if (ret) + goto out; + ret = register_pernet_subsys(&nfs4blocklayout_net_ops); + if (ret) + goto out_unregister_notifier; + return 0; + +out_unregister_notifier: + rpc_pipefs_notifier_unregister(&nfs4blocklayout_block); +out: + return ret; +} + +void bl_cleanup_pipefs(void) +{ + rpc_pipefs_notifier_unregister(&nfs4blocklayout_block); + unregister_pernet_subsys(&nfs4blocklayout_net_ops); +} diff --git a/fs/nfs/cache_lib.c b/fs/nfs/cache_lib.c new file mode 100644 index 000000000..9738a1ae9 --- /dev/null +++ b/fs/nfs/cache_lib.c @@ -0,0 +1,158 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/cache_lib.c + * + * Helper routines for the NFS client caches + * + * Copyright (c) 2009 Trond Myklebust + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "cache_lib.h" + +#define NFS_CACHE_UPCALL_PATHLEN 256 +#define NFS_CACHE_UPCALL_TIMEOUT 15 + +static char nfs_cache_getent_prog[NFS_CACHE_UPCALL_PATHLEN] = + "/sbin/nfs_cache_getent"; +static unsigned long nfs_cache_getent_timeout = NFS_CACHE_UPCALL_TIMEOUT; + +module_param_string(cache_getent, nfs_cache_getent_prog, + sizeof(nfs_cache_getent_prog), 0600); +MODULE_PARM_DESC(cache_getent, "Path to the client cache upcall program"); +module_param_named(cache_getent_timeout, nfs_cache_getent_timeout, ulong, 0600); +MODULE_PARM_DESC(cache_getent_timeout, "Timeout (in seconds) after which " + "the cache upcall is assumed to have failed"); + +int nfs_cache_upcall(struct cache_detail *cd, char *entry_name) +{ + static char *envp[] = { "HOME=/", + "TERM=linux", + "PATH=/sbin:/usr/sbin:/bin:/usr/bin", + NULL + }; + char *argv[] = { + nfs_cache_getent_prog, + cd->name, + entry_name, + NULL + }; + int ret = -EACCES; + + if (nfs_cache_getent_prog[0] == '\0') + goto out; + ret = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC); + /* + * Disable the upcall mechanism if we're getting an ENOENT or + * EACCES error. The admin can re-enable it on the fly by using + * sysfs to set the 'cache_getent' parameter once the problem + * has been fixed. + */ + if (ret == -ENOENT || ret == -EACCES) + nfs_cache_getent_prog[0] = '\0'; +out: + return ret > 0 ? 0 : ret; +} + +/* + * Deferred request handling + */ +void nfs_cache_defer_req_put(struct nfs_cache_defer_req *dreq) +{ + if (refcount_dec_and_test(&dreq->count)) + kfree(dreq); +} + +static void nfs_dns_cache_revisit(struct cache_deferred_req *d, int toomany) +{ + struct nfs_cache_defer_req *dreq; + + dreq = container_of(d, struct nfs_cache_defer_req, deferred_req); + + complete(&dreq->completion); + nfs_cache_defer_req_put(dreq); +} + +static struct cache_deferred_req *nfs_dns_cache_defer(struct cache_req *req) +{ + struct nfs_cache_defer_req *dreq; + + dreq = container_of(req, struct nfs_cache_defer_req, req); + dreq->deferred_req.revisit = nfs_dns_cache_revisit; + refcount_inc(&dreq->count); + + return &dreq->deferred_req; +} + +struct nfs_cache_defer_req *nfs_cache_defer_req_alloc(void) +{ + struct nfs_cache_defer_req *dreq; + + dreq = kzalloc_obj(*dreq); + if (dreq) { + init_completion(&dreq->completion); + refcount_set(&dreq->count, 1); + dreq->req.defer = nfs_dns_cache_defer; + } + return dreq; +} + +int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq) +{ + if (wait_for_completion_timeout(&dreq->completion, + nfs_cache_getent_timeout * HZ) == 0) + return -ETIMEDOUT; + return 0; +} + +int nfs_cache_register_sb(struct super_block *sb, struct cache_detail *cd) +{ + int ret; + struct dentry *dir; + + dir = rpc_d_lookup_sb(sb, "cache"); + ret = sunrpc_cache_register_pipefs(dir, cd->name, 0600, cd); + dput(dir); + return ret; +} + +int nfs_cache_register_net(struct net *net, struct cache_detail *cd) +{ + struct super_block *pipefs_sb; + int ret = 0; + + sunrpc_init_cache_detail(cd); + pipefs_sb = rpc_get_sb_net(net); + if (pipefs_sb) { + ret = nfs_cache_register_sb(pipefs_sb, cd); + rpc_put_sb_net(net); + if (ret) + sunrpc_destroy_cache_detail(cd); + } + return ret; +} + +void nfs_cache_unregister_sb(struct super_block *sb, struct cache_detail *cd) +{ + sunrpc_cache_unregister_pipefs(cd); +} + +void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd) +{ + struct super_block *pipefs_sb; + + pipefs_sb = rpc_get_sb_net(net); + if (pipefs_sb) { + nfs_cache_unregister_sb(pipefs_sb, cd); + rpc_put_sb_net(net); + } + sunrpc_destroy_cache_detail(cd); +} diff --git a/fs/nfs/cache_lib.h b/fs/nfs/cache_lib.h new file mode 100644 index 000000000..220ee409a --- /dev/null +++ b/fs/nfs/cache_lib.h @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Helper routines for the NFS client caches + * + * Copyright (c) 2009 Trond Myklebust + */ + +#include +#include +#include + +/* + * Deferred request handling + */ +struct nfs_cache_defer_req { + struct cache_req req; + struct cache_deferred_req deferred_req; + struct completion completion; + refcount_t count; +}; + +extern int nfs_cache_upcall(struct cache_detail *cd, char *entry_name); +extern struct nfs_cache_defer_req *nfs_cache_defer_req_alloc(void); +extern void nfs_cache_defer_req_put(struct nfs_cache_defer_req *dreq); +extern int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq); + +extern int nfs_cache_register_net(struct net *net, struct cache_detail *cd); +extern void nfs_cache_unregister_net(struct net *net, struct cache_detail *cd); +extern int nfs_cache_register_sb(struct super_block *sb, + struct cache_detail *cd); +extern void nfs_cache_unregister_sb(struct super_block *sb, + struct cache_detail *cd); diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c new file mode 100644 index 000000000..bc282b744 --- /dev/null +++ b/fs/nfs/callback.c @@ -0,0 +1,355 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/callback.c + * + * Copyright (C) 2004 Trond Myklebust + * + * NFSv4 callback handling + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "nfs4_fs.h" +#include "callback.h" +#include "internal.h" +#include "netns.h" + +#define NFSDBG_FACILITY NFSDBG_CALLBACK + +struct nfs_callback_data { + unsigned int users; + struct svc_serv *serv; +}; + +static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1]; +static DEFINE_MUTEX(nfs_callback_mutex); +static struct svc_program nfs4_callback_program; + +static int nfs4_callback_up_net(struct svc_serv *serv, struct net *net) +{ + const struct cred *cred = current_cred(); + int ret; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + ret = svc_xprt_create(serv, "tcp", net, PF_INET, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS, + cred); + if (ret <= 0) + goto out_err; + nn->nfs_callback_tcpport = ret; + dprintk("NFS: Callback listener port = %u (af %u, net %x)\n", + nn->nfs_callback_tcpport, PF_INET, net->ns.inum); + + ret = svc_xprt_create(serv, "tcp", net, PF_INET6, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS, + cred); + if (ret > 0) { + nn->nfs_callback_tcpport6 = ret; + dprintk("NFS: Callback listener port = %u (af %u, net %x)\n", + nn->nfs_callback_tcpport6, PF_INET6, net->ns.inum); + } else if (ret != -EAFNOSUPPORT) + goto out_err; + return 0; + +out_err: + return (ret) ? ret : -ENOMEM; +} + +/* + * This is the NFSv4 callback kernel thread. + */ +static int +nfs4_callback_svc(void *vrqstp) +{ + struct svc_rqst *rqstp = vrqstp; + + svc_thread_init_status(rqstp, 0); + + set_freezable(); + + while (!svc_thread_should_stop(rqstp)) + svc_recv(rqstp, 0); + + svc_exit_thread(rqstp); + return 0; +} + +static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, + struct svc_serv *serv) +{ + if (minorversion) + /* + * Save the svc_serv in the transport so that it can + * be referenced when the session backchannel is initialized + */ + xprt->bc_serv = serv; +} + +static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt, + struct svc_serv *serv) +{ + int nrservs = nfs_callback_nr_threads; + int ret; + + nfs_callback_bc_serv(minorversion, xprt, serv); + + if (nrservs < NFS4_MIN_NR_CALLBACK_THREADS) + nrservs = NFS4_MIN_NR_CALLBACK_THREADS; + + if (serv->sv_nrthreads == nrservs) + return 0; + + ret = svc_set_num_threads(serv, 0, nrservs); + if (ret) { + svc_set_num_threads(serv, 0, 0); + return ret; + } + dprintk("nfs_callback_up: service started\n"); + return 0; +} + +static void nfs_callback_down_net(u32 minorversion, struct svc_serv *serv, struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + if (--nn->cb_users[minorversion]) + return; + + dprintk("NFS: destroy per-net callback data; net=%x\n", net->ns.inum); + svc_xprt_destroy_all(serv, net, false); +} + +static int nfs_callback_up_net(int minorversion, struct svc_serv *serv, + struct net *net, struct rpc_xprt *xprt) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + int ret; + + if (nn->cb_users[minorversion]++) + return 0; + + dprintk("NFS: create per-net callback data; net=%x\n", net->ns.inum); + + ret = svc_bind(serv, net); + if (ret < 0) { + printk(KERN_WARNING "NFS: bind callback service failed\n"); + goto err; + } + + ret = 0; + if (minorversion == 0) + ret = nfs4_callback_up_net(serv, net); + else if (xprt->ops->bc_setup) + set_bc_enabled(serv); + else + ret = -EPROTONOSUPPORT; + + if (ret < 0) { + printk(KERN_ERR "NFS: callback service start failed\n"); + goto err; + } + return 0; + +err: + nn->cb_users[minorversion]--; + dprintk("NFS: Couldn't create callback socket: err = %d; " + "net = %x\n", ret, net->ns.inum); + return ret; +} + +static struct svc_serv *nfs_callback_create_svc(int minorversion) +{ + struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + int (*threadfn)(void *data); + struct svc_serv *serv; + + /* + * Check whether we're already up and running. + */ + if (cb_info->serv) + return cb_info->serv; + + /* + * Sanity check: if there's no task, + * we should be the first user ... + */ + if (cb_info->users) + printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n", + cb_info->users); + + threadfn = nfs4_callback_svc; + serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, + threadfn); + if (!serv) { + printk(KERN_ERR "nfs_callback_create_svc: create service failed\n"); + return ERR_PTR(-ENOMEM); + } + cb_info->serv = serv; + dprintk("nfs_callback_create_svc: service created\n"); + return serv; +} + +/* + * Bring up the callback thread if it is not already up. + */ +int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) +{ + struct svc_serv *serv; + struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + int ret; + struct net *net = xprt->xprt_net; + + mutex_lock(&nfs_callback_mutex); + + serv = nfs_callback_create_svc(minorversion); + if (IS_ERR(serv)) { + ret = PTR_ERR(serv); + goto err_create; + } + + ret = nfs_callback_up_net(minorversion, serv, net, xprt); + if (ret < 0) + goto err_net; + + ret = nfs_callback_start_svc(minorversion, xprt, serv); + if (ret < 0) + goto err_start; + + cb_info->users++; +err_net: + if (!cb_info->users) { + xprt_svc_shutdown_bc(xprt); + svc_set_num_threads(cb_info->serv, 0, 0); + xprt_svc_destroy_nullify_bc(xprt, &cb_info->serv); + } +err_create: + mutex_unlock(&nfs_callback_mutex); + return ret; + +err_start: + nfs_callback_down_net(minorversion, serv, net); + dprintk("NFS: Couldn't create server thread; err = %d\n", ret); + goto err_net; +} + +/* + * Kill the callback thread if it's no longer being used. + */ +void nfs_callback_down(int minorversion, struct net *net, struct rpc_xprt *xprt) +{ + struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + struct svc_serv *serv; + + mutex_lock(&nfs_callback_mutex); + serv = cb_info->serv; + xprt_svc_shutdown_bc(xprt); + nfs_callback_down_net(minorversion, serv, net); + cb_info->users--; + if (cb_info->users == 0) { + svc_set_num_threads(serv, 0, 0); + dprintk("nfs_callback_down: service destroyed\n"); + xprt_svc_destroy_nullify_bc(xprt, &cb_info->serv); + } + mutex_unlock(&nfs_callback_mutex); +} + +/* Boolean check of RPC_AUTH_GSS principal */ +int +check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp) +{ + char *p = rqstp->rq_cred.cr_principal; + + if (rqstp->rq_authop->flavour != RPC_AUTH_GSS) + return 1; + + /* No RPC_AUTH_GSS on NFSv4.1 back channel yet */ + if (clp->cl_minorversion != 0) + return 0; + /* + * It might just be a normal user principal, in which case + * userspace won't bother to tell us the name at all. + */ + if (p == NULL) + return 0; + + /* + * Did we get the acceptor from userland during the SETCLIENID + * negotiation? + */ + if (clp->cl_acceptor) + return !strcmp(p, clp->cl_acceptor); + + /* + * Otherwise try to verify it using the cl_hostname. Note that this + * doesn't work if a non-canonical hostname was used in the devname. + */ + + /* Expect a GSS_C_NT_HOSTBASED_NAME like "nfs@serverhostname" */ + + if (memcmp(p, "nfs@", 4) != 0) + return 0; + p += 4; + if (strcmp(p, clp->cl_hostname) != 0) + return 0; + return 1; +} + +/* + * pg_authenticate method for nfsv4 callback threads. + * + * The authflavor has been negotiated, so an incorrect flavor is a server + * bug. Deny packets with incorrect authflavor. + * + * All other checking done after NFS decoding where the nfs_client can be + * found in nfs4_callback_compound + */ +static enum svc_auth_status nfs_callback_authenticate(struct svc_rqst *rqstp) +{ + rqstp->rq_auth_stat = rpc_autherr_badcred; + + switch (rqstp->rq_authop->flavour) { + case RPC_AUTH_NULL: + if (rqstp->rq_proc != CB_NULL) + return SVC_DENIED; + break; + case RPC_AUTH_GSS: + /* No RPC_AUTH_GSS support yet in NFSv4.1 */ + if (svc_is_backchannel(rqstp)) + return SVC_DENIED; + } + + rqstp->rq_auth_stat = rpc_auth_ok; + return SVC_OK; +} + +/* + * Define NFS4 callback program + */ +static const struct svc_version *nfs4_callback_version[] = { + [1] = &nfs4_callback_version1, + [4] = &nfs4_callback_version4, +}; + +static struct svc_program nfs4_callback_program = { + .pg_prog = NFS4_CALLBACK, /* RPC service number */ + .pg_nvers = ARRAY_SIZE(nfs4_callback_version), /* Number of entries */ + .pg_vers = nfs4_callback_version, /* version table */ + .pg_name = "NFSv4 callback", /* service name */ + .pg_class = "nfs", /* authentication class */ + .pg_authenticate = nfs_callback_authenticate, + .pg_init_request = svc_generic_init_request, + .pg_rpcbind_set = svc_generic_rpcbind_set, +}; diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h new file mode 100644 index 000000000..2a721c422 --- /dev/null +++ b/fs/nfs/callback.h @@ -0,0 +1,204 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * linux/fs/nfs/callback.h + * + * Copyright (C) 2004 Trond Myklebust + * + * NFSv4 callback definitions + */ +#ifndef __LINUX_FS_NFS_CALLBACK_H +#define __LINUX_FS_NFS_CALLBACK_H +#include + +#define NFS4_CALLBACK 0x40000000 +#define NFS4_CALLBACK_XDRSIZE 2048 +#define NFS4_CALLBACK_BUFSIZE (1024 + NFS4_CALLBACK_XDRSIZE) + +enum nfs4_callback_procnum { + CB_NULL = 0, + CB_COMPOUND = 1, +}; + +struct nfs4_slot; +struct cb_process_state { + struct nfs_client *clp; + struct nfs4_slot *slot; + struct net *net; + u32 minorversion; + __be32 drc_status; + unsigned int referring_calls; +}; + +struct cb_compound_hdr_arg { + unsigned int taglen; + const char *tag; + unsigned int minorversion; + unsigned int cb_ident; /* v4.0 callback identifier */ + unsigned nops; +}; + +struct cb_compound_hdr_res { + __be32 *status; + unsigned int taglen; + const char *tag; + __be32 *nops; +}; + +struct cb_getattrargs { + struct nfs_fh fh; + uint32_t bitmap[3]; +}; + +struct cb_getattrres { + __be32 status; + uint32_t bitmap[3]; + uint64_t size; + uint64_t change_attr; + struct timespec64 atime; + struct timespec64 ctime; + struct timespec64 mtime; +}; + +struct cb_recallargs { + struct nfs_fh fh; + nfs4_stateid stateid; + uint32_t truncate; +}; + +struct referring_call { + uint32_t rc_sequenceid; + uint32_t rc_slotid; +}; + +struct referring_call_list { + struct nfs4_sessionid rcl_sessionid; + uint32_t rcl_nrefcalls; + struct referring_call *rcl_refcalls; +}; + +struct cb_sequenceargs { + struct sockaddr *csa_addr; + struct nfs4_sessionid csa_sessionid; + uint32_t csa_sequenceid; + uint32_t csa_slotid; + uint32_t csa_highestslotid; + uint32_t csa_cachethis; + uint32_t csa_nrclists; + struct referring_call_list *csa_rclists; +}; + +struct cb_sequenceres { + __be32 csr_status; + struct nfs4_sessionid csr_sessionid; + uint32_t csr_sequenceid; + uint32_t csr_slotid; + uint32_t csr_highestslotid; + uint32_t csr_target_highestslotid; +}; + +extern __be32 nfs4_callback_sequence(void *argp, void *resp, + struct cb_process_state *cps); + +#define RCA4_TYPE_MASK_RDATA_DLG 0 +#define RCA4_TYPE_MASK_WDATA_DLG 1 +#define RCA4_TYPE_MASK_DIR_DLG 2 +#define RCA4_TYPE_MASK_FILE_LAYOUT 3 +#define RCA4_TYPE_MASK_BLK_LAYOUT 4 +#define RCA4_TYPE_MASK_OBJ_LAYOUT_MIN 8 +#define RCA4_TYPE_MASK_OBJ_LAYOUT_MAX 9 +#define RCA4_TYPE_MASK_OTHER_LAYOUT_MIN 12 +#define RCA4_TYPE_MASK_OTHER_LAYOUT_MAX 15 +#define PNFS_FF_RCA4_TYPE_MASK_READ 16 +#define PNFS_FF_RCA4_TYPE_MASK_RW 17 +#define RCA4_TYPE_MASK_ALL 0x3f31f + +struct cb_recallanyargs { + uint32_t craa_objs_to_keep; + uint32_t craa_type_mask; +}; + +extern __be32 nfs4_callback_recallany(void *argp, void *resp, + struct cb_process_state *cps); + +struct cb_recallslotargs { + uint32_t crsa_target_highest_slotid; +}; +extern __be32 nfs4_callback_recallslot(void *argp, void *resp, + struct cb_process_state *cps); + +struct cb_layoutrecallargs { + uint32_t cbl_recall_type; + uint32_t cbl_layout_type; + uint32_t cbl_layoutchanged; + union { + struct { + struct nfs_fh cbl_fh; + struct pnfs_layout_range cbl_range; + nfs4_stateid cbl_stateid; + }; + struct nfs_fsid cbl_fsid; + }; +}; + +extern __be32 nfs4_callback_layoutrecall(void *argp, void *resp, + struct cb_process_state *cps); + +struct cb_devicenotifyitem { + uint32_t cbd_notify_type; + uint32_t cbd_layout_type; + struct nfs4_deviceid cbd_dev_id; + uint32_t cbd_immediate; +}; + +struct cb_devicenotifyargs { + uint32_t ndevs; + struct cb_devicenotifyitem *devs; +}; + +extern __be32 nfs4_callback_devicenotify(void *argp, void *resp, + struct cb_process_state *cps); + +struct cb_notify_lock_args { + struct nfs_fh cbnl_fh; + struct nfs_lowner cbnl_owner; + bool cbnl_valid; +}; + +extern __be32 nfs4_callback_notify_lock(void *argp, void *resp, + struct cb_process_state *cps); +#ifdef CONFIG_NFS_V4_2 +struct cb_offloadargs { + struct nfs_fh coa_fh; + nfs4_stateid coa_stateid; + uint32_t error; + uint64_t wr_count; + struct nfs_writeverf wr_writeverf; +}; + +extern __be32 nfs4_callback_offload(void *args, void *dummy, + struct cb_process_state *cps); +#endif /* CONFIG_NFS_V4_2 */ +extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *); +extern __be32 nfs4_callback_getattr(void *argp, void *resp, + struct cb_process_state *cps); +extern __be32 nfs4_callback_recall(void *argp, void *resp, + struct cb_process_state *cps); +#if IS_ENABLED(CONFIG_NFS_V4) +extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt); +extern void nfs_callback_down(int minorversion, struct net *net, + struct rpc_xprt *xprt); +#endif /* CONFIG_NFS_V4 */ +/* + * nfs41: Callbacks are expected to not cause substantial latency, + * so we limit their concurrency to 1 by setting up the maximum number + * of slots for the backchannel. + */ +#define NFS41_BC_MIN_CALLBACKS 1 +#define NFS41_BC_MAX_CALLBACKS 1 + +#define NFS4_MIN_NR_CALLBACK_THREADS 1 + +extern unsigned int nfs_callback_set_tcpport; +extern unsigned short nfs_callback_nr_threads; + +#endif /* __LINUX_FS_NFS_CALLBACK_H */ diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c new file mode 100644 index 000000000..10f2354ba --- /dev/null +++ b/fs/nfs/callback_proc.c @@ -0,0 +1,756 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/callback_proc.c + * + * Copyright (C) 2004 Trond Myklebust + * + * NFSv4 callback procedures + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "internal.h" +#include "pnfs.h" +#include "nfs4session.h" +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_CALLBACK + +__be32 nfs4_callback_getattr(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_getattrargs *args = argp; + struct cb_getattrres *res = resp; + struct nfs_delegation *delegation; + struct inode *inode; + + res->status = htonl(NFS4ERR_OP_NOT_IN_SESSION); + if (!cps->clp) /* Always set for v4.0. Set in cb_sequence for v4.1 */ + goto out; + + memset(res->bitmap, 0, sizeof(res->bitmap)); + res->status = htonl(NFS4ERR_BADHANDLE); + + dprintk_rcu("NFS: GETATTR callback request from %s\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + inode = nfs_delegation_find_inode(cps->clp, &args->fh); + if (IS_ERR(inode)) { + if (inode == ERR_PTR(-EAGAIN)) + res->status = htonl(NFS4ERR_DELAY); + trace_nfs4_cb_getattr(cps->clp, &args->fh, NULL, + -ntohl(res->status)); + goto out; + } + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + goto out_iput; + if ((delegation->type & FMODE_WRITE) == 0) { + nfs_put_delegation(delegation); + goto out_iput; + } + res->change_attr = delegation->change_attr; + nfs_put_delegation(delegation); + + res->size = i_size_read(inode); + if (nfs_have_writebacks(inode)) + res->change_attr++; + res->atime = inode_get_atime(inode); + res->ctime = inode_get_ctime(inode); + res->mtime = inode_get_mtime(inode); + res->bitmap[0] = (FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE) & + args->bitmap[0]; + res->bitmap[1] = (FATTR4_WORD1_TIME_ACCESS | + FATTR4_WORD1_TIME_METADATA | + FATTR4_WORD1_TIME_MODIFY) & args->bitmap[1]; + res->bitmap[2] = (FATTR4_WORD2_TIME_DELEG_ACCESS | + FATTR4_WORD2_TIME_DELEG_MODIFY) & args->bitmap[2]; + res->status = 0; +out_iput: + trace_nfs4_cb_getattr(cps->clp, &args->fh, inode, -ntohl(res->status)); + nfs_iput_and_deactive(inode); +out: + dprintk("%s: exit with status = %d\n", __func__, ntohl(res->status)); + return res->status; +} + +__be32 nfs4_callback_recall(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_recallargs *args = argp; + struct inode *inode; + __be32 res; + + res = htonl(NFS4ERR_OP_NOT_IN_SESSION); + if (!cps->clp) /* Always set for v4.0. Set in cb_sequence for v4.1 */ + goto out; + + dprintk_rcu("NFS: RECALL callback request from %s\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + res = htonl(NFS4ERR_BADHANDLE); + inode = nfs_delegation_find_inode(cps->clp, &args->fh); + if (IS_ERR(inode)) { + if (inode == ERR_PTR(-EAGAIN)) + res = htonl(NFS4ERR_DELAY); + trace_nfs4_cb_recall(cps->clp, &args->fh, NULL, + &args->stateid, -ntohl(res)); + goto out; + } + /* Set up a helper thread to actually return the delegation */ + switch (nfs_async_inode_return_delegation(inode, &args->stateid)) { + case 0: + res = 0; + break; + case -ENOENT: + res = htonl(NFS4ERR_BAD_STATEID); + break; + default: + res = htonl(NFS4ERR_RESOURCE); + } + trace_nfs4_cb_recall(cps->clp, &args->fh, inode, + &args->stateid, -ntohl(res)); + nfs_iput_and_deactive(inode); +out: + dprintk("%s: exit with status = %d\n", __func__, ntohl(res)); + return res; +} + +/* + * Lookup a layout inode by stateid + * + * Note: returns a refcount on the inode and superblock + */ +static struct inode *nfs_layout_find_inode_by_stateid(struct nfs_client *clp, + const nfs4_stateid *stateid) + __must_hold(RCU) +{ + struct nfs_server *server; + struct inode *inode; + struct pnfs_layout_hdr *lo; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) { + if (!pnfs_layout_is_valid(lo)) + continue; + if (!nfs4_stateid_match_other(stateid, &lo->plh_stateid)) + continue; + if (nfs_sb_active(server->super)) + inode = igrab(lo->plh_inode); + else + inode = ERR_PTR(-EAGAIN); + rcu_read_unlock(); + if (inode) + return inode; + nfs_sb_deactive(server->super); + return ERR_PTR(-EAGAIN); + } + } + rcu_read_unlock(); + return ERR_PTR(-ENOENT); +} + +/* + * Lookup a layout inode by filehandle. + * + * Note: returns a refcount on the inode and superblock + * + */ +static struct inode *nfs_layout_find_inode_by_fh(struct nfs_client *clp, + const struct nfs_fh *fh) +{ + struct nfs_server *server; + struct nfs_inode *nfsi; + struct inode *inode; + struct pnfs_layout_hdr *lo; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) { + nfsi = NFS_I(lo->plh_inode); + if (nfs_compare_fh(fh, &nfsi->fh)) + continue; + if (nfsi->layout != lo) + continue; + if (nfs_sb_active(server->super)) + inode = igrab(lo->plh_inode); + else + inode = ERR_PTR(-EAGAIN); + rcu_read_unlock(); + if (inode) + return inode; + nfs_sb_deactive(server->super); + return ERR_PTR(-EAGAIN); + } + } + rcu_read_unlock(); + return ERR_PTR(-ENOENT); +} + +static struct inode *nfs_layout_find_inode(struct nfs_client *clp, + const struct nfs_fh *fh, + const nfs4_stateid *stateid) +{ + struct inode *inode; + + inode = nfs_layout_find_inode_by_stateid(clp, stateid); + if (inode == ERR_PTR(-ENOENT)) + inode = nfs_layout_find_inode_by_fh(clp, fh); + return inode; +} + +/* + * Enforce RFC5661 section 12.5.5.2.1. (Layout Recall and Return Sequencing) + */ +static u32 pnfs_check_callback_stateid(struct pnfs_layout_hdr *lo, + const nfs4_stateid *new, + struct cb_process_state *cps) +{ + u32 oldseq, newseq; + + /* Is the stateid not initialised? */ + if (!pnfs_layout_is_valid(lo)) + return NFS4ERR_NOMATCHING_LAYOUT; + + /* Mismatched stateid? */ + if (!nfs4_stateid_match_other(&lo->plh_stateid, new)) + return NFS4ERR_BAD_STATEID; + + newseq = be32_to_cpu(new->seqid); + /* Are we already in a layout recall situation? */ + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + return NFS4ERR_DELAY; + + /* + * Check that the stateid matches what we think it should be. + * Note that if the server sent us a list of referring calls, + * and we know that those have completed, then we trust the + * stateid argument is correct. + */ + oldseq = be32_to_cpu(lo->plh_stateid.seqid); + if (newseq > oldseq + 1 && !cps->referring_calls) + return NFS4ERR_DELAY; + + /* Crazy server! */ + if (newseq <= oldseq) + return NFS4ERR_OLD_STATEID; + + return NFS_OK; +} + +static u32 initiate_file_draining(struct nfs_client *clp, + struct cb_layoutrecallargs *args, + struct cb_process_state *cps) +{ + struct inode *ino; + struct pnfs_layout_hdr *lo; + u32 rv = NFS4ERR_NOMATCHING_LAYOUT; + LIST_HEAD(free_me_list); + bool return_range = false; + + ino = nfs_layout_find_inode(clp, &args->cbl_fh, &args->cbl_stateid); + if (IS_ERR(ino)) { + if (ino == ERR_PTR(-EAGAIN)) + rv = NFS4ERR_DELAY; + goto out_noput; + } + + pnfs_layoutcommit_inode(ino, false); + + + spin_lock(&ino->i_lock); + lo = NFS_I(ino)->layout; + if (!lo) { + spin_unlock(&ino->i_lock); + goto out; + } + pnfs_get_layout_hdr(lo); + rv = pnfs_check_callback_stateid(lo, &args->cbl_stateid, cps); + if (rv != NFS_OK) + goto unlock; + + /* + * Enforce RFC5661 Section 12.5.5.2.1.5 (Bulk Recall and Return) + */ + if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags)) { + rv = NFS4ERR_DELAY; + goto unlock; + } + + pnfs_set_layout_stateid(lo, &args->cbl_stateid, NULL, true); + switch (pnfs_mark_matching_lsegs_return(lo, &free_me_list, + &args->cbl_range, + be32_to_cpu(args->cbl_stateid.seqid))) { + case 0: + case -EBUSY: + /* There are layout segments that need to be returned */ + rv = NFS4_OK; + break; + case -ENOENT: + set_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags); + /* Embrace your forgetfulness! */ + rv = NFS4ERR_NOMATCHING_LAYOUT; + + return_range = true; + } +unlock: + spin_unlock(&ino->i_lock); + if (return_range && NFS_SERVER(ino)->pnfs_curr_ld->return_range) + NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo, + &args->cbl_range); + pnfs_free_lseg_list(&free_me_list); + /* Free all lsegs that are attached to commit buckets */ + nfs_commit_inode(ino, 0); + pnfs_put_layout_hdr(lo); +out: + nfs_iput_and_deactive(ino); +out_noput: + trace_nfs4_cb_layoutrecall_file(clp, &args->cbl_fh, ino, + &args->cbl_stateid, -rv); + return rv; +} + +static u32 initiate_bulk_draining(struct nfs_client *clp, + struct cb_layoutrecallargs *args) +{ + int stat; + + if (args->cbl_recall_type == RETURN_FSID) + stat = pnfs_layout_destroy_byfsid(clp, &args->cbl_fsid, + PNFS_LAYOUT_BULK_RETURN); + else + stat = pnfs_layout_destroy_byclid(clp, PNFS_LAYOUT_BULK_RETURN); + if (stat != 0) + return NFS4ERR_DELAY; + return NFS4ERR_NOMATCHING_LAYOUT; +} + +static u32 do_callback_layoutrecall(struct nfs_client *clp, + struct cb_layoutrecallargs *args, + struct cb_process_state *cps) +{ + if (args->cbl_recall_type == RETURN_FILE) + return initiate_file_draining(clp, args, cps); + return initiate_bulk_draining(clp, args); +} + +__be32 nfs4_callback_layoutrecall(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_layoutrecallargs *args = argp; + u32 res = NFS4ERR_OP_NOT_IN_SESSION; + + if (cps->clp) + res = do_callback_layoutrecall(cps->clp, args, cps); + return cpu_to_be32(res); +} + +static void pnfs_recall_all_layouts(struct nfs_client *clp, + struct cb_process_state *cps) +{ + struct cb_layoutrecallargs args; + + /* Pretend we got a CB_LAYOUTRECALL(ALL) */ + memset(&args, 0, sizeof(args)); + args.cbl_recall_type = RETURN_ALL; + /* FIXME we ignore errors, what should we do? */ + do_callback_layoutrecall(clp, &args, cps); +} + +__be32 nfs4_callback_devicenotify(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_devicenotifyargs *args = argp; + const struct pnfs_layoutdriver_type *ld = NULL; + uint32_t i; + __be32 res = 0; + + if (!cps->clp) { + res = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION); + goto out; + } + + for (i = 0; i < args->ndevs; i++) { + struct cb_devicenotifyitem *dev = &args->devs[i]; + + if (!ld || ld->id != dev->cbd_layout_type) { + pnfs_put_layoutdriver(ld); + ld = pnfs_find_layoutdriver(dev->cbd_layout_type); + if (!ld) + continue; + } + nfs4_delete_deviceid(ld, cps->clp, &dev->cbd_dev_id); + } + pnfs_put_layoutdriver(ld); +out: + kfree(args->devs); + return res; +} + +/* + * Validate the sequenceID sent by the server. + * Return success if the sequenceID is one more than what we last saw on + * this slot, accounting for wraparound. Increments the slot's sequence. + * + * We don't yet implement a duplicate request cache, instead we set the + * back channel ca_maxresponsesize_cached to zero. This is OK for now + * since we only currently implement idempotent callbacks anyway. + * + * We have a single slot backchannel at this time, so we don't bother + * checking the used_slots bit array on the table. The lower layer guarantees + * a single outstanding callback request at a time. + */ +static __be32 +validate_seqid(const struct nfs4_slot_table *tbl, const struct nfs4_slot *slot, + const struct cb_sequenceargs * args) +{ + __be32 ret; + + ret = cpu_to_be32(NFS4ERR_BADSLOT); + if (args->csa_slotid > tbl->server_highest_slotid) + goto out_err; + + /* Replay */ + if (args->csa_sequenceid == slot->seq_nr) { + ret = cpu_to_be32(NFS4ERR_DELAY); + if (nfs4_test_locked_slot(tbl, slot->slot_nr)) + goto out_err; + + /* Signal process_op to set this error on next op */ + ret = cpu_to_be32(NFS4ERR_RETRY_UNCACHED_REP); + if (args->csa_cachethis == 0) + goto out_err; + + /* Liar! We never allowed you to set csa_cachethis != 0 */ + ret = cpu_to_be32(NFS4ERR_SEQ_FALSE_RETRY); + goto out_err; + } + + /* Note: wraparound relies on seq_nr being of type u32 */ + /* Misordered request */ + ret = cpu_to_be32(NFS4ERR_SEQ_MISORDERED); + if (args->csa_sequenceid != slot->seq_nr + 1) + goto out_err; + + return cpu_to_be32(NFS4_OK); + +out_err: + trace_nfs4_cb_seqid_err(args, ret); + return ret; +} + +/* + * For each referring call triple, check the session's slot table for + * a match. If the slot is in use and the sequence numbers match, the + * client is still waiting for a response to the original request. + */ +static int referring_call_exists(struct nfs_client *clp, + uint32_t nrclists, + struct referring_call_list *rclists, + spinlock_t *lock) + __releases(lock) + __acquires(lock) +{ + int status = 0; + int found = 0; + int i, j; + struct nfs4_session *session; + struct nfs4_slot_table *tbl; + struct referring_call_list *rclist; + struct referring_call *ref; + + /* + * XXX When client trunking is implemented, this becomes + * a session lookup from within the loop + */ + session = clp->cl_session; + tbl = &session->fc_slot_table; + + for (i = 0; i < nrclists; i++) { + rclist = &rclists[i]; + if (memcmp(session->sess_id.data, + rclist->rcl_sessionid.data, + NFS4_MAX_SESSIONID_LEN) != 0) + continue; + + for (j = 0; j < rclist->rcl_nrefcalls; j++) { + ref = &rclist->rcl_refcalls[j]; + spin_unlock(lock); + status = nfs4_slot_wait_on_seqid(tbl, ref->rc_slotid, + ref->rc_sequenceid, HZ >> 1) < 0; + spin_lock(lock); + if (status) + goto out; + found++; + } + } + +out: + return status < 0 ? status : found; +} + +__be32 nfs4_callback_sequence(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_sequenceargs *args = argp; + struct cb_sequenceres *res = resp; + struct nfs4_slot_table *tbl; + struct nfs4_slot *slot; + struct nfs_client *clp; + int ret; + int i; + __be32 status = htonl(NFS4ERR_BADSESSION); + + clp = nfs4_find_client_sessionid(cps->net, args->csa_addr, + &args->csa_sessionid, cps->minorversion); + if (clp == NULL) + goto out; + + if (!(clp->cl_session->flags & SESSION4_BACK_CHAN)) + goto out; + + tbl = &clp->cl_session->bc_slot_table; + + /* Set up res before grabbing the spinlock */ + memcpy(&res->csr_sessionid, &args->csa_sessionid, + sizeof(res->csr_sessionid)); + res->csr_sequenceid = args->csa_sequenceid; + res->csr_slotid = args->csa_slotid; + + spin_lock(&tbl->slot_tbl_lock); + /* state manager is resetting the session */ + if (test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) { + status = htonl(NFS4ERR_DELAY); + /* Return NFS4ERR_BADSESSION if we're draining the session + * in order to reset it. + */ + if (test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state)) + status = htonl(NFS4ERR_BADSESSION); + goto out_unlock; + } + + status = htonl(NFS4ERR_BADSLOT); + slot = nfs4_lookup_slot(tbl, args->csa_slotid); + if (IS_ERR(slot)) + goto out_unlock; + + res->csr_highestslotid = tbl->server_highest_slotid; + res->csr_target_highestslotid = tbl->target_highest_slotid; + + status = validate_seqid(tbl, slot, args); + if (status) + goto out_unlock; + if (!nfs4_try_to_lock_slot(tbl, slot)) { + status = htonl(NFS4ERR_DELAY); + goto out_unlock; + } + cps->slot = slot; + + /* The ca_maxresponsesize_cached is 0 with no DRC */ + if (args->csa_cachethis != 0) { + status = htonl(NFS4ERR_REP_TOO_BIG_TO_CACHE); + goto out_unlock; + } + + /* + * Check for pending referring calls. If a match is found, a + * related callback was received before the response to the original + * call. + */ + ret = referring_call_exists(clp, args->csa_nrclists, args->csa_rclists, + &tbl->slot_tbl_lock); + if (ret < 0) { + status = htonl(NFS4ERR_DELAY); + goto out_unlock; + } + cps->referring_calls = ret; + + /* + * RFC5661 20.9.3 + * If CB_SEQUENCE returns an error, then the state of the slot + * (sequence ID, cached reply) MUST NOT change. + */ + slot->seq_nr = args->csa_sequenceid; +out_unlock: + spin_unlock(&tbl->slot_tbl_lock); + +out: + cps->clp = clp; /* put in nfs4_callback_compound */ + for (i = 0; i < args->csa_nrclists; i++) + kfree(args->csa_rclists[i].rcl_refcalls); + kfree(args->csa_rclists); + + if (status == htonl(NFS4ERR_RETRY_UNCACHED_REP)) { + cps->drc_status = status; + status = 0; + } else + res->csr_status = status; + + trace_nfs4_cb_sequence(args, res, status); + return status; +} + +static bool +validate_bitmap_values(unsigned int mask) +{ + return (mask & ~RCA4_TYPE_MASK_ALL) == 0; +} + +__be32 nfs4_callback_recallany(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_recallanyargs *args = argp; + __be32 status; + fmode_t flags = 0; + bool schedule_manager = false; + + status = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION); + if (!cps->clp) /* set in cb_sequence */ + goto out; + + dprintk_rcu("NFS: RECALL_ANY callback request from %s\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + status = cpu_to_be32(NFS4ERR_INVAL); + if (!validate_bitmap_values(args->craa_type_mask)) + goto out; + + status = cpu_to_be32(NFS4_OK); + if (args->craa_type_mask & BIT(RCA4_TYPE_MASK_RDATA_DLG)) + flags = FMODE_READ; + if (args->craa_type_mask & BIT(RCA4_TYPE_MASK_WDATA_DLG)) + flags |= FMODE_WRITE; + if (flags) + nfs_expire_unused_delegation_types(cps->clp, flags); + + if (args->craa_type_mask & BIT(RCA4_TYPE_MASK_FILE_LAYOUT)) + pnfs_recall_all_layouts(cps->clp, cps); + + if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_READ)) { + set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &cps->clp->cl_state); + schedule_manager = true; + } + if (args->craa_type_mask & BIT(PNFS_FF_RCA4_TYPE_MASK_RW)) { + set_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &cps->clp->cl_state); + schedule_manager = true; + } + if (schedule_manager) + nfs4_schedule_state_manager(cps->clp); + +out: + dprintk("%s: exit with status = %d\n", __func__, ntohl(status)); + return status; +} + +/* Reduce the fore channel's max_slots to the target value */ +__be32 nfs4_callback_recallslot(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_recallslotargs *args = argp; + struct nfs4_slot_table *fc_tbl; + __be32 status; + + status = htonl(NFS4ERR_OP_NOT_IN_SESSION); + if (!cps->clp) /* set in cb_sequence */ + goto out; + + dprintk_rcu("NFS: CB_RECALL_SLOT request from %s target highest slotid %u\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR), + args->crsa_target_highest_slotid); + + fc_tbl = &cps->clp->cl_session->fc_slot_table; + + status = htonl(NFS4_OK); + + nfs41_set_target_slotid(fc_tbl, args->crsa_target_highest_slotid); + nfs41_notify_server(cps->clp); +out: + dprintk("%s: exit with status = %d\n", __func__, ntohl(status)); + return status; +} + +__be32 nfs4_callback_notify_lock(void *argp, void *resp, + struct cb_process_state *cps) +{ + struct cb_notify_lock_args *args = argp; + + if (!cps->clp) /* set in cb_sequence */ + return htonl(NFS4ERR_OP_NOT_IN_SESSION); + + dprintk_rcu("NFS: CB_NOTIFY_LOCK request from %s\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + /* Don't wake anybody if the string looked bogus */ + if (args->cbnl_valid) + __wake_up(&cps->clp->cl_lock_waitq, TASK_NORMAL, 0, args); + + return htonl(NFS4_OK); +} +#ifdef CONFIG_NFS_V4_2 +static void nfs4_copy_cb_args(struct nfs4_copy_state *cp_state, + struct cb_offloadargs *args) +{ + cp_state->count = args->wr_count; + cp_state->error = args->error; + if (!args->error) { + cp_state->verf.committed = args->wr_writeverf.committed; + memcpy(&cp_state->verf.verifier.data[0], + &args->wr_writeverf.verifier.data[0], + NFS4_VERIFIER_SIZE); + } +} + +__be32 nfs4_callback_offload(void *data, void *dummy, + struct cb_process_state *cps) +{ + struct cb_offloadargs *args = data; + struct nfs_server *server; + struct nfs4_copy_state *copy, *tmp_copy; + bool found = false; + + copy = kzalloc_obj(struct nfs4_copy_state); + if (!copy) + return cpu_to_be32(NFS4ERR_DELAY); + + spin_lock(&cps->clp->cl_lock); + rcu_read_lock(); + list_for_each_entry_rcu(server, &cps->clp->cl_superblocks, + client_link) { + list_for_each_entry(tmp_copy, &server->ss_copies, copies) { + if (memcmp(args->coa_stateid.other, + tmp_copy->stateid.other, + sizeof(args->coa_stateid.other))) + continue; + nfs4_copy_cb_args(tmp_copy, args); + complete(&tmp_copy->completion); + found = true; + goto out; + } + } +out: + rcu_read_unlock(); + if (!found) { + memcpy(©->stateid, &args->coa_stateid, NFS4_STATEID_SIZE); + nfs4_copy_cb_args(copy, args); + list_add_tail(©->copies, &cps->clp->pending_cb_stateids); + } else + kfree(copy); + spin_unlock(&cps->clp->cl_lock); + + trace_nfs4_cb_offload(&args->coa_fh, &args->coa_stateid, + args->wr_count, args->error, + args->wr_writeverf.committed); + return 0; +} +#endif /* CONFIG_NFS_V4_2 */ diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c new file mode 100644 index 000000000..eec604055 --- /dev/null +++ b/fs/nfs/callback_xdr.c @@ -0,0 +1,1111 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/callback_xdr.c + * + * Copyright (C) 2004 Trond Myklebust + * + * NFSv4 callback encode/decode procedures + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include "nfs4_fs.h" +#include "callback.h" +#include "internal.h" +#include "nfs4session.h" +#include "nfs4trace.h" + +#define CB_OP_TAGLEN_MAXSZ (512) +#define CB_OP_HDR_RES_MAXSZ (2 * 4) // opcode, status +#define CB_OP_GETATTR_BITMAP_MAXSZ (4 * 4) // bitmap length, 3 bitmaps +#define CB_OP_GETATTR_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ + \ + CB_OP_GETATTR_BITMAP_MAXSZ + \ + /* change, size, atime, ctime, + * mtime, deleg_atime, deleg_mtime */\ + (2 + 2 + 3 + 3 + 3 + 3 + 3) * 4) +#define CB_OP_RECALL_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) + +#define CB_OP_LAYOUTRECALL_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_DEVICENOTIFY_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_SEQUENCE_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ + \ + NFS4_MAX_SESSIONID_LEN + \ + (1 + 3) * 4) // seqid, 3 slotids +#define CB_OP_RECALLANY_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_RECALLSLOT_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_NOTIFY_LOCK_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#ifdef CONFIG_NFS_V4_2 +#define CB_OP_OFFLOAD_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#endif /* CONFIG_NFS_V4_2 */ + +#define NFSDBG_FACILITY NFSDBG_CALLBACK + +/* Internal error code */ +#define NFS4ERR_RESOURCE_HDR 11050 + +struct callback_op { + __be32 (*process_op)(void *, void *, struct cb_process_state *); + __be32 (*decode_args)(struct svc_rqst *, struct xdr_stream *, void *); + __be32 (*encode_res)(struct svc_rqst *, struct xdr_stream *, + const void *); + long res_maxsize; +}; + +static struct callback_op callback_ops[]; + +static __be32 nfs4_callback_null(struct svc_rqst *rqstp) +{ + return htonl(NFS4_OK); +} + +/* + * svc_process_common() looks for an XDR encoder to know when + * not to drop a Reply. + */ +static bool nfs4_encode_void(struct svc_rqst *rqstp, struct xdr_stream *xdr) +{ + return true; +} + +static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, + const char **str, size_t maxlen) +{ + ssize_t err; + + err = xdr_stream_decode_opaque_inline(xdr, (void **)str, maxlen); + if (err < 0) + return cpu_to_be32(NFS4ERR_RESOURCE); + *len = err; + return 0; +} + +static __be32 decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + fh->size = ntohl(*p); + if (fh->size > NFS4_FHSIZE) + return htonl(NFS4ERR_BADHANDLE); + p = xdr_inline_decode(xdr, fh->size); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + memcpy_and_pad(fh->data, sizeof(fh->data), p, fh->size, 0); + return 0; +} + +static __be32 decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +{ + __be32 *p; + unsigned int attrlen; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + attrlen = ntohl(*p); + p = xdr_inline_decode(xdr, attrlen << 2); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + if (likely(attrlen > 0)) + bitmap[0] = ntohl(*p++); + if (attrlen > 1) + bitmap[1] = ntohl(*p++); + if (attrlen > 2) + bitmap[2] = ntohl(*p); + return 0; +} + +static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS4_STATEID_SIZE); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + memcpy(stateid->data, p, NFS4_STATEID_SIZE); + return 0; +} + +static __be32 decode_delegation_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_DELEGATION_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) +{ + __be32 *p; + __be32 status; + + status = decode_string(xdr, &hdr->taglen, &hdr->tag, CB_OP_TAGLEN_MAXSZ); + if (unlikely(status != 0)) + return status; + p = xdr_inline_decode(xdr, 12); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + hdr->minorversion = ntohl(*p++); + /* Check for minor version support */ + if (hdr->minorversion <= NFS4_MAX_MINOR_VERSION) { + hdr->cb_ident = ntohl(*p++); /* ignored by v4.1 and v4.2 */ + } else { + pr_warn_ratelimited("NFS: %s: NFSv4 server callback with " + "illegal minor version %u!\n", + __func__, hdr->minorversion); + return htonl(NFS4ERR_MINOR_VERS_MISMATCH); + } + hdr->nops = ntohl(*p); + return 0; +} + +static __be32 decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) +{ + __be32 *p; + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE_HDR); + *op = ntohl(*p); + return 0; +} + +static __be32 decode_getattr_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, void *argp) +{ + struct cb_getattrargs *args = argp; + __be32 status; + + status = decode_fh(xdr, &args->fh); + if (unlikely(status != 0)) + return status; + return decode_bitmap(xdr, args->bitmap); +} + +static __be32 decode_recall_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, void *argp) +{ + struct cb_recallargs *args = argp; + __be32 *p; + __be32 status; + + status = decode_delegation_stateid(xdr, &args->stateid); + if (unlikely(status != 0)) + return status; + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + args->truncate = ntohl(*p); + return decode_fh(xdr, &args->fh); +} + +static __be32 decode_layout_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_LAYOUT_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static __be32 decode_layoutrecall_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, void *argp) +{ + struct cb_layoutrecallargs *args = argp; + __be32 *p; + __be32 status = 0; + uint32_t iomode; + + p = xdr_inline_decode(xdr, 4 * sizeof(uint32_t)); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + + args->cbl_layout_type = ntohl(*p++); + /* Depite the spec's xdr, iomode really belongs in the FILE switch, + * as it is unusable and ignored with the other types. + */ + iomode = ntohl(*p++); + args->cbl_layoutchanged = ntohl(*p++); + args->cbl_recall_type = ntohl(*p++); + + if (args->cbl_recall_type == RETURN_FILE) { + args->cbl_range.iomode = iomode; + status = decode_fh(xdr, &args->cbl_fh); + if (unlikely(status != 0)) + return status; + + p = xdr_inline_decode(xdr, 2 * sizeof(uint64_t)); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + p = xdr_decode_hyper(p, &args->cbl_range.offset); + p = xdr_decode_hyper(p, &args->cbl_range.length); + return decode_layout_stateid(xdr, &args->cbl_stateid); + } else if (args->cbl_recall_type == RETURN_FSID) { + p = xdr_inline_decode(xdr, 2 * sizeof(uint64_t)); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + p = xdr_decode_hyper(p, &args->cbl_fsid.major); + p = xdr_decode_hyper(p, &args->cbl_fsid.minor); + } else if (args->cbl_recall_type != RETURN_ALL) + return htonl(NFS4ERR_BADXDR); + return 0; +} + +static +__be32 decode_devicenotify_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + void *argp) +{ + struct cb_devicenotifyargs *args = argp; + uint32_t tmp, n, i; + __be32 *p; + __be32 status = 0; + + /* Num of device notifications */ + p = xdr_inline_decode(xdr, sizeof(uint32_t)); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto out; + } + n = ntohl(*p++); + if (n == 0) + goto out; + + args->devs = kmalloc_objs(*args->devs, n); + if (!args->devs) { + status = htonl(NFS4ERR_DELAY); + goto out; + } + + /* Decode each dev notification */ + for (i = 0; i < n; i++) { + struct cb_devicenotifyitem *dev = &args->devs[i]; + + p = xdr_inline_decode(xdr, (4 * sizeof(uint32_t)) + + NFS4_DEVICEID4_SIZE); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto err; + } + + tmp = ntohl(*p++); /* bitmap size */ + if (tmp != 1) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + dev->cbd_notify_type = ntohl(*p++); + if (dev->cbd_notify_type != NOTIFY_DEVICEID4_CHANGE && + dev->cbd_notify_type != NOTIFY_DEVICEID4_DELETE) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + + tmp = ntohl(*p++); /* opaque size */ + if (((dev->cbd_notify_type == NOTIFY_DEVICEID4_CHANGE) && + (tmp != NFS4_DEVICEID4_SIZE + 8)) || + ((dev->cbd_notify_type == NOTIFY_DEVICEID4_DELETE) && + (tmp != NFS4_DEVICEID4_SIZE + 4))) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + dev->cbd_layout_type = ntohl(*p++); + memcpy(dev->cbd_dev_id.data, p, NFS4_DEVICEID4_SIZE); + p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); + + if (dev->cbd_layout_type == NOTIFY_DEVICEID4_CHANGE) { + p = xdr_inline_decode(xdr, sizeof(uint32_t)); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto err; + } + dev->cbd_immediate = ntohl(*p++); + } else { + dev->cbd_immediate = 0; + } + + dprintk("%s: type %d layout 0x%x immediate %d\n", + __func__, dev->cbd_notify_type, dev->cbd_layout_type, + dev->cbd_immediate); + } + args->ndevs = n; + dprintk("%s: ndevs %d\n", __func__, args->ndevs); + return 0; +err: + kfree(args->devs); +out: + args->devs = NULL; + args->ndevs = 0; + dprintk("%s: status %d ndevs %d\n", + __func__, ntohl(status), args->ndevs); + return status; +} + +static __be32 decode_sessionid(struct xdr_stream *xdr, + struct nfs4_sessionid *sid) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS4_MAX_SESSIONID_LEN); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + + memcpy(sid->data, p, NFS4_MAX_SESSIONID_LEN); + return 0; +} + +static __be32 decode_rc_list(struct xdr_stream *xdr, + struct referring_call_list *rc_list) +{ + __be32 *p; + int i; + __be32 status; + + status = decode_sessionid(xdr, &rc_list->rcl_sessionid); + if (status) + goto out; + + status = htonl(NFS4ERR_RESOURCE); + p = xdr_inline_decode(xdr, sizeof(uint32_t)); + if (unlikely(p == NULL)) + goto out; + + rc_list->rcl_nrefcalls = ntohl(*p++); + if (rc_list->rcl_nrefcalls) { + if (unlikely(rc_list->rcl_nrefcalls > xdr->buf->len)) + goto out; + p = xdr_inline_decode(xdr, + rc_list->rcl_nrefcalls * 2 * sizeof(uint32_t)); + if (unlikely(p == NULL)) + goto out; + rc_list->rcl_refcalls = kmalloc_objs(*rc_list->rcl_refcalls, + rc_list->rcl_nrefcalls); + if (unlikely(rc_list->rcl_refcalls == NULL)) + goto out; + for (i = 0; i < rc_list->rcl_nrefcalls; i++) { + rc_list->rcl_refcalls[i].rc_sequenceid = ntohl(*p++); + rc_list->rcl_refcalls[i].rc_slotid = ntohl(*p++); + } + } + status = 0; + +out: + return status; +} + +static __be32 decode_cb_sequence_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + void *argp) +{ + struct cb_sequenceargs *args = argp; + __be32 *p; + int i; + __be32 status; + + status = decode_sessionid(xdr, &args->csa_sessionid); + if (status) + return status; + + p = xdr_inline_decode(xdr, 5 * sizeof(uint32_t)); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + + args->csa_addr = svc_addr(rqstp); + args->csa_sequenceid = ntohl(*p++); + args->csa_slotid = ntohl(*p++); + args->csa_highestslotid = ntohl(*p++); + args->csa_cachethis = ntohl(*p++); + args->csa_nrclists = ntohl(*p++); + args->csa_rclists = NULL; + if (args->csa_nrclists) { + args->csa_rclists = kmalloc_objs(*args->csa_rclists, + args->csa_nrclists); + if (unlikely(args->csa_rclists == NULL)) + return htonl(NFS4ERR_RESOURCE); + + for (i = 0; i < args->csa_nrclists; i++) { + status = decode_rc_list(xdr, &args->csa_rclists[i]); + if (status) { + args->csa_nrclists = i; + goto out_free; + } + } + } + return 0; + +out_free: + for (i = 0; i < args->csa_nrclists; i++) + kfree(args->csa_rclists[i].rcl_refcalls); + kfree(args->csa_rclists); + return status; +} + +static __be32 decode_recallany_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + void *argp) +{ + struct cb_recallanyargs *args = argp; + uint32_t bitmap[3]; + __be32 *p, status; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + args->craa_objs_to_keep = ntohl(*p++); + status = decode_bitmap(xdr, bitmap); + if (unlikely(status)) + return status; + args->craa_type_mask = bitmap[0]; + + return 0; +} + +static __be32 decode_recallslot_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + void *argp) +{ + struct cb_recallslotargs *args = argp; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + args->crsa_target_highest_slotid = ntohl(*p++); + return 0; +} + +static __be32 decode_lockowner(struct xdr_stream *xdr, struct cb_notify_lock_args *args) +{ + __be32 *p; + unsigned int len; + + p = xdr_inline_decode(xdr, 12); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + + p = xdr_decode_hyper(p, &args->cbnl_owner.clientid); + len = be32_to_cpu(*p); + + p = xdr_inline_decode(xdr, len); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + + /* Only try to decode if the length is right */ + if (len == 20) { + p += 2; /* skip "lock id:" */ + args->cbnl_owner.s_dev = be32_to_cpu(*p++); + xdr_decode_hyper(p, &args->cbnl_owner.id); + args->cbnl_valid = true; + } else { + args->cbnl_owner.s_dev = 0; + args->cbnl_owner.id = 0; + args->cbnl_valid = false; + } + return 0; +} + +static __be32 decode_notify_lock_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, void *argp) +{ + struct cb_notify_lock_args *args = argp; + __be32 status; + + status = decode_fh(xdr, &args->cbnl_fh); + if (unlikely(status != 0)) + return status; + return decode_lockowner(xdr, args); +} + +#ifdef CONFIG_NFS_V4_2 +static __be32 decode_write_response(struct xdr_stream *xdr, + struct cb_offloadargs *args) +{ + __be32 *p; + + /* skip the always zero field */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out; + p++; + + /* decode count, stable_how, verifier */ + p = xdr_inline_decode(xdr, 8 + 4); + if (unlikely(!p)) + goto out; + p = xdr_decode_hyper(p, &args->wr_count); + args->wr_writeverf.committed = be32_to_cpup(p); + p = xdr_inline_decode(xdr, NFS4_VERIFIER_SIZE); + if (likely(p)) { + memcpy(&args->wr_writeverf.verifier.data[0], p, + NFS4_VERIFIER_SIZE); + return 0; + } +out: + return htonl(NFS4ERR_RESOURCE); +} + +static __be32 decode_offload_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct cb_offloadargs *args = data; + __be32 *p; + __be32 status; + + /* decode fh */ + status = decode_fh(xdr, &args->coa_fh); + if (unlikely(status != 0)) + return status; + + /* decode stateid */ + status = decode_stateid(xdr, &args->coa_stateid); + if (unlikely(status != 0)) + return status; + + /* decode status */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out; + args->error = ntohl(*p++); + if (!args->error) { + status = decode_write_response(xdr, args); + if (unlikely(status != 0)) + return status; + } else { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + goto out; + p = xdr_decode_hyper(p, &args->wr_count); + } + return 0; +out: + return htonl(NFS4ERR_RESOURCE); +} +#endif /* CONFIG_NFS_V4_2 */ +static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) +{ + if (unlikely(xdr_stream_encode_opaque(xdr, str, len) < 0)) + return cpu_to_be32(NFS4ERR_RESOURCE); + return 0; +} + +static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, size_t sz) +{ + if (xdr_stream_encode_uint32_array(xdr, bitmap, sz) < 0) + return cpu_to_be32(NFS4ERR_RESOURCE); + return 0; +} + +static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) +{ + __be32 *p; + + if (!(bitmap[0] & FATTR4_WORD0_CHANGE)) + return 0; + p = xdr_reserve_space(xdr, 8); + if (unlikely(!p)) + return htonl(NFS4ERR_RESOURCE); + p = xdr_encode_hyper(p, change); + return 0; +} + +static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) +{ + __be32 *p; + + if (!(bitmap[0] & FATTR4_WORD0_SIZE)) + return 0; + p = xdr_reserve_space(xdr, 8); + if (unlikely(!p)) + return htonl(NFS4ERR_RESOURCE); + p = xdr_encode_hyper(p, size); + return 0; +} + +static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec64 *time) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 12); + if (unlikely(!p)) + return htonl(NFS4ERR_RESOURCE); + p = xdr_encode_hyper(p, time->tv_sec); + *p = htonl(time->tv_nsec); + return 0; +} + +static __be32 encode_attr_atime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec64 *time) +{ + if (!(bitmap[1] & FATTR4_WORD1_TIME_ACCESS)) + return 0; + return encode_attr_time(xdr,time); +} + +static __be32 encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec64 *time) +{ + if (!(bitmap[1] & FATTR4_WORD1_TIME_METADATA)) + return 0; + return encode_attr_time(xdr,time); +} + +static __be32 encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec64 *time) +{ + if (!(bitmap[1] & FATTR4_WORD1_TIME_MODIFY)) + return 0; + return encode_attr_time(xdr,time); +} + +static __be32 encode_attr_delegatime(struct xdr_stream *xdr, + const uint32_t *bitmap, + const struct timespec64 *time) +{ + if (!(bitmap[2] & FATTR4_WORD2_TIME_DELEG_ACCESS)) + return 0; + return encode_attr_time(xdr,time); +} + +static __be32 encode_attr_delegmtime(struct xdr_stream *xdr, + const uint32_t *bitmap, + const struct timespec64 *time) +{ + if (!(bitmap[2] & FATTR4_WORD2_TIME_DELEG_MODIFY)) + return 0; + return encode_attr_time(xdr,time); +} + +static __be32 encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound_hdr_res *hdr) +{ + __be32 status; + + hdr->status = xdr_reserve_space(xdr, 4); + if (unlikely(hdr->status == NULL)) + return htonl(NFS4ERR_RESOURCE); + status = encode_string(xdr, hdr->taglen, hdr->tag); + if (unlikely(status != 0)) + return status; + hdr->nops = xdr_reserve_space(xdr, 4); + if (unlikely(hdr->nops == NULL)) + return htonl(NFS4ERR_RESOURCE); + return 0; +} + +static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 8); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE_HDR); + *p++ = htonl(op); + *p = res; + return 0; +} + +static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, + const void *resp) +{ + const struct cb_getattrres *res = resp; + __be32 *savep = NULL; + __be32 status = res->status; + + if (unlikely(status != 0)) + goto out; + status = encode_attr_bitmap(xdr, res->bitmap, ARRAY_SIZE(res->bitmap)); + if (unlikely(status != 0)) + goto out; + status = cpu_to_be32(NFS4ERR_RESOURCE); + savep = xdr_reserve_space(xdr, sizeof(*savep)); + if (unlikely(!savep)) + goto out; + status = encode_attr_change(xdr, res->bitmap, res->change_attr); + if (unlikely(status != 0)) + goto out; + status = encode_attr_size(xdr, res->bitmap, res->size); + if (unlikely(status != 0)) + goto out; + status = encode_attr_atime(xdr, res->bitmap, &res->atime); + if (unlikely(status != 0)) + goto out; + status = encode_attr_ctime(xdr, res->bitmap, &res->ctime); + if (unlikely(status != 0)) + goto out; + status = encode_attr_mtime(xdr, res->bitmap, &res->mtime); + if (unlikely(status != 0)) + goto out; + status = encode_attr_delegatime(xdr, res->bitmap, &res->atime); + if (unlikely(status != 0)) + goto out; + status = encode_attr_delegmtime(xdr, res->bitmap, &res->mtime); + *savep = htonl((unsigned int)((char *)xdr->p - (char *)(savep+1))); +out: + return status; +} + +static __be32 encode_sessionid(struct xdr_stream *xdr, + const struct nfs4_sessionid *sid) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, NFS4_MAX_SESSIONID_LEN); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + + memcpy(p, sid, NFS4_MAX_SESSIONID_LEN); + return 0; +} + +static __be32 encode_cb_sequence_res(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + const void *resp) +{ + const struct cb_sequenceres *res = resp; + __be32 *p; + __be32 status = res->csr_status; + + if (unlikely(status != 0)) + return status; + + status = encode_sessionid(xdr, &res->csr_sessionid); + if (status) + return status; + + p = xdr_reserve_space(xdr, 4 * sizeof(uint32_t)); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_RESOURCE); + + *p++ = htonl(res->csr_sequenceid); + *p++ = htonl(res->csr_slotid); + *p++ = htonl(res->csr_highestslotid); + *p++ = htonl(res->csr_target_highestslotid); + return 0; +} + +static __be32 +preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op) +{ + if (op_nr == OP_CB_SEQUENCE) { + if (nop != 0) + return htonl(NFS4ERR_SEQUENCE_POS); + } else { + if (nop == 0) + return htonl(NFS4ERR_OP_NOT_IN_SESSION); + } + + switch (op_nr) { + case OP_CB_GETATTR: + case OP_CB_RECALL: + case OP_CB_SEQUENCE: + case OP_CB_RECALL_ANY: + case OP_CB_RECALL_SLOT: + case OP_CB_LAYOUTRECALL: + case OP_CB_NOTIFY_DEVICEID: + case OP_CB_NOTIFY_LOCK: + *op = &callback_ops[op_nr]; + break; + + case OP_CB_NOTIFY: + case OP_CB_PUSH_DELEG: + case OP_CB_RECALLABLE_OBJ_AVAIL: + case OP_CB_WANTS_CANCELLED: + return htonl(NFS4ERR_NOTSUPP); + + default: + return htonl(NFS4ERR_OP_ILLEGAL); + } + + return htonl(NFS_OK); +} + +static void nfs4_callback_free_slot(struct nfs4_session *session, + struct nfs4_slot *slot) +{ + struct nfs4_slot_table *tbl = &session->bc_slot_table; + + spin_lock(&tbl->slot_tbl_lock); + /* + * Let the state manager know callback processing done. + * A single slot, so highest used slotid is either 0 or -1 + */ + nfs4_free_slot(tbl, slot); + spin_unlock(&tbl->slot_tbl_lock); +} + +static void nfs4_cb_free_slot(struct cb_process_state *cps) +{ + if (cps->slot) { + nfs4_callback_free_slot(cps->clp->cl_session, cps->slot); + cps->slot = NULL; + } +} + +#ifdef CONFIG_NFS_V4_2 +static __be32 +preprocess_nfs42_op(int nop, unsigned int op_nr, struct callback_op **op) +{ + __be32 status = preprocess_nfs41_op(nop, op_nr, op); + if (status != htonl(NFS4ERR_OP_ILLEGAL)) + return status; + + if (op_nr == OP_CB_OFFLOAD) { + *op = &callback_ops[op_nr]; + return htonl(NFS_OK); + } else + return htonl(NFS4ERR_NOTSUPP); + return htonl(NFS4ERR_OP_ILLEGAL); +} +#else /* CONFIG_NFS_V4_2 */ +static __be32 +preprocess_nfs42_op(int nop, unsigned int op_nr, struct callback_op **op) +{ + return htonl(NFS4ERR_MINOR_VERS_MISMATCH); +} +#endif /* CONFIG_NFS_V4_2 */ + +static __be32 +preprocess_nfs4_op(unsigned int op_nr, struct callback_op **op) +{ + switch (op_nr) { + case OP_CB_GETATTR: + case OP_CB_RECALL: + *op = &callback_ops[op_nr]; + break; + default: + return htonl(NFS4ERR_OP_ILLEGAL); + } + + return htonl(NFS_OK); +} + +static __be32 process_op(int nop, struct svc_rqst *rqstp, + struct cb_process_state *cps) +{ + struct xdr_stream *xdr_out = &rqstp->rq_res_stream; + struct callback_op *op = &callback_ops[0]; + unsigned int op_nr; + __be32 status; + long maxlen; + __be32 res; + + status = decode_op_hdr(&rqstp->rq_arg_stream, &op_nr); + if (unlikely(status)) + return status; + + switch (cps->minorversion) { + case 0: + status = preprocess_nfs4_op(op_nr, &op); + break; + case 1: + status = preprocess_nfs41_op(nop, op_nr, &op); + break; + case 2: + status = preprocess_nfs42_op(nop, op_nr, &op); + break; + default: + status = htonl(NFS4ERR_MINOR_VERS_MISMATCH); + } + + if (status == htonl(NFS4ERR_OP_ILLEGAL)) + op_nr = OP_CB_ILLEGAL; + if (status) + goto encode_hdr; + + if (cps->drc_status) { + status = cps->drc_status; + goto encode_hdr; + } + + maxlen = xdr_out->end - xdr_out->p; + if (maxlen > 0 && maxlen < PAGE_SIZE) { + status = op->decode_args(rqstp, &rqstp->rq_arg_stream, + rqstp->rq_argp); + if (likely(status == 0)) + status = op->process_op(rqstp->rq_argp, rqstp->rq_resp, + cps); + } else + status = htonl(NFS4ERR_RESOURCE); + +encode_hdr: + res = encode_op_hdr(xdr_out, op_nr, status); + if (unlikely(res)) + return res; + if (op->encode_res != NULL && status == 0) + status = op->encode_res(rqstp, xdr_out, rqstp->rq_resp); + return status; +} + +/* + * Decode, process and encode a COMPOUND + */ +static __be32 nfs4_callback_compound(struct svc_rqst *rqstp) +{ + struct cb_compound_hdr_arg hdr_arg = { 0 }; + struct cb_compound_hdr_res hdr_res = { NULL }; + struct cb_process_state cps = { + .drc_status = 0, + .clp = NULL, + .net = SVC_NET(rqstp), + }; + unsigned int nops = 0; + __be32 status; + + status = decode_compound_hdr_arg(&rqstp->rq_arg_stream, &hdr_arg); + if (status == htonl(NFS4ERR_RESOURCE)) + return rpc_garbage_args; + + if (hdr_arg.minorversion == 0) { + cps.clp = nfs4_find_client_ident(SVC_NET(rqstp), hdr_arg.cb_ident); + if (!cps.clp) { + trace_nfs_cb_no_clp(rqstp->rq_xid, hdr_arg.cb_ident); + goto out_invalidcred; + } + if (!check_gss_callback_principal(cps.clp, rqstp)) { + trace_nfs_cb_badprinc(rqstp->rq_xid, hdr_arg.cb_ident); + nfs_put_client(cps.clp); + goto out_invalidcred; + } + svc_xprt_set_valid(rqstp->rq_xprt); + } + + cps.minorversion = hdr_arg.minorversion; + hdr_res.taglen = hdr_arg.taglen; + hdr_res.tag = hdr_arg.tag; + if (encode_compound_hdr_res(&rqstp->rq_res_stream, &hdr_res) != 0) { + if (cps.clp) + nfs_put_client(cps.clp); + return rpc_system_err; + } + while (status == 0 && nops != hdr_arg.nops) { + status = process_op(nops, rqstp, &cps); + nops++; + } + + /* Buffer overflow in decode_ops_hdr or encode_ops_hdr. Return + * resource error in cb_compound status without returning op */ + if (unlikely(status == htonl(NFS4ERR_RESOURCE_HDR))) { + status = htonl(NFS4ERR_RESOURCE); + nops--; + } + + if (svc_is_backchannel(rqstp) && cps.clp) { + rqstp->bc_to_initval = cps.clp->cl_rpcclient->cl_timeout->to_initval; + rqstp->bc_to_retries = cps.clp->cl_rpcclient->cl_timeout->to_retries; + } + + *hdr_res.status = status; + *hdr_res.nops = htonl(nops); + nfs4_cb_free_slot(&cps); + nfs_put_client(cps.clp); + return rpc_success; + +out_invalidcred: + pr_warn_ratelimited("NFS: NFSv4 callback contains invalid cred\n"); + rqstp->rq_auth_stat = rpc_autherr_badcred; + return rpc_success; +} + +static int +nfs_callback_dispatch(struct svc_rqst *rqstp) +{ + const struct svc_procedure *procp = rqstp->rq_procinfo; + + *rqstp->rq_accept_statp = procp->pc_func(rqstp); + return 1; +} + +/* + * Define NFS4 callback COMPOUND ops. + */ +static struct callback_op callback_ops[] = { + [0] = { + .res_maxsize = CB_OP_HDR_RES_MAXSZ, + }, + [OP_CB_GETATTR] = { + .process_op = nfs4_callback_getattr, + .decode_args = decode_getattr_args, + .encode_res = encode_getattr_res, + .res_maxsize = CB_OP_GETATTR_RES_MAXSZ, + }, + [OP_CB_RECALL] = { + .process_op = nfs4_callback_recall, + .decode_args = decode_recall_args, + .res_maxsize = CB_OP_RECALL_RES_MAXSZ, + }, + [OP_CB_LAYOUTRECALL] = { + .process_op = nfs4_callback_layoutrecall, + .decode_args = decode_layoutrecall_args, + .res_maxsize = CB_OP_LAYOUTRECALL_RES_MAXSZ, + }, + [OP_CB_NOTIFY_DEVICEID] = { + .process_op = nfs4_callback_devicenotify, + .decode_args = decode_devicenotify_args, + .res_maxsize = CB_OP_DEVICENOTIFY_RES_MAXSZ, + }, + [OP_CB_SEQUENCE] = { + .process_op = nfs4_callback_sequence, + .decode_args = decode_cb_sequence_args, + .encode_res = encode_cb_sequence_res, + .res_maxsize = CB_OP_SEQUENCE_RES_MAXSZ, + }, + [OP_CB_RECALL_ANY] = { + .process_op = nfs4_callback_recallany, + .decode_args = decode_recallany_args, + .res_maxsize = CB_OP_RECALLANY_RES_MAXSZ, + }, + [OP_CB_RECALL_SLOT] = { + .process_op = nfs4_callback_recallslot, + .decode_args = decode_recallslot_args, + .res_maxsize = CB_OP_RECALLSLOT_RES_MAXSZ, + }, + [OP_CB_NOTIFY_LOCK] = { + .process_op = nfs4_callback_notify_lock, + .decode_args = decode_notify_lock_args, + .res_maxsize = CB_OP_NOTIFY_LOCK_RES_MAXSZ, + }, +#ifdef CONFIG_NFS_V4_2 + [OP_CB_OFFLOAD] = { + .process_op = nfs4_callback_offload, + .decode_args = decode_offload_args, + .res_maxsize = CB_OP_OFFLOAD_RES_MAXSZ, + }, +#endif /* CONFIG_NFS_V4_2 */ +}; + +/* + * Define NFS4 callback procedures + */ +static const struct svc_procedure nfs4_callback_procedures1[] = { + [CB_NULL] = { + .pc_func = nfs4_callback_null, + .pc_encode = nfs4_encode_void, + .pc_xdrressize = 1, + .pc_name = "NULL", + }, + [CB_COMPOUND] = { + .pc_func = nfs4_callback_compound, + .pc_encode = nfs4_encode_void, + .pc_argsize = 256, + .pc_argzero = 256, + .pc_ressize = 256, + .pc_xdrressize = NFS4_CALLBACK_BUFSIZE, + .pc_name = "COMPOUND", + } +}; + +const struct svc_version nfs4_callback_version1 = { + .vs_vers = 1, + .vs_nproc = ARRAY_SIZE(nfs4_callback_procedures1), + .vs_proc = nfs4_callback_procedures1, + .vs_xdrsize = NFS4_CALLBACK_XDRSIZE, + .vs_dispatch = nfs_callback_dispatch, + .vs_hidden = true, + .vs_need_cong_ctrl = true, +}; + +const struct svc_version nfs4_callback_version4 = { + .vs_vers = 4, + .vs_nproc = ARRAY_SIZE(nfs4_callback_procedures1), + .vs_proc = nfs4_callback_procedures1, + .vs_xdrsize = NFS4_CALLBACK_XDRSIZE, + .vs_dispatch = nfs_callback_dispatch, + .vs_hidden = true, + .vs_need_cong_ctrl = true, +}; diff --git a/fs/nfs/client.c b/fs/nfs/client.c new file mode 100644 index 000000000..4dcb91ab3 --- /dev/null +++ b/fs/nfs/client.c @@ -0,0 +1,1542 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* client.c: NFS client sharing and management code + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "iostat.h" +#include "internal.h" +#include "fscache.h" +#include "pnfs.h" +#include "nfs.h" +#include "netns.h" +#include "sysfs.h" +#include "nfs42.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq); +static DEFINE_RWLOCK(nfs_version_lock); + +static struct nfs_subversion *nfs_version_mods[5] = { + [2] = NULL, + [3] = NULL, + [4] = NULL, +}; + +/* + * RPC cruft for NFS + */ +static const struct rpc_version *nfs_version[5] = { + [2] = NULL, + [3] = NULL, + [4] = NULL, +}; + +const struct rpc_program nfs_program = { + .name = "nfs", + .number = NFS_PROGRAM, + .nrvers = ARRAY_SIZE(nfs_version), + .version = nfs_version, + .pipe_dir_name = NFS_PIPE_DIRNAME, +}; + +static struct nfs_subversion *__find_nfs_version(unsigned int version) +{ + struct nfs_subversion *nfs; + + read_lock(&nfs_version_lock); + nfs = nfs_version_mods[version]; + read_unlock(&nfs_version_lock); + return nfs; +} + +struct nfs_subversion *find_nfs_version(unsigned int version) +{ + struct nfs_subversion *nfs = __find_nfs_version(version); + + if (!nfs && request_module("nfsv%d", version) == 0) + nfs = __find_nfs_version(version); + + if (!nfs) + return ERR_PTR(-EPROTONOSUPPORT); + + if (!get_nfs_version(nfs)) + return ERR_PTR(-EAGAIN); + + return nfs; +} + +int get_nfs_version(struct nfs_subversion *nfs) +{ + return try_module_get(nfs->owner); +} +EXPORT_SYMBOL_GPL(get_nfs_version); + +void put_nfs_version(struct nfs_subversion *nfs) +{ + module_put(nfs->owner); +} + +void register_nfs_version(struct nfs_subversion *nfs) +{ + write_lock(&nfs_version_lock); + + nfs_version_mods[nfs->rpc_ops->version] = nfs; + nfs_version[nfs->rpc_ops->version] = nfs->rpc_vers; + + write_unlock(&nfs_version_lock); +} +EXPORT_SYMBOL_GPL(register_nfs_version); + +void unregister_nfs_version(struct nfs_subversion *nfs) +{ + write_lock(&nfs_version_lock); + + nfs_version[nfs->rpc_ops->version] = NULL; + nfs_version_mods[nfs->rpc_ops->version] = NULL; + + write_unlock(&nfs_version_lock); +} +EXPORT_SYMBOL_GPL(unregister_nfs_version); + +/* + * Allocate a shared client record + * + * Since these are allocated/deallocated very rarely, we don't + * bother putting them in a slab cache... + */ +struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) +{ + struct nfs_client *clp; + int err = -ENOMEM; + + if ((clp = kzalloc_obj(*clp)) == NULL) + goto error_0; + + clp->cl_minorversion = cl_init->minorversion; + clp->cl_nfs_mod = cl_init->nfs_mod; + if (!get_nfs_version(clp->cl_nfs_mod)) + goto error_dealloc; + + clp->rpc_ops = clp->cl_nfs_mod->rpc_ops; + + refcount_set(&clp->cl_count, 1); + clp->cl_cons_state = NFS_CS_INITING; + + memcpy(&clp->cl_addr, cl_init->addr, cl_init->addrlen); + clp->cl_addrlen = cl_init->addrlen; + + if (cl_init->hostname) { + err = -ENOMEM; + clp->cl_hostname = kstrdup(cl_init->hostname, GFP_KERNEL); + if (!clp->cl_hostname) + goto error_cleanup; + } + + INIT_LIST_HEAD(&clp->cl_superblocks); + clp->cl_rpcclient = ERR_PTR(-EINVAL); + + clp->cl_flags = cl_init->init_flags; + clp->cl_proto = cl_init->proto; + clp->cl_nconnect = cl_init->nconnect; + clp->cl_max_connect = cl_init->max_connect ? cl_init->max_connect : 1; + clp->cl_net = get_net_track(cl_init->net, &clp->cl_ns_tracker, GFP_KERNEL); + +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + seqlock_init(&clp->cl_boot_lock); + ktime_get_real_ts64(&clp->cl_nfssvc_boot); + nfs_uuid_init(&clp->cl_uuid); + INIT_WORK(&clp->cl_local_probe_work, nfs_local_probe_async_work); +#endif /* CONFIG_NFS_LOCALIO */ + + clp->cl_principal = "*"; + clp->cl_xprtsec = cl_init->xprtsec; + return clp; + +error_cleanup: + put_nfs_version(clp->cl_nfs_mod); +error_dealloc: + kfree(clp); +error_0: + return ERR_PTR(err); +} +EXPORT_SYMBOL_GPL(nfs_alloc_client); + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_cleanup_cb_ident_idr(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + idr_destroy(&nn->cb_ident_idr); +} + +/* nfs_client_lock held */ +static void nfs_cb_idr_remove_locked(struct nfs_client *clp) +{ + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + + if (clp->cl_cb_ident) + idr_remove(&nn->cb_ident_idr, clp->cl_cb_ident); +} + +static void pnfs_init_server(struct nfs_server *server) +{ + rpc_init_wait_queue(&server->roc_rpcwaitq, "pNFS ROC"); +} + +#else +static void nfs_cleanup_cb_ident_idr(struct net *net) +{ +} + +static void nfs_cb_idr_remove_locked(struct nfs_client *clp) +{ +} + +static void pnfs_init_server(struct nfs_server *server) +{ +} + +#endif /* CONFIG_NFS_V4 */ + +/* + * Destroy a shared client record + */ +void nfs_free_client(struct nfs_client *clp) +{ + nfs_localio_disable_client(clp); + + /* -EIO all pending I/O */ + if (!IS_ERR(clp->cl_rpcclient)) + rpc_shutdown_client(clp->cl_rpcclient); + + put_net_track(clp->cl_net, &clp->cl_ns_tracker); + put_nfs_version(clp->cl_nfs_mod); + kfree(clp->cl_hostname); + kfree(clp->cl_acceptor); + kfree_rcu(clp, rcu); +} +EXPORT_SYMBOL_GPL(nfs_free_client); + +/* + * Release a reference to a shared client record + */ +void nfs_put_client(struct nfs_client *clp) +{ + struct nfs_net *nn; + + if (!clp) + return; + + nn = net_generic(clp->cl_net, nfs_net_id); + + if (refcount_dec_and_lock(&clp->cl_count, &nn->nfs_client_lock)) { + list_del(&clp->cl_share_link); + nfs_cb_idr_remove_locked(clp); + spin_unlock(&nn->nfs_client_lock); + + WARN_ON_ONCE(!list_empty(&clp->cl_superblocks)); + + clp->rpc_ops->free_client(clp); + } +} +EXPORT_SYMBOL_GPL(nfs_put_client); + +/* + * Find an nfs_client on the list that matches the initialisation data + * that is supplied. + */ +static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *data) +{ + struct nfs_client *clp; + const struct sockaddr *sap = (struct sockaddr *)data->addr; + struct nfs_net *nn = net_generic(data->net, nfs_net_id); + int error; + +again: + list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) { + const struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr; + /* Don't match clients that failed to initialise properly */ + if (clp->cl_cons_state < 0) + continue; + + /* If a client is still initializing then we need to wait */ + if (clp->cl_cons_state > NFS_CS_READY) { + refcount_inc(&clp->cl_count); + spin_unlock(&nn->nfs_client_lock); + error = nfs_wait_client_init_complete(clp); + nfs_put_client(clp); + spin_lock(&nn->nfs_client_lock); + if (error < 0) + return ERR_PTR(error); + goto again; + } + + /* Different NFS versions cannot share the same nfs_client */ + if (clp->rpc_ops != data->nfs_mod->rpc_ops) + continue; + + if (clp->cl_proto != data->proto) + continue; + /* Match nfsv4 minorversion */ + if (clp->cl_minorversion != data->minorversion) + continue; + + /* Match request for a dedicated DS */ + if (test_bit(NFS_CS_DS, &data->init_flags) != + test_bit(NFS_CS_DS, &clp->cl_flags)) + continue; + + /* Match the full socket address */ + if (!rpc_cmp_addr_port(sap, clap)) + /* Match all xprt_switch full socket addresses */ + if (IS_ERR(clp->cl_rpcclient) || + !rpc_clnt_xprt_switch_has_addr(clp->cl_rpcclient, + sap)) + continue; + + /* Match the xprt security policy */ + if (clp->cl_xprtsec.policy != data->xprtsec.policy) + continue; + if (clp->cl_xprtsec.policy == RPC_XPRTSEC_TLS_X509) { + if (clp->cl_xprtsec.cert_serial != + data->xprtsec.cert_serial) + continue; + if (clp->cl_xprtsec.privkey_serial != + data->xprtsec.privkey_serial) + continue; + } + + refcount_inc(&clp->cl_count); + return clp; + } + return NULL; +} + +/* + * Return true if @clp is done initializing, false if still working on it. + * + * Use nfs_client_init_status to check if it was successful. + */ +bool nfs_client_init_is_complete(const struct nfs_client *clp) +{ + return clp->cl_cons_state <= NFS_CS_READY; +} +EXPORT_SYMBOL_GPL(nfs_client_init_is_complete); + +/* + * Return 0 if @clp was successfully initialized, -errno otherwise. + * + * This must be called *after* nfs_client_init_is_complete() returns true, + * otherwise it will pop WARN_ON_ONCE and return -EINVAL + */ +int nfs_client_init_status(const struct nfs_client *clp) +{ + /* called without checking nfs_client_init_is_complete */ + if (clp->cl_cons_state > NFS_CS_READY) { + WARN_ON_ONCE(1); + return -EINVAL; + } + return clp->cl_cons_state; +} +EXPORT_SYMBOL_GPL(nfs_client_init_status); + +int nfs_wait_client_init_complete(const struct nfs_client *clp) +{ + return wait_event_killable(nfs_client_active_wq, + nfs_client_init_is_complete(clp)); +} +EXPORT_SYMBOL_GPL(nfs_wait_client_init_complete); + +/* + * Found an existing client. Make sure it's ready before returning. + */ +static struct nfs_client * +nfs_found_client(const struct nfs_client_initdata *cl_init, + struct nfs_client *clp) +{ + int error; + + error = nfs_wait_client_init_complete(clp); + if (error < 0) { + nfs_put_client(clp); + return ERR_PTR(-ERESTARTSYS); + } + + if (clp->cl_cons_state < NFS_CS_READY) { + error = clp->cl_cons_state; + nfs_put_client(clp); + return ERR_PTR(error); + } + + smp_rmb(); + return clp; +} + +/* + * Look up a client by IP address and protocol version + * - creates a new record if one doesn't yet exist + */ +struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_init) +{ + struct nfs_client *clp, *new = NULL; + struct nfs_net *nn = net_generic(cl_init->net, nfs_net_id); + const struct nfs_rpc_ops *rpc_ops = cl_init->nfs_mod->rpc_ops; + + if (cl_init->hostname == NULL) { + WARN_ON(1); + return ERR_PTR(-EINVAL); + } + + /* see if the client already exists */ + do { + spin_lock(&nn->nfs_client_lock); + + clp = nfs_match_client(cl_init); + if (clp) { + spin_unlock(&nn->nfs_client_lock); + if (new) + new->rpc_ops->free_client(new); + if (IS_ERR(clp)) + return clp; + return nfs_found_client(cl_init, clp); + } + if (new) { + list_add_tail(&new->cl_share_link, + &nn->nfs_client_list); + spin_unlock(&nn->nfs_client_lock); + new = rpc_ops->init_client(new, cl_init); + if (!IS_ERR(new)) + nfs_local_probe_async(new); + return new; + } + + spin_unlock(&nn->nfs_client_lock); + + new = rpc_ops->alloc_client(cl_init); + } while (!IS_ERR(new)); + + return new; +} +EXPORT_SYMBOL_GPL(nfs_get_client); + +/* + * Mark a server as ready or failed + */ +void nfs_mark_client_ready(struct nfs_client *clp, int state) +{ + smp_wmb(); + clp->cl_cons_state = state; + wake_up_all(&nfs_client_active_wq); +} +EXPORT_SYMBOL_GPL(nfs_mark_client_ready); + +/* + * Initialise the timeout values for a connection + */ +void nfs_init_timeout_values(struct rpc_timeout *to, int proto, + int timeo, int retrans) +{ + to->to_initval = timeo * HZ / 10; + to->to_retries = retrans; + + switch (proto) { + case XPRT_TRANSPORT_TCP: + case XPRT_TRANSPORT_TCP_TLS: + case XPRT_TRANSPORT_RDMA: + if (retrans == NFS_UNSPEC_RETRANS) + to->to_retries = NFS_DEF_TCP_RETRANS; + if (timeo == NFS_UNSPEC_TIMEO || to->to_initval == 0) + to->to_initval = NFS_DEF_TCP_TIMEO * HZ / 10; + if (to->to_initval > NFS_MAX_TCP_TIMEOUT) + to->to_initval = NFS_MAX_TCP_TIMEOUT; + to->to_increment = to->to_initval; + to->to_maxval = to->to_initval + (to->to_increment * to->to_retries); + if (to->to_maxval > NFS_MAX_TCP_TIMEOUT) + to->to_maxval = NFS_MAX_TCP_TIMEOUT; + if (to->to_maxval < to->to_initval) + to->to_maxval = to->to_initval; + to->to_exponential = 0; + break; + case XPRT_TRANSPORT_UDP: + if (retrans == NFS_UNSPEC_RETRANS) + to->to_retries = NFS_DEF_UDP_RETRANS; + if (timeo == NFS_UNSPEC_TIMEO || to->to_initval == 0) + to->to_initval = NFS_DEF_UDP_TIMEO * HZ / 10; + if (to->to_initval > NFS_MAX_UDP_TIMEOUT) + to->to_initval = NFS_MAX_UDP_TIMEOUT; + to->to_maxval = NFS_MAX_UDP_TIMEOUT; + to->to_exponential = 1; + break; + default: + BUG(); + } +} +EXPORT_SYMBOL_GPL(nfs_init_timeout_values); + +/* + * Create an RPC client handle + */ +int nfs_create_rpc_client(struct nfs_client *clp, + const struct nfs_client_initdata *cl_init, + rpc_authflavor_t flavor) +{ + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + struct rpc_clnt *clnt = NULL; + struct rpc_create_args args = { + .net = clp->cl_net, + .protocol = clp->cl_proto, + .nconnect = clp->cl_nconnect, + .address = (struct sockaddr *)&clp->cl_addr, + .addrsize = clp->cl_addrlen, + .timeout = cl_init->timeparms, + .servername = clp->cl_hostname, + .nodename = cl_init->nodename, + .program = &nfs_program, + .stats = &nn->rpcstats, + .version = clp->rpc_ops->version, + .authflavor = flavor, + .cred = cl_init->cred, + .xprtsec = cl_init->xprtsec, + .connect_timeout = cl_init->connect_timeout, + .reconnect_timeout = cl_init->reconnect_timeout, + }; + + if (test_bit(NFS_CS_DISCRTRY, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_DISCRTRY; + if (test_bit(NFS_CS_NO_RETRANS_TIMEOUT, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_NO_RETRANS_TIMEOUT; + if (test_bit(NFS_CS_NORESVPORT, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_NONPRIVPORT; + if (test_bit(NFS_CS_INFINITE_SLOTS, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_INFINITE_SLOTS; + if (test_bit(NFS_CS_NOPING, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_NOPING; + if (test_bit(NFS_CS_REUSEPORT, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_REUSEPORT; + if (test_bit(NFS_CS_NETUNREACH_FATAL, &clp->cl_flags)) + args.flags |= RPC_CLNT_CREATE_NETUNREACH_FATAL; + + if (!IS_ERR(clp->cl_rpcclient)) + return 0; + + clnt = rpc_create(&args); + if (IS_ERR(clnt)) { + dprintk("%s: cannot create RPC client. Error = %ld\n", + __func__, PTR_ERR(clnt)); + return PTR_ERR(clnt); + } + + clnt->cl_principal = clp->cl_principal; + clp->cl_rpcclient = clnt; + clnt->cl_max_connect = clp->cl_max_connect; + return 0; +} +EXPORT_SYMBOL_GPL(nfs_create_rpc_client); + +/* + * Version 2 or 3 client destruction + */ +static void nfs_destroy_server(struct nfs_server *server) +{ + if (server->nlm_host) + nlmclnt_done(server->nlm_host); +} + +/* + * Version 2 or 3 lockd setup + */ +static int nfs_start_lockd(struct nfs_server *server) +{ + struct nlm_host *host; + struct nfs_client *clp = server->nfs_client; + struct nlmclnt_initdata nlm_init = { + .hostname = clp->cl_hostname, + .address = (struct sockaddr *)&clp->cl_addr, + .addrlen = clp->cl_addrlen, + .nfs_version = clp->rpc_ops->version, + .noresvport = server->flags & NFS_MOUNT_NORESVPORT ? + 1 : 0, + .net = clp->cl_net, + .nlmclnt_ops = clp->cl_nfs_mod->rpc_ops->nlmclnt_ops, + .cred = server->cred, + }; + + if (nlm_init.nfs_version > 3) + return 0; + if ((server->flags & NFS_MOUNT_LOCAL_FLOCK) && + (server->flags & NFS_MOUNT_LOCAL_FCNTL)) + return 0; + + switch (clp->cl_proto) { + default: + nlm_init.protocol = IPPROTO_TCP; + break; +#ifndef CONFIG_NFS_DISABLE_UDP_SUPPORT + case XPRT_TRANSPORT_UDP: + nlm_init.protocol = IPPROTO_UDP; +#endif + } + + host = nlmclnt_init(&nlm_init); + if (IS_ERR(host)) + return PTR_ERR(host); + + server->nlm_host = host; + server->destroy = nfs_destroy_server; + nfs_sysfs_link_rpc_client(server, nlmclnt_rpc_clnt(host), NULL); + return 0; +} + +/* + * Create a general RPC client + */ +int nfs_init_server_rpcclient(struct nfs_server *server, + const struct rpc_timeout *timeo, + rpc_authflavor_t pseudoflavour) +{ + struct nfs_client *clp = server->nfs_client; + + server->client = rpc_clone_client_set_auth(clp->cl_rpcclient, + pseudoflavour); + if (IS_ERR(server->client)) { + dprintk("%s: couldn't create rpc_client!\n", __func__); + return PTR_ERR(server->client); + } + + memcpy(&server->client->cl_timeout_default, + timeo, + sizeof(server->client->cl_timeout_default)); + server->client->cl_timeout = &server->client->cl_timeout_default; + server->client->cl_softrtry = 0; + if (server->flags & NFS_MOUNT_SOFTERR) + server->client->cl_softerr = 1; + if (server->flags & NFS_MOUNT_SOFT) + server->client->cl_softrtry = 1; + + nfs_sysfs_link_rpc_client(server, server->client, NULL); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_init_server_rpcclient); + +/** + * nfs_init_client - Initialise an NFS2 or NFS3 client + * + * @clp: nfs_client to initialise + * @cl_init: Initialisation parameters + * + * Returns pointer to an NFS client, or an ERR_PTR value. + */ +struct nfs_client *nfs_init_client(struct nfs_client *clp, + const struct nfs_client_initdata *cl_init) +{ + int error; + + /* the client is already initialised */ + if (clp->cl_cons_state == NFS_CS_READY) + return clp; + + /* + * Create a client RPC handle for doing FSSTAT with UNIX auth only + * - RFC 2623, sec 2.3.2 + */ + error = nfs_create_rpc_client(clp, cl_init, RPC_AUTH_UNIX); + nfs_mark_client_ready(clp, error == 0 ? NFS_CS_READY : error); + if (error < 0) { + nfs_put_client(clp); + clp = ERR_PTR(error); + } + return clp; +} +EXPORT_SYMBOL_GPL(nfs_init_client); + +static void nfs4_server_set_init_caps(struct nfs_server *server) +{ +#if IS_ENABLED(CONFIG_NFS_V4) + /* Set the basic capabilities */ + server->caps = server->nfs_client->cl_mvops->init_caps; + if (server->flags & NFS_MOUNT_NORDIRPLUS) + server->caps &= ~NFS_CAP_READDIRPLUS; + if (server->nfs_client->cl_proto == XPRT_TRANSPORT_RDMA) + server->caps &= ~NFS_CAP_READ_PLUS; + + /* + * Don't use NFS uid/gid mapping if we're using AUTH_SYS or lower + * authentication. + */ + if (nfs4_disable_idmapping && + server->client->cl_auth->au_flavor == RPC_AUTH_UNIX) + server->caps |= NFS_CAP_UIDGID_NOMAP; +#endif +} + +void nfs_server_set_init_caps(struct nfs_server *server) +{ + switch (server->nfs_client->rpc_ops->version) { + case 2: + server->caps = NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS; + break; + case 3: + server->caps = NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS; + if (!(server->flags & NFS_MOUNT_NORDIRPLUS)) + server->caps |= NFS_CAP_READDIRPLUS; + break; + default: + nfs4_server_set_init_caps(server); + break; + } +} +EXPORT_SYMBOL_GPL(nfs_server_set_init_caps); + +/* + * Create a version 2 or 3 client + */ +static int nfs_init_server(struct nfs_server *server, + const struct fs_context *fc) +{ + const struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct rpc_timeout timeparms; + struct nfs_client_initdata cl_init = { + .hostname = ctx->nfs_server.hostname, + .addr = &ctx->nfs_server._address, + .addrlen = ctx->nfs_server.addrlen, + .nfs_mod = ctx->nfs_mod, + .proto = ctx->nfs_server.protocol, + .net = fc->net_ns, + .timeparms = &timeparms, + .cred = server->cred, + .nconnect = ctx->nfs_server.nconnect, + .init_flags = (1UL << NFS_CS_REUSEPORT), + .xprtsec = ctx->xprtsec, + }; + struct nfs_client *clp; + int error; + + nfs_init_timeout_values(&timeparms, ctx->nfs_server.protocol, + ctx->timeo, ctx->retrans); + if (ctx->flags & NFS_MOUNT_NORESVPORT) + set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); + + if (ctx->flags & NFS_MOUNT_NETUNREACH_FATAL) + __set_bit(NFS_CS_NETUNREACH_FATAL, &cl_init.init_flags); + + /* Allocate or find a client reference we can use */ + clp = nfs_get_client(&cl_init); + if (IS_ERR(clp)) + return PTR_ERR(clp); + + server->nfs_client = clp; + nfs_sysfs_add_server(server); + nfs_sysfs_link_rpc_client(server, clp->cl_rpcclient, "_state"); + + /* Initialise the client representation from the mount data */ + server->flags = ctx->flags; + server->options = ctx->options; + + switch (clp->rpc_ops->version) { + case 2: + server->fattr_valid = NFS_ATTR_FATTR_V2; + break; + case 3: + server->fattr_valid = NFS_ATTR_FATTR_V3; + break; + default: + server->fattr_valid = NFS_ATTR_FATTR_V4; + } + + if (ctx->bsize) { + server->bsize = ctx->bsize; + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_BSIZE; + } + if (ctx->rsize) { + server->rsize = nfs_io_size(ctx->rsize, clp->cl_proto); + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_RSIZE; + } + if (ctx->wsize) { + server->wsize = nfs_io_size(ctx->wsize, clp->cl_proto); + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_WSIZE; + } + + server->acregmin = ctx->acregmin * HZ; + server->acregmax = ctx->acregmax * HZ; + server->acdirmin = ctx->acdirmin * HZ; + server->acdirmax = ctx->acdirmax * HZ; + + /* Start lockd here, before we might error out */ + error = nfs_start_lockd(server); + if (error < 0) + goto error; + + server->port = ctx->nfs_server.port; + server->auth_info = ctx->auth_info; + + error = nfs_init_server_rpcclient(server, &timeparms, + ctx->selected_flavor); + if (error < 0) + goto error; + + nfs_server_set_init_caps(server); + + /* Preserve the values of mount_server-related mount options */ + if (ctx->mount_server.addrlen) { + memcpy(&server->mountd_address, &ctx->mount_server.address, + ctx->mount_server.addrlen); + server->mountd_addrlen = ctx->mount_server.addrlen; + } + server->mountd_version = ctx->mount_server.version; + server->mountd_port = ctx->mount_server.port; + server->mountd_protocol = ctx->mount_server.protocol; + + server->namelen = ctx->namlen; + return 0; + +error: + server->nfs_client = NULL; + nfs_put_client(clp); + return error; +} + +/* + * Load up the server record from information gained in an fsinfo record + */ +static void nfs_server_set_fsinfo(struct nfs_server *server, + struct nfs_fsinfo *fsinfo) +{ + struct nfs_client *clp = server->nfs_client; + unsigned long max_rpc_payload, raw_max_rpc_payload; + + /* Work out a lot of parameters */ + if (server->rsize == 0) + server->rsize = nfs_io_size(fsinfo->rtpref, clp->cl_proto); + if (server->wsize == 0) + server->wsize = nfs_io_size(fsinfo->wtpref, clp->cl_proto); + + if (fsinfo->rtmax >= 512 && server->rsize > fsinfo->rtmax) + server->rsize = nfs_io_size(fsinfo->rtmax, clp->cl_proto); + if (fsinfo->wtmax >= 512 && server->wsize > fsinfo->wtmax) + server->wsize = nfs_io_size(fsinfo->wtmax, clp->cl_proto); + + raw_max_rpc_payload = rpc_max_payload(server->client); + max_rpc_payload = nfs_block_size(raw_max_rpc_payload, NULL); + + if (server->rsize > max_rpc_payload) + server->rsize = max_rpc_payload; + if (server->rsize > NFS_MAX_FILE_IO_SIZE) + server->rsize = NFS_MAX_FILE_IO_SIZE; + server->rpages = (server->rsize + PAGE_SIZE - 1) >> PAGE_SHIFT; + + if (server->wsize > max_rpc_payload) + server->wsize = max_rpc_payload; + if (server->wsize > NFS_MAX_FILE_IO_SIZE) + server->wsize = NFS_MAX_FILE_IO_SIZE; + + server->wtmult = nfs_block_bits(fsinfo->wtmult, NULL); + + server->dtsize = nfs_block_size(fsinfo->dtpref, NULL); + if (server->dtsize > NFS_MAX_FILE_IO_SIZE) + server->dtsize = NFS_MAX_FILE_IO_SIZE; + if (server->dtsize > server->rsize) + server->dtsize = server->rsize; + + if (server->flags & NFS_MOUNT_NOAC) { + server->acregmin = server->acregmax = 0; + server->acdirmin = server->acdirmax = 0; + } + + server->maxfilesize = fsinfo->maxfilesize; + + server->change_attr_type = fsinfo->change_attr_type; + + server->clone_blksize = fsinfo->clone_blksize; + /* We're airborne Set socket buffersize */ + rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100); + +#ifdef CONFIG_NFS_V4_2 + /* + * Defaults until limited by the session parameters. + */ + server->gxasize = min_t(unsigned int, raw_max_rpc_payload, + XATTR_SIZE_MAX); + server->sxasize = min_t(unsigned int, raw_max_rpc_payload, + XATTR_SIZE_MAX); + server->lxasize = min_t(unsigned int, raw_max_rpc_payload, + nfs42_listxattr_xdrsize(XATTR_LIST_MAX)); + + if (fsinfo->xattr_support) + server->caps |= NFS_CAP_XATTR; + else + server->caps &= ~NFS_CAP_XATTR; +#endif +} + +/* + * Probe filesystem information, including the FSID on v2/v3 + */ +static int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs_fattr *fattr) +{ + struct nfs_pathconf pathinfo = { }; + struct nfs_fsinfo fsinfo; + struct nfs_client *clp = server->nfs_client; + int error; + + if (clp->rpc_ops->set_capabilities != NULL) { + error = clp->rpc_ops->set_capabilities(server, mntfh); + if (error < 0) + return error; + } + + fsinfo.fattr = fattr; + fsinfo.nlayouttypes = 0; + memset(fsinfo.layouttype, 0, sizeof(fsinfo.layouttype)); + error = clp->rpc_ops->fsinfo(server, mntfh, &fsinfo); + if (error < 0) + return error; + + nfs_server_set_fsinfo(server, &fsinfo); + + pathinfo.fattr = fattr; + nfs_fattr_init(fattr); + + if (clp->rpc_ops->version < 4 || server->namelen == 0) { + if (clp->rpc_ops->pathconf(server, mntfh, &pathinfo) >= 0) { + if (server->namelen == 0) + server->namelen = pathinfo.max_namelen; + if (clp->rpc_ops->version < 4) { + unsigned int caps = server->caps; + + caps &= ~(NFS_CAP_CASE_INSENSITIVE | + NFS_CAP_CASE_NONPRESERVING); + if (pathinfo.case_insensitive) + caps |= NFS_CAP_CASE_INSENSITIVE; + if (!pathinfo.case_preserving) + caps |= NFS_CAP_CASE_NONPRESERVING; + server->caps = caps; + } + } else if (clp->rpc_ops->version < 4) { + server->caps &= ~(NFS_CAP_CASE_INSENSITIVE | + NFS_CAP_CASE_NONPRESERVING); + } + } + + if (clp->rpc_ops->discover_trunking != NULL && + (server->caps & NFS_CAP_FS_LOCATIONS && + (server->flags & NFS_MOUNT_TRUNK_DISCOVERY))) { + error = clp->rpc_ops->discover_trunking(server, mntfh); + if (error < 0) + return error; + } + + return 0; +} + +/* + * Grab the destination's particulars, including lease expiry time. + * + * Returns zero if probe succeeded and retrieved FSID matches the FSID + * we have cached. + */ +int nfs_probe_server(struct nfs_server *server, struct nfs_fh *mntfh) +{ + struct nfs_fattr *fattr; + int error; + + fattr = nfs_alloc_fattr(); + if (fattr == NULL) + return -ENOMEM; + + /* Sanity: the probe won't work if the destination server + * does not recognize the migrated FH. */ + error = nfs_probe_fsinfo(server, mntfh, fattr); + + nfs_free_fattr(fattr); + return error; +} +EXPORT_SYMBOL_GPL(nfs_probe_server); + +/* + * Copy useful information when duplicating a server record + */ +void nfs_server_copy_userdata(struct nfs_server *target, struct nfs_server *source) +{ + target->flags = source->flags; + target->automount_inherit = source->automount_inherit; + if (source->automount_inherit & NFS_AUTOMOUNT_INHERIT_BSIZE) + target->bsize = source->bsize; + if (source->automount_inherit & NFS_AUTOMOUNT_INHERIT_RSIZE) + target->rsize = source->rsize; + if (source->automount_inherit & NFS_AUTOMOUNT_INHERIT_WSIZE) + target->wsize = source->wsize; + target->acregmin = source->acregmin; + target->acregmax = source->acregmax; + target->acdirmin = source->acdirmin; + target->acdirmax = source->acdirmax; + target->options = source->options; + target->auth_info = source->auth_info; + target->port = source->port; +} +EXPORT_SYMBOL_GPL(nfs_server_copy_userdata); + +void nfs_server_insert_lists(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + + spin_lock(&nn->nfs_client_lock); + list_add_tail_rcu(&server->client_link, &clp->cl_superblocks); + list_add_tail(&server->master_link, &nn->nfs_volume_list); + clear_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state); + spin_unlock(&nn->nfs_client_lock); + +} +EXPORT_SYMBOL_GPL(nfs_server_insert_lists); + +void nfs_server_remove_lists(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs_net *nn; + + if (clp == NULL) + return; + nn = net_generic(clp->cl_net, nfs_net_id); + spin_lock(&nn->nfs_client_lock); + list_del_rcu(&server->client_link); + if (list_empty(&clp->cl_superblocks)) + set_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state); + list_del(&server->master_link); + spin_unlock(&nn->nfs_client_lock); + + synchronize_rcu(); +} +EXPORT_SYMBOL_GPL(nfs_server_remove_lists); + +static DEFINE_IDA(s_sysfs_ids); + +/* + * Allocate and initialise a server record + */ +struct nfs_server *nfs_alloc_server(void) +{ + struct nfs_server *server; + + server = kzalloc_obj(struct nfs_server); + if (!server) + return NULL; + + server->s_sysfs_id = ida_alloc(&s_sysfs_ids, GFP_KERNEL); + if (server->s_sysfs_id < 0) + goto free_server; + + server->client = server->client_acl = ERR_PTR(-EINVAL); + + /* Zero out the NFS state stuff */ + INIT_LIST_HEAD(&server->client_link); + INIT_LIST_HEAD(&server->master_link); + INIT_LIST_HEAD(&server->delegations); + spin_lock_init(&server->delegations_lock); + INIT_LIST_HEAD(&server->delegations_return); + INIT_LIST_HEAD(&server->delegations_lru); + INIT_LIST_HEAD(&server->delegations_delayed); + INIT_LIST_HEAD(&server->layouts); + INIT_LIST_HEAD(&server->state_owners_lru); + INIT_LIST_HEAD(&server->ss_copies); + INIT_LIST_HEAD(&server->ss_src_copies); + + atomic_set(&server->active, 0); + atomic_long_set(&server->nr_active_delegations, 0); + + server->io_stats = nfs_alloc_iostats(); + if (!server->io_stats) { + ida_free(&s_sysfs_ids, server->s_sysfs_id); + goto free_server; + } + + server->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED; + + init_waitqueue_head(&server->write_congestion_wait); + atomic_long_set(&server->writeback, 0); + + atomic64_set(&server->owner_ctr, 0); + + pnfs_init_server(server); + rpc_init_wait_queue(&server->uoc_rpcwaitq, "NFS UOC"); + + return server; + +free_server: + kfree(server); + return NULL; +} +EXPORT_SYMBOL_GPL(nfs_alloc_server); + +static void delayed_free(struct rcu_head *p) +{ + struct nfs_server *server = container_of(p, struct nfs_server, rcu); + + nfs_free_iostats(server->io_stats); + kfree(server); +} + +/* + * Free up a server record + */ +void nfs_free_server(struct nfs_server *server) +{ + nfs_server_remove_lists(server); + + if (server->destroy != NULL) + server->destroy(server); + + if (!IS_ERR(server->client_acl)) + rpc_shutdown_client(server->client_acl); + if (!IS_ERR(server->client)) + rpc_shutdown_client(server->client); + + nfs_put_client(server->nfs_client); + + if (server->kobj.state_initialized) { + nfs_sysfs_remove_server(server); + kobject_put(&server->kobj); + } + ida_free(&s_sysfs_ids, server->s_sysfs_id); + + put_cred(server->cred); + nfs_release_automount_timer(); + call_rcu(&server->rcu, delayed_free); +} +EXPORT_SYMBOL_GPL(nfs_free_server); + +/* + * Create a version 2 or 3 volume record + * - keyed on server and FSID + */ +struct nfs_server *nfs_create_server(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_server *server; + struct nfs_fattr *fattr; + int error; + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + server->cred = get_cred(fc->cred); + + error = -ENOMEM; + fattr = nfs_alloc_fattr(); + if (fattr == NULL) + goto error; + + /* Get a client representation */ + error = nfs_init_server(server, fc); + if (error < 0) + goto error; + + /* Probe the root fh to retrieve its FSID */ + error = nfs_probe_fsinfo(server, ctx->mntfh, fattr); + if (error < 0) + goto error; + if (server->nfs_client->rpc_ops->version == 3) { + if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) + server->namelen = NFS3_MAXNAMLEN; + if (!(ctx->flags & NFS_MOUNT_NORDIRPLUS)) + server->caps |= NFS_CAP_READDIRPLUS; + } else { + if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) + server->namelen = NFS2_MAXNAMLEN; + } + /* Linux 'subtree_check' borkenness mandates this setting */ + server->fh_expire_type = NFS_FH_VOL_RENAME; + + if (!(fattr->valid & NFS_ATTR_FATTR)) { + error = ctx->nfs_mod->rpc_ops->getattr(server, ctx->mntfh, + fattr, NULL); + if (error < 0) { + dprintk("nfs_create_server: getattr error = %d\n", -error); + goto error; + } + } + memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid)); + + dprintk("Server FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + nfs_server_insert_lists(server); + server->mount_time = jiffies; + nfs_free_fattr(fattr); + return server; + +error: + nfs_free_fattr(fattr); + nfs_free_server(server); + return ERR_PTR(error); +} +EXPORT_SYMBOL_GPL(nfs_create_server); + +/* + * Clone an NFS2, NFS3 or NFS4 server record + */ +struct nfs_server *nfs_clone_server(struct nfs_server *source, + struct nfs_fh *fh, + struct nfs_fattr *fattr, + rpc_authflavor_t flavor) +{ + struct nfs_server *server; + int error; + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + server->cred = get_cred(source->cred); + + /* Copy data from the source */ + server->nfs_client = source->nfs_client; + server->destroy = source->destroy; + refcount_inc(&server->nfs_client->cl_count); + nfs_server_copy_userdata(server, source); + + server->fsid = fattr->fsid; + + nfs_sysfs_add_server(server); + + nfs_sysfs_link_rpc_client(server, + server->nfs_client->cl_rpcclient, "_state"); + + error = nfs_init_server_rpcclient(server, + source->client->cl_timeout, + flavor); + if (error < 0) + goto out_free_server; + + nfs_server_set_init_caps(server); + + /* probe the filesystem info for this server filesystem */ + error = nfs_probe_server(server, fh); + if (error < 0) + goto out_free_server; + + if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN) + server->namelen = NFS4_MAXNAMLEN; + + error = nfs_start_lockd(server); + if (error < 0) + goto out_free_server; + + nfs_server_insert_lists(server); + server->mount_time = jiffies; + + return server; + +out_free_server: + nfs_free_server(server); + return ERR_PTR(error); +} +EXPORT_SYMBOL_GPL(nfs_clone_server); + +void nfs_clients_init(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + INIT_LIST_HEAD(&nn->nfs_client_list); + INIT_LIST_HEAD(&nn->nfs_volume_list); +#if IS_ENABLED(CONFIG_NFS_V4) + idr_init(&nn->cb_ident_idr); + INIT_LIST_HEAD(&nn->nfs4_data_server_cache); + spin_lock_init(&nn->nfs4_data_server_lock); +#endif /* CONFIG_NFS_V4 */ + spin_lock_init(&nn->nfs_client_lock); + nn->boot_time = ktime_get_real(); + memset(&nn->rpcstats, 0, sizeof(nn->rpcstats)); + nn->rpcstats.program = &nfs_program; + + nfs_netns_sysfs_setup(nn, net); +} + +void nfs_clients_exit(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + nfs_netns_sysfs_destroy(nn); + nfs_cleanup_cb_ident_idr(net); + WARN_ON_ONCE(!list_empty(&nn->nfs_client_list)); + WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list)); +#if IS_ENABLED(CONFIG_NFS_V4) + WARN_ON_ONCE(!list_empty(&nn->nfs4_data_server_cache)); +#endif /* CONFIG_NFS_V4 */ +} + +#ifdef CONFIG_PROC_FS +static void *nfs_server_list_start(struct seq_file *p, loff_t *pos); +static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos); +static void nfs_server_list_stop(struct seq_file *p, void *v); +static int nfs_server_list_show(struct seq_file *m, void *v); + +static const struct seq_operations nfs_server_list_ops = { + .start = nfs_server_list_start, + .next = nfs_server_list_next, + .stop = nfs_server_list_stop, + .show = nfs_server_list_show, +}; + +static void *nfs_volume_list_start(struct seq_file *p, loff_t *pos); +static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos); +static void nfs_volume_list_stop(struct seq_file *p, void *v); +static int nfs_volume_list_show(struct seq_file *m, void *v); + +static const struct seq_operations nfs_volume_list_ops = { + .start = nfs_volume_list_start, + .next = nfs_volume_list_next, + .stop = nfs_volume_list_stop, + .show = nfs_volume_list_show, +}; + +/* + * set up the iterator to start reading from the server list and return the first item + */ +static void *nfs_server_list_start(struct seq_file *m, loff_t *_pos) + __acquires(&nn->nfs_client_lock) +{ + struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id); + + /* lock the list against modification */ + spin_lock(&nn->nfs_client_lock); + return seq_list_start_head(&nn->nfs_client_list, *_pos); +} + +/* + * move to next server + */ +static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos) +{ + struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id); + + return seq_list_next(v, &nn->nfs_client_list, pos); +} + +/* + * clean up after reading from the transports list + */ +static void nfs_server_list_stop(struct seq_file *p, void *v) + __releases(&nn->nfs_client_lock) +{ + struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id); + + spin_unlock(&nn->nfs_client_lock); +} + +/* + * display a header line followed by a load of call lines + */ +static int nfs_server_list_show(struct seq_file *m, void *v) +{ + struct nfs_client *clp; + struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id); + + /* display header on line 1 */ + if (v == &nn->nfs_client_list) { + seq_puts(m, "NV SERVER PORT USE HOSTNAME\n"); + return 0; + } + + /* display one transport per line on subsequent lines */ + clp = list_entry(v, struct nfs_client, cl_share_link); + + /* Check if the client is initialized */ + if (clp->cl_cons_state != NFS_CS_READY) + return 0; + + rcu_read_lock(); + seq_printf(m, "v%u %s %s %3d %s\n", + clp->rpc_ops->version, + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_PORT), + refcount_read(&clp->cl_count), + clp->cl_hostname); + rcu_read_unlock(); + + return 0; +} + +/* + * set up the iterator to start reading from the volume list and return the first item + */ +static void *nfs_volume_list_start(struct seq_file *m, loff_t *_pos) + __acquires(&nn->nfs_client_lock) +{ + struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id); + + /* lock the list against modification */ + spin_lock(&nn->nfs_client_lock); + return seq_list_start_head(&nn->nfs_volume_list, *_pos); +} + +/* + * move to next volume + */ +static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos) +{ + struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id); + + return seq_list_next(v, &nn->nfs_volume_list, pos); +} + +/* + * clean up after reading from the transports list + */ +static void nfs_volume_list_stop(struct seq_file *p, void *v) + __releases(&nn->nfs_client_lock) +{ + struct nfs_net *nn = net_generic(seq_file_net(p), nfs_net_id); + + spin_unlock(&nn->nfs_client_lock); +} + +/* + * display a header line followed by a load of call lines + */ +static int nfs_volume_list_show(struct seq_file *m, void *v) +{ + struct nfs_server *server; + struct nfs_client *clp; + char dev[13]; // 8 for 2^24, 1 for ':', 3 for 2^8, 1 for '\0' + char fsid[34]; // 2 * 16 for %llx, 1 for ':', 1 for '\0' + struct nfs_net *nn = net_generic(seq_file_net(m), nfs_net_id); + + /* display header on line 1 */ + if (v == &nn->nfs_volume_list) { + seq_puts(m, "NV SERVER PORT DEV FSID" + " FSC\n"); + return 0; + } + /* display one transport per line on subsequent lines */ + server = list_entry(v, struct nfs_server, master_link); + clp = server->nfs_client; + + snprintf(dev, sizeof(dev), "%u:%u", + MAJOR(server->s_dev), MINOR(server->s_dev)); + + snprintf(fsid, sizeof(fsid), "%llx:%llx", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + rcu_read_lock(); + seq_printf(m, "v%u %s %s %-12s %-33s %s\n", + clp->rpc_ops->version, + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_HEX_PORT), + dev, + fsid, + nfs_server_fscache_state(server)); + rcu_read_unlock(); + + return 0; +} + +int nfs_fs_proc_net_init(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + struct proc_dir_entry *p; + + nn->proc_nfsfs = proc_net_mkdir(net, "nfsfs", net->proc_net); + if (!nn->proc_nfsfs) + goto error_0; + + /* a file of servers with which we're dealing */ + p = proc_create_net("servers", S_IFREG|S_IRUGO, nn->proc_nfsfs, + &nfs_server_list_ops, sizeof(struct seq_net_private)); + if (!p) + goto error_1; + + /* a file of volumes that we have mounted */ + p = proc_create_net("volumes", S_IFREG|S_IRUGO, nn->proc_nfsfs, + &nfs_volume_list_ops, sizeof(struct seq_net_private)); + if (!p) + goto error_1; + return 0; + +error_1: + remove_proc_subtree("nfsfs", net->proc_net); +error_0: + return -ENOMEM; +} + +void nfs_fs_proc_net_exit(struct net *net) +{ + remove_proc_subtree("nfsfs", net->proc_net); +} + +/* + * initialise the /proc/fs/nfsfs/ directory + */ +int __init nfs_fs_proc_init(void) +{ + if (!proc_mkdir("fs/nfsfs", NULL)) + goto error_0; + + /* a file of servers with which we're dealing */ + if (!proc_symlink("fs/nfsfs/servers", NULL, "../../net/nfsfs/servers")) + goto error_1; + + /* a file of volumes that we have mounted */ + if (!proc_symlink("fs/nfsfs/volumes", NULL, "../../net/nfsfs/volumes")) + goto error_1; + + return 0; +error_1: + remove_proc_subtree("fs/nfsfs", NULL); +error_0: + return -ENOMEM; +} + +/* + * clean up the /proc/fs/nfsfs/ directory + */ +void nfs_fs_proc_exit(void) +{ + remove_proc_subtree("fs/nfsfs", NULL); + ida_destroy(&s_sysfs_ids); +} + +#endif /* CONFIG_PROC_FS */ diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c new file mode 100644 index 000000000..9546d2195 --- /dev/null +++ b/fs/nfs/delegation.c @@ -0,0 +1,1619 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/delegation.c + * + * Copyright (C) 2004 Trond Myklebust + * + * NFS file delegation management + * + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "nfs4_fs.h" +#include "nfs4session.h" +#include "delegation.h" +#include "internal.h" +#include "nfs4trace.h" + +#define NFS_DEFAULT_DELEGATION_WATERMARK (5000U) + +static unsigned nfs_delegation_watermark = NFS_DEFAULT_DELEGATION_WATERMARK; +module_param_named(delegation_watermark, nfs_delegation_watermark, uint, 0644); + +bool directory_delegations = true; +module_param(directory_delegations, bool, 0644); +MODULE_PARM_DESC(directory_delegations, + "Enable the use of directory delegations, defaults to on."); + +static struct hlist_head *nfs_delegation_hash(struct nfs_server *server, + const struct nfs_fh *fhandle) +{ + return server->delegation_hash_table + + (nfs_fhandle_hash(fhandle) & server->delegation_hash_mask); +} + +static void __nfs_free_delegation(struct nfs_delegation *delegation) +{ + put_cred(delegation->cred); + delegation->cred = NULL; + kfree_rcu(delegation, rcu); +} + +static void nfs_mark_delegation_revoked(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + bool put_ref = false; + + if (test_and_set_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) + return; + + delegation->stateid.type = NFS4_INVALID_STATEID_TYPE; + atomic_long_dec(&server->nr_active_delegations); + if (!test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) + nfs_clear_verifier_delegated(delegation->inode); + + spin_lock(&server->delegations_lock); + if (!list_empty(&delegation->entry)) { + list_del_init(&delegation->entry); + put_ref = true; + } + spin_unlock(&server->delegations_lock); + + if (put_ref) + nfs_put_delegation(delegation); +} + +void nfs_put_delegation(struct nfs_delegation *delegation) +{ + if (refcount_dec_and_test(&delegation->refcount)) + __nfs_free_delegation(delegation); +} + +/** + * nfs_mark_delegation_referenced - set delegation's REFERENCED flag + * @delegation: delegation to process + * + */ +void nfs_mark_delegation_referenced(struct nfs_delegation *delegation) +{ + set_bit(NFS_DELEGATION_REFERENCED, &delegation->flags); +} + +static void nfs_mark_return_delegation(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + spin_lock(&server->delegations_lock); + if (list_empty(&delegation->entry)) + refcount_inc(&delegation->refcount); + list_move_tail(&delegation->entry, &server->delegations_return); + spin_unlock(&server->delegations_lock); + + set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state); +} + +static bool nfs4_is_valid_delegation(const struct nfs_delegation *delegation, + fmode_t type) +{ + if (delegation != NULL && (delegation->type & type) == type && + !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) && + !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) + return true; + return false; +} + +struct nfs_delegation *nfs4_get_valid_delegation(const struct inode *inode) +{ + struct nfs_delegation *delegation; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (!nfs4_is_valid_delegation(delegation, 0) || + !refcount_inc_not_zero(&delegation->refcount)) + delegation = NULL; + rcu_read_unlock(); + + return delegation; +} + +static int nfs4_do_check_delegation(struct inode *inode, fmode_t type, + int flags, bool mark) +{ + struct nfs_delegation *delegation; + int ret = 0; + + type &= FMODE_READ|FMODE_WRITE; + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (nfs4_is_valid_delegation(delegation, type)) { + if (mark) + nfs_mark_delegation_referenced(delegation); + ret = 1; + if ((flags & NFS_DELEGATION_FLAG_TIME) && + !test_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags)) + ret = 0; + } + rcu_read_unlock(); + return ret; +} +/** + * nfs4_have_delegation - check if inode has a delegation, mark it + * NFS_DELEGATION_REFERENCED if there is one. + * @inode: inode to check + * @type: delegation types to check for + * @flags: various modifiers + * + * Returns one if inode has the indicated delegation, otherwise zero. + */ +int nfs4_have_delegation(struct inode *inode, fmode_t type, int flags) +{ + if (S_ISDIR(inode->i_mode) && !directory_delegations) + nfs4_inode_set_return_delegation_on_close(inode); + return nfs4_do_check_delegation(inode, type, flags, true); +} + +/* + * nfs4_check_delegation - check if inode has a delegation, do not mark + * NFS_DELEGATION_REFERENCED if it has one. + */ +int nfs4_check_delegation(struct inode *inode, fmode_t type) +{ + return nfs4_do_check_delegation(inode, type, 0, false); +} + +static int nfs_delegation_claim_locks(struct nfs4_state *state, const nfs4_stateid *stateid) +{ + struct inode *inode = state->inode; + struct nfs_inode *nfsi = NFS_I(inode); + struct file_lock *fl; + struct file_lock_context *flctx = locks_inode_context(inode); + struct list_head *list; + int status = 0; + + if (flctx == NULL) + goto out; + + list = &flctx->flc_posix; + + /* Guard against reclaim and new lock/unlock calls */ + down_write(&nfsi->rwsem); + spin_lock(&flctx->flc_lock); +restart: + for_each_file_lock(fl, list) { + if (nfs_file_open_context(fl->c.flc_file)->state != state) + continue; + spin_unlock(&flctx->flc_lock); + status = nfs4_lock_delegation_recall(fl, state, stateid); + if (status < 0) { + up_write(&nfsi->rwsem); + goto out; + } + spin_lock(&flctx->flc_lock); + } + if (list == &flctx->flc_posix) { + list = &flctx->flc_flock; + goto restart; + } + spin_unlock(&flctx->flc_lock); + up_write(&nfsi->rwsem); +out: + return status; +} + +static int nfs_delegation_claim_opens(struct inode *inode, + const nfs4_stateid *stateid, fmode_t type) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_open_context *ctx; + struct nfs4_state_owner *sp; + struct nfs4_state *state; + int err; + +again: + rcu_read_lock(); + list_for_each_entry_rcu(ctx, &nfsi->open_files, list) { + state = ctx->state; + if (state == NULL) + continue; + if (!test_bit(NFS_DELEGATED_STATE, &state->flags)) + continue; + if (!nfs4_valid_open_stateid(state)) + continue; + if (!nfs4_stateid_match(&state->stateid, stateid)) + continue; + if (!get_nfs_open_context(ctx)) + continue; + rcu_read_unlock(); + sp = state->owner; + /* Block nfs4_proc_unlck */ + mutex_lock(&sp->so_delegreturn_mutex); + err = nfs4_open_delegation_recall(ctx, state, stateid); + if (!err) + err = nfs_delegation_claim_locks(state, stateid); + mutex_unlock(&sp->so_delegreturn_mutex); + put_nfs_open_context(ctx); + if (err != 0) + return err; + goto again; + } + rcu_read_unlock(); + return 0; +} + +/** + * nfs_inode_reclaim_delegation - process a delegation reclaim request + * @inode: inode to process + * @cred: credential to use for request + * @type: delegation type + * @stateid: delegation stateid + * @pagemod_limit: write delegation "space_limit" + * @deleg_type: raw delegation type + * + */ +void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred, + fmode_t type, const nfs4_stateid *stateid, + unsigned long pagemod_limit, u32 deleg_type) +{ + struct nfs_delegation *delegation; + const struct cred *oldcred = NULL; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (!delegation) { + rcu_read_unlock(); + nfs_inode_set_delegation(inode, cred, type, stateid, + pagemod_limit, deleg_type); + return; + } + + spin_lock(&delegation->lock); + nfs4_stateid_copy(&delegation->stateid, stateid); + delegation->type = type; + delegation->pagemod_limit = pagemod_limit; + oldcred = delegation->cred; + delegation->cred = get_cred(cred); + switch (deleg_type) { + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + set_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags); + break; + default: + clear_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags); + } + clear_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags); + if (test_and_clear_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) + atomic_long_inc(&NFS_SERVER(inode)->nr_active_delegations); + spin_unlock(&delegation->lock); + rcu_read_unlock(); + put_cred(oldcred); + trace_nfs4_reclaim_delegation(inode, type); +} + +static int nfs_do_return_delegation(struct inode *inode, + struct nfs_delegation *delegation, + int issync) +{ + const struct cred *cred; + int res = 0; + + if (!test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) { + spin_lock(&delegation->lock); + cred = get_cred(delegation->cred); + spin_unlock(&delegation->lock); + res = nfs4_proc_delegreturn(inode, cred, &delegation->stateid, + delegation, issync); + put_cred(cred); + } + return res; +} + +static struct inode *nfs_delegation_grab_inode(struct nfs_delegation *delegation) +{ + struct inode *inode = NULL; + + spin_lock(&delegation->lock); + if (delegation->inode != NULL) + inode = igrab(delegation->inode); + spin_unlock(&delegation->lock); + return inode; +} + +static struct nfs_delegation * +nfs_start_delegation_return(struct nfs_inode *nfsi) +{ + struct nfs_delegation *delegation; + bool return_now = false; + + rcu_read_lock(); + delegation = rcu_dereference(nfsi->delegation); + if (!delegation || !refcount_inc_not_zero(&delegation->refcount)) { + rcu_read_unlock(); + return NULL; + } + rcu_read_unlock(); + + spin_lock(&delegation->lock); + if (delegation->inode && + !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) + return_now = true; + spin_unlock(&delegation->lock); + + if (!return_now) { + nfs_put_delegation(delegation); + return NULL; + } + nfs_clear_verifier_delegated(&nfsi->vfs_inode); + return delegation; +} + +static bool +nfs_detach_delegations_locked(struct nfs_inode *nfsi, + struct nfs_delegation *delegation, + struct nfs_client *clp) +{ + lockdep_assert_held(&clp->cl_lock); + + trace_nfs4_detach_delegation(&nfsi->vfs_inode, delegation->type); + + spin_lock(&delegation->lock); + if (!delegation->inode) { + spin_unlock(&delegation->lock); + return false; + } + hlist_del_init_rcu(&delegation->hash); + list_del_rcu(&delegation->super_list); + delegation->inode = NULL; + rcu_assign_pointer(nfsi->delegation, NULL); + spin_unlock(&delegation->lock); + clear_bit(NFS_INO_REQ_DIR_DELEG, &nfsi->flags); + return true; +} + +static bool nfs_detach_delegation(struct nfs_inode *nfsi, + struct nfs_delegation *delegation, + struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs_delegation *deleg_cur; + bool ret = false; + + spin_lock(&clp->cl_lock); + deleg_cur = rcu_dereference_protected(nfsi->delegation, + lockdep_is_held(&clp->cl_lock)); + if (delegation == deleg_cur) + ret = nfs_detach_delegations_locked(nfsi, delegation, clp); + spin_unlock(&clp->cl_lock); + return ret; +} + +static void +nfs_update_delegation_cred(struct nfs_delegation *delegation, + const struct cred *cred) +{ + const struct cred *old; + + if (cred_fscmp(delegation->cred, cred) != 0) { + old = xchg(&delegation->cred, get_cred(cred)); + put_cred(old); + } +} + +static void +nfs_update_inplace_delegation(struct nfs_server *server, + struct nfs_delegation *delegation, + const struct nfs_delegation *update) +{ + if (nfs4_stateid_is_newer(&update->stateid, &delegation->stateid)) { + delegation->stateid.seqid = update->stateid.seqid; + smp_wmb(); + delegation->type = update->type; + delegation->pagemod_limit = update->pagemod_limit; + if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) { + delegation->change_attr = update->change_attr; + nfs_update_delegation_cred(delegation, update->cred); + /* smp_mb__before_atomic() is implicit due to xchg() */ + clear_bit(NFS_DELEGATION_REVOKED, &delegation->flags); + atomic_long_inc(&server->nr_active_delegations); + } + } +} + +/** + * nfs_inode_set_delegation - set up a delegation on an inode + * @inode: inode to which delegation applies + * @cred: cred to use for subsequent delegation processing + * @type: delegation type + * @stateid: delegation stateid + * @pagemod_limit: write delegation "space_limit" + * @deleg_type: raw delegation type + * + * Returns zero on success, or a negative errno value. + */ +int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred, + fmode_t type, const nfs4_stateid *stateid, + unsigned long pagemod_limit, u32 deleg_type) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_client *clp = server->nfs_client; + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_delegation *delegation, *old_delegation; + struct nfs_delegation *freeme = NULL; + int status = 0; + + delegation = kmalloc_obj(*delegation, GFP_KERNEL_ACCOUNT); + if (delegation == NULL) + return -ENOMEM; + nfs4_stateid_copy(&delegation->stateid, stateid); + refcount_set(&delegation->refcount, 1); + delegation->type = type; + delegation->pagemod_limit = pagemod_limit; + delegation->change_attr = inode_peek_iversion_raw(inode); + delegation->cred = get_cred(cred); + delegation->inode = inode; + delegation->flags = 1<entry); + switch (deleg_type) { + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + delegation->flags |= BIT(NFS_DELEGATION_DELEGTIME); + } + delegation->test_gen = 0; + spin_lock_init(&delegation->lock); + + spin_lock(&clp->cl_lock); + old_delegation = rcu_dereference_protected(nfsi->delegation, + lockdep_is_held(&clp->cl_lock)); + if (old_delegation == NULL) + goto add_new; + /* Is this an update of the existing delegation? */ + if (nfs4_stateid_match_other(&old_delegation->stateid, + &delegation->stateid)) { + spin_lock(&old_delegation->lock); + nfs_update_inplace_delegation(server, old_delegation, + delegation); + spin_unlock(&old_delegation->lock); + goto out; + } + if (!test_bit(NFS_DELEGATION_REVOKED, &old_delegation->flags)) { + /* + * Deal with broken servers that hand out two + * delegations for the same file. + * Allow for upgrades to a WRITE delegation, but + * nothing else. + */ + dfprintk(FILE, "%s: server %s handed out " + "a duplicate delegation!\n", + __func__, clp->cl_hostname); + if (delegation->type == old_delegation->type || + !(delegation->type & FMODE_WRITE)) { + freeme = delegation; + delegation = NULL; + goto out; + } + if (test_and_set_bit(NFS_DELEGATION_RETURNING, + &old_delegation->flags)) + goto out; + } + if (!nfs_detach_delegations_locked(nfsi, old_delegation, clp)) + goto out; + freeme = old_delegation; +add_new: + /* + * If we didn't revalidate the change attribute before setting + * the delegation, then pre-emptively ask for a full attribute + * cache revalidation. + */ + spin_lock(&inode->i_lock); + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_CHANGE) + nfs_set_cache_invalid(inode, + NFS_INO_INVALID_ATIME | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_MTIME | NFS_INO_INVALID_SIZE | + NFS_INO_INVALID_BLOCKS | NFS_INO_INVALID_NLINK | + NFS_INO_INVALID_OTHER | NFS_INO_INVALID_DATA | + NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL | + NFS_INO_INVALID_XATTR); + spin_unlock(&inode->i_lock); + + list_add_tail_rcu(&delegation->super_list, &server->delegations); + hlist_add_head_rcu(&delegation->hash, + nfs_delegation_hash(server, &NFS_I(inode)->fh)); + rcu_assign_pointer(nfsi->delegation, delegation); + delegation = NULL; + + atomic_long_inc(&server->nr_active_delegations); + + trace_nfs4_set_delegation(inode, type); + + /* If we hold writebacks and have delegated mtime then update */ + if (deleg_type == NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG && + nfs_have_writebacks(inode)) + nfs_update_delegated_mtime(inode); +out: + spin_unlock(&clp->cl_lock); + if (delegation != NULL) + __nfs_free_delegation(delegation); + if (freeme != NULL) { + nfs_do_return_delegation(inode, freeme, 0); + nfs_mark_delegation_revoked(server, freeme); + nfs_put_delegation(freeme); + } + return status; +} + +/* + * Basic procedure for returning a delegation to the server. + * If @issync is set, wait until state recovery has finished. Otherwise + * return -EAGAIN to the caller if we need more time. + */ +static int nfs_end_delegation_return(struct inode *inode, + struct nfs_delegation *delegation, bool issync) +{ + struct nfs_server *server = NFS_SERVER(inode); + unsigned int mode = O_WRONLY | O_RDWR; + int err = 0; + + /* Directory delegations don't require any state recovery */ + if (!S_ISREG(inode->i_mode)) + goto out_return; + + if (!issync) + mode |= O_NONBLOCK; + /* Recall of any remaining application leases */ + err = break_lease(inode, mode); + + while (err == 0) { + if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) + break; + err = nfs_delegation_claim_opens(inode, &delegation->stateid, + delegation->type); + if (!err) + break; + if (err != -EAGAIN) + goto abort; + if (!issync) + goto delay; + + /* + * Guard against state recovery + */ + err = nfs4_wait_clnt_recover(server->nfs_client); + } + +out_return: + return nfs_do_return_delegation(inode, delegation, issync); +delay: + spin_lock(&server->delegations_lock); + if (list_empty(&delegation->entry)) + refcount_inc(&delegation->refcount); + list_move_tail(&delegation->entry, &server->delegations_return); + spin_unlock(&server->delegations_lock); + set_bit(NFS4CLNT_DELEGRETURN_DELAYED, &server->nfs_client->cl_state); +abort: + clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags); + return err; +} + +static int nfs_return_one_delegation(struct nfs_server *server) +{ + struct nfs_delegation *delegation; + struct inode *inode; + int err = 0; + + spin_lock(&server->delegations_lock); + delegation = list_first_entry_or_null(&server->delegations_return, + struct nfs_delegation, entry); + if (!delegation) { + spin_unlock(&server->delegations_lock); + return 0; /* no more delegations */ + } + list_del_init(&delegation->entry); + spin_unlock(&server->delegations_lock); + + spin_lock(&delegation->lock); + inode = delegation->inode; + if (!inode || !igrab(inode)) { + spin_unlock(&delegation->lock); + goto out_put_delegation; + } + if (test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) || + test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { + spin_unlock(&delegation->lock); + goto out_put_inode; + } + spin_unlock(&delegation->lock); + + nfs_clear_verifier_delegated(inode); + + err = nfs_end_delegation_return(inode, delegation, false); + +out_put_inode: + iput(inode); +out_put_delegation: + nfs_put_delegation(delegation); + if (err) + return err; + return 1; /* keep going */ +} + +static int nfs_server_return_marked_delegations(struct nfs_server *server, + void __always_unused *data) +{ + int err; + + while ((err = nfs_return_one_delegation(server)) > 0) + cond_resched(); + return err; +} + +static inline bool nfs_delegations_over_limit(struct nfs_server *server) +{ + return !list_empty_careful(&server->delegations_lru) && + atomic_long_read(&server->nr_active_delegations) > + nfs_delegation_watermark; +} + +static void nfs_delegations_return_from_lru(struct nfs_server *server) +{ + struct nfs_delegation *d, *n; + unsigned int pass = 0; + bool moved = false; + +retry: + spin_lock(&server->delegations_lock); + list_for_each_entry_safe(d, n, &server->delegations_lru, entry) { + if (!nfs_delegations_over_limit(server)) + break; + if (pass == 0 && test_bit(NFS_DELEGATION_REFERENCED, &d->flags)) + continue; + list_move_tail(&d->entry, &server->delegations_return); + moved = true; + } + spin_unlock(&server->delegations_lock); + + /* + * If we are still over the limit, try to reclaim referenced delegations + * as well. + */ + if (pass == 0 && nfs_delegations_over_limit(server)) { + pass++; + goto retry; + } + + if (moved) { + set_bit(NFS4CLNT_DELEGRETURN, &server->nfs_client->cl_state); + nfs4_schedule_state_manager(server->nfs_client); + } +} + +static void nfs_delegation_add_lru(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + spin_lock(&server->delegations_lock); + if (list_empty(&delegation->entry)) { + list_add_tail(&delegation->entry, &server->delegations_lru); + refcount_inc(&delegation->refcount); + } + spin_unlock(&server->delegations_lock); + + if (nfs_delegations_over_limit(server)) + nfs_delegations_return_from_lru(server); +} + +static bool nfs_server_clear_delayed_delegations(struct nfs_server *server) +{ + bool ret = false; + + if (list_empty_careful(&server->delegations_delayed)) + return false; + + spin_lock(&server->delegations_lock); + if (!list_empty(&server->delegations_delayed)) { + list_splice_tail_init(&server->delegations_delayed, + &server->delegations_return); + ret = true; + } + spin_unlock(&server->delegations_lock); + + return ret; +} + +static bool nfs_client_clear_delayed_delegations(struct nfs_client *clp) +{ + struct nfs_server *server; + bool ret = false; + + if (!test_and_clear_bit(NFS4CLNT_DELEGRETURN_DELAYED, &clp->cl_state)) + return false; + + rcu_read_lock(); + list_for_each_entry_rcu (server, &clp->cl_superblocks, client_link) { + if (nfs_server_clear_delayed_delegations(server)) + ret = true; + } + rcu_read_unlock(); + + if (ret) + set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state); + return ret; +} + +/** + * nfs_client_return_marked_delegations - return previously marked delegations + * @clp: nfs_client to process + * + * Note that this function is designed to be called by the state + * manager thread. For this reason, it cannot flush the dirty data, + * since that could deadlock in case of a state recovery error. + * + * Returns zero on success, or a negative errno value. + */ +int nfs_client_return_marked_delegations(struct nfs_client *clp) +{ + int err = nfs_client_for_each_server( + clp, nfs_server_return_marked_delegations, NULL); + if (err) + return err; + /* If a return was delayed, sleep to prevent hard looping */ + if (nfs_client_clear_delayed_delegations(clp)) + ssleep(1); + return 0; +} + +/** + * nfs_inode_evict_delegation - return delegation, don't reclaim opens + * @inode: inode to process + * + * Does not protect against delegation reclaims, therefore really only safe + * to be called from nfs4_clear_inode(). Guaranteed to always free + * the delegation structure. + */ +void nfs_inode_evict_delegation(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_delegation *delegation; + + rcu_read_lock(); + delegation = rcu_dereference(nfsi->delegation); + if (delegation && !nfs_detach_delegation(nfsi, delegation, server)) + delegation = NULL; + rcu_read_unlock(); + + if (!delegation) + return; + + set_bit(NFS_DELEGATION_RETURNING, &delegation->flags); + nfs_do_return_delegation(inode, delegation, 1); + nfs_mark_delegation_revoked(server, delegation); + nfs_put_delegation(delegation); +} + +/** + * nfs4_inode_return_delegation - synchronously return a delegation + * @inode: inode to process + * + * This routine will always flush any dirty data to disk on the + * assumption that if we need to return the delegation, then + * we should stop caching. + * + * Returns zero on success, or a negative errno value. + */ +void nfs4_inode_return_delegation(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_delegation *delegation; + + delegation = nfs_start_delegation_return(nfsi); + if (!delegation) + return; + + /* Synchronous recall of any application leases */ + break_lease(inode, O_WRONLY | O_RDWR); + if (S_ISREG(inode->i_mode)) + nfs_wb_all(inode); + nfs_end_delegation_return(inode, delegation, true); + nfs_put_delegation(delegation); +} + +/** + * nfs4_inode_set_return_delegation_on_close - asynchronously return a delegation + * @inode: inode to process + * + * This routine is called to request that the delegation be returned as soon + * as the file is closed. If the file is already closed, the delegation is + * immediately returned. + */ +void nfs4_inode_set_return_delegation_on_close(struct inode *inode) +{ + struct nfs_delegation *delegation; + bool return_now = false; + + if (!inode) + return; + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return; + + spin_lock(&delegation->lock); + if (!delegation->inode) + goto out_unlock; + if (list_empty(&NFS_I(inode)->open_files) && + !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) + return_now = true; + else + set_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags); +out_unlock: + spin_unlock(&delegation->lock); + if (return_now) { + nfs_clear_verifier_delegated(inode); + nfs_end_delegation_return(inode, delegation, false); + } + nfs_put_delegation(delegation); +} + +/** + * nfs4_inode_return_delegation_on_close - asynchronously return a delegation + * @inode: inode to process + * + * This routine is called on file close in order to determine if the + * inode delegation needs to be returned immediately. + */ +void nfs4_inode_return_delegation_on_close(struct inode *inode) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_delegation *delegation; + bool return_now = false; + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return; + + if (test_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags)) { + spin_lock(&delegation->lock); + if (delegation->inode && + list_empty(&NFS_I(inode)->open_files) && + !test_and_set_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { + clear_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags); + return_now = true; + } + spin_unlock(&delegation->lock); + } + + if (return_now) { + nfs_clear_verifier_delegated(inode); + nfs_end_delegation_return(inode, delegation, false); + } else { + nfs_delegation_add_lru(server, delegation); + } + nfs_put_delegation(delegation); +} + +/** + * nfs4_inode_make_writeable + * @inode: pointer to inode + * + * Make the inode writeable by returning the delegation if necessary + */ +void nfs4_inode_make_writeable(struct inode *inode) +{ + struct nfs_delegation *delegation; + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return; + + if (!nfs4_has_session(NFS_SERVER(inode)->nfs_client) || + !(delegation->type & FMODE_WRITE)) + nfs4_inode_return_delegation(inode); + nfs_put_delegation(delegation); +} + +static void +nfs_mark_return_if_closed_delegation(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + struct inode *inode; + + if (!list_empty_careful(&server->delegations_return) || + test_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags)) + return; + spin_lock(&delegation->lock); + inode = delegation->inode; + if (!inode) + goto out; + if (list_empty(&NFS_I(inode)->open_files)) + nfs_mark_return_delegation(server, delegation); + else + set_bit(NFS_DELEGATION_RETURN_IF_CLOSED, &delegation->flags); +out: + spin_unlock(&delegation->lock); +} + +static bool nfs_server_mark_return_all_delegations(struct nfs_server *server) +{ + struct nfs_delegation *delegation; + bool ret = false; + + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + nfs_mark_return_delegation(server, delegation); + ret = true; + } + return ret; +} + +static void nfs_delegation_run_state_manager(struct nfs_client *clp) +{ + if (test_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) + nfs4_schedule_state_manager(clp); +} + +/** + * nfs_expire_all_delegations + * @clp: client to process + * + */ +void nfs_expire_all_delegations(struct nfs_client *clp) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs_server_mark_return_all_delegations(server); + rcu_read_unlock(); + + nfs_delegation_run_state_manager(clp); +} + +/** + * nfs_server_return_all_delegations - return delegations for one superblock + * @server: pointer to nfs_server to process + * + */ +void nfs_server_return_all_delegations(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + bool need_wait; + + if (clp == NULL) + return; + + rcu_read_lock(); + need_wait = nfs_server_mark_return_all_delegations(server); + rcu_read_unlock(); + + if (need_wait) { + nfs4_schedule_state_manager(clp); + nfs4_wait_clnt_recover(clp); + } +} + +static void nfs_mark_return_unused_delegation_types(struct nfs_server *server, + fmode_t flags) +{ + struct nfs_delegation *delegation; + + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + if ((delegation->type == (FMODE_READ|FMODE_WRITE)) && !(flags & FMODE_WRITE)) + continue; + if (delegation->type & flags) + nfs_mark_return_if_closed_delegation(server, delegation); + } +} + +void nfs_expire_unused_delegation_types(struct nfs_client *clp, fmode_t flags) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs_mark_return_unused_delegation_types(server, flags); + rcu_read_unlock(); + + nfs_delegation_run_state_manager(clp); +} + +static void nfs_revoke_delegation(struct inode *inode, + const nfs4_stateid *stateid) +{ + struct nfs_delegation *delegation; + nfs4_stateid tmp; + bool ret = false; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation == NULL) + goto out; + if (stateid == NULL) { + nfs4_stateid_copy(&tmp, &delegation->stateid); + stateid = &tmp; + } else { + if (!nfs4_stateid_match_other(stateid, &delegation->stateid)) + goto out; + spin_lock(&delegation->lock); + if (stateid->seqid) { + if (nfs4_stateid_is_newer(&delegation->stateid, stateid)) { + spin_unlock(&delegation->lock); + goto out; + } + delegation->stateid.seqid = stateid->seqid; + } + spin_unlock(&delegation->lock); + } + nfs_mark_delegation_revoked(NFS_SERVER(inode), delegation); + ret = true; +out: + rcu_read_unlock(); + if (ret) + nfs_inode_find_state_and_recover(inode, stateid); +} + +void nfs_delegation_mark_returned(struct inode *inode, + const nfs4_stateid *stateid) +{ + struct nfs_delegation *delegation; + + if (!inode) + return; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (!delegation) + goto out_rcu_unlock; + + spin_lock(&delegation->lock); + if (!nfs4_stateid_match_other(stateid, &delegation->stateid)) + goto out_spin_unlock; + if (stateid->seqid) { + /* If delegation->stateid is newer, dont mark as returned */ + if (nfs4_stateid_is_newer(&delegation->stateid, stateid)) + goto out_clear_returning; + if (delegation->stateid.seqid != stateid->seqid) + delegation->stateid.seqid = stateid->seqid; + } + + nfs_mark_delegation_revoked(NFS_SERVER(inode), delegation); + clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags); + spin_unlock(&delegation->lock); + if (nfs_detach_delegation(NFS_I(inode), delegation, NFS_SERVER(inode))) + nfs_put_delegation(delegation); + goto out_rcu_unlock; + +out_clear_returning: + clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags); +out_spin_unlock: + spin_unlock(&delegation->lock); +out_rcu_unlock: + rcu_read_unlock(); + + nfs_inode_find_state_and_recover(inode, stateid); +} + +/** + * nfs_remove_bad_delegation - handle delegations that are unusable + * @inode: inode to process + * @stateid: the delegation's stateid + * + * If the server ACK-ed our FREE_STATEID then clean + * up the delegation, else mark and keep the revoked state. + */ +void nfs_remove_bad_delegation(struct inode *inode, + const nfs4_stateid *stateid) +{ + if (stateid && stateid->type == NFS4_FREED_STATEID_TYPE) + nfs_delegation_mark_returned(inode, stateid); + else + nfs_revoke_delegation(inode, stateid); +} +EXPORT_SYMBOL_GPL(nfs_remove_bad_delegation); + +static bool nfs_mark_return_unreferenced_delegations(struct nfs_server *server) +{ + struct nfs_delegation *d, *n; + bool marked = false; + + spin_lock(&server->delegations_lock); + list_for_each_entry_safe(d, n, &server->delegations_lru, entry) { + if (test_and_clear_bit(NFS_DELEGATION_REFERENCED, &d->flags)) + continue; + list_move_tail(&d->entry, &server->delegations_return); + marked = true; + } + spin_unlock(&server->delegations_lock); + + return marked; +} + +/** + * nfs_expire_unreferenced_delegations - Eliminate unused delegations + * @clp: nfs_client to process + * + */ +void nfs_expire_unreferenced_delegations(struct nfs_client *clp) +{ + struct nfs_server *server; + bool marked = false; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + marked |= nfs_mark_return_unreferenced_delegations(server); + rcu_read_unlock(); + + if (marked) { + set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state); + nfs4_schedule_state_manager(clp); + } +} + +/** + * nfs_async_inode_return_delegation - asynchronously return a delegation + * @inode: inode to process + * @stateid: state ID information + * + * Returns zero on success, or a negative errno value. + */ +int nfs_async_inode_return_delegation(struct inode *inode, + const nfs4_stateid *stateid) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_client *clp = server->nfs_client; + struct nfs_delegation *delegation; + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return -ENOENT; + + if (stateid != NULL && + !clp->cl_mvops->match_stateid(&delegation->stateid, stateid)) { + nfs_put_delegation(delegation); + return -ENOENT; + } + + nfs_mark_return_delegation(server, delegation); + nfs_put_delegation(delegation); + + /* If there are any application leases or delegations, recall them */ + break_lease(inode, O_WRONLY | O_RDWR | O_NONBLOCK); + + nfs_delegation_run_state_manager(clp); + return 0; +} + +static struct inode * +nfs_delegation_find_inode_server(struct nfs_server *server, + const struct nfs_fh *fhandle) +{ + struct hlist_head *head = nfs_delegation_hash(server, fhandle); + struct nfs_delegation *delegation; + struct super_block *freeme = NULL; + struct inode *res = NULL; + + hlist_for_each_entry_rcu(delegation, head, hash) { + spin_lock(&delegation->lock); + if (delegation->inode != NULL && + !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) && + nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) { + if (nfs_sb_active(server->super)) { + freeme = server->super; + res = igrab(delegation->inode); + } + spin_unlock(&delegation->lock); + if (res != NULL) + return res; + if (freeme) { + rcu_read_unlock(); + nfs_sb_deactive(freeme); + rcu_read_lock(); + } + return ERR_PTR(-EAGAIN); + } + spin_unlock(&delegation->lock); + } + return ERR_PTR(-ENOENT); +} + +/** + * nfs_delegation_find_inode - retrieve the inode associated with a delegation + * @clp: client state handle + * @fhandle: filehandle from a delegation recall + * + * Returns pointer to inode matching "fhandle," or NULL if a matching inode + * cannot be found. + */ +struct inode *nfs_delegation_find_inode(struct nfs_client *clp, + const struct nfs_fh *fhandle) +{ + struct nfs_server *server; + struct inode *res; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + res = nfs_delegation_find_inode_server(server, fhandle); + if (res != ERR_PTR(-ENOENT)) { + rcu_read_unlock(); + return res; + } + } + rcu_read_unlock(); + return ERR_PTR(-ENOENT); +} + +static void nfs_delegation_mark_reclaim_server(struct nfs_server *server) +{ + struct nfs_delegation *delegation; + + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + /* + * If the delegation may have been admin revoked, then we + * cannot reclaim it. + */ + if (test_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags)) + continue; + set_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags); + } +} + +/** + * nfs_delegation_mark_reclaim - mark all delegations as needing to be reclaimed + * @clp: nfs_client to process + * + */ +void nfs_delegation_mark_reclaim(struct nfs_client *clp) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs_delegation_mark_reclaim_server(server); + rcu_read_unlock(); +} + +static int nfs_server_reap_unclaimed_delegations(struct nfs_server *server, + void __always_unused *data) +{ + struct nfs_delegation *delegation; + struct inode *inode; +restart: + rcu_read_lock(); + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + if (test_bit(NFS_DELEGATION_RETURNING, + &delegation->flags) || + test_bit(NFS_DELEGATION_NEED_RECLAIM, + &delegation->flags) == 0) + continue; + inode = nfs_delegation_grab_inode(delegation); + if (inode == NULL) + continue; + delegation = nfs_start_delegation_return(NFS_I(inode)); + rcu_read_unlock(); + if (delegation != NULL) { + if (nfs_detach_delegation(NFS_I(inode), delegation, + server)) { + nfs_mark_delegation_revoked(server, delegation); + nfs_put_delegation(delegation); + } + /* Match nfs_start_delegation_return */ + nfs_put_delegation(delegation); + } + iput(inode); + cond_resched(); + goto restart; + } + rcu_read_unlock(); + return 0; +} + +/** + * nfs_delegation_reap_unclaimed - reap unclaimed delegations after reboot recovery is done + * @clp: nfs_client to process + * + */ +void nfs_delegation_reap_unclaimed(struct nfs_client *clp) +{ + nfs_client_for_each_server(clp, nfs_server_reap_unclaimed_delegations, + NULL); +} + +static inline bool nfs4_server_rebooted(const struct nfs_client *clp) +{ + return (clp->cl_state & (BIT(NFS4CLNT_CHECK_LEASE) | + BIT(NFS4CLNT_LEASE_EXPIRED) | + BIT(NFS4CLNT_SESSION_RESET))) != 0; +} + +static void nfs_mark_test_expired_delegation(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + if (delegation->stateid.type == NFS4_INVALID_STATEID_TYPE) + return; + clear_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags); + set_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags); + set_bit(NFS4SERV_DELEGATION_EXPIRED, &server->delegation_flags); + set_bit(NFS4CLNT_DELEGATION_EXPIRED, &server->nfs_client->cl_state); +} + +static void nfs_inode_mark_test_expired_delegation(struct nfs_server *server, + struct inode *inode) +{ + struct nfs_delegation *delegation; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation) + nfs_mark_test_expired_delegation(server, delegation); + rcu_read_unlock(); + +} + +static void nfs_delegation_mark_test_expired_server(struct nfs_server *server) +{ + struct nfs_delegation *delegation; + + list_for_each_entry_rcu(delegation, &server->delegations, super_list) + nfs_mark_test_expired_delegation(server, delegation); +} + +/** + * nfs_mark_test_expired_all_delegations - mark all delegations for testing + * @clp: nfs_client to process + * + * Iterates through all the delegations associated with this server and + * marks them as needing to be checked for validity. + */ +void nfs_mark_test_expired_all_delegations(struct nfs_client *clp) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs_delegation_mark_test_expired_server(server); + rcu_read_unlock(); +} + +/** + * nfs_test_expired_all_delegations - test all delegations for a client + * @clp: nfs_client to process + * + * Helper for handling "recallable state revoked" status from server. + */ +void nfs_test_expired_all_delegations(struct nfs_client *clp) +{ + nfs_mark_test_expired_all_delegations(clp); + nfs4_schedule_state_manager(clp); +} + +static void +nfs_delegation_test_free_expired(struct inode *inode, + nfs4_stateid *stateid, + const struct cred *cred) +{ + struct nfs_server *server = NFS_SERVER(inode); + const struct nfs4_minor_version_ops *ops = server->nfs_client->cl_mvops; + int status; + + if (!cred) + return; + status = ops->test_and_free_expired(server, stateid, cred); + if (status == -NFS4ERR_EXPIRED || status == -NFS4ERR_BAD_STATEID) + nfs_remove_bad_delegation(inode, stateid); +} + +static int nfs_server_reap_expired_delegations(struct nfs_server *server, + void __always_unused *data) +{ + struct nfs_delegation *delegation; + struct inode *inode; + const struct cred *cred; + nfs4_stateid stateid; + unsigned long gen = ++server->delegation_gen; + + if (!test_and_clear_bit(NFS4SERV_DELEGATION_EXPIRED, + &server->delegation_flags)) + return 0; +restart: + rcu_read_lock(); + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + if (test_bit(NFS_DELEGATION_RETURNING, + &delegation->flags) || + test_bit(NFS_DELEGATION_TEST_EXPIRED, + &delegation->flags) == 0 || + delegation->test_gen == gen) + continue; + inode = nfs_delegation_grab_inode(delegation); + if (inode == NULL) + continue; + spin_lock(&delegation->lock); + cred = get_cred_rcu(delegation->cred); + nfs4_stateid_copy(&stateid, &delegation->stateid); + spin_unlock(&delegation->lock); + delegation->test_gen = gen; + clear_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags); + rcu_read_unlock(); + nfs_delegation_test_free_expired(inode, &stateid, cred); + put_cred(cred); + if (!nfs4_server_rebooted(server->nfs_client)) { + iput(inode); + cond_resched(); + goto restart; + } + nfs_inode_mark_test_expired_delegation(server,inode); + set_bit(NFS4SERV_DELEGATION_EXPIRED, &server->delegation_flags); + set_bit(NFS4CLNT_DELEGATION_EXPIRED, + &server->nfs_client->cl_state); + iput(inode); + return -EAGAIN; + } + rcu_read_unlock(); + return 0; +} + +/** + * nfs_reap_expired_delegations - reap expired delegations + * @clp: nfs_client to process + * + * Iterates through all the delegations associated with this server and + * checks if they have may have been revoked. This function is usually + * expected to be called in cases where the server may have lost its + * lease. + */ +void nfs_reap_expired_delegations(struct nfs_client *clp) +{ + nfs_client_for_each_server(clp, nfs_server_reap_expired_delegations, + NULL); +} + +void nfs_inode_find_delegation_state_and_recover(struct inode *inode, + const nfs4_stateid *stateid) +{ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + struct nfs_delegation *delegation; + bool found = false; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation && + nfs4_stateid_match_or_older(&delegation->stateid, stateid) && + !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) { + nfs_mark_test_expired_delegation(NFS_SERVER(inode), delegation); + found = true; + } + rcu_read_unlock(); + if (found) + nfs4_schedule_state_manager(clp); +} + +/** + * nfs_delegations_present - check for existence of delegations + * @clp: client state handle + * + * Returns one if there are any nfs_delegation structures attached + * to this nfs_client. + */ +int nfs_delegations_present(struct nfs_client *clp) +{ + struct nfs_server *server; + int ret = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + if (atomic_long_read(&server->nr_active_delegations) > 0) { + ret = 1; + break; + } + rcu_read_unlock(); + return ret; +} + +/** + * nfs4_refresh_delegation_stateid - Update delegation stateid seqid + * @dst: stateid to refresh + * @inode: inode to check + * + * Returns "true" and updates "dst->seqid" * if inode had a delegation + * that matches our delegation stateid. Otherwise "false" is returned. + */ +bool nfs4_refresh_delegation_stateid(nfs4_stateid *dst, struct inode *inode) +{ + struct nfs_delegation *delegation; + bool ret = false; + if (!inode) + goto out; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation != NULL && + nfs4_stateid_match_other(dst, &delegation->stateid) && + nfs4_stateid_is_newer(&delegation->stateid, dst) && + !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags)) { + dst->seqid = delegation->stateid.seqid; + ret = true; + } + rcu_read_unlock(); +out: + return ret; +} + +/** + * nfs4_copy_delegation_stateid - Copy inode's state ID information + * @inode: inode to check + * @flags: delegation type requirement + * @dst: stateid data structure to fill in + * @cred: optional argument to retrieve credential + * + * Returns "true" and fills in "dst->data" * if inode had a delegation, + * otherwise "false" is returned. + */ +bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags, + nfs4_stateid *dst, const struct cred **cred) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_delegation *delegation; + bool ret = false; + + flags &= FMODE_READ|FMODE_WRITE; + rcu_read_lock(); + delegation = rcu_dereference(nfsi->delegation); + if (!delegation) + goto out; + spin_lock(&delegation->lock); + ret = nfs4_is_valid_delegation(delegation, flags); + if (ret) { + nfs4_stateid_copy(dst, &delegation->stateid); + nfs_mark_delegation_referenced(delegation); + if (cred) + *cred = get_cred(delegation->cred); + } + spin_unlock(&delegation->lock); +out: + rcu_read_unlock(); + return ret; +} + +/** + * nfs4_delegation_flush_on_close - Check if we must flush file on close + * @inode: inode to check + * + * This function checks the number of outstanding writes to the file + * against the delegation 'space_limit' field to see if + * the spec requires us to flush the file on close. + */ +bool nfs4_delegation_flush_on_close(const struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_delegation *delegation; + bool ret = true; + + rcu_read_lock(); + delegation = rcu_dereference(nfsi->delegation); + if (delegation == NULL || !(delegation->type & FMODE_WRITE)) + goto out; + if (atomic_long_read(&nfsi->nrequests) < delegation->pagemod_limit) + ret = false; +out: + rcu_read_unlock(); + return ret; +} + +int nfs4_delegation_hash_alloc(struct nfs_server *server) +{ + int delegation_buckets, i; + + delegation_buckets = roundup_pow_of_two(nfs_delegation_watermark / 16); + server->delegation_hash_mask = delegation_buckets - 1; + server->delegation_hash_table = kmalloc_objs(*server->delegation_hash_table, + delegation_buckets); + if (!server->delegation_hash_table) + return -ENOMEM; + for (i = 0; i < delegation_buckets; i++) + INIT_HLIST_HEAD(&server->delegation_hash_table[i]); + return 0; +} diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h new file mode 100644 index 000000000..fba469995 --- /dev/null +++ b/fs/nfs/delegation.h @@ -0,0 +1,141 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * linux/fs/nfs/delegation.h + * + * Copyright (c) Trond Myklebust + * + * Definitions pertaining to NFS delegated files + */ +#ifndef FS_NFS_DELEGATION_H +#define FS_NFS_DELEGATION_H + +#if IS_ENABLED(CONFIG_NFS_V4) +/* + * NFSv4 delegation + */ +struct nfs_delegation { + struct hlist_node hash; + struct list_head super_list; + const struct cred *cred; + struct inode *inode; + nfs4_stateid stateid; + fmode_t type; + unsigned long pagemod_limit; + __u64 change_attr; + unsigned long test_gen; + unsigned long flags; + refcount_t refcount; + spinlock_t lock; + struct list_head entry; + struct rcu_head rcu; +}; + +enum { + NFS_DELEGATION_NEED_RECLAIM = 0, + NFS_DELEGATION_RETURN_IF_CLOSED, + NFS_DELEGATION_REFERENCED, + NFS_DELEGATION_RETURNING, + NFS_DELEGATION_REVOKED, + NFS_DELEGATION_TEST_EXPIRED, + NFS_DELEGATION_DELEGTIME, +}; + +int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred, + fmode_t type, const nfs4_stateid *stateid, + unsigned long pagemod_limit, u32 deleg_type); +void nfs_inode_reclaim_delegation(struct inode *inode, const struct cred *cred, + fmode_t type, const nfs4_stateid *stateid, + unsigned long pagemod_limit, u32 deleg_type); +void nfs4_inode_return_delegation(struct inode *inode); +void nfs4_inode_return_delegation_on_close(struct inode *inode); +void nfs4_inode_set_return_delegation_on_close(struct inode *inode); +int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); +void nfs_inode_evict_delegation(struct inode *inode); + +struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle); +void nfs_server_return_all_delegations(struct nfs_server *); +void nfs_expire_all_delegations(struct nfs_client *clp); +void nfs_expire_unused_delegation_types(struct nfs_client *clp, fmode_t flags); +void nfs_expire_unreferenced_delegations(struct nfs_client *clp); +int nfs_client_return_marked_delegations(struct nfs_client *clp); +int nfs_delegations_present(struct nfs_client *clp); +void nfs_remove_bad_delegation(struct inode *inode, const nfs4_stateid *stateid); +void nfs_delegation_mark_returned(struct inode *inode, const nfs4_stateid *stateid); + +void nfs_delegation_mark_reclaim(struct nfs_client *clp); +void nfs_delegation_reap_unclaimed(struct nfs_client *clp); + +void nfs_mark_test_expired_all_delegations(struct nfs_client *clp); +void nfs_test_expired_all_delegations(struct nfs_client *clp); +void nfs_reap_expired_delegations(struct nfs_client *clp); + +/* NFSv4 delegation-related procedures */ +int nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred, + const nfs4_stateid *stateid, + struct nfs_delegation *delegation, int issync); +int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid); +int nfs4_lock_delegation_recall(struct file_lock *fl, struct nfs4_state *state, const nfs4_stateid *stateid); +bool nfs4_copy_delegation_stateid(struct inode *inode, fmode_t flags, nfs4_stateid *dst, const struct cred **cred); +bool nfs4_refresh_delegation_stateid(nfs4_stateid *dst, struct inode *inode); + +struct nfs_delegation *nfs4_get_valid_delegation(const struct inode *inode); +void nfs_put_delegation(struct nfs_delegation *delegation); +void nfs_mark_delegation_referenced(struct nfs_delegation *delegation); +int nfs4_have_delegation(struct inode *inode, fmode_t type, int flags); +int nfs4_check_delegation(struct inode *inode, fmode_t type); +bool nfs4_delegation_flush_on_close(const struct inode *inode); +void nfs_inode_find_delegation_state_and_recover(struct inode *inode, + const nfs4_stateid *stateid); +void nfs4_inode_make_writeable(struct inode *inode); + +#endif + +#define NFS_DELEGATION_FLAG_TIME BIT(1) + +void nfs_update_delegated_atime(struct inode *inode); +void nfs_update_delegated_mtime(struct inode *inode); +void nfs_update_delegated_mtime_locked(struct inode *inode); + +static inline int nfs_have_read_or_write_delegation(struct inode *inode) +{ + return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, 0); +} + +static inline int nfs_have_write_delegation(struct inode *inode) +{ + return NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE, 0); +} + +static inline int nfs_have_delegated_attributes(struct inode *inode) +{ + return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, 0); +} + +static inline int nfs_have_delegated_atime(struct inode *inode) +{ + return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, + NFS_DELEGATION_FLAG_TIME); +} + +static inline int nfs_have_delegated_mtime(struct inode *inode) +{ + return NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE, + NFS_DELEGATION_FLAG_TIME); +} + +extern bool directory_delegations; + +static inline void nfs_request_directory_delegation(struct inode *inode) +{ + if (S_ISDIR(inode->i_mode)) + set_bit(NFS_INO_REQ_DIR_DELEG, &NFS_I(inode)->flags); +} + +static inline bool nfs_have_directory_delegation(struct inode *inode) +{ + return S_ISDIR(inode->i_mode) && nfs_have_delegated_attributes(inode); +} + +int nfs4_delegation_hash_alloc(struct nfs_server *server); + +#endif diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c new file mode 100644 index 000000000..36f2e8588 --- /dev/null +++ b/fs/nfs/dir.c @@ -0,0 +1,3433 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/dir.c + * + * Copyright (C) 1992 Rick Sladkey + * + * nfs directory handling functions + * + * 10 Apr 1996 Added silly rename for unlink --okir + * 28 Sep 1996 Improved directory cache --okir + * 23 Aug 1997 Claus Heine claus@momo.math.rwth-aachen.de + * Re-implemented silly rename for unlink, newly implemented + * silly rename for nfs_rename() following the suggestions + * of Olaf Kirch (okir) found in this file. + * Following Linus comments on my original hack, this version + * depends only on the dcache stuff and doesn't touch the inode + * layer (iput() and friends). + * 6 Jun 1999 Cache readdir lookups in the page cache. -DaveM + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "delegation.h" +#include "iostat.h" +#include "internal.h" +#include "fscache.h" + +#include "nfstrace.h" + +/* #define NFS_DEBUG_VERBOSE 1 */ + +static int nfs_opendir(struct inode *, struct file *); +static int nfs_closedir(struct inode *, struct file *); +static int nfs_readdir(struct file *, struct dir_context *); +static int nfs_fsync_dir(struct file *, loff_t, loff_t, int); +static loff_t nfs_llseek_dir(struct file *, loff_t, int); +static void nfs_readdir_clear_array(struct folio *); +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags); + +const struct file_operations nfs_dir_operations = { + .llseek = nfs_llseek_dir, + .read = generic_read_dir, + .iterate_shared = nfs_readdir, + .open = nfs_opendir, + .release = nfs_closedir, + .fsync = nfs_fsync_dir, +}; + +const struct address_space_operations nfs_dir_aops = { + .free_folio = nfs_readdir_clear_array, +}; + +#define NFS_INIT_DTSIZE SZ_64K + +static struct nfs_open_dir_context * +alloc_nfs_open_dir_context(struct inode *dir) +{ + struct nfs_inode *nfsi = NFS_I(dir); + struct nfs_open_dir_context *ctx; + + ctx = kzalloc_obj(*ctx, GFP_KERNEL_ACCOUNT); + if (ctx != NULL) { + ctx->attr_gencount = nfsi->attr_gencount; + ctx->dtsize = min(NFS_SERVER(dir)->dtsize, NFS_INIT_DTSIZE); + spin_lock(&dir->i_lock); + if (list_empty(&nfsi->open_files) && + (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) + nfs_set_cache_invalid(dir, + NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); + list_add_tail_rcu(&ctx->list, &nfsi->open_files); + memcpy(ctx->verf, nfsi->cookieverf, sizeof(ctx->verf)); + spin_unlock(&dir->i_lock); + return ctx; + } + return ERR_PTR(-ENOMEM); +} + +static void put_nfs_open_dir_context(struct inode *dir, struct nfs_open_dir_context *ctx) +{ + spin_lock(&dir->i_lock); + list_del_rcu(&ctx->list); + spin_unlock(&dir->i_lock); + kfree_rcu(ctx, rcu_head); +} + +/* + * Open file + */ +static int +nfs_opendir(struct inode *inode, struct file *filp) +{ + int res = 0; + struct nfs_open_dir_context *ctx; + + dfprintk(FILE, "NFS: open dir(%pD2)\n", filp); + + nfs_inc_stats(inode, NFSIOS_VFSOPEN); + + ctx = alloc_nfs_open_dir_context(inode); + if (IS_ERR(ctx)) { + res = PTR_ERR(ctx); + goto out; + } + filp->private_data = ctx; +out: + return res; +} + +static int +nfs_closedir(struct inode *inode, struct file *filp) +{ + put_nfs_open_dir_context(file_inode(filp), filp->private_data); + return 0; +} + +struct nfs_cache_array_entry { + u64 cookie; + u64 ino; + const char *name; + unsigned int name_len; + unsigned char d_type; +}; + +struct nfs_cache_array { + u64 change_attr; + u64 last_cookie; + unsigned int size; + unsigned char folio_full : 1, + folio_is_eof : 1, + cookies_are_ordered : 1; + struct nfs_cache_array_entry array[] __counted_by(size); +}; + +struct nfs_readdir_descriptor { + struct file *file; + struct folio *folio; + struct dir_context *ctx; + pgoff_t folio_index; + pgoff_t folio_index_max; + u64 dir_cookie; + u64 last_cookie; + loff_t current_index; + + __be32 verf[NFS_DIR_VERIFIER_SIZE]; + unsigned long dir_verifier; + unsigned long timestamp; + unsigned long gencount; + unsigned long attr_gencount; + unsigned int cache_entry_index; + unsigned int buffer_fills; + unsigned int dtsize; + bool clear_cache; + bool plus; + bool eob; + bool eof; +}; + +static void nfs_set_dtsize(struct nfs_readdir_descriptor *desc, unsigned int sz) +{ + struct nfs_server *server = NFS_SERVER(file_inode(desc->file)); + unsigned int maxsize = server->dtsize; + + if (sz > maxsize) + sz = maxsize; + if (sz < NFS_MIN_FILE_IO_SIZE) + sz = NFS_MIN_FILE_IO_SIZE; + desc->dtsize = sz; +} + +static void nfs_shrink_dtsize(struct nfs_readdir_descriptor *desc) +{ + nfs_set_dtsize(desc, desc->dtsize >> 1); +} + +static void nfs_grow_dtsize(struct nfs_readdir_descriptor *desc) +{ + nfs_set_dtsize(desc, desc->dtsize << 1); +} + +static void nfs_readdir_folio_init_array(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + struct nfs_cache_array *array; + + array = kmap_local_folio(folio, 0); + array->change_attr = change_attr; + array->last_cookie = last_cookie; + array->size = 0; + array->folio_full = 0; + array->folio_is_eof = 0; + array->cookies_are_ordered = 1; + kunmap_local(array); +} + +/* + * we are freeing strings created by nfs_add_to_readdir_array() + */ +static void nfs_readdir_clear_array(struct folio *folio) +{ + struct nfs_cache_array *array; + unsigned int i; + + array = kmap_local_folio(folio, 0); + for (i = 0; i < array->size; i++) + kfree(array->array[i].name); + array->size = 0; + kunmap_local(array); +} + +static void nfs_readdir_folio_reinit_array(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + nfs_readdir_clear_array(folio); + nfs_readdir_folio_init_array(folio, last_cookie, change_attr); +} + +static struct folio * +nfs_readdir_folio_array_alloc(u64 last_cookie, gfp_t gfp_flags) +{ + struct folio *folio = folio_alloc(gfp_flags, 0); + if (folio) + nfs_readdir_folio_init_array(folio, last_cookie, 0); + return folio; +} + +static void nfs_readdir_folio_array_free(struct folio *folio) +{ + if (folio) { + nfs_readdir_clear_array(folio); + folio_put(folio); + } +} + +static u64 nfs_readdir_array_index_cookie(struct nfs_cache_array *array) +{ + return array->size == 0 ? array->last_cookie : array->array[0].cookie; +} + +static void nfs_readdir_array_set_eof(struct nfs_cache_array *array) +{ + array->folio_is_eof = 1; + array->folio_full = 1; +} + +static bool nfs_readdir_array_is_full(struct nfs_cache_array *array) +{ + return array->folio_full; +} + +/* + * the caller is responsible for freeing qstr.name + * when called by nfs_readdir_add_to_array, the strings will be freed in + * nfs_clear_readdir_array() + */ +static const char *nfs_readdir_copy_name(const char *name, unsigned int len) +{ + const char *ret = kmemdup_nul(name, len, GFP_KERNEL); + + /* + * Avoid a kmemleak false positive. The pointer to the name is stored + * in a page cache page which kmemleak does not scan. + */ + if (ret != NULL) + kmemleak_not_leak(ret); + return ret; +} + +static size_t nfs_readdir_array_maxentries(void) +{ + return (PAGE_SIZE - sizeof(struct nfs_cache_array)) / + sizeof(struct nfs_cache_array_entry); +} + +/* + * Check that the next array entry lies entirely within the page bounds + */ +static int nfs_readdir_array_can_expand(struct nfs_cache_array *array) +{ + if (array->folio_full) + return -ENOSPC; + if (array->size == nfs_readdir_array_maxentries()) { + array->folio_full = 1; + return -ENOSPC; + } + return 0; +} + +static int nfs_readdir_folio_array_append(struct folio *folio, + const struct nfs_entry *entry, + u64 *cookie) +{ + struct nfs_cache_array *array; + struct nfs_cache_array_entry *cache_entry; + const char *name; + int ret = -ENOMEM; + + name = nfs_readdir_copy_name(entry->name, entry->len); + + array = kmap_local_folio(folio, 0); + if (!name) + goto out; + ret = nfs_readdir_array_can_expand(array); + if (ret) { + kfree(name); + goto out; + } + + array->size++; + cache_entry = &array->array[array->size - 1]; + cache_entry->cookie = array->last_cookie; + cache_entry->ino = entry->ino; + cache_entry->d_type = entry->d_type; + cache_entry->name_len = entry->len; + cache_entry->name = name; + array->last_cookie = entry->cookie; + if (array->last_cookie <= cache_entry->cookie) + array->cookies_are_ordered = 0; + if (entry->eof != 0) + nfs_readdir_array_set_eof(array); +out: + *cookie = array->last_cookie; + kunmap_local(array); + return ret; +} + +#define NFS_READDIR_COOKIE_MASK (U32_MAX >> 14) +/* + * Hash algorithm allowing content addressible access to sequences + * of directory cookies. Content is addressed by the value of the + * cookie index of the first readdir entry in a page. + * + * We select only the first 18 bits to avoid issues with excessive + * memory use for the page cache XArray. 18 bits should allow the caching + * of 262144 pages of sequences of readdir entries. Since each page holds + * 127 readdir entries for a typical 64-bit system, that works out to a + * cache of ~ 33 million entries per directory. + */ +static pgoff_t nfs_readdir_folio_cookie_hash(u64 cookie) +{ + if (cookie == 0) + return 0; + return hash_64(cookie, 18); +} + +static bool nfs_readdir_folio_validate(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + struct nfs_cache_array *array = kmap_local_folio(folio, 0); + int ret = true; + + if (array->change_attr != change_attr) + ret = false; + if (nfs_readdir_array_index_cookie(array) != last_cookie) + ret = false; + kunmap_local(array); + return ret; +} + +static void nfs_readdir_folio_unlock_and_put(struct folio *folio) +{ + folio_unlock(folio); + folio_put(folio); +} + +static void nfs_readdir_folio_init_and_validate(struct folio *folio, u64 cookie, + u64 change_attr) +{ + if (folio_test_uptodate(folio)) { + if (nfs_readdir_folio_validate(folio, cookie, change_attr)) + return; + nfs_readdir_clear_array(folio); + } + nfs_readdir_folio_init_array(folio, cookie, change_attr); + folio_mark_uptodate(folio); +} + +static struct folio *nfs_readdir_folio_get_locked(struct address_space *mapping, + u64 cookie, u64 change_attr) +{ + pgoff_t index = nfs_readdir_folio_cookie_hash(cookie); + struct folio *folio; + + folio = filemap_grab_folio(mapping, index); + if (IS_ERR(folio)) + return NULL; + nfs_readdir_folio_init_and_validate(folio, cookie, change_attr); + return folio; +} + +static u64 nfs_readdir_folio_last_cookie(struct folio *folio) +{ + struct nfs_cache_array *array; + u64 ret; + + array = kmap_local_folio(folio, 0); + ret = array->last_cookie; + kunmap_local(array); + return ret; +} + +static bool nfs_readdir_folio_needs_filling(struct folio *folio) +{ + struct nfs_cache_array *array; + bool ret; + + array = kmap_local_folio(folio, 0); + ret = !nfs_readdir_array_is_full(array); + kunmap_local(array); + return ret; +} + +static void nfs_readdir_folio_set_eof(struct folio *folio) +{ + struct nfs_cache_array *array; + + array = kmap_local_folio(folio, 0); + nfs_readdir_array_set_eof(array); + kunmap_local(array); +} + +static struct folio *nfs_readdir_folio_get_next(struct address_space *mapping, + u64 cookie, u64 change_attr) +{ + pgoff_t index = nfs_readdir_folio_cookie_hash(cookie); + struct folio *folio; + + folio = __filemap_get_folio(mapping, index, + FGP_LOCK|FGP_CREAT|FGP_NOFS|FGP_NOWAIT, + mapping_gfp_mask(mapping)); + if (IS_ERR(folio)) + return NULL; + nfs_readdir_folio_init_and_validate(folio, cookie, change_attr); + if (nfs_readdir_folio_last_cookie(folio) != cookie) + nfs_readdir_folio_reinit_array(folio, cookie, change_attr); + return folio; +} + +static inline +int is_32bit_api(void) +{ +#ifdef CONFIG_COMPAT + return in_compat_syscall(); +#else + return (BITS_PER_LONG == 32); +#endif +} + +static +bool nfs_readdir_use_cookie(const struct file *filp) +{ + if ((filp->f_mode & FMODE_32BITHASH) || + (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api())) + return false; + return true; +} + +static void nfs_readdir_seek_next_array(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) +{ + if (array->folio_full) { + desc->last_cookie = array->last_cookie; + desc->current_index += array->size; + desc->cache_entry_index = 0; + desc->folio_index++; + } else + desc->last_cookie = nfs_readdir_array_index_cookie(array); +} + +static void nfs_readdir_rewind_search(struct nfs_readdir_descriptor *desc) +{ + desc->current_index = 0; + desc->last_cookie = 0; + desc->folio_index = 0; +} + +static int nfs_readdir_search_for_pos(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) +{ + loff_t diff = desc->ctx->pos - desc->current_index; + unsigned int index; + + if (diff < 0) + goto out_eof; + if (diff >= array->size) { + if (array->folio_is_eof) + goto out_eof; + nfs_readdir_seek_next_array(array, desc); + return -EAGAIN; + } + + index = (unsigned int)diff; + desc->dir_cookie = array->array[index].cookie; + desc->cache_entry_index = index; + return 0; +out_eof: + desc->eof = true; + return -EBADCOOKIE; +} + +static bool nfs_readdir_array_cookie_in_range(struct nfs_cache_array *array, + u64 cookie) +{ + if (!array->cookies_are_ordered) + return true; + /* Optimisation for monotonically increasing cookies */ + if (cookie >= array->last_cookie) + return false; + if (array->size && cookie < array->array[0].cookie) + return false; + return true; +} + +static int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) +{ + unsigned int i; + int status = -EAGAIN; + + if (!nfs_readdir_array_cookie_in_range(array, desc->dir_cookie)) + goto check_eof; + + for (i = 0; i < array->size; i++) { + if (array->array[i].cookie == desc->dir_cookie) { + if (nfs_readdir_use_cookie(desc->file)) + desc->ctx->pos = desc->dir_cookie; + else + desc->ctx->pos = desc->current_index + i; + desc->cache_entry_index = i; + return 0; + } + } +check_eof: + if (array->folio_is_eof) { + status = -EBADCOOKIE; + if (desc->dir_cookie == array->last_cookie) + desc->eof = true; + } else + nfs_readdir_seek_next_array(array, desc); + return status; +} + +static int nfs_readdir_search_array(struct nfs_readdir_descriptor *desc) +{ + struct nfs_cache_array *array; + int status; + + array = kmap_local_folio(desc->folio, 0); + + if (desc->dir_cookie == 0) + status = nfs_readdir_search_for_pos(array, desc); + else + status = nfs_readdir_search_for_cookie(array, desc); + + kunmap_local(array); + return status; +} + +/* Fill a page with xdr information before transferring to the cache page */ +static int nfs_readdir_xdr_filler(struct nfs_readdir_descriptor *desc, + __be32 *verf, u64 cookie, + struct page **pages, size_t bufsize, + __be32 *verf_res) +{ + struct inode *inode = file_inode(desc->file); + struct nfs_readdir_arg arg = { + .dentry = file_dentry(desc->file), + .cred = desc->file->f_cred, + .verf = verf, + .cookie = cookie, + .pages = pages, + .page_len = bufsize, + .plus = desc->plus, + }; + struct nfs_readdir_res res = { + .verf = verf_res, + }; + unsigned long timestamp, gencount; + int error; + + again: + timestamp = jiffies; + gencount = nfs_inc_attr_generation_counter(); + desc->dir_verifier = nfs_save_change_attribute(inode); + error = NFS_PROTO(inode)->readdir(&arg, &res); + if (error < 0) { + /* We requested READDIRPLUS, but the server doesn't grok it */ + if (error == -ENOTSUPP && desc->plus) { + NFS_SERVER(inode)->caps &= ~NFS_CAP_READDIRPLUS; + desc->plus = arg.plus = false; + goto again; + } + goto error; + } + desc->timestamp = timestamp; + desc->gencount = gencount; +error: + return error; +} + +static int xdr_decode(struct nfs_readdir_descriptor *desc, + struct nfs_entry *entry, struct xdr_stream *xdr) +{ + struct inode *inode = file_inode(desc->file); + int error; + + error = NFS_PROTO(inode)->decode_dirent(xdr, entry, desc->plus); + if (error) + return error; + entry->fattr->time_start = desc->timestamp; + entry->fattr->gencount = desc->gencount; + return 0; +} + +/* Match file and dirent using either filehandle or fileid + * Note: caller is responsible for checking the fsid + */ +static +int nfs_same_file(struct dentry *dentry, struct nfs_entry *entry) +{ + struct inode *inode; + struct nfs_inode *nfsi; + + if (d_really_is_negative(dentry)) + return 0; + + inode = d_inode(dentry); + if (is_bad_inode(inode) || NFS_STALE(inode)) + return 0; + + nfsi = NFS_I(inode); + if (entry->fattr->fileid != inode->i_ino) + return 0; + if (entry->fh->size && nfs_compare_fh(entry->fh, &nfsi->fh) != 0) + return 0; + return 1; +} + +#define NFS_READDIR_CACHE_USAGE_THRESHOLD (8UL) + +static bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx, + unsigned int cache_hits, + unsigned int cache_misses) +{ + if (!nfs_server_capable(dir, NFS_CAP_READDIRPLUS)) + return false; + if (NFS_SERVER(dir)->flags & NFS_MOUNT_FORCE_RDIRPLUS) + return true; + if (ctx->pos == 0 || + cache_hits + cache_misses > NFS_READDIR_CACHE_USAGE_THRESHOLD) + return true; + return false; +} + +/* + * This function is called by the getattr code to request the + * use of readdirplus to accelerate any future lookups in the same + * directory. + */ +void nfs_readdir_record_entry_cache_hit(struct inode *dir) +{ + struct nfs_inode *nfsi = NFS_I(dir); + struct nfs_open_dir_context *ctx; + + if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) && + S_ISDIR(dir->i_mode)) { + rcu_read_lock(); + list_for_each_entry_rcu (ctx, &nfsi->open_files, list) + atomic_inc(&ctx->cache_hits); + rcu_read_unlock(); + } +} + +/* + * This function is mainly for use by nfs_getattr(). + * + * If this is an 'ls -l', we want to force use of readdirplus. + */ +void nfs_readdir_record_entry_cache_miss(struct inode *dir) +{ + struct nfs_inode *nfsi = NFS_I(dir); + struct nfs_open_dir_context *ctx; + + if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) && + S_ISDIR(dir->i_mode)) { + rcu_read_lock(); + list_for_each_entry_rcu (ctx, &nfsi->open_files, list) + atomic_inc(&ctx->cache_misses); + rcu_read_unlock(); + } +} + +static void nfs_lookup_advise_force_readdirplus(struct inode *dir, + unsigned int flags) +{ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + return; + if (flags & (LOOKUP_EXCL | LOOKUP_PARENT | LOOKUP_REVAL)) + return; + nfs_readdir_record_entry_cache_miss(dir); +} + +static +void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry, + unsigned long dir_verifier) +{ + struct qstr filename = QSTR_INIT(entry->name, entry->len); + struct dentry *dentry; + struct dentry *alias; + struct inode *inode; + int status; + + if (!(entry->fattr->valid & NFS_ATTR_FATTR_FILEID)) + return; + if (!(entry->fattr->valid & NFS_ATTR_FATTR_FSID)) + return; + if (filename.len == 0) + return; + /* Validate that the name doesn't contain any illegal '\0' */ + if (strnlen(filename.name, filename.len) != filename.len) + return; + /* ...or '/' */ + if (strnchr(filename.name, filename.len, '/')) + return; + if (filename.name[0] == '.') { + if (filename.len == 1) + return; + if (filename.len == 2 && filename.name[1] == '.') + return; + } + filename.hash = full_name_hash(parent, filename.name, filename.len); + + dentry = d_lookup(parent, &filename); +again: + if (!dentry) { + dentry = d_alloc_parallel(parent, &filename); + if (IS_ERR(dentry)) + return; + } + if (!d_in_lookup(dentry)) { + /* Is there a mountpoint here? If so, just exit */ + if (!nfs_fsid_equal(&NFS_SB(dentry->d_sb)->fsid, + &entry->fattr->fsid)) + goto out; + if (nfs_same_file(dentry, entry)) { + if (!entry->fh->size) + goto out; + nfs_set_verifier(dentry, dir_verifier); + status = nfs_refresh_inode(d_inode(dentry), entry->fattr); + if (!status) + nfs_setsecurity(d_inode(dentry), entry->fattr); + trace_nfs_readdir_lookup_revalidate(d_inode(parent), + dentry, 0, status); + goto out; + } else { + trace_nfs_readdir_lookup_revalidate_failed( + d_inode(parent), dentry, 0); + d_invalidate(dentry); + dput(dentry); + dentry = NULL; + goto again; + } + } + if (!entry->fh->size) { + d_lookup_done(dentry); + goto out; + } + + nfs_set_verifier(dentry, dir_verifier); + inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr); + alias = d_splice_alias(inode, dentry); + d_lookup_done(dentry); + if (alias) { + if (IS_ERR(alias)) + goto out; + nfs_set_verifier(alias, dir_verifier); + dput(dentry); + dentry = alias; + } + trace_nfs_readdir_lookup(d_inode(parent), dentry, 0); +out: + dput(dentry); +} + +static int nfs_readdir_entry_decode(struct nfs_readdir_descriptor *desc, + struct nfs_entry *entry, + struct xdr_stream *stream) +{ + int ret; + + if (entry->fattr->label) + entry->fattr->label->len = NFS4_MAXLABELLEN; + ret = xdr_decode(desc, entry, stream); + if (ret || !desc->plus) + return ret; + nfs_prime_dcache(file_dentry(desc->file), entry, desc->dir_verifier); + return 0; +} + +/* Perform conversion from xdr to cache array */ +static int nfs_readdir_folio_filler(struct nfs_readdir_descriptor *desc, + struct nfs_entry *entry, + struct page **xdr_pages, unsigned int buflen, + struct folio **arrays, size_t narrays, + u64 change_attr) +{ + struct address_space *mapping = desc->file->f_mapping; + struct folio *new, *folio = *arrays; + struct xdr_stream stream; + struct folio *scratch; + struct xdr_buf buf; + u64 cookie; + int status; + + scratch = folio_alloc(GFP_KERNEL, 0); + if (scratch == NULL) + return -ENOMEM; + + xdr_init_decode_pages(&stream, &buf, xdr_pages, buflen); + xdr_set_scratch_folio(&stream, scratch); + + do { + status = nfs_readdir_entry_decode(desc, entry, &stream); + if (status != 0) + break; + + status = nfs_readdir_folio_array_append(folio, entry, &cookie); + if (status != -ENOSPC) + continue; + + if (folio->mapping != mapping) { + if (!--narrays) + break; + new = nfs_readdir_folio_array_alloc(cookie, GFP_KERNEL); + if (!new) + break; + arrays++; + *arrays = folio = new; + } else { + new = nfs_readdir_folio_get_next(mapping, cookie, + change_attr); + if (!new) + break; + if (folio != *arrays) + nfs_readdir_folio_unlock_and_put(folio); + folio = new; + } + desc->folio_index_max++; + status = nfs_readdir_folio_array_append(folio, entry, &cookie); + } while (!status && !entry->eof); + + switch (status) { + case -EBADCOOKIE: + if (!entry->eof) + break; + nfs_readdir_folio_set_eof(folio); + fallthrough; + case -EAGAIN: + status = 0; + break; + case -ENOSPC: + status = 0; + if (!desc->plus) + break; + while (!nfs_readdir_entry_decode(desc, entry, &stream)) + ; + } + + if (folio != *arrays) + nfs_readdir_folio_unlock_and_put(folio); + + folio_put(scratch); + return status; +} + +static void nfs_readdir_free_pages(struct page **pages, size_t npages) +{ + while (npages--) + put_page(pages[npages]); + kfree(pages); +} + +/* + * nfs_readdir_alloc_pages() will allocate pages that must be freed with a call + * to nfs_readdir_free_pages() + */ +static struct page **nfs_readdir_alloc_pages(size_t npages) +{ + struct page **pages; + size_t i; + + pages = kmalloc_objs(*pages, npages); + if (!pages) + return NULL; + for (i = 0; i < npages; i++) { + struct page *page = alloc_page(GFP_KERNEL); + if (page == NULL) + goto out_freepages; + pages[i] = page; + } + return pages; + +out_freepages: + nfs_readdir_free_pages(pages, i); + return NULL; +} + +static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, + __be32 *verf_arg, __be32 *verf_res, + struct folio **arrays, size_t narrays) +{ + u64 change_attr; + struct page **pages; + struct folio *folio = *arrays; + struct nfs_entry *entry; + size_t array_size; + struct inode *inode = file_inode(desc->file); + unsigned int dtsize = desc->dtsize; + unsigned int pglen; + int status = -ENOMEM; + + entry = kzalloc_obj(*entry); + if (!entry) + return -ENOMEM; + entry->cookie = nfs_readdir_folio_last_cookie(folio); + entry->fh = nfs_alloc_fhandle(); + entry->fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + entry->server = NFS_SERVER(inode); + if (entry->fh == NULL || entry->fattr == NULL) + goto out; + + array_size = (dtsize + PAGE_SIZE - 1) >> PAGE_SHIFT; + pages = nfs_readdir_alloc_pages(array_size); + if (!pages) + goto out; + + change_attr = inode_peek_iversion_raw(inode); + status = nfs_readdir_xdr_filler(desc, verf_arg, entry->cookie, pages, + dtsize, verf_res); + if (status < 0) + goto free_pages; + + pglen = status; + if (pglen != 0) + status = nfs_readdir_folio_filler(desc, entry, pages, pglen, + arrays, narrays, change_attr); + else + nfs_readdir_folio_set_eof(folio); + desc->buffer_fills++; + +free_pages: + nfs_readdir_free_pages(pages, array_size); +out: + nfs_free_fattr(entry->fattr); + nfs_free_fhandle(entry->fh); + kfree(entry); + return status; +} + +static void nfs_readdir_folio_put(struct nfs_readdir_descriptor *desc) +{ + folio_put(desc->folio); + desc->folio = NULL; +} + +static void +nfs_readdir_folio_unlock_and_put_cached(struct nfs_readdir_descriptor *desc) +{ + folio_unlock(desc->folio); + nfs_readdir_folio_put(desc); +} + +static struct folio * +nfs_readdir_folio_get_cached(struct nfs_readdir_descriptor *desc) +{ + struct address_space *mapping = desc->file->f_mapping; + u64 change_attr = inode_peek_iversion_raw(mapping->host); + u64 cookie = desc->last_cookie; + struct folio *folio; + + folio = nfs_readdir_folio_get_locked(mapping, cookie, change_attr); + if (!folio) + return NULL; + if (desc->clear_cache && !nfs_readdir_folio_needs_filling(folio)) + nfs_readdir_folio_reinit_array(folio, cookie, change_attr); + return folio; +} + +/* + * Returns 0 if desc->dir_cookie was found on page desc->page_index + * and locks the page to prevent removal from the page cache. + */ +static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc) +{ + struct inode *inode = file_inode(desc->file); + struct nfs_inode *nfsi = NFS_I(inode); + __be32 verf[NFS_DIR_VERIFIER_SIZE]; + int res; + + desc->folio = nfs_readdir_folio_get_cached(desc); + if (!desc->folio) + return -ENOMEM; + if (nfs_readdir_folio_needs_filling(desc->folio)) { + /* Grow the dtsize if we had to go back for more pages */ + if (desc->folio_index == desc->folio_index_max) + nfs_grow_dtsize(desc); + desc->folio_index_max = desc->folio_index; + trace_nfs_readdir_cache_fill(desc->file, nfsi->cookieverf, + desc->last_cookie, + desc->folio->index, desc->dtsize); + res = nfs_readdir_xdr_to_array(desc, nfsi->cookieverf, verf, + &desc->folio, 1); + if (res < 0) { + nfs_readdir_folio_unlock_and_put_cached(desc); + trace_nfs_readdir_cache_fill_done(inode, res); + if (res == -EBADCOOKIE || res == -ENOTSYNC) { + invalidate_inode_pages2(desc->file->f_mapping); + nfs_readdir_rewind_search(desc); + trace_nfs_readdir_invalidate_cache_range( + inode, 0, MAX_LFS_FILESIZE); + return -EAGAIN; + } + return res; + } + /* + * Set the cookie verifier if the page cache was empty + */ + if (desc->last_cookie == 0 && + memcmp(nfsi->cookieverf, verf, sizeof(nfsi->cookieverf))) { + memcpy(nfsi->cookieverf, verf, + sizeof(nfsi->cookieverf)); + invalidate_inode_pages2_range(desc->file->f_mapping, 1, + -1); + trace_nfs_readdir_invalidate_cache_range( + inode, 1, MAX_LFS_FILESIZE); + } + desc->clear_cache = false; + } + res = nfs_readdir_search_array(desc); + if (res == 0) + return 0; + nfs_readdir_folio_unlock_and_put_cached(desc); + return res; +} + +/* Search for desc->dir_cookie from the beginning of the page cache */ +static int readdir_search_pagecache(struct nfs_readdir_descriptor *desc) +{ + int res; + + do { + res = find_and_lock_cache_page(desc); + } while (res == -EAGAIN); + return res; +} + +#define NFS_READDIR_CACHE_MISS_THRESHOLD (16UL) + +/* + * Once we've found the start of the dirent within a page: fill 'er up... + */ +static void nfs_do_filldir(struct nfs_readdir_descriptor *desc, + const __be32 *verf) +{ + struct file *file = desc->file; + struct nfs_cache_array *array; + unsigned int i; + bool first_emit = !desc->dir_cookie; + + array = kmap_local_folio(desc->folio, 0); + for (i = desc->cache_entry_index; i < array->size; i++) { + struct nfs_cache_array_entry *ent; + + /* + * nfs_readdir_handle_cache_misses return force clear at + * (cache_misses > NFS_READDIR_CACHE_MISS_THRESHOLD) for + * readdir heuristic, NFS_READDIR_CACHE_MISS_THRESHOLD + 1 + * entries need be emitted here. + */ + if (first_emit && i > NFS_READDIR_CACHE_MISS_THRESHOLD + 2) { + desc->eob = true; + break; + } + + ent = &array->array[i]; + if (!dir_emit(desc->ctx, ent->name, ent->name_len, + ent->ino, ent->d_type)) { + desc->eob = true; + break; + } + memcpy(desc->verf, verf, sizeof(desc->verf)); + if (i == array->size - 1) { + desc->dir_cookie = array->last_cookie; + nfs_readdir_seek_next_array(array, desc); + } else { + desc->dir_cookie = array->array[i + 1].cookie; + desc->last_cookie = array->array[0].cookie; + } + if (nfs_readdir_use_cookie(file)) + desc->ctx->pos = desc->dir_cookie; + else + desc->ctx->pos++; + } + if (array->folio_is_eof) + desc->eof = !desc->eob; + + kunmap_local(array); + dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %llu\n", + (unsigned long long)desc->dir_cookie); +} + +/* + * If we cannot find a cookie in our cache, we suspect that this is + * because it points to a deleted file, so we ask the server to return + * whatever it thinks is the next entry. We then feed this to filldir. + * If all goes well, we should then be able to find our way round the + * cache on the next call to readdir_search_pagecache(); + * + * NOTE: we cannot add the anonymous page to the pagecache because + * the data it contains might not be page aligned. Besides, + * we should already have a complete representation of the + * directory in the page cache by the time we get here. + */ +static int uncached_readdir(struct nfs_readdir_descriptor *desc) +{ + struct folio **arrays; + size_t i, sz = 512; + __be32 verf[NFS_DIR_VERIFIER_SIZE]; + int status = -ENOMEM; + + dfprintk(DIRCACHE, "NFS: uncached_readdir() searching for cookie %llu\n", + (unsigned long long)desc->dir_cookie); + + arrays = kzalloc_objs(*arrays, sz); + if (!arrays) + goto out; + arrays[0] = nfs_readdir_folio_array_alloc(desc->dir_cookie, GFP_KERNEL); + if (!arrays[0]) + goto out; + + desc->folio_index = 0; + desc->cache_entry_index = 0; + desc->last_cookie = desc->dir_cookie; + desc->folio_index_max = 0; + + trace_nfs_readdir_uncached(desc->file, desc->verf, desc->last_cookie, + -1, desc->dtsize); + + status = nfs_readdir_xdr_to_array(desc, desc->verf, verf, arrays, sz); + if (status < 0) { + trace_nfs_readdir_uncached_done(file_inode(desc->file), status); + goto out_free; + } + + for (i = 0; !desc->eob && i < sz && arrays[i]; i++) { + desc->folio = arrays[i]; + nfs_do_filldir(desc, verf); + } + desc->folio = NULL; + + /* + * Grow the dtsize if we have to go back for more pages, + * or shrink it if we're reading too many. + */ + if (!desc->eof) { + if (!desc->eob) + nfs_grow_dtsize(desc); + else if (desc->buffer_fills == 1 && + i < (desc->folio_index_max >> 1)) + nfs_shrink_dtsize(desc); + } +out_free: + for (i = 0; i < sz && arrays[i]; i++) + nfs_readdir_folio_array_free(arrays[i]); +out: + if (!nfs_readdir_use_cookie(desc->file)) + nfs_readdir_rewind_search(desc); + desc->folio_index_max = -1; + kfree(arrays); + dfprintk(DIRCACHE, "NFS: %s: returns %d\n", __func__, status); + return status; +} + +static bool nfs_readdir_handle_cache_misses(struct inode *inode, + struct nfs_readdir_descriptor *desc, + unsigned int cache_misses, + bool force_clear) +{ + if (desc->ctx->pos == 0 || !desc->plus) + return false; + if (cache_misses <= NFS_READDIR_CACHE_MISS_THRESHOLD && !force_clear) + return false; + trace_nfs_readdir_force_readdirplus(inode); + return true; +} + +/* The file offset position represents the dirent entry number. A + last cookie cache takes care of the common case of reading the + whole directory. + */ +static int nfs_readdir(struct file *file, struct dir_context *ctx) +{ + struct dentry *dentry = file_dentry(file); + struct inode *inode = d_inode(dentry); + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_open_dir_context *dir_ctx = file->private_data; + struct nfs_readdir_descriptor *desc; + unsigned int cache_hits, cache_misses; + bool force_clear; + int res; + + dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n", + file, (long long)ctx->pos); + nfs_inc_stats(inode, NFSIOS_VFSGETDENTS); + + /* + * ctx->pos points to the dirent entry number. + * *desc->dir_cookie has the cookie for the next entry. We have + * to either find the entry with the appropriate number or + * revalidate the cookie. + */ + nfs_revalidate_mapping(inode, file->f_mapping); + + res = -ENOMEM; + desc = kzalloc_obj(*desc); + if (!desc) + goto out; + desc->file = file; + desc->ctx = ctx; + desc->folio_index_max = -1; + + spin_lock(&file->f_lock); + desc->dir_cookie = dir_ctx->dir_cookie; + desc->folio_index = dir_ctx->page_index; + desc->last_cookie = dir_ctx->last_cookie; + desc->attr_gencount = dir_ctx->attr_gencount; + desc->eof = dir_ctx->eof; + nfs_set_dtsize(desc, dir_ctx->dtsize); + memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf)); + cache_hits = atomic_xchg(&dir_ctx->cache_hits, 0); + cache_misses = atomic_xchg(&dir_ctx->cache_misses, 0); + force_clear = dir_ctx->force_clear; + spin_unlock(&file->f_lock); + + if (desc->eof) { + res = 0; + goto out_free; + } + + desc->plus = nfs_use_readdirplus(inode, ctx, cache_hits, cache_misses); + force_clear = nfs_readdir_handle_cache_misses(inode, desc, cache_misses, + force_clear); + desc->clear_cache = force_clear; + + do { + res = readdir_search_pagecache(desc); + + if (res == -EBADCOOKIE) { + res = 0; + /* This means either end of directory */ + if (desc->dir_cookie && !desc->eof) { + /* Or that the server has 'lost' a cookie */ + res = uncached_readdir(desc); + if (res == 0) + continue; + if (res == -EBADCOOKIE || res == -ENOTSYNC) + res = 0; + } + break; + } + if (res == -ETOOSMALL && desc->plus) { + nfs_zap_caches(inode); + desc->plus = false; + desc->eof = false; + continue; + } + if (res < 0) + break; + + nfs_do_filldir(desc, nfsi->cookieverf); + nfs_readdir_folio_unlock_and_put_cached(desc); + if (desc->folio_index == desc->folio_index_max) + desc->clear_cache = force_clear; + } while (!desc->eob && !desc->eof); + + spin_lock(&file->f_lock); + dir_ctx->dir_cookie = desc->dir_cookie; + dir_ctx->last_cookie = desc->last_cookie; + dir_ctx->attr_gencount = desc->attr_gencount; + dir_ctx->page_index = desc->folio_index; + dir_ctx->force_clear = force_clear; + dir_ctx->eof = desc->eof; + dir_ctx->dtsize = desc->dtsize; + memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf)); + spin_unlock(&file->f_lock); +out_free: + kfree(desc); + +out: + dfprintk(FILE, "NFS: readdir(%pD2) returns %d\n", file, res); + return res; +} + +static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence) +{ + struct nfs_open_dir_context *dir_ctx = filp->private_data; + + dfprintk(FILE, "NFS: llseek dir(%pD2, %lld, %d)\n", + filp, offset, whence); + + switch (whence) { + default: + return -EINVAL; + case SEEK_SET: + if (offset < 0) + return -EINVAL; + spin_lock(&filp->f_lock); + break; + case SEEK_CUR: + if (offset == 0) + return filp->f_pos; + spin_lock(&filp->f_lock); + offset += filp->f_pos; + if (offset < 0) { + spin_unlock(&filp->f_lock); + return -EINVAL; + } + } + if (offset != filp->f_pos) { + filp->f_pos = offset; + dir_ctx->page_index = 0; + if (!nfs_readdir_use_cookie(filp)) { + dir_ctx->dir_cookie = 0; + dir_ctx->last_cookie = 0; + } else { + dir_ctx->dir_cookie = offset; + dir_ctx->last_cookie = offset; + } + dir_ctx->eof = false; + } + spin_unlock(&filp->f_lock); + return offset; +} + +/* + * All directory operations under NFS are synchronous, so fsync() + * is a dummy operation. + */ +static int nfs_fsync_dir(struct file *filp, loff_t start, loff_t end, + int datasync) +{ + dfprintk(FILE, "NFS: fsync dir(%pD2) datasync %d\n", filp, datasync); + + nfs_inc_stats(file_inode(filp), NFSIOS_VFSFSYNC); + return 0; +} + +/** + * nfs_force_lookup_revalidate - Mark the directory as having changed + * @dir: pointer to directory inode + * + * This forces the revalidation code in nfs_lookup_revalidate() to do a + * full lookup on all child dentries of 'dir' whenever a change occurs + * on the server that might have invalidated our dcache. + * + * Note that we reserve bit '0' as a tag to let us know when a dentry + * was revalidated while holding a delegation on its inode. + * + * The caller should be holding dir->i_lock + */ +void nfs_force_lookup_revalidate(struct inode *dir) +{ + NFS_I(dir)->cache_change_attribute += 2; +} +EXPORT_SYMBOL_GPL(nfs_force_lookup_revalidate); + +/** + * nfs_verify_change_attribute - Detects NFS remote directory changes + * @dir: pointer to parent directory inode + * @verf: previously saved change attribute + * + * Return "false" if the verifiers doesn't match the change attribute. + * This would usually indicate that the directory contents have changed on + * the server, and that any dentries need revalidating. + */ +static bool nfs_verify_change_attribute(struct inode *dir, unsigned long verf) +{ + return (verf & ~1UL) == nfs_save_change_attribute(dir); +} + +static void nfs_set_verifier_delegated(unsigned long *verf) +{ + *verf |= 1UL; +} + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_unset_verifier_delegated(unsigned long *verf) +{ + *verf &= ~1UL; +} +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ + +static bool nfs_test_verifier_delegated(unsigned long verf) +{ + return verf & 1; +} + +static bool nfs_verifier_is_delegated(struct dentry *dentry) +{ + return nfs_test_verifier_delegated(dentry->d_time); +} + +static void nfs_set_verifier_locked(struct dentry *dentry, unsigned long verf) +{ + struct inode *inode = d_inode(dentry); + struct inode *dir = d_inode_rcu(dentry->d_parent); + + if (!dir || !nfs_verify_change_attribute(dir, verf)) + return; + if (NFS_PROTO(dir)->have_delegation(dir, FMODE_READ, 0) || + (inode && NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, 0))) + nfs_set_verifier_delegated(&verf); + dentry->d_time = verf; +} + +/** + * nfs_set_verifier - save a parent directory verifier in the dentry + * @dentry: pointer to dentry + * @verf: verifier to save + * + * Saves the parent directory verifier in @dentry. If the inode has + * a delegation, we also tag the dentry as having been revalidated + * while holding a delegation so that we know we don't have to + * look it up again after a directory change. + */ +void nfs_set_verifier(struct dentry *dentry, unsigned long verf) +{ + + spin_lock(&dentry->d_lock); + nfs_set_verifier_locked(dentry, verf); + spin_unlock(&dentry->d_lock); +} +EXPORT_SYMBOL_GPL(nfs_set_verifier); + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_clear_verifier_file(struct inode *inode) +{ + struct dentry *alias; + struct inode *dir; + + for_each_alias(alias, inode) { + spin_lock(&alias->d_lock); + dir = d_inode_rcu(alias->d_parent); + if (!dir || + !NFS_PROTO(dir)->have_delegation(dir, FMODE_READ, 0)) + nfs_unset_verifier_delegated(&alias->d_time); + spin_unlock(&alias->d_lock); + } +} + +static void nfs_clear_verifier_directory(struct inode *dir) +{ + struct dentry *this_parent; + struct dentry *dentry; + struct inode *inode; + + if (hlist_empty(&dir->i_dentry)) + return; + this_parent = + hlist_entry(dir->i_dentry.first, struct dentry, d_alias); + + spin_lock(&this_parent->d_lock); + nfs_unset_verifier_delegated(&this_parent->d_time); + dentry = d_first_child(this_parent); + hlist_for_each_entry_from(dentry, d_sib) { + if (unlikely(dentry->d_flags & DCACHE_DENTRY_CURSOR)) + continue; + inode = d_inode_rcu(dentry); + if (inode && + NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, 0)) + continue; + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + nfs_unset_verifier_delegated(&dentry->d_time); + spin_unlock(&dentry->d_lock); + } + spin_unlock(&this_parent->d_lock); +} + +/** + * nfs_clear_verifier_delegated - clear the dir verifier delegation tag + * @inode: pointer to inode + * + * Iterates through the dentries in the inode alias list and clears + * the tag used to indicate that the dentry has been revalidated + * while holding a delegation. + * This function is intended for use when the delegation is being + * returned or revoked. + */ +void nfs_clear_verifier_delegated(struct inode *inode) +{ + if (!inode) + return; + spin_lock(&inode->i_lock); + if (S_ISREG(inode->i_mode)) + nfs_clear_verifier_file(inode); + else if (S_ISDIR(inode->i_mode)) + nfs_clear_verifier_directory(inode); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_clear_verifier_delegated); +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ + +static int nfs_dentry_verify_change(struct inode *dir, struct dentry *dentry) +{ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE) && + d_really_is_negative(dentry)) + return dentry->d_time == inode_peek_iversion_raw(dir); + return nfs_verify_change_attribute(dir, dentry->d_time); +} + +/* + * A check for whether or not the parent directory has changed. + * In the case it has, we assume that the dentries are untrustworthy + * and may need to be looked up again. + * If rcu_walk prevents us from performing a full check, return 0. + */ +static int nfs_check_verifier(struct inode *dir, struct dentry *dentry, + int rcu_walk) +{ + if (IS_ROOT(dentry)) + return 1; + if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) + return 0; + if (!nfs_dentry_verify_change(dir, dentry)) + return 0; + + /* Revalidate nfsi->cache_change_attribute before we declare a match */ + if (nfs_mapping_need_revalidate_inode(dir)) { + if (rcu_walk) + return 0; + if (__nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0) + return 0; + } + if (!nfs_dentry_verify_change(dir, dentry)) + return 0; + return 1; +} + +/* + * Use intent information to check whether or not we're going to do + * an O_EXCL create using this path component. + */ +static int nfs_is_exclusive_create(struct inode *dir, unsigned int flags) +{ + if (NFS_PROTO(dir)->version == 2) + return 0; + return (flags & (LOOKUP_CREATE | LOOKUP_EXCL)) == + (LOOKUP_CREATE | LOOKUP_EXCL); +} + +/* + * Inode and filehandle revalidation for lookups. + * + * We force revalidation in the cases where the VFS sets LOOKUP_REVAL, + * or if the intent information indicates that we're about to open this + * particular file and the "nocto" mount flag is not set. + * + */ +static +int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags) +{ + struct nfs_server *server = NFS_SERVER(inode); + int ret; + + if (IS_AUTOMOUNT(inode)) + return 0; + + if (flags & LOOKUP_OPEN) { + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + /* A NFSv4 OPEN will revalidate later */ + if (server->caps & NFS_CAP_ATOMIC_OPEN) + goto out; + fallthrough; + case S_IFDIR: + if (server->flags & NFS_MOUNT_NOCTO) + break; + /* NFS close-to-open cache consistency validation */ + goto out_force; + } + } + + /* VFS wants an on-the-wire revalidation */ + if (flags & LOOKUP_REVAL) + goto out_force; +out: + if (inode->i_nlink > 0 || + (inode->i_nlink == 0 && + test_bit(NFS_INO_PRESERVE_UNLINKED, &NFS_I(inode)->flags))) + return 0; + else + return -ESTALE; +out_force: + if (flags & LOOKUP_RCU) + return -ECHILD; + ret = __nfs_revalidate_inode(server, inode); + if (ret != 0) + return ret; + goto out; +} + +static void nfs_mark_dir_for_revalidate(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE); + spin_unlock(&inode->i_lock); +} + +/* + * We judge how long we want to trust negative + * dentries by looking at the parent inode mtime. + * + * If parent mtime has changed, we revalidate, else we wait for a + * period corresponding to the parent's attribute cache timeout value. + * + * If LOOKUP_RCU prevents us from performing a full check, return 1 + * suggesting a reval is needed. + * + * Note that when creating a new file, or looking up a rename target, + * then it shouldn't be necessary to revalidate a negative dentry. + */ +static inline +int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET)) + return 0; + if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) + return 1; + /* Case insensitive server? Revalidate negative dentries */ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + return 1; + return !nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU); +} + +static int +nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, + struct inode *inode, int error) +{ + switch (error) { + case 1: + break; + case -ETIMEDOUT: + if (inode && (IS_ROOT(dentry) || + NFS_SERVER(inode)->flags & NFS_MOUNT_SOFTREVAL)) + error = 1; + break; + case -ESTALE: + case -ENOENT: + error = 0; + fallthrough; + default: + /* + * We can't d_drop the root of a disconnected tree: + * its d_hash is on the s_anon list and d_drop() would hide + * it from shrink_dcache_for_unmount(), leading to busy + * inodes on unmount and further oopses. + */ + if (inode && IS_ROOT(dentry)) + error = 1; + break; + } + trace_nfs_lookup_revalidate_exit(dir, dentry, 0, error); + return error; +} + +static int +nfs_lookup_revalidate_negative(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + int ret = 1; + if (nfs_neg_need_reval(dir, dentry, flags)) { + if (flags & LOOKUP_RCU) + return -ECHILD; + ret = 0; + } + return nfs_lookup_revalidate_done(dir, dentry, NULL, ret); +} + +static int +nfs_lookup_revalidate_delegated(struct inode *dir, struct dentry *dentry, + struct inode *inode) +{ + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + return nfs_lookup_revalidate_done(dir, dentry, inode, 1); +} + +static int nfs_lookup_revalidate_dentry(struct inode *dir, const struct qstr *name, + struct dentry *dentry, + struct inode *inode, unsigned int flags) +{ + struct nfs_fh *fhandle; + struct nfs_fattr *fattr; + unsigned long dir_verifier; + int ret; + + trace_nfs_lookup_revalidate_enter(dir, dentry, flags); + + ret = -ENOMEM; + fhandle = nfs_alloc_fhandle(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fhandle == NULL || fattr == NULL) + goto out; + + dir_verifier = nfs_save_change_attribute(dir); + ret = NFS_PROTO(dir)->lookup(dir, dentry, name, fhandle, fattr); + if (ret < 0) + goto out; + + /* Request help from readdirplus */ + nfs_lookup_advise_force_readdirplus(dir, flags); + + ret = 0; + if (nfs_compare_fh(NFS_FH(inode), fhandle)) + goto out; + if (nfs_refresh_inode(inode, fattr) < 0) + goto out; + + nfs_setsecurity(inode, fattr); + nfs_set_verifier(dentry, dir_verifier); + + ret = 1; +out: + nfs_free_fattr(fattr); + nfs_free_fhandle(fhandle); + + /* + * If the lookup failed despite the dentry change attribute being + * a match, then we should revalidate the directory cache. + */ + if (!ret && nfs_dentry_verify_change(dir, dentry)) + nfs_mark_dir_for_revalidate(dir); + return nfs_lookup_revalidate_done(dir, dentry, inode, ret); +} + +/* + * This is called every time the dcache has a lookup hit, + * and we should check whether we can really trust that + * lookup. + * + * NOTE! The hit can be a negative hit too, don't assume + * we have an inode! + * + * If the parent directory is seen to have changed, we throw out the + * cached dentry and do a new lookup. + */ +static int +nfs_do_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) +{ + struct inode *inode; + int error = 0; + + nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE); + inode = d_inode(dentry); + + if (!inode) + return nfs_lookup_revalidate_negative(dir, dentry, flags); + + if (is_bad_inode(inode)) { + dfprintk(LOOKUPCACHE, "%s: %pd2 has dud inode\n", + __func__, dentry); + goto out_bad; + } + + if ((flags & LOOKUP_RENAME_TARGET) && d_count(dentry) < 2 && + nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + goto out_bad; + + if (nfs_verifier_is_delegated(dentry)) + return nfs_lookup_revalidate_delegated(dir, dentry, inode); + + /* Force a full look up iff the parent directory has changed */ + if (!(flags & (LOOKUP_EXCL | LOOKUP_REVAL)) && + nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) { + error = nfs_lookup_verify_inode(inode, flags); + if (error) { + if (error == -ESTALE) + nfs_mark_dir_for_revalidate(dir); + goto out_bad; + } + goto out_valid; + } + + if (flags & LOOKUP_RCU) + return -ECHILD; + + if (NFS_STALE(inode)) + goto out_bad; + + return nfs_lookup_revalidate_dentry(dir, name, dentry, inode, flags); +out_valid: + return nfs_lookup_revalidate_done(dir, dentry, inode, 1); +out_bad: + if (flags & LOOKUP_RCU) + return -ECHILD; + return nfs_lookup_revalidate_done(dir, dentry, inode, error); +} + +static int +__nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) +{ + if (flags & LOOKUP_RCU) { + if (dentry->d_fsdata == NFS_FSDATA_BLOCKED) + return -ECHILD; + } else { + /* Wait for unlink to complete - see unblock_revalidate() */ + wait_var_event(&dentry->d_fsdata, + smp_load_acquire(&dentry->d_fsdata) + != NFS_FSDATA_BLOCKED); + } + return 0; +} + +static int nfs_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) +{ + if (__nfs_lookup_revalidate(dentry, flags)) + return -ECHILD; + return nfs_do_lookup_revalidate(dir, name, dentry, flags); +} + +static void block_revalidate(struct dentry *dentry) +{ + /* old devname - just in case */ + kfree(dentry->d_fsdata); + + /* Any new reference that could lead to an open + * will take ->d_lock in lookup_open() -> d_lookup(). + * Holding this lock ensures we cannot race with + * __nfs_lookup_revalidate() and removes and need + * for further barriers. + */ + lockdep_assert_held(&dentry->d_lock); + + dentry->d_fsdata = NFS_FSDATA_BLOCKED; +} + +static void unblock_revalidate(struct dentry *dentry) +{ + store_release_wake_up(&dentry->d_fsdata, NULL); +} + +/* + * A weaker form of d_revalidate for revalidating just the d_inode(dentry) + * when we don't really care about the dentry name. This is called when a + * pathwalk ends on a dentry that was not found via a normal lookup in the + * parent dir (e.g.: ".", "..", procfs symlinks or mountpoint traversals). + * + * In this situation, we just want to verify that the inode itself is OK + * since the dentry might have changed on the server. + */ +static int nfs_weak_revalidate(struct dentry *dentry, unsigned int flags) +{ + struct inode *inode = d_inode(dentry); + int error = 0; + + /* + * I believe we can only get a negative dentry here in the case of a + * procfs-style symlink. Just assume it's correct for now, but we may + * eventually need to do something more here. + */ + if (!inode) { + dfprintk(LOOKUPCACHE, "%s: %pd2 has negative inode\n", + __func__, dentry); + return 1; + } + + if (is_bad_inode(inode)) { + dfprintk(LOOKUPCACHE, "%s: %pd2 has dud inode\n", + __func__, dentry); + return 0; + } + + error = nfs_lookup_verify_inode(inode, flags); + dfprintk(LOOKUPCACHE, "NFS: %s: inode %llu is %s\n", + __func__, inode->i_ino, error ? "invalid" : "valid"); + return !error; +} + +/* + * This is called from dput() when d_count is going to 0. + */ +static int nfs_dentry_delete(const struct dentry *dentry) +{ + dfprintk(VFS, "NFS: dentry_delete(%pd2, %x)\n", + dentry, dentry->d_flags); + + /* Unhash any dentry with a stale inode */ + if (d_really_is_positive(dentry) && NFS_STALE(d_inode(dentry))) + return 1; + + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { + /* Unhash it, so that ->d_iput() would be called */ + return 1; + } + if (!(dentry->d_sb->s_flags & SB_ACTIVE)) { + /* Unhash it, so that ancestors of killed async unlink + * files will be cleaned up during umount */ + return 1; + } + return 0; + +} + +/* Ensure that we revalidate inode->i_nlink */ +static void nfs_drop_nlink(struct inode *inode, unsigned long gencount) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + spin_lock(&inode->i_lock); + /* drop the inode if we're reasonably sure this is the last link */ + if (inode->i_nlink > 0 && gencount == nfsi->attr_gencount) + drop_nlink(inode); + nfsi->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid( + inode, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_NLINK); + spin_unlock(&inode->i_lock); +} + +/* + * Called when the dentry loses inode. + * We use it to clean up silly-renamed files. + */ +static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) +{ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { + unsigned long gencount = READ_ONCE(NFS_I(inode)->attr_gencount); + nfs_complete_unlink(dentry, inode); + nfs_drop_nlink(inode, gencount); + } + iput(inode); +} + +static void nfs_d_release(struct dentry *dentry) +{ + /* free cached devname value, if it survived that far */ + if (unlikely(dentry->d_fsdata)) { + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + WARN_ON(1); + else + kfree(dentry->d_fsdata); + } +} + +const struct dentry_operations nfs_dentry_operations = { + .d_revalidate = nfs_lookup_revalidate, + .d_weak_revalidate = nfs_weak_revalidate, + .d_delete = nfs_dentry_delete, + .d_iput = nfs_dentry_iput, + .d_automount = nfs_d_automount, + .d_release = nfs_d_release, +}; +EXPORT_SYMBOL_GPL(nfs_dentry_operations); + +struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) +{ + struct dentry *res; + struct inode *inode = NULL; + struct nfs_fh *fhandle = NULL; + struct nfs_fattr *fattr = NULL; + unsigned long dir_verifier; + int error; + + dfprintk(VFS, "NFS: lookup(%pd2)\n", dentry); + nfs_inc_stats(dir, NFSIOS_VFSLOOKUP); + + if (unlikely(dentry->d_name.len > NFS_SERVER(dir)->namelen)) + return ERR_PTR(-ENAMETOOLONG); + + /* + * If we're doing an exclusive create, optimize away the lookup + * but don't hash the dentry. + */ + if (nfs_is_exclusive_create(dir, flags) || flags & LOOKUP_RENAME_TARGET) + return NULL; + + res = ERR_PTR(-ENOMEM); + fhandle = nfs_alloc_fhandle(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(dir)); + if (fhandle == NULL || fattr == NULL) + goto out; + + dir_verifier = nfs_save_change_attribute(dir); + trace_nfs_lookup_enter(dir, dentry, flags); + error = NFS_PROTO(dir)->lookup(dir, dentry, &dentry->d_name, + fhandle, fattr); + if (error == -ENOENT) { + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + dir_verifier = inode_peek_iversion_raw(dir); + goto no_entry; + } + if (error < 0) { + res = ERR_PTR(error); + goto out; + } + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); + res = ERR_CAST(inode); + if (IS_ERR(res)) + goto out; + + /* Notify readdir to use READDIRPLUS */ + nfs_lookup_advise_force_readdirplus(dir, flags); + +no_entry: + nfs_set_verifier(dentry, dir_verifier); + res = d_splice_alias(inode, dentry); + if (res != NULL) { + if (IS_ERR(res)) + goto out; + nfs_set_verifier(res, dir_verifier); + dentry = res; + } +out: + trace_nfs_lookup_exit(dir, dentry, flags, PTR_ERR_OR_ZERO(res)); + nfs_free_fattr(fattr); + nfs_free_fhandle(fhandle); + return res; +} +EXPORT_SYMBOL_GPL(nfs_lookup); + +void nfs_d_prune_case_insensitive_aliases(struct inode *inode) +{ + /* Case insensitive server? Revalidate dentries */ + if (inode && nfs_server_capable(inode, NFS_CAP_CASE_INSENSITIVE)) + d_prune_aliases(inode); +} +EXPORT_SYMBOL_GPL(nfs_d_prune_case_insensitive_aliases); + +#if IS_ENABLED(CONFIG_NFS_V4) +static int nfs4_lookup_revalidate(struct inode *, const struct qstr *, + struct dentry *, unsigned int); + +const struct dentry_operations nfs4_dentry_operations = { + .d_revalidate = nfs4_lookup_revalidate, + .d_weak_revalidate = nfs_weak_revalidate, + .d_delete = nfs_dentry_delete, + .d_iput = nfs_dentry_iput, + .d_automount = nfs_d_automount, + .d_release = nfs_d_release, +}; +EXPORT_SYMBOL_GPL(nfs4_dentry_operations); + +static struct nfs_open_context *create_nfs_open_context(struct dentry *dentry, int open_flags, struct file *filp) +{ + return alloc_nfs_open_context(dentry, flags_to_mode(open_flags), filp); +} + +static int do_open(struct inode *inode, struct file *filp) +{ + nfs_fscache_open_file(inode, filp); + return 0; +} + +static int nfs_finish_open(struct nfs_open_context *ctx, + struct dentry *dentry, + struct file *file, unsigned open_flags) +{ + int err; + + err = finish_open(file, dentry, do_open); + if (err) + goto out; + if (S_ISREG(file_inode(file)->i_mode)) + nfs_file_set_open_context(file, ctx); + else + err = -EOPENSTALE; +out: + return err; +} + +int nfs_atomic_open(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned open_flags, + umode_t mode) +{ + struct nfs_open_context *ctx; + struct dentry *res; + struct iattr attr = { .ia_valid = ATTR_OPEN }; + struct inode *inode; + unsigned int lookup_flags = 0; + unsigned long dir_verifier; + bool switched = false; + int created = 0; + int err; + + /* Expect a negative dentry */ + BUG_ON(d_inode(dentry)); + + dfprintk(VFS, "NFS: atomic_open(%s/%llu), %pd\n", + dir->i_sb->s_id, dir->i_ino, dentry); + + err = nfs_check_flags(open_flags); + if (err) + return err; + + /* NFS only supports OPEN on regular files */ + if ((open_flags & O_DIRECTORY)) { + if (!d_in_lookup(dentry)) { + /* + * Hashed negative dentry with O_DIRECTORY: dentry was + * revalidated and is fine, no need to perform lookup + * again + */ + return -ENOENT; + } + lookup_flags = LOOKUP_OPEN|LOOKUP_DIRECTORY; + goto no_open; + } + + if (dentry->d_name.len > NFS_SERVER(dir)->namelen) + return -ENAMETOOLONG; + + if (open_flags & O_CREAT) { + struct nfs_server *server = NFS_SERVER(dir); + + if (!(server->attr_bitmask[2] & FATTR4_WORD2_MODE_UMASK)) + mode &= ~current_umask(); + + attr.ia_valid |= ATTR_MODE; + attr.ia_mode = mode; + } + if (open_flags & O_TRUNC) { + attr.ia_valid |= ATTR_SIZE; + attr.ia_size = 0; + } + + if (!(open_flags & O_CREAT) && !d_in_lookup(dentry)) { + d_drop(dentry); + switched = true; + dentry = d_alloc_parallel(dentry->d_parent, + &dentry->d_name); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + if (unlikely(!d_in_lookup(dentry))) + return finish_no_open(file, dentry); + } + + ctx = create_nfs_open_context(dentry, open_flags, file); + err = PTR_ERR(ctx); + if (IS_ERR(ctx)) + goto out; + + trace_nfs_atomic_open_enter(dir, ctx, open_flags); + inode = NFS_PROTO(dir)->open_context(dir, ctx, open_flags, &attr, &created); + if (created) + file->f_mode |= FMODE_CREATED; + if (IS_ERR(inode)) { + err = PTR_ERR(inode); + trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); + put_nfs_open_context(ctx); + d_drop(dentry); + switch (err) { + case -ENOENT: + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + dir_verifier = inode_peek_iversion_raw(dir); + else + dir_verifier = nfs_save_change_attribute(dir); + nfs_set_verifier(dentry, dir_verifier); + d_splice_alias(NULL, dentry); + break; + case -EISDIR: + case -ENOTDIR: + if (open_flags & __O_REGULAR) { + err = -EFTYPE; + break; + } + goto no_open; + case -ELOOP: + if (!(open_flags & O_NOFOLLOW)) + goto no_open; + break; + /* case -EINVAL: */ + default: + break; + } + goto out; + } + file->f_mode |= FMODE_CAN_ODIRECT; + + err = nfs_finish_open(ctx, ctx->dentry, file, open_flags); + trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); + put_nfs_open_context(ctx); +out: + if (unlikely(switched)) { + d_lookup_done(dentry); + dput(dentry); + } + return err; + +no_open: + res = nfs_lookup(dir, dentry, lookup_flags); + if (!res) { + inode = d_inode(dentry); + if ((lookup_flags & LOOKUP_DIRECTORY) && inode && + !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) + res = ERR_PTR(-ENOTDIR); + else if (inode && S_ISREG(inode->i_mode)) + res = ERR_PTR(-EOPENSTALE); + } else if (!IS_ERR(res)) { + inode = d_inode(res); + if ((lookup_flags & LOOKUP_DIRECTORY) && inode && + !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) { + dput(res); + res = ERR_PTR(-ENOTDIR); + } else if (inode && S_ISREG(inode->i_mode)) { + dput(res); + res = ERR_PTR(-EOPENSTALE); + } + } + if (switched) { + d_lookup_done(dentry); + if (!res) + res = dentry; + else + dput(dentry); + } + return finish_no_open(file, res); +} +EXPORT_SYMBOL_GPL(nfs_atomic_open); + +static int +nfs4_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) +{ + struct inode *inode; + + if (__nfs_lookup_revalidate(dentry, flags)) + return -ECHILD; + + trace_nfs_lookup_revalidate_enter(dir, dentry, flags); + + if (!(flags & LOOKUP_OPEN) || (flags & LOOKUP_DIRECTORY)) + goto full_reval; + if (d_mountpoint(dentry)) + goto full_reval; + + inode = d_inode(dentry); + + /* We can't create new files in nfs_open_revalidate(), so we + * optimize away revalidation of negative dentries. + */ + if (inode == NULL) + goto full_reval; + + if (nfs_verifier_is_delegated(dentry) || + nfs_have_directory_delegation(inode)) + return nfs_lookup_revalidate_delegated(dir, dentry, inode); + + /* NFS only supports OPEN on regular files */ + if (!S_ISREG(inode->i_mode)) + goto full_reval; + + /* We cannot do exclusive creation on a positive dentry */ + if (flags & (LOOKUP_EXCL | LOOKUP_REVAL)) + goto reval_dentry; + + /* Check if the directory changed */ + if (!nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) + goto reval_dentry; + + /* Let f_op->open() actually open (and revalidate) the file */ + return 1; +reval_dentry: + if (flags & LOOKUP_RCU) + return -ECHILD; + return nfs_lookup_revalidate_dentry(dir, name, dentry, inode, flags); + +full_reval: + return nfs_do_lookup_revalidate(dir, name, dentry, flags); +} + +#endif /* CONFIG_NFS_V4 */ + +int nfs_atomic_open_v23(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned int open_flags, + umode_t mode) +{ + struct dentry *res = NULL; + /* Same as look+open from lookup_open(), but with different O_TRUNC + * handling. + */ + int error = 0; + + if (dentry->d_name.len > NFS_SERVER(dir)->namelen) + return -ENAMETOOLONG; + + if (open_flags & O_CREAT) { + error = nfs_do_create(dir, dentry, mode, open_flags); + if (!error) { + file->f_mode |= FMODE_CREATED; + return finish_open(file, dentry, NULL); + } else if (error != -EEXIST || open_flags & O_EXCL) + return error; + } + if (d_in_lookup(dentry)) { + /* The only flags nfs_lookup considers are + * LOOKUP_EXCL and LOOKUP_RENAME_TARGET, and + * we want those to be zero so the lookup isn't skipped. + */ + res = nfs_lookup(dir, dentry, 0); + } + return finish_no_open(file, res); + +} +EXPORT_SYMBOL_GPL(nfs_atomic_open_v23); + +struct dentry * +nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct dentry *parent = dget_parent(dentry); + struct inode *dir = d_inode(parent); + struct inode *inode; + struct dentry *d; + int error; + + d_drop(dentry); + + if (fhandle->size == 0) { + error = NFS_PROTO(dir)->lookup(dir, dentry, &dentry->d_name, + fhandle, fattr); + if (error) + goto out_error; + } + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + if (!(fattr->valid & NFS_ATTR_FATTR)) { + struct nfs_server *server = NFS_SB(dentry->d_sb); + error = server->nfs_client->rpc_ops->getattr(server, fhandle, + fattr, NULL); + if (error < 0) + goto out_error; + } + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); + d = d_splice_alias(inode, dentry); +out: + dput(parent); + return d; +out_error: + d = ERR_PTR(error); + goto out; +} +EXPORT_SYMBOL_GPL(nfs_add_or_obtain); + +/* + * Code common to create, mkdir, and mknod. + */ +int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct dentry *d; + + d = nfs_add_or_obtain(dentry, fhandle, fattr); + if (IS_ERR(d)) + return PTR_ERR(d); + + /* Callers don't care */ + dput(d); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_instantiate); + +/* + * Following a failed create operation, we drop the dentry rather + * than retain a negative dentry. This avoids a problem in the event + * that the operation succeeded on the server, but an error in the + * reply path made it appear to have failed. + */ +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags) +{ + struct iattr attr; + int error; + + open_flags |= O_CREAT; + + dfprintk(VFS, "NFS: create(%s/%llu), %pd\n", + dir->i_sb->s_id, dir->i_ino, dentry); + + attr.ia_mode = mode; + attr.ia_valid = ATTR_MODE; + if (open_flags & O_TRUNC) { + attr.ia_size = 0; + attr.ia_valid |= ATTR_SIZE; + } + + trace_nfs_create_enter(dir, dentry, open_flags); + error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags); + trace_nfs_create_exit(dir, dentry, open_flags, error); + if (error != 0) + goto out_err; + return 0; +out_err: + d_drop(dentry); + return error; +} + +int nfs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode) +{ + return nfs_do_create(dir, dentry, mode, O_EXCL); +} +EXPORT_SYMBOL_GPL(nfs_create); + +/* + * See comments for nfs_proc_create regarding failed operations. + */ +int +nfs_mknod(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) +{ + struct iattr attr; + int status; + + dfprintk(VFS, "NFS: mknod(%s/%llu), %pd\n", + dir->i_sb->s_id, dir->i_ino, dentry); + + attr.ia_mode = mode; + attr.ia_valid = ATTR_MODE; + + trace_nfs_mknod_enter(dir, dentry); + status = NFS_PROTO(dir)->mknod(dir, dentry, &attr, rdev); + trace_nfs_mknod_exit(dir, dentry, status); + if (status != 0) + goto out_err; + return 0; +out_err: + d_drop(dentry); + return status; +} +EXPORT_SYMBOL_GPL(nfs_mknod); + +/* + * See comments for nfs_proc_create regarding failed operations. + */ +struct dentry *nfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode) +{ + struct iattr attr; + struct dentry *ret; + + dfprintk(VFS, "NFS: mkdir(%s/%llu), %pd\n", + dir->i_sb->s_id, dir->i_ino, dentry); + + attr.ia_valid = ATTR_MODE; + attr.ia_mode = mode; + + trace_nfs_mkdir_enter(dir, dentry); + ret = NFS_PROTO(dir)->mkdir(dir, dentry, &attr); + trace_nfs_mkdir_exit(dir, dentry, PTR_ERR_OR_ZERO(ret)); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_mkdir); + +static void nfs_dentry_handle_enoent(struct dentry *dentry) +{ + if (simple_positive(dentry)) + d_delete(dentry); +} + +static void nfs_dentry_remove_handle_error(struct inode *dir, + struct dentry *dentry, int error) +{ + switch (error) { + case -ENOENT: + if (d_really_is_positive(dentry)) + d_delete(dentry); + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + break; + case 0: + nfs_d_prune_case_insensitive_aliases(d_inode(dentry)); + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + } +} + +int nfs_rmdir(struct inode *dir, struct dentry *dentry) +{ + int error; + + dfprintk(VFS, "NFS: rmdir(%s/%llu), %pd\n", + dir->i_sb->s_id, dir->i_ino, dentry); + + trace_nfs_rmdir_enter(dir, dentry); + if (d_really_is_positive(dentry)) { + down_write(&NFS_I(d_inode(dentry))->rmdir_sem); + error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + /* Ensure the VFS deletes this inode */ + switch (error) { + case 0: + clear_nlink(d_inode(dentry)); + break; + case -ENOENT: + nfs_dentry_handle_enoent(dentry); + } + up_write(&NFS_I(d_inode(dentry))->rmdir_sem); + } else + error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + nfs_dentry_remove_handle_error(dir, dentry, error); + trace_nfs_rmdir_exit(dir, dentry, error); + + return error; +} +EXPORT_SYMBOL_GPL(nfs_rmdir); + +/* + * Remove a file after making sure there are no pending writes, + * and after checking that the file has only one user. + * + * We invalidate the attribute cache and free the inode prior to the operation + * to avoid possible races if the server reuses the inode. + */ +static int nfs_safe_remove(struct dentry *dentry) +{ + struct inode *dir = d_inode(dentry->d_parent); + struct inode *inode = d_inode(dentry); + int error = -EBUSY; + + dfprintk(VFS, "NFS: safe_remove(%pd2)\n", dentry); + + /* If the dentry was sillyrenamed, we simply call d_delete() */ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { + error = 0; + goto out; + } + + trace_nfs_remove_enter(dir, dentry); + if (inode != NULL) { + unsigned long gencount = READ_ONCE(NFS_I(inode)->attr_gencount); + + error = NFS_PROTO(dir)->remove(dir, dentry); + if (error == 0) + nfs_drop_nlink(inode, gencount); + } else + error = NFS_PROTO(dir)->remove(dir, dentry); + if (error == -ENOENT) + nfs_dentry_handle_enoent(dentry); + trace_nfs_remove_exit(dir, dentry, error); +out: + return error; +} + +/* We do silly rename. In case sillyrename() returns -EBUSY, the inode + * belongs to an active ".nfs..." file and we return -EBUSY. + * + * If sillyrename() returns 0, we do nothing, otherwise we unlink. + */ +int nfs_unlink(struct inode *dir, struct dentry *dentry) +{ + int error; + + dfprintk(VFS, "NFS: unlink(%s/%llu, %pd)\n", dir->i_sb->s_id, + dir->i_ino, dentry); + + trace_nfs_unlink_enter(dir, dentry); + spin_lock(&dentry->d_lock); + if (d_count(dentry) > 1 && !test_bit(NFS_INO_PRESERVE_UNLINKED, + &NFS_I(d_inode(dentry))->flags)) { + spin_unlock(&dentry->d_lock); + /* Start asynchronous writeout of the inode */ + write_inode_now(d_inode(dentry), 0); + error = nfs_sillyrename(dir, dentry); + goto out; + } + /* We must prevent any concurrent open until the unlink + * completes. ->d_revalidate will wait for ->d_fsdata + * to clear. We set it here to ensure no lookup succeeds until + * the unlink is complete on the server. + */ + error = -ETXTBSY; + if (WARN_ON(dentry->d_flags & DCACHE_NFSFS_RENAMED) || + WARN_ON(dentry->d_fsdata == NFS_FSDATA_BLOCKED)) { + spin_unlock(&dentry->d_lock); + goto out; + } + block_revalidate(dentry); + + spin_unlock(&dentry->d_lock); + error = nfs_safe_remove(dentry); + nfs_dentry_remove_handle_error(dir, dentry, error); + unblock_revalidate(dentry); +out: + trace_nfs_unlink_exit(dir, dentry, error); + return error; +} +EXPORT_SYMBOL_GPL(nfs_unlink); + +/* + * To create a symbolic link, most file systems instantiate a new inode, + * add a page to it containing the path, then write it out to the disk + * using prepare_write/commit_write. + * + * Unfortunately the NFS client can't create the in-core inode first + * because it needs a file handle to create an in-core inode (see + * fs/nfs/inode.c:nfs_fhget). We only have a file handle *after* the + * symlink request has completed on the server. + * + * So instead we allocate a raw page, copy the symname into it, then do + * the SYMLINK request with the page as the buffer. If it succeeds, we + * now have a new file handle and can instantiate an in-core NFS inode + * and move the raw page into its mapping. + */ +int nfs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *symname) +{ + struct folio *folio; + char *kaddr; + struct iattr attr; + unsigned int pathlen = strlen(symname); + int error; + + dfprintk(VFS, "NFS: symlink(%s/%llu, %pd, %s)\n", dir->i_sb->s_id, + dir->i_ino, dentry, symname); + + if (pathlen > PAGE_SIZE) + return -ENAMETOOLONG; + + attr.ia_mode = S_IFLNK | S_IRWXUGO; + attr.ia_valid = ATTR_MODE; + + folio = folio_alloc(GFP_USER, 0); + if (!folio) + return -ENOMEM; + + kaddr = folio_address(folio); + memcpy(kaddr, symname, pathlen); + if (pathlen < PAGE_SIZE) + memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen); + + trace_nfs_symlink_enter(dir, dentry); + error = NFS_PROTO(dir)->symlink(dir, dentry, folio, pathlen, &attr); + trace_nfs_symlink_exit(dir, dentry, error); + if (error != 0) { + dfprintk(VFS, "NFS: symlink(%s/%llu, %pd, %s) error %d\n", + dir->i_sb->s_id, dir->i_ino, + dentry, symname, error); + d_drop(dentry); + folio_put(folio); + return error; + } + + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + + /* + * No big deal if we can't add this page to the page cache here. + * READLINK will get the missing page from the server if needed. + */ + if (filemap_add_folio(d_inode(dentry)->i_mapping, folio, 0, + GFP_KERNEL) == 0) { + folio_mark_uptodate(folio); + folio_unlock(folio); + } + + folio_put(folio); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_symlink); + +int +nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) +{ + struct inode *inode = d_inode(old_dentry); + int error; + + dfprintk(VFS, "NFS: link(%pd2 -> %pd2)\n", + old_dentry, dentry); + + trace_nfs_link_enter(inode, dir, dentry); + d_drop(dentry); + if (S_ISREG(inode->i_mode)) + nfs_sync_inode(inode); + error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name); + if (error == 0) { + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + ihold(inode); + d_add(dentry, inode); + } + trace_nfs_link_exit(inode, dir, dentry, error); + return error; +} +EXPORT_SYMBOL_GPL(nfs_link); + +static void +nfs_unblock_rename(struct rpc_task *task, struct nfs_renamedata *data) +{ + struct dentry *new_dentry = data->new_dentry; + + unblock_revalidate(new_dentry); +} + +static bool nfs_rename_is_unsafe_cross_dir(struct dentry *old_dentry, + struct dentry *new_dentry) +{ + struct nfs_server *server = NFS_SB(old_dentry->d_sb); + + if (old_dentry->d_parent != new_dentry->d_parent) + return false; + if (server->fh_expire_type & NFS_FH_RENAME_UNSAFE) + return !(server->fh_expire_type & NFS_FH_NOEXPIRE_WITH_OPEN); + return true; +} + +/* + * RENAME + * FIXME: Some nfsds, like the Linux user space nfsd, may generate a + * different file handle for the same inode after a rename (e.g. when + * moving to a different directory). A fail-safe method to do so would + * be to look up old_dir/old_name, create a link to new_dir/new_name and + * rename the old file using the sillyrename stuff. This way, the original + * file in old_dir will go away when the last process iput()s the inode. + * + * FIXED. + * + * It actually works quite well. One needs to have the possibility for + * at least one ".nfs..." file in each directory the file ever gets + * moved or linked to which happens automagically with the new + * implementation that only depends on the dcache stuff instead of + * using the inode layer + * + * Unfortunately, things are a little more complicated than indicated + * above. For a cross-directory move, we want to make sure we can get + * rid of the old inode after the operation. This means there must be + * no pending writes (if it's a file), and the use count must be 1. + * If these conditions are met, we can drop the dentries before doing + * the rename. + */ +int nfs_rename(struct mnt_idmap *idmap, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) +{ + struct inode *old_inode = d_inode(old_dentry); + struct inode *new_inode = d_inode(new_dentry); + unsigned long new_gencount = 0; + struct dentry *dentry = NULL; + struct rpc_task *task; + bool must_unblock = false; + int error = -EBUSY; + + if (flags) + return -EINVAL; + + dfprintk(VFS, "NFS: rename(%pd2 -> %pd2, ct=%d)\n", + old_dentry, new_dentry, + d_count(new_dentry)); + + trace_nfs_rename_enter(old_dir, old_dentry, new_dir, new_dentry); + /* + * For non-directories, check whether the target is busy and if so, + * make a copy of the dentry and then do a silly-rename. If the + * silly-rename succeeds, the copied dentry is hashed and becomes + * the new target. + */ + if (new_inode && !S_ISDIR(new_inode->i_mode)) { + /* We must prevent any concurrent open until the unlink + * completes. ->d_revalidate will wait for ->d_fsdata + * to clear. We set it here to ensure no lookup succeeds until + * the unlink is complete on the server. + */ + error = -ETXTBSY; + if (WARN_ON(new_dentry->d_flags & DCACHE_NFSFS_RENAMED) || + WARN_ON(new_dentry->d_fsdata == NFS_FSDATA_BLOCKED)) + goto out; + + spin_lock(&new_dentry->d_lock); + if (d_count(new_dentry) > 2) { + int err; + + spin_unlock(&new_dentry->d_lock); + + /* copy the target dentry's name */ + dentry = d_alloc(new_dentry->d_parent, + &new_dentry->d_name); + if (!dentry) + goto out; + + /* silly-rename the existing target ... */ + err = nfs_sillyrename(new_dir, new_dentry); + if (err) + goto out; + + new_dentry = dentry; + new_inode = NULL; + } else { + block_revalidate(new_dentry); + must_unblock = true; + new_gencount = NFS_I(new_inode)->attr_gencount; + spin_unlock(&new_dentry->d_lock); + } + + } + + if (S_ISREG(old_inode->i_mode) && + nfs_rename_is_unsafe_cross_dir(old_dentry, new_dentry)) + nfs_sync_inode(old_inode); + task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, + must_unblock ? nfs_unblock_rename : NULL); + if (IS_ERR(task)) { + if (must_unblock) + unblock_revalidate(new_dentry); + error = PTR_ERR(task); + goto out; + } + + error = rpc_wait_for_completion_task(task); + if (error != 0) { + ((struct nfs_renamedata *)task->tk_calldata)->cancelled = 1; + /* Paired with the atomic_dec_and_test() barrier in rpc_do_put_task() */ + smp_wmb(); + } else + error = task->tk_status; + rpc_put_task(task); + /* Ensure the inode attributes are revalidated */ + if (error == 0) { + spin_lock(&old_inode->i_lock); + NFS_I(old_inode)->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid(old_inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); + spin_unlock(&old_inode->i_lock); + } +out: + trace_nfs_rename_exit(old_dir, old_dentry, + new_dir, new_dentry, error); + if (!error) { + if (new_inode != NULL) + nfs_drop_nlink(new_inode, new_gencount); + /* + * The d_move() should be here instead of in an async RPC completion + * handler because we need the proper locks to move the dentry. If + * we're interrupted by a signal, the async RPC completion handler + * should mark the directories for revalidation. + */ + d_move(old_dentry, new_dentry); + nfs_set_verifier(old_dentry, + nfs_save_change_attribute(new_dir)); + } else if (error == -ENOENT) + nfs_dentry_handle_enoent(old_dentry); + + /* new dentry created? */ + if (dentry) + dput(dentry); + return error; +} +EXPORT_SYMBOL_GPL(nfs_rename); + +static DEFINE_SPINLOCK(nfs_access_lru_lock); +static LIST_HEAD(nfs_access_lru_list); +static atomic_long_t nfs_access_nr_entries; + +static unsigned long nfs_access_max_cachesize = 4*1024*1024; +module_param(nfs_access_max_cachesize, ulong, 0644); +MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length"); + +static void nfs_access_free_entry(struct nfs_access_entry *entry) +{ + put_group_info(entry->group_info); + kfree_rcu(entry, rcu_head); + smp_mb__before_atomic(); + atomic_long_dec(&nfs_access_nr_entries); + smp_mb__after_atomic(); +} + +static void nfs_access_free_list(struct list_head *head) +{ + struct nfs_access_entry *cache; + + while (!list_empty(head)) { + cache = list_entry(head->next, struct nfs_access_entry, lru); + list_del(&cache->lru); + nfs_access_free_entry(cache); + } +} + +static unsigned long +nfs_do_access_cache_scan(unsigned int nr_to_scan) +{ + LIST_HEAD(head); + struct nfs_inode *nfsi, *next; + struct nfs_access_entry *cache; + long freed = 0; + + spin_lock(&nfs_access_lru_lock); + list_for_each_entry_safe(nfsi, next, &nfs_access_lru_list, access_cache_inode_lru) { + struct inode *inode; + + if (nr_to_scan-- == 0) + break; + inode = &nfsi->vfs_inode; + spin_lock(&inode->i_lock); + if (list_empty(&nfsi->access_cache_entry_lru)) + goto remove_lru_entry; + cache = list_entry(nfsi->access_cache_entry_lru.next, + struct nfs_access_entry, lru); + list_move(&cache->lru, &head); + rb_erase(&cache->rb_node, &nfsi->access_cache); + freed++; + if (!list_empty(&nfsi->access_cache_entry_lru)) + list_move_tail(&nfsi->access_cache_inode_lru, + &nfs_access_lru_list); + else { +remove_lru_entry: + list_del_init(&nfsi->access_cache_inode_lru); + smp_mb__before_atomic(); + clear_bit(NFS_INO_ACL_LRU_SET, &nfsi->flags); + smp_mb__after_atomic(); + } + spin_unlock(&inode->i_lock); + } + spin_unlock(&nfs_access_lru_lock); + nfs_access_free_list(&head); + return freed; +} + +unsigned long +nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc) +{ + int nr_to_scan = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; + + if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL) + return SHRINK_STOP; + return nfs_do_access_cache_scan(nr_to_scan); +} + + +unsigned long +nfs_access_cache_count(struct shrinker *shrink, struct shrink_control *sc) +{ + return vfs_pressure_ratio(atomic_long_read(&nfs_access_nr_entries)); +} + +static void +nfs_access_cache_enforce_limit(void) +{ + long nr_entries = atomic_long_read(&nfs_access_nr_entries); + unsigned long diff; + unsigned int nr_to_scan; + + if (nr_entries < 0 || nr_entries <= nfs_access_max_cachesize) + return; + nr_to_scan = 100; + diff = nr_entries - nfs_access_max_cachesize; + if (diff < nr_to_scan) + nr_to_scan = diff; + nfs_do_access_cache_scan(nr_to_scan); +} + +static void __nfs_access_zap_cache(struct nfs_inode *nfsi, struct list_head *head) +{ + struct rb_root *root_node = &nfsi->access_cache; + struct rb_node *n; + struct nfs_access_entry *entry; + + /* Unhook entries from the cache */ + while ((n = rb_first(root_node)) != NULL) { + entry = rb_entry(n, struct nfs_access_entry, rb_node); + rb_erase(n, root_node); + list_move(&entry->lru, head); + } + nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS; +} + +void nfs_access_zap_cache(struct inode *inode) +{ + LIST_HEAD(head); + + if (test_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags) == 0) + return; + /* Remove from global LRU init */ + spin_lock(&nfs_access_lru_lock); + if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) + list_del_init(&NFS_I(inode)->access_cache_inode_lru); + + spin_lock(&inode->i_lock); + __nfs_access_zap_cache(NFS_I(inode), &head); + spin_unlock(&inode->i_lock); + spin_unlock(&nfs_access_lru_lock); + nfs_access_free_list(&head); +} +EXPORT_SYMBOL_GPL(nfs_access_zap_cache); + +static int access_cmp(const struct cred *a, const struct nfs_access_entry *b) +{ + struct group_info *ga, *gb; + int g; + + if (uid_lt(a->fsuid, b->fsuid)) + return -1; + if (uid_gt(a->fsuid, b->fsuid)) + return 1; + + if (gid_lt(a->fsgid, b->fsgid)) + return -1; + if (gid_gt(a->fsgid, b->fsgid)) + return 1; + + ga = a->group_info; + gb = b->group_info; + if (ga == gb) + return 0; + if (ga == NULL) + return -1; + if (gb == NULL) + return 1; + if (ga->ngroups < gb->ngroups) + return -1; + if (ga->ngroups > gb->ngroups) + return 1; + + for (g = 0; g < ga->ngroups; g++) { + if (gid_lt(ga->gid[g], gb->gid[g])) + return -1; + if (gid_gt(ga->gid[g], gb->gid[g])) + return 1; + } + return 0; +} + +static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, const struct cred *cred) +{ + struct rb_node *n = NFS_I(inode)->access_cache.rb_node; + + while (n != NULL) { + struct nfs_access_entry *entry = + rb_entry(n, struct nfs_access_entry, rb_node); + int cmp = access_cmp(cred, entry); + + if (cmp < 0) + n = n->rb_left; + else if (cmp > 0) + n = n->rb_right; + else + return entry; + } + return NULL; +} + +static u64 nfs_access_login_time(const struct task_struct *task, + const struct cred *cred) +{ + const struct task_struct *parent; + const struct cred *pcred; + u64 ret; + + rcu_read_lock(); + for (;;) { + parent = rcu_dereference(task->real_parent); + pcred = __task_cred(parent); + if (parent == task || cred_fscmp(pcred, cred) != 0) + break; + task = parent; + } + ret = task->start_time; + rcu_read_unlock(); + return ret; +} + +static int nfs_access_get_cached_locked(struct inode *inode, const struct cred *cred, u32 *mask, bool may_block) +{ + struct nfs_inode *nfsi = NFS_I(inode); + u64 login_time = nfs_access_login_time(current, cred); + struct nfs_access_entry *cache; + bool retry = true; + int err; + + spin_lock(&inode->i_lock); + for(;;) { + if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) + goto out_zap; + cache = nfs_access_search_rbtree(inode, cred); + err = -ENOENT; + if (cache == NULL) + goto out; + /* Found an entry, is our attribute cache valid? */ + if (!nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) + break; + if (!retry) + break; + err = -ECHILD; + if (!may_block) + goto out; + spin_unlock(&inode->i_lock); + err = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (err) + return err; + spin_lock(&inode->i_lock); + retry = false; + } + err = -ENOENT; + if ((s64)(login_time - cache->timestamp) > 0) + goto out; + *mask = cache->mask; + list_move_tail(&cache->lru, &nfsi->access_cache_entry_lru); + err = 0; +out: + spin_unlock(&inode->i_lock); + return err; +out_zap: + spin_unlock(&inode->i_lock); + nfs_access_zap_cache(inode); + return -ENOENT; +} + +static int nfs_access_get_cached_rcu(struct inode *inode, const struct cred *cred, u32 *mask) +{ + /* Only check the most recently returned cache entry, + * but do it without locking. + */ + struct nfs_inode *nfsi = NFS_I(inode); + u64 login_time = nfs_access_login_time(current, cred); + struct nfs_access_entry *cache; + int err = -ECHILD; + struct list_head *lh; + + rcu_read_lock(); + if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) + goto out; + lh = rcu_dereference(list_tail_rcu(&nfsi->access_cache_entry_lru)); + cache = list_entry(lh, struct nfs_access_entry, lru); + if (lh == &nfsi->access_cache_entry_lru || + access_cmp(cred, cache) != 0) + cache = NULL; + if (cache == NULL) + goto out; + if ((s64)(login_time - cache->timestamp) > 0) + goto out; + if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) + goto out; + *mask = cache->mask; + err = 0; +out: + rcu_read_unlock(); + return err; +} + +int nfs_access_get_cached(struct inode *inode, const struct cred *cred, + u32 *mask, bool may_block) +{ + int status; + + status = nfs_access_get_cached_rcu(inode, cred, mask); + if (status != 0) + status = nfs_access_get_cached_locked(inode, cred, mask, + may_block); + + return status; +} +EXPORT_SYMBOL_GPL(nfs_access_get_cached); + +static void nfs_access_add_rbtree(struct inode *inode, + struct nfs_access_entry *set, + const struct cred *cred) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct rb_root *root_node = &nfsi->access_cache; + struct rb_node **p = &root_node->rb_node; + struct rb_node *parent = NULL; + struct nfs_access_entry *entry; + int cmp; + + spin_lock(&inode->i_lock); + while (*p != NULL) { + parent = *p; + entry = rb_entry(parent, struct nfs_access_entry, rb_node); + cmp = access_cmp(cred, entry); + + if (cmp < 0) + p = &parent->rb_left; + else if (cmp > 0) + p = &parent->rb_right; + else + goto found; + } + rb_link_node(&set->rb_node, parent, p); + rb_insert_color(&set->rb_node, root_node); + list_add_tail(&set->lru, &nfsi->access_cache_entry_lru); + spin_unlock(&inode->i_lock); + return; +found: + rb_replace_node(parent, &set->rb_node, root_node); + list_add_tail(&set->lru, &nfsi->access_cache_entry_lru); + list_del(&entry->lru); + spin_unlock(&inode->i_lock); + nfs_access_free_entry(entry); +} + +void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set, + const struct cred *cred) +{ + struct nfs_access_entry *cache = kmalloc_obj(*cache); + if (cache == NULL) + return; + RB_CLEAR_NODE(&cache->rb_node); + cache->fsuid = cred->fsuid; + cache->fsgid = cred->fsgid; + cache->group_info = get_group_info(cred->group_info); + cache->mask = set->mask; + cache->timestamp = ktime_get_ns(); + + /* The above field assignments must be visible + * before this item appears on the lru. We cannot easily + * use rcu_assign_pointer, so just force the memory barrier. + */ + smp_wmb(); + nfs_access_add_rbtree(inode, cache, cred); + + /* Update accounting */ + smp_mb__before_atomic(); + atomic_long_inc(&nfs_access_nr_entries); + smp_mb__after_atomic(); + + /* Add inode to global LRU list */ + if (!test_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) { + spin_lock(&nfs_access_lru_lock); + if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) + list_add_tail(&NFS_I(inode)->access_cache_inode_lru, + &nfs_access_lru_list); + spin_unlock(&nfs_access_lru_lock); + } + nfs_access_cache_enforce_limit(); +} +EXPORT_SYMBOL_GPL(nfs_access_add_cache); + +#define NFS_MAY_READ (NFS_ACCESS_READ) +#define NFS_MAY_WRITE (NFS_ACCESS_MODIFY | \ + NFS_ACCESS_EXTEND | \ + NFS_ACCESS_DELETE) +#define NFS_FILE_MAY_WRITE (NFS_ACCESS_MODIFY | \ + NFS_ACCESS_EXTEND) +#define NFS_DIR_MAY_WRITE NFS_MAY_WRITE +#define NFS_MAY_LOOKUP (NFS_ACCESS_LOOKUP) +#define NFS_MAY_EXECUTE (NFS_ACCESS_EXECUTE) +static int +nfs_access_calc_mask(u32 access_result, umode_t umode) +{ + int mask = 0; + + if (access_result & NFS_MAY_READ) + mask |= MAY_READ; + if (S_ISDIR(umode)) { + if ((access_result & NFS_DIR_MAY_WRITE) == NFS_DIR_MAY_WRITE) + mask |= MAY_WRITE; + if ((access_result & NFS_MAY_LOOKUP) == NFS_MAY_LOOKUP) + mask |= MAY_EXEC; + } else if (S_ISREG(umode)) { + if ((access_result & NFS_FILE_MAY_WRITE) == NFS_FILE_MAY_WRITE) + mask |= MAY_WRITE; + if ((access_result & NFS_MAY_EXECUTE) == NFS_MAY_EXECUTE) + mask |= MAY_EXEC; + } else if (access_result & NFS_MAY_WRITE) + mask |= MAY_WRITE; + return mask; +} + +void nfs_access_set_mask(struct nfs_access_entry *entry, u32 access_result) +{ + entry->mask = access_result; +} +EXPORT_SYMBOL_GPL(nfs_access_set_mask); + +static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask) +{ + struct nfs_access_entry cache; + bool may_block = (mask & MAY_NOT_BLOCK) == 0; + int cache_mask = -1; + int status; + + trace_nfs_access_enter(inode); + + status = nfs_access_get_cached(inode, cred, &cache.mask, may_block); + if (status == 0) + goto out_cached; + + status = -ECHILD; + if (!may_block) + goto out; + + /* + * Determine which access bits we want to ask for... + */ + cache.mask = NFS_ACCESS_READ | NFS_ACCESS_MODIFY | NFS_ACCESS_EXTEND | + nfs_access_xattr_mask(NFS_SERVER(inode)); + if (S_ISDIR(inode->i_mode)) + cache.mask |= NFS_ACCESS_DELETE | NFS_ACCESS_LOOKUP; + else + cache.mask |= NFS_ACCESS_EXECUTE; + status = NFS_PROTO(inode)->access(inode, &cache, cred); + if (status != 0) { + if (status == -ESTALE) { + if (!S_ISDIR(inode->i_mode)) + nfs_set_inode_stale(inode); + else + nfs_zap_caches(inode); + } + goto out; + } + nfs_access_add_cache(inode, &cache, cred); +out_cached: + cache_mask = nfs_access_calc_mask(cache.mask, inode->i_mode); + if ((mask & ~cache_mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0) + status = -EACCES; +out: + trace_nfs_access_exit(inode, mask, cache_mask, status); + return status; +} + +static int nfs_open_permission_mask(int openflags) +{ + int mask = 0; + + if (openflags & __FMODE_EXEC) { + /* ONLY check exec rights */ + mask = MAY_EXEC; + } else { + if ((openflags & O_ACCMODE) != O_WRONLY) + mask |= MAY_READ; + if ((openflags & O_ACCMODE) != O_RDONLY) + mask |= MAY_WRITE; + if (openflags & O_TRUNC) + mask |= MAY_WRITE; + } + + return mask; +} + +int nfs_may_open(struct inode *inode, const struct cred *cred, int openflags) +{ + return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags)); +} +EXPORT_SYMBOL_GPL(nfs_may_open); + +static int nfs_execute_ok(struct inode *inode, int mask) +{ + struct nfs_server *server = NFS_SERVER(inode); + int ret = 0; + + if (S_ISDIR(inode->i_mode)) + return 0; + if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_MODE)) { + if (mask & MAY_NOT_BLOCK) + return -ECHILD; + ret = __nfs_revalidate_inode(server, inode); + } + if (ret == 0 && !execute_ok(inode)) + ret = -EACCES; + return ret; +} + +int nfs_permission(struct mnt_idmap *idmap, + struct inode *inode, + int mask) +{ + const struct cred *cred = current_cred(); + int res = 0; + + nfs_inc_stats(inode, NFSIOS_VFSACCESS); + + if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) + goto out; + /* Is this sys_access() ? */ + if (mask & (MAY_ACCESS | MAY_CHDIR)) + goto force_lookup; + + switch (inode->i_mode & S_IFMT) { + case S_IFLNK: + goto out; + case S_IFREG: + if ((mask & MAY_OPEN) && + nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN)) + return 0; + break; + case S_IFDIR: + /* + * Optimize away all write operations, since the server + * will check permissions when we perform the op. + */ + if ((mask & MAY_WRITE) && !(mask & MAY_READ)) + goto out; + } + +force_lookup: + if (!NFS_PROTO(inode)->access) + goto out_notsup; + + res = nfs_do_access(inode, cred, mask); +out: + if (!res && (mask & MAY_EXEC)) + res = nfs_execute_ok(inode, mask); + + dfprintk(VFS, "NFS: permission(%s/%llu), mask=0x%x, res=%d\n", + inode->i_sb->s_id, inode->i_ino, mask, res); + return res; +out_notsup: + if (mask & MAY_NOT_BLOCK) + return -ECHILD; + + res = nfs_revalidate_inode(inode, NFS_INO_INVALID_MODE | + NFS_INO_INVALID_OTHER); + if (res == 0) + res = generic_permission(&nop_mnt_idmap, inode, mask); + goto out; +} +EXPORT_SYMBOL_GPL(nfs_permission); diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c new file mode 100644 index 000000000..e626c7249 --- /dev/null +++ b/fs/nfs/direct.c @@ -0,0 +1,1112 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/direct.c + * + * Copyright (C) 2003 by Chuck Lever + * + * High-performance uncached I/O for the Linux NFS client + * + * There are important applications whose performance or correctness + * depends on uncached access to file data. Database clusters + * (multiple copies of the same instance running on separate hosts) + * implement their own cache coherency protocol that subsumes file + * system cache protocols. Applications that process datasets + * considerably larger than the client's memory do not always benefit + * from a local cache. A streaming video server, for instance, has no + * need to cache the contents of a file. + * + * When an application requests uncached I/O, all read and write requests + * are made directly to the server; data stored or fetched via these + * requests is not cached in the Linux page cache. The client does not + * correct unaligned requests from applications. All requested bytes are + * held on permanent storage before a direct write system call returns to + * an application. + * + * Solaris implements an uncached I/O facility called directio() that + * is used for backups and sequential I/O to very large files. Solaris + * also supports uncaching whole NFS partitions with "-o forcedirectio," + * an undocumented mount option. + * + * Designed by Jeff Kimmel, Chuck Lever, and Trond Myklebust, with + * help from Andrew Morton. + * + * 18 Dec 2001 Initial implementation for 2.4 --cel + * 08 Jul 2002 Version for 2.4.19, with bug fixes --trondmy + * 08 Jun 2003 Port to 2.5 APIs --cel + * 31 Mar 2004 Handle direct I/O without VFS support --cel + * 15 Sep 2004 Parallel async reads --cel + * 04 May 2005 support O_DIRECT with aio --cel + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include +#include + +#include "delegation.h" +#include "internal.h" +#include "iostat.h" +#include "pnfs.h" +#include "fscache.h" +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static struct kmem_cache *nfs_direct_cachep; + +static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops; +static const struct nfs_commit_completion_ops nfs_direct_commit_completion_ops; +static void nfs_direct_write_complete(struct nfs_direct_req *dreq); +static void nfs_direct_write_schedule_work(struct work_struct *work); + +static inline void get_dreq(struct nfs_direct_req *dreq) +{ + atomic_inc(&dreq->io_count); +} + +static inline int put_dreq(struct nfs_direct_req *dreq) +{ + return atomic_dec_and_test(&dreq->io_count); +} + +static void +nfs_direct_handle_truncated(struct nfs_direct_req *dreq, + const struct nfs_pgio_header *hdr, + ssize_t dreq_len) +{ + if (!(test_bit(NFS_IOHDR_ERROR, &hdr->flags) || + test_bit(NFS_IOHDR_EOF, &hdr->flags))) + return; + if (dreq->max_count >= dreq_len) { + dreq->max_count = dreq_len; + if (dreq->count > dreq_len) + dreq->count = dreq_len; + } + + if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && !dreq->error) + dreq->error = hdr->error; +} + +static void +nfs_direct_count_bytes(struct nfs_direct_req *dreq, + const struct nfs_pgio_header *hdr) +{ + loff_t hdr_end = hdr->io_start + hdr->good_bytes; + ssize_t dreq_len = 0; + + if (hdr_end > dreq->io_start) + dreq_len = hdr_end - dreq->io_start; + + nfs_direct_handle_truncated(dreq, hdr, dreq_len); + + if (dreq_len > dreq->max_count) + dreq_len = dreq->max_count; + + if (dreq->count < dreq_len) + dreq->count = dreq_len; +} + +static void nfs_direct_truncate_request(struct nfs_direct_req *dreq, + struct nfs_page *req) +{ + loff_t offs = req_offset(req); + size_t req_start = (size_t)(offs - dreq->io_start); + + if (req_start < dreq->max_count) + dreq->max_count = req_start; + if (req_start < dreq->count) + dreq->count = req_start; +} + +static void nfs_direct_file_adjust_size_locked(struct inode *inode, + loff_t offset, size_t count) +{ + loff_t newsize = offset + (loff_t)count; + loff_t oldsize = i_size_read(inode); + + if (newsize > oldsize) { + i_size_write(inode, newsize); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; + trace_nfs_size_grow(inode, newsize); + nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); + } +} + +/** + * nfs_swap_rw - NFS address space operation for swap I/O + * @iocb: target I/O control block + * @iter: I/O buffer + * + * Perform IO to the swap-file. This is much like direct IO. + */ +int nfs_swap_rw(struct kiocb *iocb, struct iov_iter *iter) +{ + ssize_t ret; + + if (iov_iter_rw(iter) == READ) + ret = nfs_file_direct_read(iocb, iter, true); + else + ret = nfs_file_direct_write(iocb, iter, true); + if (ret < 0) + return ret; + return 0; +} + +static void nfs_direct_release_pages(struct page **pages, unsigned int npages) +{ + unsigned int i; + for (i = 0; i < npages; i++) + put_page(pages[i]); +} + +void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo, + struct nfs_direct_req *dreq) +{ + cinfo->inode = dreq->inode; + cinfo->mds = &dreq->mds_cinfo; + cinfo->ds = &dreq->ds_cinfo; + cinfo->dreq = dreq; + cinfo->completion_ops = &nfs_direct_commit_completion_ops; +} + +static inline struct nfs_direct_req *nfs_direct_req_alloc(void) +{ + struct nfs_direct_req *dreq; + + dreq = kmem_cache_zalloc(nfs_direct_cachep, GFP_KERNEL); + if (!dreq) + return NULL; + + kref_init(&dreq->kref); + kref_get(&dreq->kref); + init_completion(&dreq->completion); + INIT_LIST_HEAD(&dreq->mds_cinfo.list); + pnfs_init_ds_commit_info(&dreq->ds_cinfo); + INIT_WORK(&dreq->work, nfs_direct_write_schedule_work); + spin_lock_init(&dreq->lock); + + return dreq; +} + +static void nfs_direct_req_free(struct kref *kref) +{ + struct nfs_direct_req *dreq = container_of(kref, struct nfs_direct_req, kref); + + pnfs_release_ds_info(&dreq->ds_cinfo, dreq->inode); + if (dreq->l_ctx != NULL) + nfs_put_lock_context(dreq->l_ctx); + if (dreq->ctx != NULL) + put_nfs_open_context(dreq->ctx); + kmem_cache_free(nfs_direct_cachep, dreq); +} + +static void nfs_direct_req_release(struct nfs_direct_req *dreq) +{ + kref_put(&dreq->kref, nfs_direct_req_free); +} + +ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq, loff_t offset) +{ + loff_t start = offset - dreq->io_start; + return dreq->max_count - start; +} +EXPORT_SYMBOL_GPL(nfs_dreq_bytes_left); + +/* + * Collects and returns the final error value/byte-count. + */ +static ssize_t nfs_direct_wait(struct nfs_direct_req *dreq) +{ + ssize_t result = -EIOCBQUEUED; + + /* Async requests don't wait here */ + if (dreq->iocb) + goto out; + + result = wait_for_completion_killable(&dreq->completion); + + if (!result) { + result = dreq->count; + WARN_ON_ONCE(dreq->count < 0); + } + if (!result) + result = dreq->error; + +out: + return (ssize_t) result; +} + +/* + * Synchronous I/O uses a stack-allocated iocb. Thus we can't trust + * the iocb is still valid here if this is a synchronous request. + */ +static void nfs_direct_complete(struct nfs_direct_req *dreq) +{ + struct inode *inode = dreq->inode; + + inode_dio_end(inode); + + if (dreq->iocb) { + long res = (long) dreq->error; + if (dreq->count != 0) { + res = (long) dreq->count; + WARN_ON_ONCE(dreq->count < 0); + } + dreq->iocb->ki_complete(dreq->iocb, res); + } + + complete(&dreq->completion); + + nfs_direct_req_release(dreq); +} + +static void nfs_direct_read_completion(struct nfs_pgio_header *hdr) +{ + unsigned long bytes = 0; + struct nfs_direct_req *dreq = hdr->dreq; + + spin_lock(&dreq->lock); + if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) { + spin_unlock(&dreq->lock); + goto out_put; + } + + nfs_direct_count_bytes(dreq, hdr); + spin_unlock(&dreq->lock); + + nfs_update_delegated_atime(dreq->inode); + + while (!list_empty(&hdr->pages)) { + struct nfs_page *req = nfs_list_entry(hdr->pages.next); + struct page *page = req->wb_page; + + if (!PageCompound(page) && bytes < hdr->good_bytes && + (dreq->flags == NFS_ODIRECT_SHOULD_DIRTY)) + set_page_dirty(page); + bytes += req->wb_bytes; + nfs_list_remove_request(req); + nfs_release_request(req); + } +out_put: + if (put_dreq(dreq)) + nfs_direct_complete(dreq); + hdr->release(hdr); +} + +static void nfs_read_sync_pgio_error(struct list_head *head, int error) +{ + struct nfs_page *req; + + while (!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_release_request(req); + } +} + +static void nfs_direct_pgio_init(struct nfs_pgio_header *hdr) +{ + get_dreq(hdr->dreq); + set_bit(NFS_IOHDR_ODIRECT, &hdr->flags); +} + +static const struct nfs_pgio_completion_ops nfs_direct_read_completion_ops = { + .error_cleanup = nfs_read_sync_pgio_error, + .init_hdr = nfs_direct_pgio_init, + .completion = nfs_direct_read_completion, +}; + +/* + * For each rsize'd chunk of the user's buffer, dispatch an NFS READ + * operation. If nfs_readdata_alloc() or get_user_pages() fails, + * bail and stop sending more reads. Read length accounting is + * handled automatically by nfs_direct_read_result(). Otherwise, if + * no requests have been sent, just return an error. + */ + +static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq, + struct iov_iter *iter, + loff_t pos) +{ + struct nfs_pageio_descriptor desc; + struct inode *inode = dreq->inode; + ssize_t result = -EINVAL; + size_t requested_bytes = 0; + size_t rsize = max_t(size_t, NFS_SERVER(inode)->rsize, PAGE_SIZE); + + nfs_pageio_init_read(&desc, dreq->inode, false, + &nfs_direct_read_completion_ops); + get_dreq(dreq); + desc.pg_dreq = dreq; + inode_dio_begin(inode); + + while (iov_iter_count(iter)) { + struct page **pagevec; + size_t bytes; + size_t pgbase; + unsigned npages, i; + + result = iov_iter_get_pages_alloc2(iter, &pagevec, + rsize, &pgbase); + if (result < 0) + break; + + bytes = result; + npages = (result + pgbase + PAGE_SIZE - 1) / PAGE_SIZE; + for (i = 0; i < npages; i++) { + struct nfs_page *req; + unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); + /* XXX do we need to do the eof zeroing found in async_filler? */ + req = nfs_page_create_from_page(dreq->ctx, pagevec[i], + pgbase, pos, req_len); + if (IS_ERR(req)) { + result = PTR_ERR(req); + break; + } + if (!nfs_pageio_add_request(&desc, req)) { + result = desc.pg_error; + nfs_release_request(req); + break; + } + pgbase = 0; + bytes -= req_len; + requested_bytes += req_len; + pos += req_len; + } + nfs_direct_release_pages(pagevec, npages); + kvfree(pagevec); + if (result < 0) + break; + } + + nfs_pageio_complete(&desc); + + /* + * If no bytes were started, return the error, and let the + * generic layer handle the completion. + */ + if (requested_bytes == 0) { + inode_dio_end(inode); + nfs_direct_req_release(dreq); + return result < 0 ? result : -EIO; + } + + if (put_dreq(dreq)) + nfs_direct_complete(dreq); + return requested_bytes; +} + +/** + * nfs_file_direct_read - file direct read operation for NFS files + * @iocb: target I/O control block + * @iter: vector of user buffers into which to read data + * @swap: flag indicating this is swap IO, not O_DIRECT IO + * + * We use this function for direct reads instead of calling + * generic_file_aio_read() in order to avoid gfar's check to see if + * the request starts before the end of the file. For that check + * to work, we must generate a GETATTR before each direct read, and + * even then there is a window between the GETATTR and the subsequent + * READ where the file size could change. Our preference is simply + * to do all reads the application wants, and the server will take + * care of managing the end of file boundary. + * + * This function also eliminates unnecessarily updating the file's + * atime locally, as the NFS server sets the file's atime, and this + * client must read the updated atime from the server back into its + * cache. + */ +ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter, + bool swap) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct nfs_direct_req *dreq; + struct nfs_lock_context *l_ctx; + ssize_t result, requested; + size_t count = iov_iter_count(iter); + nfs_add_stats(mapping->host, NFSIOS_DIRECTREADBYTES, count); + + dfprintk(FILE, "NFS: direct read(%pD2, %zd@%Ld)\n", + file, count, (long long) iocb->ki_pos); + + result = 0; + if (!count) + goto out; + + task_io_account_read(count); + + result = -ENOMEM; + dreq = nfs_direct_req_alloc(); + if (dreq == NULL) + goto out; + + dreq->inode = inode; + dreq->max_count = count; + dreq->io_start = iocb->ki_pos; + dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); + l_ctx = nfs_get_lock_context(dreq->ctx); + if (IS_ERR(l_ctx)) { + result = PTR_ERR(l_ctx); + nfs_direct_req_release(dreq); + goto out_release; + } + dreq->l_ctx = l_ctx; + if (!is_sync_kiocb(iocb)) { + dreq->iocb = iocb; + } else if (iocb->ki_flags & IOCB_NOWAIT) { + result = -EAGAIN; + nfs_direct_req_release(dreq); + goto out_release; + } + + if (user_backed_iter(iter)) + dreq->flags = NFS_ODIRECT_SHOULD_DIRTY; + + if (!swap) { + if (iocb->ki_flags & IOCB_NOWAIT) + result = nfs_start_io_direct_nowait(inode); + else + result = nfs_start_io_direct(inode); + if (result) { + /* release the reference that would usually be + * consumed by nfs_direct_read_schedule_iovec() + */ + nfs_direct_req_release(dreq); + goto out_release; + } + } + + NFS_I(inode)->read_io += count; + requested = nfs_direct_read_schedule_iovec(dreq, iter, iocb->ki_pos); + + if (!swap) + nfs_end_io_direct(inode); + + if (requested > 0) { + result = nfs_direct_wait(dreq); + if (result > 0) { + requested -= result; + iocb->ki_pos += result; + } + iov_iter_revert(iter, requested); + } else { + result = requested; + } + +out_release: + nfs_direct_req_release(dreq); +out: + return result; +} + +static void nfs_direct_add_page_head(struct list_head *list, + struct nfs_page *req) +{ + struct nfs_page *head = req->wb_head; + + if (!list_empty(&head->wb_list) || !nfs_lock_request(head)) + return; + if (!list_empty(&head->wb_list)) { + nfs_unlock_request(head); + return; + } + list_add(&head->wb_list, list); + kref_get(&head->wb_kref); + kref_get(&head->wb_kref); +} + +static void nfs_direct_join_group(struct list_head *list, + struct nfs_commit_info *cinfo, + struct inode *inode) +{ + struct nfs_page *req, *subreq; + + list_for_each_entry(req, list, wb_list) { + if (req->wb_head != req) { + nfs_direct_add_page_head(&req->wb_list, req); + continue; + } + subreq = req->wb_this_page; + if (subreq == req) + continue; + do { + /* + * Remove subrequests from this list before freeing + * them in the call to nfs_join_page_group(). + */ + if (!list_empty(&subreq->wb_list)) { + nfs_list_remove_request(subreq); + nfs_release_request(subreq); + } + } while ((subreq = subreq->wb_this_page) != req); + nfs_join_page_group(req, cinfo, inode); + } +} + +static void +nfs_direct_write_scan_commit_list(struct inode *inode, + struct list_head *list, + struct nfs_commit_info *cinfo) +{ + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + pnfs_recover_commit_reqs(list, cinfo); + nfs_scan_commit_list(&cinfo->mds->list, list, cinfo, 0); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); +} + +static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) +{ + struct nfs_pageio_descriptor desc; + struct nfs_page *req; + LIST_HEAD(reqs); + struct nfs_commit_info cinfo; + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo); + + nfs_direct_join_group(&reqs, &cinfo, dreq->inode); + + nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo); + get_dreq(dreq); + + nfs_pageio_init_write(&desc, dreq->inode, FLUSH_STABLE, false, + &nfs_direct_write_completion_ops); + desc.pg_dreq = dreq; + + while (!list_empty(&reqs)) { + req = nfs_list_entry(reqs.next); + /* Bump the transmission count */ + req->wb_nio++; + if (!nfs_pageio_add_request(&desc, req)) { + spin_lock(&dreq->lock); + if (dreq->error < 0) { + desc.pg_error = dreq->error; + } else if (desc.pg_error != -EAGAIN) { + dreq->flags = 0; + if (!desc.pg_error) + desc.pg_error = -EIO; + dreq->error = desc.pg_error; + } else + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&dreq->lock); + break; + } + nfs_release_request(req); + } + nfs_pageio_complete(&desc); + + while (!list_empty(&reqs)) { + req = nfs_list_entry(reqs.next); + nfs_list_remove_request(req); + nfs_unlock_and_release_request(req); + if (desc.pg_error == -EAGAIN) { + nfs_mark_request_commit(req, NULL, &cinfo, 0); + } else { + spin_lock(&dreq->lock); + nfs_direct_truncate_request(dreq, req); + spin_unlock(&dreq->lock); + nfs_release_request(req); + } + } + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq); +} + +static void nfs_direct_commit_complete(struct nfs_commit_data *data) +{ + const struct nfs_writeverf *verf = data->res.verf; + struct nfs_direct_req *dreq = data->dreq; + struct nfs_commit_info cinfo; + struct nfs_page *req; + int status = data->task.tk_status; + + trace_nfs_direct_commit_complete(dreq); + + spin_lock(&dreq->lock); + if (status < 0) { + /* Errors in commit are fatal */ + dreq->error = status; + dreq->flags = NFS_ODIRECT_DONE; + } else { + status = dreq->error; + } + spin_unlock(&dreq->lock); + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + + while (!list_empty(&data->pages)) { + req = nfs_list_entry(data->pages.next); + nfs_list_remove_request(req); + if (status < 0) { + spin_lock(&dreq->lock); + nfs_direct_truncate_request(dreq, req); + spin_unlock(&dreq->lock); + nfs_release_request(req); + } else if (!nfs_write_match_verf(verf, req)) { + spin_lock(&dreq->lock); + if (dreq->flags == 0) + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&dreq->lock); + /* + * Despite the reboot, the write was successful, + * so reset wb_nio. + */ + req->wb_nio = 0; + nfs_mark_request_commit(req, NULL, &cinfo, 0); + } else + nfs_release_request(req); + nfs_unlock_and_release_request(req); + } + + if (nfs_commit_end(cinfo.mds)) + nfs_direct_write_complete(dreq); +} + +static void nfs_direct_resched_write(struct nfs_commit_info *cinfo, + struct nfs_page *req) +{ + struct nfs_direct_req *dreq = cinfo->dreq; + + trace_nfs_direct_resched_write(dreq); + + spin_lock(&dreq->lock); + if (dreq->flags != NFS_ODIRECT_DONE) + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&dreq->lock); + nfs_mark_request_commit(req, NULL, cinfo, 0); +} + +static const struct nfs_commit_completion_ops nfs_direct_commit_completion_ops = { + .completion = nfs_direct_commit_complete, + .resched_write = nfs_direct_resched_write, +}; + +static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) +{ + int res; + struct nfs_commit_info cinfo; + LIST_HEAD(mds_list); + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + nfs_commit_begin(cinfo.mds); + nfs_scan_commit(dreq->inode, &mds_list, &cinfo); + res = nfs_generic_commit_list(dreq->inode, &mds_list, 0, &cinfo); + if (res < 0) { /* res == -ENOMEM */ + spin_lock(&dreq->lock); + if (dreq->flags == 0) + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&dreq->lock); + } + if (nfs_commit_end(cinfo.mds)) + nfs_direct_write_complete(dreq); +} + +static void nfs_direct_write_clear_reqs(struct nfs_direct_req *dreq) +{ + struct nfs_commit_info cinfo; + struct nfs_page *req; + LIST_HEAD(reqs); + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo); + + while (!list_empty(&reqs)) { + req = nfs_list_entry(reqs.next); + nfs_list_remove_request(req); + nfs_direct_truncate_request(dreq, req); + nfs_release_request(req); + nfs_unlock_and_release_request(req); + } +} + +static void nfs_direct_write_schedule_work(struct work_struct *work) +{ + struct nfs_direct_req *dreq = container_of(work, struct nfs_direct_req, work); + int flags = dreq->flags; + + dreq->flags = 0; + switch (flags) { + case NFS_ODIRECT_DO_COMMIT: + nfs_direct_commit_schedule(dreq); + break; + case NFS_ODIRECT_RESCHED_WRITES: + nfs_direct_write_reschedule(dreq); + break; + default: + nfs_direct_write_clear_reqs(dreq); + nfs_zap_mapping(dreq->inode, dreq->inode->i_mapping); + nfs_direct_complete(dreq); + } +} + +static void nfs_direct_write_complete(struct nfs_direct_req *dreq) +{ + trace_nfs_direct_write_complete(dreq); + queue_work(nfsiod_workqueue, &dreq->work); /* Calls nfs_direct_write_schedule_work */ +} + +static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) +{ + struct nfs_direct_req *dreq = hdr->dreq; + struct nfs_commit_info cinfo; + struct inode *inode = dreq->inode; + int flags = NFS_ODIRECT_DONE; + + trace_nfs_direct_write_completion(dreq); + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + + spin_lock(&dreq->lock); + if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) { + spin_unlock(&dreq->lock); + goto out_put; + } + + nfs_direct_count_bytes(dreq, hdr); + if (test_bit(NFS_IOHDR_UNSTABLE_WRITES, &hdr->flags) && + !test_bit(NFS_IOHDR_ERROR, &hdr->flags)) { + if (!dreq->flags) + dreq->flags = NFS_ODIRECT_DO_COMMIT; + flags = dreq->flags; + } + spin_unlock(&dreq->lock); + + spin_lock(&inode->i_lock); + nfs_direct_file_adjust_size_locked(inode, dreq->io_start, dreq->count); + nfs_update_delegated_mtime_locked(dreq->inode); + spin_unlock(&inode->i_lock); + + while (!list_empty(&hdr->pages)) { + struct nfs_page *req; + + req = nfs_list_entry(hdr->pages.next); + nfs_list_remove_request(req); + if (flags == NFS_ODIRECT_DO_COMMIT) { + kref_get(&req->wb_kref); + memcpy(&req->wb_verf, &hdr->verf.verifier, + sizeof(req->wb_verf)); + nfs_mark_request_commit(req, hdr->lseg, &cinfo, + hdr->ds_commit_idx); + } else if (flags == NFS_ODIRECT_RESCHED_WRITES) { + kref_get(&req->wb_kref); + nfs_mark_request_commit(req, NULL, &cinfo, 0); + } + nfs_unlock_and_release_request(req); + } + +out_put: + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq); + hdr->release(hdr); +} + +static void nfs_write_sync_pgio_error(struct list_head *head, int error) +{ + struct nfs_page *req; + + while (!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_unlock_and_release_request(req); + } +} + +static void nfs_direct_write_reschedule_io(struct nfs_pgio_header *hdr) +{ + struct nfs_direct_req *dreq = hdr->dreq; + struct nfs_page *req; + struct nfs_commit_info cinfo; + + trace_nfs_direct_write_reschedule_io(dreq); + + nfs_init_cinfo_from_dreq(&cinfo, dreq); + spin_lock(&dreq->lock); + if (dreq->error == 0) + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + set_bit(NFS_IOHDR_REDO, &hdr->flags); + spin_unlock(&dreq->lock); + while (!list_empty(&hdr->pages)) { + req = nfs_list_entry(hdr->pages.next); + nfs_list_remove_request(req); + nfs_unlock_request(req); + nfs_mark_request_commit(req, NULL, &cinfo, 0); + } +} + +static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops = { + .error_cleanup = nfs_write_sync_pgio_error, + .init_hdr = nfs_direct_pgio_init, + .completion = nfs_direct_write_completion, + .reschedule_io = nfs_direct_write_reschedule_io, +}; + + +/* + * NB: Return the value of the first error return code. Subsequent + * errors after the first one are ignored. + */ +/* + * For each wsize'd chunk of the user's buffer, dispatch an NFS WRITE + * operation. If nfs_writedata_alloc() or get_user_pages() fails, + * bail and stop sending more writes. Write length accounting is + * handled automatically by nfs_direct_write_result(). Otherwise, if + * no requests have been sent, just return an error. + */ +static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, + struct iov_iter *iter, + loff_t pos, int ioflags) +{ + struct nfs_pageio_descriptor desc; + struct inode *inode = dreq->inode; + struct nfs_commit_info cinfo; + ssize_t result = 0; + size_t requested_bytes = 0; + size_t wsize = max_t(size_t, NFS_SERVER(inode)->wsize, PAGE_SIZE); + bool defer = false; + + trace_nfs_direct_write_schedule_iovec(dreq); + + nfs_pageio_init_write(&desc, inode, ioflags, false, + &nfs_direct_write_completion_ops); + desc.pg_dreq = dreq; + get_dreq(dreq); + inode_dio_begin(inode); + + NFS_I(inode)->write_io += iov_iter_count(iter); + while (iov_iter_count(iter)) { + struct page **pagevec; + size_t bytes; + size_t pgbase; + unsigned npages, i; + + result = iov_iter_get_pages_alloc2(iter, &pagevec, + wsize, &pgbase); + if (result < 0) + break; + + bytes = result; + npages = (result + pgbase + PAGE_SIZE - 1) / PAGE_SIZE; + for (i = 0; i < npages; i++) { + struct nfs_page *req; + unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); + + req = nfs_page_create_from_page(dreq->ctx, pagevec[i], + pgbase, pos, req_len); + if (IS_ERR(req)) { + result = PTR_ERR(req); + break; + } + + if (desc.pg_error < 0) { + nfs_free_request(req); + result = desc.pg_error; + break; + } + + pgbase = 0; + bytes -= req_len; + requested_bytes += req_len; + pos += req_len; + + if (defer) { + nfs_mark_request_commit(req, NULL, &cinfo, 0); + continue; + } + + nfs_lock_request(req); + if (nfs_pageio_add_request(&desc, req)) + continue; + + /* Exit on hard errors */ + if (desc.pg_error < 0 && desc.pg_error != -EAGAIN) { + result = desc.pg_error; + nfs_unlock_and_release_request(req); + break; + } + + /* If the error is soft, defer remaining requests */ + nfs_init_cinfo_from_dreq(&cinfo, dreq); + spin_lock(&dreq->lock); + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&dreq->lock); + nfs_unlock_request(req); + nfs_mark_request_commit(req, NULL, &cinfo, 0); + desc.pg_error = 0; + defer = true; + } + nfs_direct_release_pages(pagevec, npages); + kvfree(pagevec); + if (result < 0) + break; + } + nfs_pageio_complete(&desc); + + /* + * If no bytes were started, return the error, and let the + * generic layer handle the completion. + */ + if (requested_bytes == 0) { + inode_dio_end(inode); + nfs_direct_req_release(dreq); + return result < 0 ? result : -EIO; + } + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq); + return requested_bytes; +} + +/** + * nfs_file_direct_write - file direct write operation for NFS files + * @iocb: target I/O control block + * @iter: vector of user buffers from which to write data + * @swap: flag indicating this is swap IO, not O_DIRECT IO + * + * We use this function for direct writes instead of calling + * generic_file_aio_write() in order to avoid taking the inode + * semaphore and updating the i_size. The NFS server will set + * the new i_size and this client must read the updated size + * back into its cache. We let the server do generic write + * parameter checking and report problems. + * + * We eliminate local atime updates, see direct read above. + * + * We avoid unnecessary page cache invalidations for normal cached + * readers of this file. + * + * Note that O_APPEND is not supported for NFS direct writes, as there + * is no atomic O_APPEND write facility in the NFS protocol. + */ +ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, + bool swap) +{ + ssize_t result, requested; + size_t count; + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct nfs_direct_req *dreq; + struct nfs_lock_context *l_ctx; + loff_t pos, end; + + dfprintk(FILE, "NFS: direct write(%pD2, %zd@%Ld)\n", + file, iov_iter_count(iter), (long long) iocb->ki_pos); + + if (swap) + /* bypass generic checks */ + result = iov_iter_count(iter); + else + result = generic_write_checks(iocb, iter); + if (result <= 0) + return result; + count = result; + nfs_add_stats(mapping->host, NFSIOS_DIRECTWRITTENBYTES, count); + + pos = iocb->ki_pos; + end = (pos + iov_iter_count(iter) - 1) >> PAGE_SHIFT; + + task_io_account_write(count); + + result = -ENOMEM; + dreq = nfs_direct_req_alloc(); + if (!dreq) + goto out; + + dreq->inode = inode; + dreq->max_count = count; + dreq->io_start = pos; + dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); + l_ctx = nfs_get_lock_context(dreq->ctx); + if (IS_ERR(l_ctx)) { + result = PTR_ERR(l_ctx); + nfs_direct_req_release(dreq); + goto out_release; + } + dreq->l_ctx = l_ctx; + if (!is_sync_kiocb(iocb)) + dreq->iocb = iocb; + pnfs_init_ds_commit_info_ops(&dreq->ds_cinfo, inode); + + if (swap) { + requested = nfs_direct_write_schedule_iovec(dreq, iter, pos, + FLUSH_STABLE); + } else { + result = nfs_start_io_direct(inode); + if (result) { + /* release the reference that would usually be + * consumed by nfs_direct_write_schedule_iovec() + */ + nfs_direct_req_release(dreq); + goto out_release; + } + + requested = nfs_direct_write_schedule_iovec(dreq, iter, pos, + FLUSH_COND_STABLE); + + if (mapping->nrpages) { + invalidate_inode_pages2_range(mapping, + pos >> PAGE_SHIFT, end); + } + + nfs_end_io_direct(inode); + } + + if (requested > 0) { + result = nfs_direct_wait(dreq); + if (result > 0) { + requested -= result; + iocb->ki_pos = pos + result; + /* XXX: should check the generic_write_sync retval */ + generic_write_sync(iocb, result); + } + iov_iter_revert(iter, requested); + } else { + result = requested; + } + nfs_fscache_invalidate(inode, FSCACHE_INVAL_DIO_WRITE); +out_release: + nfs_direct_req_release(dreq); +out: + return result; +} + +/** + * nfs_init_directcache - create a slab cache for nfs_direct_req structures + * + */ +int __init nfs_init_directcache(void) +{ + nfs_direct_cachep = kmem_cache_create("nfs_direct_cache", + sizeof(struct nfs_direct_req), + 0, SLAB_RECLAIM_ACCOUNT, + NULL); + if (nfs_direct_cachep == NULL) + return -ENOMEM; + + return 0; +} + +/** + * nfs_destroy_directcache - destroy the slab cache for nfs_direct_req structures + * + */ +void nfs_destroy_directcache(void) +{ + kmem_cache_destroy(nfs_direct_cachep); +} diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c new file mode 100644 index 000000000..acd3511c0 --- /dev/null +++ b/fs/nfs/dns_resolve.c @@ -0,0 +1,480 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/dns_resolve.c + * + * Copyright (c) 2009 Trond Myklebust + * + * Resolves DNS hostnames into valid ip addresses + */ + +#include +#include +#include + +#include "dns_resolve.h" + +#ifdef CONFIG_NFS_USE_KERNEL_DNS + +#include + +ssize_t nfs_dns_resolve_name(struct net *net, char *name, size_t namelen, + struct sockaddr_storage *ss, size_t salen) +{ + struct sockaddr *sa = (struct sockaddr *)ss; + ssize_t ret; + char *ip_addr = NULL; + int ip_len; + + ip_len = dns_query(net, NULL, name, namelen, NULL, &ip_addr, NULL, + false); + if (ip_len > 0) + ret = rpc_pton(net, ip_addr, ip_len, sa, salen); + else + ret = -ESRCH; + kfree(ip_addr); + return ret; +} + +#else + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "cache_lib.h" +#include "netns.h" + +#define NFS_DNS_HASHBITS 4 +#define NFS_DNS_HASHTBL_SIZE (1 << NFS_DNS_HASHBITS) + +struct nfs_dns_ent { + struct cache_head h; + + char *hostname; + size_t namelen; + + struct sockaddr_storage addr; + size_t addrlen; + struct rcu_head rcu_head; +}; + + +static void nfs_dns_ent_update(struct cache_head *cnew, + struct cache_head *ckey) +{ + struct nfs_dns_ent *new; + struct nfs_dns_ent *key; + + new = container_of(cnew, struct nfs_dns_ent, h); + key = container_of(ckey, struct nfs_dns_ent, h); + + memcpy(&new->addr, &key->addr, key->addrlen); + new->addrlen = key->addrlen; +} + +static void nfs_dns_ent_init(struct cache_head *cnew, + struct cache_head *ckey) +{ + struct nfs_dns_ent *new; + struct nfs_dns_ent *key; + + new = container_of(cnew, struct nfs_dns_ent, h); + key = container_of(ckey, struct nfs_dns_ent, h); + + kfree(new->hostname); + new->hostname = kmemdup_nul(key->hostname, key->namelen, GFP_KERNEL); + if (new->hostname) { + new->namelen = key->namelen; + nfs_dns_ent_update(cnew, ckey); + } else { + new->namelen = 0; + new->addrlen = 0; + } +} + +static void nfs_dns_ent_free_rcu(struct rcu_head *head) +{ + struct nfs_dns_ent *item; + + item = container_of(head, struct nfs_dns_ent, rcu_head); + kfree(item->hostname); + kfree(item); +} + +static void nfs_dns_ent_put(struct kref *ref) +{ + struct nfs_dns_ent *item; + + item = container_of(ref, struct nfs_dns_ent, h.ref); + call_rcu(&item->rcu_head, nfs_dns_ent_free_rcu); +} + +static struct cache_head *nfs_dns_ent_alloc(void) +{ + struct nfs_dns_ent *item = kmalloc_obj(*item); + + if (item != NULL) { + item->hostname = NULL; + item->namelen = 0; + item->addrlen = 0; + return &item->h; + } + return NULL; +}; + +static unsigned int nfs_dns_hash(const struct nfs_dns_ent *key) +{ + return hash_str(key->hostname, NFS_DNS_HASHBITS); +} + +static void nfs_dns_request(struct cache_detail *cd, + struct cache_head *ch, + char **bpp, int *blen) +{ + struct nfs_dns_ent *key = container_of(ch, struct nfs_dns_ent, h); + + qword_add(bpp, blen, key->hostname); + (*bpp)[-1] = '\n'; +} + +static int nfs_dns_upcall(struct cache_detail *cd, + struct cache_head *ch) +{ + struct nfs_dns_ent *key = container_of(ch, struct nfs_dns_ent, h); + + if (test_and_set_bit(CACHE_PENDING, &ch->flags)) + return 0; + if (!nfs_cache_upcall(cd, key->hostname)) + return 0; + clear_bit(CACHE_PENDING, &ch->flags); + return sunrpc_cache_upcall_warn(cd, ch); +} + +static int nfs_dns_match(struct cache_head *ca, + struct cache_head *cb) +{ + struct nfs_dns_ent *a; + struct nfs_dns_ent *b; + + a = container_of(ca, struct nfs_dns_ent, h); + b = container_of(cb, struct nfs_dns_ent, h); + + if (a->namelen == 0 || a->namelen != b->namelen) + return 0; + return memcmp(a->hostname, b->hostname, a->namelen) == 0; +} + +static int nfs_dns_show(struct seq_file *m, struct cache_detail *cd, + struct cache_head *h) +{ + struct nfs_dns_ent *item; + long ttl; + + if (h == NULL) { + seq_puts(m, "# ip address hostname ttl\n"); + return 0; + } + item = container_of(h, struct nfs_dns_ent, h); + ttl = item->h.expiry_time - seconds_since_boot(); + if (ttl < 0) + ttl = 0; + + if (!test_bit(CACHE_NEGATIVE, &h->flags)) { + char buf[INET6_ADDRSTRLEN+IPV6_SCOPE_ID_LEN+1]; + + rpc_ntop((struct sockaddr *)&item->addr, buf, sizeof(buf)); + seq_printf(m, "%15s ", buf); + } else + seq_puts(m, " "); + seq_printf(m, "%15s %ld\n", item->hostname, ttl); + return 0; +} + +static struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd, + struct nfs_dns_ent *key) +{ + struct cache_head *ch; + + ch = sunrpc_cache_lookup_rcu(cd, + &key->h, + nfs_dns_hash(key)); + if (!ch) + return NULL; + return container_of(ch, struct nfs_dns_ent, h); +} + +static struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd, + struct nfs_dns_ent *new, + struct nfs_dns_ent *key) +{ + struct cache_head *ch; + + ch = sunrpc_cache_update(cd, + &new->h, &key->h, + nfs_dns_hash(key)); + if (!ch) + return NULL; + return container_of(ch, struct nfs_dns_ent, h); +} + +static int nfs_dns_parse(struct cache_detail *cd, char *buf, int buflen) +{ + char buf1[NFS_DNS_HOSTNAME_MAXLEN+1]; + struct nfs_dns_ent key, *item; + unsigned int ttl; + ssize_t len; + int ret = -EINVAL; + + if (buf[buflen-1] != '\n') + goto out; + buf[buflen-1] = '\0'; + + len = qword_get(&buf, buf1, sizeof(buf1)); + if (len <= 0) + goto out; + key.addrlen = rpc_pton(cd->net, buf1, len, + (struct sockaddr *)&key.addr, + sizeof(key.addr)); + + len = qword_get(&buf, buf1, sizeof(buf1)); + if (len <= 0) + goto out; + + key.hostname = buf1; + key.namelen = len; + memset(&key.h, 0, sizeof(key.h)); + + if (get_uint(&buf, &ttl) < 0) + goto out; + if (ttl == 0) + goto out; + key.h.expiry_time = ttl + seconds_since_boot(); + + ret = -ENOMEM; + item = nfs_dns_lookup(cd, &key); + if (item == NULL) + goto out; + + if (key.addrlen == 0) + set_bit(CACHE_NEGATIVE, &key.h.flags); + + item = nfs_dns_update(cd, &key, item); + if (item == NULL) + goto out; + + ret = 0; + cache_put(&item->h, cd); +out: + return ret; +} + +static int do_cache_lookup(struct cache_detail *cd, + struct nfs_dns_ent *key, + struct nfs_dns_ent **item, + struct nfs_cache_defer_req *dreq) +{ + int ret = -ENOMEM; + + *item = nfs_dns_lookup(cd, key); + if (*item) { + ret = cache_check(cd, &(*item)->h, &dreq->req); + if (ret) + *item = NULL; + } + return ret; +} + +static int do_cache_lookup_nowait(struct cache_detail *cd, + struct nfs_dns_ent *key, + struct nfs_dns_ent **item) +{ + int ret = -ENOMEM; + + *item = nfs_dns_lookup(cd, key); + if (!*item) + goto out_err; + ret = -ETIMEDOUT; + if (!test_bit(CACHE_VALID, &(*item)->h.flags) + || (*item)->h.expiry_time < seconds_since_boot() + || cd->flush_time > (*item)->h.last_refresh) + goto out_put; + ret = -ENOENT; + if (test_bit(CACHE_NEGATIVE, &(*item)->h.flags)) + goto out_put; + return 0; +out_put: + cache_put(&(*item)->h, cd); +out_err: + *item = NULL; + return ret; +} + +static int do_cache_lookup_wait(struct cache_detail *cd, + struct nfs_dns_ent *key, + struct nfs_dns_ent **item) +{ + struct nfs_cache_defer_req *dreq; + int ret = -ENOMEM; + + dreq = nfs_cache_defer_req_alloc(); + if (!dreq) + goto out; + ret = do_cache_lookup(cd, key, item, dreq); + if (ret == -EAGAIN) { + ret = nfs_cache_wait_for_upcall(dreq); + if (!ret) + ret = do_cache_lookup_nowait(cd, key, item); + } + nfs_cache_defer_req_put(dreq); +out: + return ret; +} + +ssize_t nfs_dns_resolve_name(struct net *net, char *name, + size_t namelen, struct sockaddr_storage *ss, size_t salen) +{ + struct nfs_dns_ent key = { + .hostname = name, + .namelen = namelen, + }; + struct nfs_dns_ent *item = NULL; + ssize_t ret; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + ret = do_cache_lookup_wait(nn->nfs_dns_resolve, &key, &item); + if (ret == 0) { + if (salen >= item->addrlen) { + memcpy(ss, &item->addr, item->addrlen); + ret = item->addrlen; + } else + ret = -EOVERFLOW; + cache_put(&item->h, nn->nfs_dns_resolve); + } else if (ret == -ENOENT) + ret = -ESRCH; + return ret; +} + +static struct cache_detail nfs_dns_resolve_template = { + .owner = THIS_MODULE, + .hash_size = NFS_DNS_HASHTBL_SIZE, + .name = "dns_resolve", + .cache_put = nfs_dns_ent_put, + .cache_upcall = nfs_dns_upcall, + .cache_request = nfs_dns_request, + .cache_parse = nfs_dns_parse, + .cache_show = nfs_dns_show, + .match = nfs_dns_match, + .init = nfs_dns_ent_init, + .update = nfs_dns_ent_update, + .alloc = nfs_dns_ent_alloc, +}; + + +int nfs_dns_resolver_cache_init(struct net *net) +{ + int err; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + nn->nfs_dns_resolve = cache_create_net(&nfs_dns_resolve_template, net); + if (IS_ERR(nn->nfs_dns_resolve)) + return PTR_ERR(nn->nfs_dns_resolve); + + err = nfs_cache_register_net(net, nn->nfs_dns_resolve); + if (err) + goto err_reg; + return 0; + +err_reg: + cache_destroy_net(nn->nfs_dns_resolve, net); + return err; +} + +void nfs_dns_resolver_cache_destroy(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + nfs_cache_unregister_net(net, nn->nfs_dns_resolve); + cache_destroy_net(nn->nfs_dns_resolve, net); +} + +static int nfs4_dns_net_init(struct net *net) +{ + return nfs_dns_resolver_cache_init(net); +} + +static void nfs4_dns_net_exit(struct net *net) +{ + nfs_dns_resolver_cache_destroy(net); +} + +static struct pernet_operations nfs4_dns_resolver_ops = { + .init = nfs4_dns_net_init, + .exit = nfs4_dns_net_exit, +}; + +static int rpc_pipefs_event(struct notifier_block *nb, unsigned long event, + void *ptr) +{ + struct super_block *sb = ptr; + struct net *net = sb->s_fs_info; + struct nfs_net *nn = net_generic(net, nfs_net_id); + struct cache_detail *cd = nn->nfs_dns_resolve; + int ret = 0; + + if (cd == NULL) + return 0; + + if (!try_module_get(THIS_MODULE)) + return 0; + + switch (event) { + case RPC_PIPEFS_MOUNT: + ret = nfs_cache_register_sb(sb, cd); + break; + case RPC_PIPEFS_UMOUNT: + nfs_cache_unregister_sb(sb, cd); + break; + default: + ret = -ENOTSUPP; + break; + } + module_put(THIS_MODULE); + return ret; +} + +static struct notifier_block nfs_dns_resolver_block = { + .notifier_call = rpc_pipefs_event, +}; + +int nfs_dns_resolver_init(void) +{ + int err; + + err = register_pernet_subsys(&nfs4_dns_resolver_ops); + if (err < 0) + goto out; + err = rpc_pipefs_notifier_register(&nfs_dns_resolver_block); + if (err < 0) + goto out1; + return 0; +out1: + unregister_pernet_subsys(&nfs4_dns_resolver_ops); +out: + return err; +} + +void nfs_dns_resolver_destroy(void) +{ + rpc_pipefs_notifier_unregister(&nfs_dns_resolver_block); + unregister_pernet_subsys(&nfs4_dns_resolver_ops); +} +#endif diff --git a/fs/nfs/dns_resolve.h b/fs/nfs/dns_resolve.h new file mode 100644 index 000000000..fe3b172c4 --- /dev/null +++ b/fs/nfs/dns_resolve.h @@ -0,0 +1,37 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Resolve DNS hostnames into valid ip addresses + */ +#ifndef __LINUX_FS_NFS_DNS_RESOLVE_H +#define __LINUX_FS_NFS_DNS_RESOLVE_H + +#define NFS_DNS_HOSTNAME_MAXLEN (128) + + +#ifdef CONFIG_NFS_USE_KERNEL_DNS +static inline int nfs_dns_resolver_init(void) +{ + return 0; +} + +static inline void nfs_dns_resolver_destroy(void) +{} + +static inline int nfs_dns_resolver_cache_init(struct net *net) +{ + return 0; +} + +static inline void nfs_dns_resolver_cache_destroy(struct net *net) +{} +#else +extern int nfs_dns_resolver_init(void); +extern void nfs_dns_resolver_destroy(void); +extern int nfs_dns_resolver_cache_init(struct net *net); +extern void nfs_dns_resolver_cache_destroy(struct net *net); +#endif + +extern ssize_t nfs_dns_resolve_name(struct net *net, char *name, + size_t namelen, struct sockaddr_storage *sa, size_t salen); + +#endif diff --git a/fs/nfs/export.c b/fs/nfs/export.c new file mode 100644 index 000000000..8fb08bce0 --- /dev/null +++ b/fs/nfs/export.c @@ -0,0 +1,166 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2015, Primary Data, Inc. All rights reserved. + * + * Tao Peng + */ +#include +#include +#include +#include + +#include "internal.h" +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +enum { + FILEID_HIGH_OFF = 0, /* inode fileid high */ + FILEID_LOW_OFF, /* inode fileid low */ + FILE_I_TYPE_OFF, /* inode type */ + EMBED_FH_OFF /* embeded server fh */ +}; + + +static struct nfs_fh *nfs_exp_embedfh(__u32 *p) +{ + return (struct nfs_fh *)(p + EMBED_FH_OFF); +} + +/* + * Let's break subtree checking for now... otherwise we'll have to embed parent fh + * but there might not be enough space. + */ +static int +nfs_encode_fh(struct inode *inode, __u32 *p, int *max_len, struct inode *parent) +{ + struct nfs_fh *server_fh = NFS_FH(inode); + struct nfs_fh *clnt_fh = nfs_exp_embedfh(p); + size_t fh_size = offsetof(struct nfs_fh, data) + server_fh->size; + int len = EMBED_FH_OFF + XDR_QUADLEN(fh_size); + + dprintk("%s: max fh len %d inode %p parent %p", + __func__, *max_len, inode, parent); + + if (*max_len < len) { + dprintk("%s: fh len %d too small, required %d\n", + __func__, *max_len, len); + *max_len = len; + return FILEID_INVALID; + } + + p[FILEID_HIGH_OFF] = inode->i_ino >> 32; + p[FILEID_LOW_OFF] = inode->i_ino; + p[FILE_I_TYPE_OFF] = inode->i_mode & S_IFMT; + p[len - 1] = 0; /* Padding */ + nfs_copy_fh(clnt_fh, server_fh); + *max_len = len; + dprintk("%s: result fh fileid %llu mode %u size %d\n", + __func__, inode->i_ino, inode->i_mode, *max_len); + return *max_len; +} + +static struct dentry * +nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type) +{ + struct nfs_fattr *fattr = NULL; + struct nfs_fh *server_fh = nfs_exp_embedfh(fid->raw); + size_t fh_size = offsetof(struct nfs_fh, data); + const struct nfs_rpc_ops *rpc_ops; + struct dentry *dentry; + struct inode *inode; + int len = EMBED_FH_OFF; + u32 *p = fid->raw; + int ret; + + /* Initial check of bounds */ + if (fh_len < len + XDR_QUADLEN(fh_size) || + fh_len > XDR_QUADLEN(NFS_MAXFHSIZE)) + return NULL; + /* Calculate embedded filehandle size */ + fh_size += server_fh->size; + len += XDR_QUADLEN(fh_size); + /* NULL translates to ESTALE */ + if (fh_len < len || fh_type != len) + return NULL; + + fattr = nfs_alloc_fattr_with_label(NFS_SB(sb)); + if (fattr == NULL) { + dentry = ERR_PTR(-ENOMEM); + goto out; + } + + fattr->fileid = ((u64)p[FILEID_HIGH_OFF] << 32) + p[FILEID_LOW_OFF]; + fattr->mode = p[FILE_I_TYPE_OFF]; + fattr->valid |= NFS_ATTR_FATTR_FILEID | NFS_ATTR_FATTR_TYPE; + + dprintk("%s: fileid %llu mode %d\n", __func__, fattr->fileid, fattr->mode); + + inode = nfs_ilookup(sb, fattr, server_fh); + if (inode) + goto out_found; + + rpc_ops = NFS_SB(sb)->nfs_client->rpc_ops; + ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, NULL); + if (ret) { + dprintk("%s: getattr failed %d\n", __func__, ret); + trace_nfs_fh_to_dentry(sb, server_fh, fattr->fileid, ret); + dentry = ERR_PTR(ret); + goto out_free_fattr; + } + + inode = nfs_fhget(sb, server_fh, fattr); + +out_found: + dentry = d_obtain_alias(inode); +out_free_fattr: + nfs_free_fattr(fattr); +out: + return dentry; +} + +static struct dentry * +nfs_get_parent(struct dentry *dentry) +{ + int ret; + struct inode *inode = d_inode(dentry), *pinode; + struct super_block *sb = inode->i_sb; + struct nfs_server *server = NFS_SB(sb); + struct nfs_fattr *fattr = NULL; + struct dentry *parent; + struct nfs_rpc_ops const *ops = server->nfs_client->rpc_ops; + struct nfs_fh fh; + + if (!ops->lookupp) + return ERR_PTR(-EACCES); + + fattr = nfs_alloc_fattr_with_label(server); + if (fattr == NULL) + return ERR_PTR(-ENOMEM); + + ret = ops->lookupp(inode, &fh, fattr); + if (ret) { + parent = ERR_PTR(ret); + goto out; + } + + pinode = nfs_fhget(sb, &fh, fattr); + parent = d_obtain_alias(pinode); +out: + nfs_free_fattr(fattr); + return parent; +} + +const struct export_operations nfs_export_ops = { + .encode_fh = nfs_encode_fh, + .fh_to_dentry = nfs_fh_to_dentry, + .get_parent = nfs_get_parent, + .flags = EXPORT_OP_NOWCC | + EXPORT_OP_NOSUBTREECHK | + EXPORT_OP_CLOSE_BEFORE_UNLINK | + EXPORT_OP_REMOTE_FS | + EXPORT_OP_NOATOMIC_ATTR | + EXPORT_OP_FLUSH_ON_CLOSE | + EXPORT_OP_NOLOCKS, +}; diff --git a/fs/nfs/file.c b/fs/nfs/file.c new file mode 100644 index 000000000..a0d8f1c1c --- /dev/null +++ b/fs/nfs/file.c @@ -0,0 +1,982 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/file.c + * + * Copyright (C) 1992 Rick Sladkey + * + * Changes Copyright (C) 1994 by Florian La Roche + * - Do not copy data too often around in the kernel. + * - In nfs_file_read the return value of kmalloc wasn't checked. + * - Put in a better version of read look-ahead buffering. Original idea + * and implementation by Wai S Kok elekokws@ee.nus.sg. + * + * Expire cache on write to a file by Wai S Kok (Oct 1994). + * + * Total rewrite of read side for new NFS buffer cache.. Linus. + * + * nfs regular file handling functions + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include "delegation.h" +#include "internal.h" +#include "iostat.h" +#include "fscache.h" +#include "pnfs.h" + +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_FILE + +static const struct vm_operations_struct nfs_file_vm_ops; + +int nfs_check_flags(int flags) +{ + if ((flags & (O_APPEND | O_DIRECT)) == (O_APPEND | O_DIRECT)) + return -EINVAL; + + return 0; +} +EXPORT_SYMBOL_GPL(nfs_check_flags); + +/* + * Open file + */ +static int +nfs_file_open(struct inode *inode, struct file *filp) +{ + int res; + + dprintk("NFS: open file(%pD2)\n", filp); + + nfs_inc_stats(inode, NFSIOS_VFSOPEN); + res = nfs_check_flags(filp->f_flags); + if (res) + return res; + + res = nfs_open(inode, filp); + if (res == 0) { + filp->f_mode |= FMODE_CAN_ODIRECT; + /* flag NOWAIT on read-only files only */ + if (!(filp->f_mode & FMODE_WRITE)) + filp->f_mode |= FMODE_NOWAIT; + } + return res; +} + +int +nfs_file_release(struct inode *inode, struct file *filp) +{ + dprintk("NFS: release(%pD2)\n", filp); + + nfs_inc_stats(inode, NFSIOS_VFSRELEASE); + nfs_file_clear_open_context(filp); + nfs_fscache_release_file(inode, filp); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_file_release); + +/** + * nfs_revalidate_file_size - Revalidate the file size + * @inode: pointer to inode struct + * @filp: pointer to struct file + * + * Revalidates the file length. This is basically a wrapper around + * nfs_revalidate_inode() that takes into account the fact that we may + * have cached writes (in which case we don't care about the server's + * idea of what the file length is), or O_DIRECT (in which case we + * shouldn't trust the cache). + */ +static int nfs_revalidate_file_size(struct inode *inode, struct file *filp) +{ + struct nfs_server *server = NFS_SERVER(inode); + + if (filp->f_flags & O_DIRECT) + goto force_reval; + if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_SIZE)) + goto force_reval; + return 0; +force_reval: + return __nfs_revalidate_inode(server, inode); +} + +loff_t nfs_file_llseek(struct file *filp, loff_t offset, int whence) +{ + dprintk("NFS: llseek file(%pD2, %lld, %d)\n", + filp, offset, whence); + + /* + * whence == SEEK_END || SEEK_DATA || SEEK_HOLE => we must revalidate + * the cached file length + */ + if (whence != SEEK_SET && whence != SEEK_CUR) { + struct inode *inode = filp->f_mapping->host; + + int retval = nfs_revalidate_file_size(inode, filp); + if (retval < 0) + return (loff_t)retval; + } + + return generic_file_llseek(filp, offset, whence); +} +EXPORT_SYMBOL_GPL(nfs_file_llseek); + +/* + * Flush all dirty pages, and check for write errors. + */ +static int +nfs_file_flush(struct file *file, fl_owner_t id) +{ + struct inode *inode = file_inode(file); + errseq_t since; + + dprintk("NFS: flush(%pD2)\n", file); + + nfs_inc_stats(inode, NFSIOS_VFSFLUSH); + if ((file->f_mode & FMODE_WRITE) == 0) + return 0; + + /* Flush writes to the server and return any errors */ + since = filemap_sample_wb_err(file->f_mapping); + nfs_wb_all(inode); + return filemap_check_wb_err(file->f_mapping, since); +} + +ssize_t +nfs_file_read(struct kiocb *iocb, struct iov_iter *to) +{ + struct inode *inode = file_inode(iocb->ki_filp); + ssize_t result; + + trace_nfs_file_read(iocb, to); + + if (iocb->ki_flags & IOCB_DIRECT) + return nfs_file_direct_read(iocb, to, false); + + /* NOWAIT only supported on direct reads */ + if (iocb->ki_flags & IOCB_NOWAIT) + return -EAGAIN; + + dprintk("NFS: read(%pD2, %zu@%lu)\n", + iocb->ki_filp, + iov_iter_count(to), (unsigned long) iocb->ki_pos); + + result = nfs_start_io_read(inode); + if (result) + return result; + + result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); + if (!result) { + result = generic_file_read_iter(iocb, to); + if (result > 0) + nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, result); + } + nfs_end_io_read(inode); + return result; +} +EXPORT_SYMBOL_GPL(nfs_file_read); + +ssize_t +nfs_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct inode *inode = file_inode(in); + ssize_t result; + + dprintk("NFS: splice_read(%pD2, %zu@%llu)\n", in, len, *ppos); + + result = nfs_start_io_read(inode); + if (result) + return result; + + result = nfs_revalidate_mapping(inode, in->f_mapping); + if (!result) { + result = filemap_splice_read(in, ppos, pipe, len, flags); + if (result > 0) + nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, result); + } + nfs_end_io_read(inode); + return result; +} +EXPORT_SYMBOL_GPL(nfs_file_splice_read); + +int +nfs_file_mmap_prepare(struct vm_area_desc *desc) +{ + struct file *file = desc->file; + struct inode *inode = file_inode(file); + int status; + + dprintk("NFS: mmap(%pD2)\n", file); + + /* Note: generic_file_mmap_prepare() returns ENOSYS on nommu systems + * so we call that before revalidating the mapping + */ + status = generic_file_mmap_prepare(desc); + if (!status) { + desc->vm_ops = &nfs_file_vm_ops; + status = nfs_revalidate_mapping(inode, file->f_mapping); + } + return status; +} +EXPORT_SYMBOL_GPL(nfs_file_mmap_prepare); + +/* + * Flush any dirty pages for this process, and check for write errors. + * The return status from this call provides a reliable indication of + * whether any write errors occurred for this process. + */ +static int +nfs_file_fsync_commit(struct file *file, int datasync) +{ + struct inode *inode = file_inode(file); + int ret, ret2; + + dprintk("NFS: fsync file(%pD2) datasync %d\n", file, datasync); + + nfs_inc_stats(inode, NFSIOS_VFSFSYNC); + ret = nfs_commit_inode(inode, FLUSH_SYNC); + ret2 = file_check_and_advance_wb_err(file); + if (ret2 < 0) + return ret2; + return ret; +} + +int +nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) +{ + struct inode *inode = file_inode(file); + struct nfs_inode *nfsi = NFS_I(inode); + long save_nredirtied = atomic_long_read(&nfsi->redirtied_pages); + long nredirtied; + int ret; + + trace_nfs_fsync_enter(inode); + + for (;;) { + ret = file_write_and_wait_range(file, start, end); + if (ret != 0) + break; + ret = nfs_file_fsync_commit(file, datasync); + if (ret != 0) + break; + ret = pnfs_sync_inode(inode, !!datasync); + if (ret != 0) + break; + nredirtied = atomic_long_read(&nfsi->redirtied_pages); + if (nredirtied == save_nredirtied) + break; + save_nredirtied = nredirtied; + } + + trace_nfs_fsync_exit(inode, ret); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_file_fsync); + +void nfs_truncate_last_folio(struct address_space *mapping, loff_t from, + loff_t to) +{ + struct folio *folio; + + if (from >= to) + return; + + folio = filemap_lock_folio(mapping, from >> PAGE_SHIFT); + if (IS_ERR(folio)) + return; + + if (folio_mkclean(folio)) + folio_mark_dirty(folio); + + if (folio_test_uptodate(folio)) { + loff_t fpos = folio_pos(folio); + size_t offset = from - fpos; + size_t end = folio_size(folio); + + if (to - fpos < end) + end = to - fpos; + folio_zero_segment(folio, offset, end); + trace_nfs_size_truncate_folio(mapping->host, to); + } + + folio_unlock(folio); + folio_put(folio); +} +EXPORT_SYMBOL_GPL(nfs_truncate_last_folio); + +/* + * Decide whether a read/modify/write cycle may be more efficient + * then a modify/write/read cycle when writing to a page in the + * page cache. + * + * Some pNFS layout drivers can only read/write at a certain block + * granularity like all block devices and therefore we must perform + * read/modify/write whenever a page hasn't read yet and the data + * to be written there is not aligned to a block boundary and/or + * smaller than the block size. + * + * The modify/write/read cycle may occur if a page is read before + * being completely filled by the writer. In this situation, the + * page must be completely written to stable storage on the server + * before it can be refilled by reading in the page from the server. + * This can lead to expensive, small, FILE_SYNC mode writes being + * done. + * + * It may be more efficient to read the page first if the file is + * open for reading in addition to writing, the page is not marked + * as Uptodate, it is not dirty or waiting to be committed, + * indicating that it was previously allocated and then modified, + * that there were valid bytes of data in that range of the file, + * and that the new data won't completely replace the old data in + * that range of the file. + */ +static bool nfs_folio_is_full_write(struct folio *folio, loff_t pos, + unsigned int len) +{ + unsigned int pglen = nfs_folio_length(folio); + unsigned int offset = offset_in_folio(folio, pos); + unsigned int end = offset + len; + + return !pglen || (end >= pglen && !offset); +} + +static bool nfs_want_read_modify_write(struct file *file, struct folio *folio, + loff_t pos, unsigned int len) +{ + /* + * Up-to-date pages, those with ongoing or full-page write + * don't need read/modify/write + */ + if (folio_test_uptodate(folio) || folio_test_private(folio) || + nfs_folio_is_full_write(folio, pos, len)) + return false; + + if (pnfs_ld_read_whole_page(file_inode(file))) + return true; + if (folio_test_dropbehind(folio)) + return false; + /* Open for reading too? */ + if (file->f_mode & FMODE_READ) + return true; + return false; +} + +/* + * This does the "real" work of the write. We must allocate and lock the + * page to be sent back to the generic routine, which then copies the + * data from user space. + * + * If the writer ends up delaying the write, the writer needs to + * increment the page use counts until he is done with the page. + */ +static int nfs_write_begin(const struct kiocb *iocb, + struct address_space *mapping, + loff_t pos, unsigned len, struct folio **foliop, + void **fsdata) +{ + struct folio *folio; + struct file *file = iocb->ki_filp; + int once_thru = 0; + int ret; + + trace_nfs_write_begin(file_inode(file), pos, len); + + dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%llu), %u@%lld)\n", + file, mapping->host->i_ino, len, (long long) pos); + nfs_truncate_last_folio(mapping, i_size_read(mapping->host), pos); + +start: + folio = write_begin_get_folio(iocb, mapping, pos >> PAGE_SHIFT, len); + if (IS_ERR(folio)) { + ret = PTR_ERR(folio); + goto out; + } + *foliop = folio; + + ret = nfs_flush_incompatible(file, folio); + if (ret) { + folio_unlock(folio); + folio_put(folio); + } else if (!once_thru && + nfs_want_read_modify_write(file, folio, pos, len)) { + once_thru = 1; + folio_clear_dropbehind(folio); + ret = nfs_read_folio(file, folio); + folio_put(folio); + if (!ret) + goto start; + } +out: + trace_nfs_write_begin_done(file_inode(file), pos, len, ret); + return ret; +} + +static int nfs_write_end(const struct kiocb *iocb, + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct folio *folio, void *fsdata) +{ + struct file *file = iocb->ki_filp; + struct nfs_open_context *ctx = nfs_file_open_context(file); + unsigned offset = offset_in_folio(folio, pos); + int status; + + trace_nfs_write_end(file_inode(file), pos, len); + dfprintk(PAGECACHE, "NFS: write_end(%pD2(%llu), %u@%lld)\n", + file, mapping->host->i_ino, len, (long long) pos); + + /* + * Zero any uninitialised parts of the page, and then mark the page + * as up to date if it turns out that we're extending the file. + */ + if (!folio_test_uptodate(folio)) { + size_t fsize = folio_size(folio); + unsigned pglen = nfs_folio_length(folio); + unsigned end = offset + copied; + + if (pglen == 0) { + folio_zero_segments(folio, 0, offset, end, fsize); + folio_mark_uptodate(folio); + } else if (end >= pglen) { + folio_zero_segment(folio, end, fsize); + if (offset == 0) + folio_mark_uptodate(folio); + } else + folio_zero_segment(folio, pglen, fsize); + } + + status = nfs_update_folio(file, folio, offset, copied); + + folio_unlock(folio); + folio_put(folio); + + if (status < 0) { + trace_nfs_write_end_done(file_inode(file), pos, len, status); + return status; + } + NFS_I(mapping->host)->write_io += copied; + + if (nfs_ctx_key_to_expire(ctx, mapping->host)) + nfs_wb_all(mapping->host); + + trace_nfs_write_end_done(file_inode(file), pos, len, copied); + return copied; +} + +/* + * Partially or wholly invalidate a page + * - Release the private state associated with a page if undergoing complete + * page invalidation + * - Called if either PG_private or PG_fscache is set on the page + * - Caller holds page lock + */ +static void nfs_invalidate_folio(struct folio *folio, size_t offset, + size_t length) +{ + struct inode *inode = folio->mapping->host; + dfprintk(PAGECACHE, "NFS: invalidate_folio(%lu, %zu, %zu)\n", + folio->index, offset, length); + + /* Cancel any unstarted writes on this page */ + if (offset != 0 || length < folio_size(folio)) + nfs_wb_folio(inode, folio); + else + nfs_wb_folio_cancel(inode, folio); + folio_wait_private_2(folio); /* [DEPRECATED] */ + trace_nfs_invalidate_folio(inode, folio_pos(folio) + offset, length); +} + +/* + * Attempt to release the private state associated with a folio + * - Called if either private or fscache flags are set on the folio + * - Caller holds folio lock + * - Return true (may release folio) or false (may not) + */ +static bool nfs_release_folio(struct folio *folio, gfp_t gfp) +{ + dfprintk(PAGECACHE, "NFS: release_folio(%p)\n", folio); + + /* If the private flag is set, then the folio is not freeable */ + if (folio_test_private(folio)) { + if ((current_gfp_context(gfp) & GFP_KERNEL) != GFP_KERNEL || + current_is_kswapd() || current_is_kcompactd()) + return false; + if (nfs_wb_folio_reclaim(folio->mapping->host, folio) < 0 || + folio_test_private(folio)) + return false; + } + return nfs_fscache_release_folio(folio, gfp); +} + +static void nfs_check_dirty_writeback(struct folio *folio, + bool *dirty, bool *writeback) +{ + struct nfs_inode *nfsi; + struct address_space *mapping = folio->mapping; + + /* + * Check if an unstable folio is currently being committed and + * if so, have the VM treat it as if the folio is under writeback + * so it will not block due to folios that will shortly be freeable. + */ + nfsi = NFS_I(mapping->host); + if (atomic_read(&nfsi->commit_info.rpcs_out)) { + *writeback = true; + return; + } + + /* + * If the private flag is set, then the folio is not freeable + * and as the inode is not being committed, it's not going to + * be cleaned in the near future so treat it as dirty + */ + if (folio_test_private(folio)) + *dirty = true; +} + +/* + * Attempt to clear the private state associated with a page when an error + * occurs that requires the cached contents of an inode to be written back or + * destroyed + * - Called if either PG_private or fscache is set on the page + * - Caller holds page lock + * - Return 0 if successful, -error otherwise + */ +static int nfs_launder_folio(struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + int ret; + + dfprintk(PAGECACHE, "NFS: launder_folio(%llu, %llu)\n", + inode->i_ino, folio_pos(folio)); + + folio_wait_private_2(folio); /* [DEPRECATED] */ + ret = nfs_wb_folio(inode, folio); + trace_nfs_launder_folio_done(inode, folio_pos(folio), + folio_size(folio), ret); + return ret; +} + +static int nfs_swap_activate(struct swap_info_struct *sis, struct file *file, + sector_t *span) +{ + unsigned long blocks; + long long isize; + int ret; + struct inode *inode = file_inode(file); + struct rpc_clnt *clnt = NFS_CLIENT(inode); + struct nfs_client *cl = NFS_SERVER(inode)->nfs_client; + + spin_lock(&inode->i_lock); + blocks = inode->i_blocks; + isize = inode->i_size; + spin_unlock(&inode->i_lock); + if (blocks*512 < isize) { + pr_warn("swap activate: swapfile has holes\n"); + return -EINVAL; + } + + ret = rpc_clnt_swap_activate(clnt); + if (ret) + return ret; + ret = add_swap_extent(sis, 0, sis->max, 0); + if (ret < 0) { + rpc_clnt_swap_deactivate(clnt); + return ret; + } + + *span = sis->pages; + + if (cl->rpc_ops->enable_swap) + cl->rpc_ops->enable_swap(inode); + + sis->flags |= SWP_FS_OPS; + return ret; +} + +static void nfs_swap_deactivate(struct file *file) +{ + struct inode *inode = file_inode(file); + struct rpc_clnt *clnt = NFS_CLIENT(inode); + struct nfs_client *cl = NFS_SERVER(inode)->nfs_client; + + rpc_clnt_swap_deactivate(clnt); + if (cl->rpc_ops->disable_swap) + cl->rpc_ops->disable_swap(file_inode(file)); +} + +const struct address_space_operations nfs_file_aops = { + .read_folio = nfs_read_folio, + .readahead = nfs_readahead, + .dirty_folio = filemap_dirty_folio, + .writepages = nfs_writepages, + .write_begin = nfs_write_begin, + .write_end = nfs_write_end, + .invalidate_folio = nfs_invalidate_folio, + .release_folio = nfs_release_folio, + .migrate_folio = nfs_migrate_folio, + .launder_folio = nfs_launder_folio, + .is_dirty_writeback = nfs_check_dirty_writeback, + .error_remove_folio = generic_error_remove_folio, + .swap_activate = nfs_swap_activate, + .swap_deactivate = nfs_swap_deactivate, + .swap_rw = nfs_swap_rw, +}; + +/* + * Notification that a PTE pointing to an NFS page is about to be made + * writable, implying that someone is about to modify the page through a + * shared-writable mapping + */ +static vm_fault_t nfs_vm_page_mkwrite(struct vm_fault *vmf) +{ + struct file *filp = vmf->vma->vm_file; + struct inode *inode = file_inode(filp); + unsigned pagelen; + vm_fault_t ret = VM_FAULT_NOPAGE; + struct address_space *mapping; + struct folio *folio = page_folio(vmf->page); + + dfprintk(PAGECACHE, "NFS: vm_page_mkwrite(%pD2(%llu), offset %lld)\n", + filp, filp->f_mapping->host->i_ino, + (long long)folio_pos(folio)); + + sb_start_pagefault(inode->i_sb); + + /* make sure the cache has finished storing the page */ + if (folio_test_private_2(folio) && /* [DEPRECATED] */ + folio_wait_private_2_killable(folio) < 0) { + ret = VM_FAULT_RETRY; + goto out; + } + + wait_on_bit_action(&NFS_I(inode)->flags, NFS_INO_INVALIDATING, + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + + folio_lock(folio); + mapping = folio->mapping; + if (mapping != inode->i_mapping) + goto out_unlock; + + folio_wait_writeback(folio); + + pagelen = nfs_folio_length(folio); + if (pagelen == 0) + goto out_unlock; + + ret = VM_FAULT_LOCKED; + if (nfs_flush_incompatible(filp, folio) == 0 && + nfs_update_folio(filp, folio, 0, pagelen) == 0) + goto out; + + ret = VM_FAULT_SIGBUS; +out_unlock: + folio_unlock(folio); +out: + sb_end_pagefault(inode->i_sb); + return ret; +} + +static const struct vm_operations_struct nfs_file_vm_ops = { + .fault = filemap_fault, + .map_pages = filemap_map_pages, + .page_mkwrite = nfs_vm_page_mkwrite, +}; + +ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + unsigned int mntflags = NFS_SERVER(inode)->flags; + ssize_t result, written; + errseq_t since; + int error; + + trace_nfs_file_write(iocb, from); + + /* + * FMODE_NOWAIT is not set for writable files + */ + if (WARN_ON_ONCE(iocb->ki_flags & IOCB_NOWAIT)) + return -EAGAIN; + + result = nfs_key_timeout_notify(file, inode); + if (result) + return result; + + if (iocb->ki_flags & IOCB_DIRECT) + return nfs_file_direct_write(iocb, from, false); + + dprintk("NFS: write(%pD2, %zu@%Ld)\n", + file, iov_iter_count(from), (long long) iocb->ki_pos); + + if (IS_SWAPFILE(inode)) + goto out_swapfile; + /* + * O_APPEND implies that we must revalidate the file length. + */ + if (iocb->ki_flags & IOCB_APPEND || iocb->ki_pos > i_size_read(inode)) { + result = nfs_revalidate_file_size(inode, file); + if (result) + return result; + } + + nfs_clear_invalid_mapping(file->f_mapping); + + since = filemap_sample_wb_err(file->f_mapping); + error = nfs_start_io_write(inode); + if (error) + return error; + result = generic_write_checks(iocb, from); + if (result > 0) + result = generic_perform_write(iocb, from); + nfs_end_io_write(inode); + if (result <= 0) + goto out; + + written = result; + nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, written); + + if (mntflags & NFS_MOUNT_WRITE_EAGER) { + result = filemap_fdatawrite_range(file->f_mapping, + iocb->ki_pos - written, + iocb->ki_pos - 1); + if (result < 0) + goto out; + } + if (mntflags & NFS_MOUNT_WRITE_WAIT) { + filemap_fdatawait_range(file->f_mapping, + iocb->ki_pos - written, + iocb->ki_pos - 1); + } + result = generic_write_sync(iocb, written); + if (result < 0) + return result; + +out: + /* Return error values */ + error = filemap_check_wb_err(file->f_mapping, since); + switch (error) { + default: + break; + case -EDQUOT: + case -EFBIG: + case -ENOSPC: + nfs_wb_all(inode); + error = file_check_and_advance_wb_err(file); + if (error < 0) + result = error; + } + return result; + +out_swapfile: + printk(KERN_INFO "NFS: attempt to write to active swap file!\n"); + return -ETXTBSY; +} +EXPORT_SYMBOL_GPL(nfs_file_write); + +static int +do_getlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) +{ + struct inode *inode = filp->f_mapping->host; + int status = 0; + unsigned int saved_type = fl->c.flc_type; + + /* Try local locking first */ + posix_test_lock(filp, fl); + if (fl->c.flc_type != F_UNLCK) { + /* found a conflict */ + goto out; + } + fl->c.flc_type = saved_type; + + if (nfs_have_read_or_write_delegation(inode)) + goto out_noconflict; + + if (is_local) + goto out_noconflict; + + status = NFS_PROTO(inode)->lock(filp, cmd, fl); +out: + return status; +out_noconflict: + fl->c.flc_type = F_UNLCK; + goto out; +} + +static int +do_unlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) +{ + struct inode *inode = filp->f_mapping->host; + struct nfs_lock_context *l_ctx; + int status; + + /* + * Flush all pending writes before doing anything + * with locks.. + */ + nfs_wb_all(inode); + + l_ctx = nfs_get_lock_context(nfs_file_open_context(filp)); + if (!IS_ERR(l_ctx)) { + status = nfs_iocounter_wait(l_ctx); + nfs_put_lock_context(l_ctx); + /* NOTE: special case + * If we're signalled while cleaning up locks on process exit, we + * still need to complete the unlock. + */ + if (status < 0 && !(fl->c.flc_flags & FL_CLOSE)) + return status; + } + + /* + * Use local locking if mounted with "-onolock" or with appropriate + * "-olocal_lock=" + */ + if (!is_local) + status = NFS_PROTO(inode)->lock(filp, cmd, fl); + else + status = locks_lock_file_wait(filp, fl); + return status; +} + +static int +do_setlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) +{ + struct inode *inode = filp->f_mapping->host; + int status; + + /* + * Flush all pending writes before doing anything + * with locks.. + */ + status = nfs_sync_mapping(filp->f_mapping); + if (status != 0) + goto out; + + /* + * Use local locking if mounted with "-onolock" or with appropriate + * "-olocal_lock=" + */ + if (!is_local) + status = NFS_PROTO(inode)->lock(filp, cmd, fl); + else + status = locks_lock_file_wait(filp, fl); + if (status < 0) + goto out; + + /* + * Invalidate cache to prevent missing any changes. If + * the file is mapped, clear the page cache as well so + * those mappings will be loaded. + * + * This makes locking act as a cache coherency point. + */ + nfs_sync_mapping(filp->f_mapping); + if (!nfs_have_read_or_write_delegation(inode)) { + nfs_zap_caches(inode); + if (mapping_mapped(filp->f_mapping)) + nfs_revalidate_mapping(inode, filp->f_mapping); + } +out: + return status; +} + +/* + * Lock a (portion of) a file + */ +int nfs_lock(struct file *filp, int cmd, struct file_lock *fl) +{ + struct inode *inode = filp->f_mapping->host; + int ret = -ENOLCK; + int is_local = 0; + + dprintk("NFS: lock(%pD2, t=%x, fl=%x, r=%lld:%lld)\n", + filp, fl->c.flc_type, fl->c.flc_flags, + (long long)fl->fl_start, (long long)fl->fl_end); + + nfs_inc_stats(inode, NFSIOS_VFSLOCK); + + if (fl->c.flc_flags & FL_RECLAIM) + return -ENOGRACE; + + if (NFS_SERVER(inode)->flags & NFS_MOUNT_LOCAL_FCNTL) + is_local = 1; + + if (NFS_PROTO(inode)->lock_check_bounds != NULL) { + ret = NFS_PROTO(inode)->lock_check_bounds(fl); + if (ret < 0) + goto out_err; + } + + if (IS_GETLK(cmd)) + ret = do_getlk(filp, cmd, fl, is_local); + else if (lock_is_unlock(fl)) + ret = do_unlk(filp, cmd, fl, is_local); + else + ret = do_setlk(filp, cmd, fl, is_local); +out_err: + return ret; +} +EXPORT_SYMBOL_GPL(nfs_lock); + +/* + * Lock a (portion of) a file + */ +int nfs_flock(struct file *filp, int cmd, struct file_lock *fl) +{ + struct inode *inode = filp->f_mapping->host; + int is_local = 0; + + dprintk("NFS: flock(%pD2, t=%x, fl=%x)\n", + filp, fl->c.flc_type, fl->c.flc_flags); + + if (!(fl->c.flc_flags & FL_FLOCK)) + return -ENOLCK; + + if (NFS_SERVER(inode)->flags & NFS_MOUNT_LOCAL_FLOCK) + is_local = 1; + + /* We're simulating flock() locks using posix locks on the server */ + if (lock_is_unlock(fl)) + return do_unlk(filp, cmd, fl, is_local); + return do_setlk(filp, cmd, fl, is_local); +} +EXPORT_SYMBOL_GPL(nfs_flock); + +const struct file_operations nfs_file_operations = { + .llseek = nfs_file_llseek, + .read_iter = nfs_file_read, + .write_iter = nfs_file_write, + .mmap_prepare = nfs_file_mmap_prepare, + .open = nfs_file_open, + .flush = nfs_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = iter_file_splice_write, + .check_flags = nfs_check_flags, + .fop_flags = FOP_DONTCACHE, +}; +EXPORT_SYMBOL_GPL(nfs_file_operations); diff --git a/fs/nfs/filelayout/Makefile b/fs/nfs/filelayout/Makefile new file mode 100644 index 000000000..de056312d --- /dev/null +++ b/fs/nfs/filelayout/Makefile @@ -0,0 +1,6 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Makefile for the pNFS Files Layout Driver kernel module +# +obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o +nfs_layout_nfsv41_files-y := filelayout.o filelayoutdev.o diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c new file mode 100644 index 000000000..72e20b56f --- /dev/null +++ b/fs/nfs/filelayout/filelayout.c @@ -0,0 +1,1155 @@ +/* + * Module for the pnfs nfs4 file layout driver. + * Defines all I/O and Policy interface operations, plus code + * to register itself with the pNFS client. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#include +#include +#include +#include + +#include + +#include "../nfs4session.h" +#include "../internal.h" +#include "../delegation.h" +#include "filelayout.h" +#include "../nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Dean Hildebrand "); +MODULE_DESCRIPTION("The NFSv4 file layout driver"); + +#define FILELAYOUT_POLL_RETRY_MAX (15*HZ) +static const struct pnfs_commit_ops filelayout_commit_ops; + +static loff_t +filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg, + loff_t offset) +{ + u32 stripe_width = flseg->stripe_unit * flseg->dsaddr->stripe_count; + u64 stripe_no; + u32 rem; + + offset -= flseg->pattern_offset; + stripe_no = div_u64(offset, stripe_width); + div_u64_rem(offset, flseg->stripe_unit, &rem); + + return stripe_no * flseg->stripe_unit + rem; +} + +/* This function is used by the layout driver to calculate the + * offset of the file on the dserver based on whether the + * layout type is STRIPE_DENSE or STRIPE_SPARSE + */ +static loff_t +filelayout_get_dserver_offset(struct pnfs_layout_segment *lseg, loff_t offset) +{ + struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); + + switch (flseg->stripe_type) { + case STRIPE_SPARSE: + return offset; + + case STRIPE_DENSE: + return filelayout_get_dense_offset(flseg, offset); + } + + BUG(); +} + +static void filelayout_reset_write(struct nfs_pgio_header *hdr) +{ + struct rpc_task *task = &hdr->task; + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + dprintk("%s Reset task %5u for i/o through MDS " + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, + hdr->task.tk_pid, + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + task->tk_status = pnfs_write_done_resend_to_mds(hdr); + } +} + +static void filelayout_reset_read(struct nfs_pgio_header *hdr) +{ + struct rpc_task *task = &hdr->task; + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + dprintk("%s Reset task %5u for i/o through MDS " + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, + hdr->task.tk_pid, + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + task->tk_status = pnfs_read_done_resend_to_mds(hdr); + } +} + +static int filelayout_async_handle_error(struct rpc_task *task, + struct nfs4_state *state, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_layout_hdr *lo = lseg->pls_layout; + struct inode *inode = lo->plh_inode; + struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(lseg); + struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table; + + if (task->tk_status >= 0) + return 0; + + switch (task->tk_status) { + /* DS session errors */ + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + dprintk("%s ERROR %d, Reset session. Exchangeid " + "flags 0x%x\n", __func__, task->tk_status, + clp->cl_exchange_flags); + nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); + break; + case -NFS4ERR_DELAY: + case -NFS4ERR_GRACE: + rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX); + break; + case -NFS4ERR_RETRY_UNCACHED_REP: + break; + /* Invalidate Layout errors */ + case -NFS4ERR_ACCESS: + case -NFS4ERR_PNFS_NO_LAYOUT: + case -ESTALE: /* mapped NFS4ERR_STALE */ + case -EBADHANDLE: /* mapped NFS4ERR_BADHANDLE */ + case -EISDIR: /* mapped NFS4ERR_ISDIR */ + case -NFS4ERR_FHEXPIRED: + case -NFS4ERR_WRONG_TYPE: + dprintk("%s Invalid layout error %d\n", __func__, + task->tk_status); + /* + * Destroy layout so new i/o will get a new layout. + * Layout will not be destroyed until all current lseg + * references are put. Mark layout as invalid to resend failed + * i/o and all i/o waiting on the slot table to the MDS until + * layout is destroyed and a new valid layout is obtained. + */ + pnfs_destroy_layout(NFS_I(inode)); + rpc_wake_up(&tbl->slot_tbl_waitq); + goto reset; + /* RPC connection errors */ + case -ECONNREFUSED: + case -EHOSTDOWN: + case -EHOSTUNREACH: + case -ENETUNREACH: + case -EIO: + case -ETIMEDOUT: + case -EPIPE: + case -EPROTO: + case -ENODEV: + dprintk("%s DS connection error %d\n", __func__, + task->tk_status); + nfs4_mark_deviceid_unavailable(devid); + pnfs_error_mark_layout_for_return(inode, lseg); + pnfs_set_lo_fail(lseg); + rpc_wake_up(&tbl->slot_tbl_waitq); + fallthrough; + default: +reset: + dprintk("%s Retry through MDS. Error %d\n", __func__, + task->tk_status); + return -NFS4ERR_RESET_TO_MDS; + } + task->tk_status = 0; + return -EAGAIN; +} + +/* NFS_PROTO call done callback routines */ + +static int filelayout_read_done_cb(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + int err; + + trace_nfs4_pnfs_read(hdr, task->tk_status); + err = filelayout_async_handle_error(task, hdr->args.context->state, + hdr->ds_clp, hdr->lseg); + + switch (err) { + case -NFS4ERR_RESET_TO_MDS: + filelayout_reset_read(hdr); + return task->tk_status; + case -EAGAIN: + rpc_restart_call_prepare(task); + return -EAGAIN; + } + + return 0; +} + +/* + * We reference the rpc_cred of the first WRITE that triggers the need for + * a LAYOUTCOMMIT, and use it to send the layoutcommit compound. + * rfc5661 is not clear about which credential should be used. + */ +static void +filelayout_set_layoutcommit(struct nfs_pgio_header *hdr) +{ + loff_t end_offs = 0; + + if (FILELAYOUT_LSEG(hdr->lseg)->commit_through_mds || + hdr->res.verf->committed == NFS_FILE_SYNC) + return; + if (hdr->res.verf->committed == NFS_DATA_SYNC) + end_offs = hdr->mds_offset + (loff_t)hdr->res.count; + + /* Note: if the write is unstable, don't set end_offs until commit */ + pnfs_set_layoutcommit(hdr->inode, hdr->lseg, end_offs); + dprintk("%s inode %llu pls_end_pos %lu\n", __func__, hdr->inode->i_ino, + (unsigned long) NFS_I(hdr->inode)->layout->plh_lwb); +} + +bool +filelayout_test_devid_unavailable(struct nfs4_deviceid_node *node) +{ + return filelayout_test_devid_invalid(node) || + nfs4_test_deviceid_unavailable(node); +} + +static bool +filelayout_reset_to_mds(struct pnfs_layout_segment *lseg) +{ + struct nfs4_deviceid_node *node = FILELAYOUT_DEVID_NODE(lseg); + + return filelayout_test_devid_unavailable(node); +} + +/* + * Call ops for the async read/write cases + * In the case of dense layouts, the offset needs to be reset to its + * original value. + */ +static void filelayout_read_prepare(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) { + rpc_exit(task, -EIO); + return; + } + if (filelayout_reset_to_mds(hdr->lseg)) { + dprintk("%s task %u reset io to MDS\n", __func__, task->tk_pid); + filelayout_reset_read(hdr); + rpc_exit(task, 0); + return; + } + hdr->pgio_done_cb = filelayout_read_done_cb; + + if (nfs4_setup_sequence(hdr->ds_clp, + &hdr->args.seq_args, + &hdr->res.seq_res, + task)) + return; + if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context, + hdr->args.lock_context, FMODE_READ) == -EIO) + rpc_exit(task, -EIO); /* lost lock, terminate I/O */ +} + +static void filelayout_read_call_done(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && + task->tk_status == 0) { + nfs41_sequence_done(task, &hdr->res.seq_res); + return; + } + + /* Note this may cause RPC to be resent */ + hdr->mds_ops->rpc_call_done(task, data); +} + +static void filelayout_read_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + rpc_count_iostats(task, NFS_SERVER(hdr->inode)->client->cl_metrics); +} + +static int filelayout_write_done_cb(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + int err; + + trace_nfs4_pnfs_write(hdr, task->tk_status); + err = filelayout_async_handle_error(task, hdr->args.context->state, + hdr->ds_clp, hdr->lseg); + + switch (err) { + case -NFS4ERR_RESET_TO_MDS: + filelayout_reset_write(hdr); + return task->tk_status; + case -EAGAIN: + rpc_restart_call_prepare(task); + return -EAGAIN; + } + + filelayout_set_layoutcommit(hdr); + + /* zero out the fattr */ + hdr->fattr.valid = 0; + if (task->tk_status >= 0) + nfs_writeback_update_inode(hdr); + + return 0; +} + +static int filelayout_commit_done_cb(struct rpc_task *task, + struct nfs_commit_data *data) +{ + int err; + + trace_nfs4_pnfs_commit_ds(data, task->tk_status); + err = filelayout_async_handle_error(task, NULL, data->ds_clp, + data->lseg); + + switch (err) { + case -NFS4ERR_RESET_TO_MDS: + pnfs_generic_prepare_to_resend_writes(data); + return -EAGAIN; + case -EAGAIN: + rpc_restart_call_prepare(task); + return -EAGAIN; + } + + pnfs_set_layoutcommit(data->inode, data->lseg, data->lwb); + + return 0; +} + +static void filelayout_write_prepare(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) { + rpc_exit(task, -EIO); + return; + } + if (filelayout_reset_to_mds(hdr->lseg)) { + dprintk("%s task %u reset io to MDS\n", __func__, task->tk_pid); + filelayout_reset_write(hdr); + rpc_exit(task, 0); + return; + } + if (nfs4_setup_sequence(hdr->ds_clp, + &hdr->args.seq_args, + &hdr->res.seq_res, + task)) + return; + if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context, + hdr->args.lock_context, FMODE_WRITE) == -EIO) + rpc_exit(task, -EIO); /* lost lock, terminate I/O */ +} + +static void filelayout_write_call_done(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && + task->tk_status == 0) { + nfs41_sequence_done(task, &hdr->res.seq_res); + return; + } + + /* Note this may cause RPC to be resent */ + hdr->mds_ops->rpc_call_done(task, data); +} + +static void filelayout_write_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + rpc_count_iostats(task, NFS_SERVER(hdr->inode)->client->cl_metrics); +} + +static void filelayout_commit_prepare(struct rpc_task *task, void *data) +{ + struct nfs_commit_data *wdata = data; + + nfs4_setup_sequence(wdata->ds_clp, + &wdata->args.seq_args, + &wdata->res.seq_res, + task); +} + +static void filelayout_commit_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_commit_data *cdata = data; + + rpc_count_iostats(task, NFS_SERVER(cdata->inode)->client->cl_metrics); +} + +static const struct rpc_call_ops filelayout_read_call_ops = { + .rpc_call_prepare = filelayout_read_prepare, + .rpc_call_done = filelayout_read_call_done, + .rpc_count_stats = filelayout_read_count_stats, + .rpc_release = pnfs_generic_rw_release, +}; + +static const struct rpc_call_ops filelayout_write_call_ops = { + .rpc_call_prepare = filelayout_write_prepare, + .rpc_call_done = filelayout_write_call_done, + .rpc_count_stats = filelayout_write_count_stats, + .rpc_release = pnfs_generic_rw_release, +}; + +static const struct rpc_call_ops filelayout_commit_call_ops = { + .rpc_call_prepare = filelayout_commit_prepare, + .rpc_call_done = pnfs_generic_write_commit_done, + .rpc_count_stats = filelayout_commit_count_stats, + .rpc_release = pnfs_generic_commit_release, +}; + +static enum pnfs_try_status +filelayout_read_pagelist(struct nfs_pgio_header *hdr) +{ + struct pnfs_layout_segment *lseg = hdr->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + loff_t offset = hdr->args.offset; + u32 j, idx; + struct nfs_fh *fh; + + dprintk("--> %s ino %llu pgbase %u req %zu@%llu\n", + __func__, hdr->inode->i_ino, + hdr->args.pgbase, (size_t)hdr->args.count, offset); + + /* Retrieve the correct rpc_client for the byte range */ + j = nfs4_fl_calc_j_index(lseg, offset); + idx = nfs4_fl_calc_ds_index(lseg, j); + ds = nfs4_fl_prepare_ds(lseg, idx); + if (!ds) + return PNFS_NOT_ATTEMPTED; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode); + if (IS_ERR(ds_clnt)) + return PNFS_NOT_ATTEMPTED; + + dprintk("%s USE DS: %s cl_count %d\n", __func__, + ds->ds_remotestr, refcount_read(&ds->ds_clp->cl_count)); + + /* No multipath support. Use first DS */ + refcount_inc(&ds->ds_clp->cl_count); + hdr->ds_clp = ds->ds_clp; + hdr->ds_commit_idx = idx; + fh = nfs4_fl_select_ds_fh(lseg, j); + if (fh) + hdr->args.fh = fh; + + hdr->args.offset = filelayout_get_dserver_offset(lseg, offset); + hdr->mds_offset = offset; + + /* Perform an asynchronous read to ds */ + nfs_initiate_pgio(ds_clnt, hdr, hdr->cred, + NFS_PROTO(hdr->inode), &filelayout_read_call_ops, + 0, RPC_TASK_SOFTCONN, NULL); + return PNFS_ATTEMPTED; +} + +/* Perform async writes. */ +static enum pnfs_try_status +filelayout_write_pagelist(struct nfs_pgio_header *hdr, int sync) +{ + struct pnfs_layout_segment *lseg = hdr->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + loff_t offset = hdr->args.offset; + u32 j, idx; + struct nfs_fh *fh; + + /* Retrieve the correct rpc_client for the byte range */ + j = nfs4_fl_calc_j_index(lseg, offset); + idx = nfs4_fl_calc_ds_index(lseg, j); + ds = nfs4_fl_prepare_ds(lseg, idx); + if (!ds) + return PNFS_NOT_ATTEMPTED; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, hdr->inode); + if (IS_ERR(ds_clnt)) + return PNFS_NOT_ATTEMPTED; + + dprintk("%s ino %llu sync %d req %zu@%llu DS: %s cl_count %d\n", + __func__, hdr->inode->i_ino, sync, (size_t) hdr->args.count, + offset, ds->ds_remotestr, refcount_read(&ds->ds_clp->cl_count)); + + hdr->pgio_done_cb = filelayout_write_done_cb; + refcount_inc(&ds->ds_clp->cl_count); + hdr->ds_clp = ds->ds_clp; + hdr->ds_commit_idx = idx; + fh = nfs4_fl_select_ds_fh(lseg, j); + if (fh) + hdr->args.fh = fh; + hdr->args.offset = filelayout_get_dserver_offset(lseg, offset); + + /* Perform an asynchronous write */ + nfs_initiate_pgio(ds_clnt, hdr, hdr->cred, + NFS_PROTO(hdr->inode), &filelayout_write_call_ops, + sync, RPC_TASK_SOFTCONN, NULL); + return PNFS_ATTEMPTED; +} + +static int +filelayout_check_deviceid(struct pnfs_layout_hdr *lo, + struct nfs4_filelayout_segment *fl, + gfp_t gfp_flags) +{ + struct nfs4_deviceid_node *d; + struct nfs4_file_layout_dsaddr *dsaddr; + int status = -EINVAL; + + /* Is the deviceid already set? If so, we're good. */ + if (fl->dsaddr != NULL) + return 0; + + /* find and reference the deviceid */ + d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), &fl->deviceid, + lo->plh_lc_cred, gfp_flags); + if (d == NULL) + goto out; + + dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node); + /* Found deviceid is unavailable */ + if (filelayout_test_devid_unavailable(&dsaddr->id_node)) + goto out_put; + + if (fl->first_stripe_index >= dsaddr->stripe_count) { + dprintk("%s Bad first_stripe_index %u\n", + __func__, fl->first_stripe_index); + goto out_put; + } + + if ((fl->stripe_type == STRIPE_SPARSE && + fl->num_fh > 1 && fl->num_fh != dsaddr->ds_num) || + (fl->stripe_type == STRIPE_DENSE && + fl->num_fh != dsaddr->stripe_count)) { + dprintk("%s num_fh %u not valid for given packing\n", + __func__, fl->num_fh); + goto out_put; + } + status = 0; + + /* + * Atomic compare and xchange to ensure we don't scribble + * over a non-NULL pointer. + */ + if (cmpxchg(&fl->dsaddr, NULL, dsaddr) != NULL) + goto out_put; +out: + return status; +out_put: + nfs4_fl_put_deviceid(dsaddr); + goto out; +} + +/* + * filelayout_check_layout() + * + * Make sure layout segment parameters are sane WRT the device. + * At this point no generic layer initialization of the lseg has occurred, + * and nothing has been added to the layout_hdr cache. + * + */ +static int +filelayout_check_layout(struct pnfs_layout_hdr *lo, + struct nfs4_filelayout_segment *fl, + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) +{ + int status = -EINVAL; + + dprintk("--> %s\n", __func__); + + if (fl->pattern_offset > lgr->range.offset) { + dprintk("%s pattern_offset %lld too large\n", + __func__, fl->pattern_offset); + goto out; + } + + if (!fl->stripe_unit) { + dprintk("%s Invalid stripe unit (%u)\n", + __func__, fl->stripe_unit); + goto out; + } + + status = 0; +out: + dprintk("--> %s returns %d\n", __func__, status); + return status; +} + +static void _filelayout_free_lseg(struct nfs4_filelayout_segment *fl) +{ + int i; + + if (fl->fh_array) { + for (i = 0; i < fl->num_fh; i++) { + if (!fl->fh_array[i]) + break; + kfree(fl->fh_array[i]); + } + kfree(fl->fh_array); + } + kfree(fl); +} + +static int +filelayout_decode_layout(struct pnfs_layout_hdr *flo, + struct nfs4_filelayout_segment *fl, + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) +{ + struct xdr_stream stream; + struct xdr_buf buf; + struct folio *scratch; + __be32 *p; + uint32_t nfl_util; + int i; + + dprintk("%s: set_layout_map Begin\n", __func__); + + scratch = folio_alloc(gfp_flags, 0); + if (!scratch) + return -ENOMEM; + + xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, lgr->layoutp->len); + xdr_set_scratch_folio(&stream, scratch); + + /* 20 = ufl_util (4), first_stripe_index (4), pattern_offset (8), + * num_fh (4) */ + p = xdr_inline_decode(&stream, NFS4_DEVICEID4_SIZE + 20); + if (unlikely(!p)) + goto out_err; + + memcpy(&fl->deviceid, p, sizeof(fl->deviceid)); + p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); + nfs4_print_deviceid(&fl->deviceid); + + nfl_util = be32_to_cpup(p++); + if (nfl_util & NFL4_UFLG_COMMIT_THRU_MDS) + fl->commit_through_mds = 1; + if (nfl_util & NFL4_UFLG_DENSE) + fl->stripe_type = STRIPE_DENSE; + else + fl->stripe_type = STRIPE_SPARSE; + fl->stripe_unit = nfl_util & ~NFL4_UFLG_MASK; + + fl->first_stripe_index = be32_to_cpup(p++); + p = xdr_decode_hyper(p, &fl->pattern_offset); + fl->num_fh = be32_to_cpup(p++); + + dprintk("%s: nfl_util 0x%X num_fh %u fsi %u po %llu\n", + __func__, nfl_util, fl->num_fh, fl->first_stripe_index, + fl->pattern_offset); + + /* Note that a zero value for num_fh is legal for STRIPE_SPARSE. + * Futher checking is done in filelayout_check_layout */ + if (fl->num_fh > + max(NFS4_PNFS_MAX_STRIPE_CNT, NFS4_PNFS_MAX_MULTI_CNT)) + goto out_err; + + if (fl->num_fh > 0) { + fl->fh_array = kzalloc_objs(fl->fh_array[0], fl->num_fh, + gfp_flags); + if (!fl->fh_array) + goto out_err; + } + + for (i = 0; i < fl->num_fh; i++) { + /* Do we want to use a mempool here? */ + fl->fh_array[i] = kmalloc_obj(struct nfs_fh, gfp_flags); + if (!fl->fh_array[i]) + goto out_err; + + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err; + fl->fh_array[i]->size = be32_to_cpup(p++); + if (fl->fh_array[i]->size > NFS_MAXFHSIZE) { + printk(KERN_ERR "NFS: Too big fh %d received %d\n", + i, fl->fh_array[i]->size); + goto out_err; + } + + p = xdr_inline_decode(&stream, fl->fh_array[i]->size); + if (unlikely(!p)) + goto out_err; + memcpy(fl->fh_array[i]->data, p, fl->fh_array[i]->size); + dprintk("DEBUG: %s: fh len %d\n", __func__, + fl->fh_array[i]->size); + } + + folio_put(scratch); + return 0; + +out_err: + folio_put(scratch); + return -EIO; +} + +static void +filelayout_free_lseg(struct pnfs_layout_segment *lseg) +{ + struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); + + dprintk("--> %s\n", __func__); + if (fl->dsaddr != NULL) + nfs4_fl_put_deviceid(fl->dsaddr); + /* This assumes a single RW lseg */ + if (lseg->pls_range.iomode == IOMODE_RW) { + struct nfs4_filelayout *flo; + struct inode *inode; + + flo = FILELAYOUT_FROM_HDR(lseg->pls_layout); + inode = flo->generic_hdr.plh_inode; + spin_lock(&inode->i_lock); + pnfs_generic_ds_cinfo_release_lseg(&flo->commit_info, lseg); + spin_unlock(&inode->i_lock); + } + _filelayout_free_lseg(fl); +} + +static struct pnfs_layout_segment * +filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) +{ + struct nfs4_filelayout_segment *fl; + int rc; + + dprintk("--> %s\n", __func__); + fl = kzalloc_obj(*fl, gfp_flags); + if (!fl) + return NULL; + + rc = filelayout_decode_layout(layoutid, fl, lgr, gfp_flags); + if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, gfp_flags)) { + _filelayout_free_lseg(fl); + return NULL; + } + return &fl->generic_hdr; +} + +static bool +filelayout_lseg_is_striped(const struct nfs4_filelayout_segment *flseg) +{ + if (flseg->dsaddr) + return flseg->dsaddr->stripe_count > 1; + return flseg->num_fh > 1; +} + +/* + * filelayout_pg_test(). Called by nfs_can_coalesce_requests() + * + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t +filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + unsigned int size; + u64 p_stripe, r_stripe; + u32 stripe_offset; + u64 segment_offset = pgio->pg_lseg->pls_range.offset; + u32 stripe_unit = FILELAYOUT_LSEG(pgio->pg_lseg)->stripe_unit; + + /* calls nfs_generic_pg_test */ + size = pnfs_generic_pg_test(pgio, prev, req); + if (!size) + return 0; + else if (!filelayout_lseg_is_striped(FILELAYOUT_LSEG(pgio->pg_lseg))) + return size; + + /* see if req and prev are in the same stripe */ + if (prev) { + p_stripe = (u64)req_offset(prev) - segment_offset; + r_stripe = (u64)req_offset(req) - segment_offset; + do_div(p_stripe, stripe_unit); + do_div(r_stripe, stripe_unit); + + if (p_stripe != r_stripe) + return 0; + } + + /* calculate remaining bytes in the current stripe */ + div_u64_rem((u64)req_offset(req) - segment_offset, + stripe_unit, + &stripe_offset); + WARN_ON_ONCE(stripe_offset > stripe_unit); + if (stripe_offset >= stripe_unit) + return 0; + return min(stripe_unit - (unsigned int)stripe_offset, size); +} + +static struct pnfs_layout_segment * +fl_pnfs_update_layout(struct inode *ino, + struct nfs_open_context *ctx, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + bool strict_iomode, + gfp_t gfp_flags) +{ + struct pnfs_layout_segment *lseg = NULL; + struct pnfs_layout_hdr *lo; + struct nfs4_filelayout_segment *fl; + int status; + + lseg = pnfs_update_layout(ino, ctx, pos, count, iomode, strict_iomode, + gfp_flags); + if (IS_ERR(lseg)) { + /* Fall back to MDS on recoverable errors */ + if (!nfs_error_is_fatal_on_server(PTR_ERR(lseg))) + lseg = NULL; + goto out; + } else if (!lseg) + goto out; + + lo = NFS_I(ino)->layout; + fl = FILELAYOUT_LSEG(lseg); + + status = filelayout_check_deviceid(lo, fl, gfp_flags); + if (status) { + pnfs_error_mark_layout_for_return(ino, lseg); + pnfs_set_lo_fail(lseg); + pnfs_put_lseg(lseg); + lseg = NULL; + } +out: + return lseg; +} + +static void +filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + pnfs_generic_pg_check_layout(pgio, req); + if (!pgio->pg_lseg) { + pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, + nfs_req_openctx(req), + req_offset(req), + req->wb_bytes, + IOMODE_READ, + false, + nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + return; + } + } + /* If no lseg, fall back to read through mds */ + if (pgio->pg_lseg == NULL) + nfs_pageio_reset_read_mds(pgio); +} + +static void +filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + pnfs_generic_pg_check_layout(pgio, req); + if (!pgio->pg_lseg) { + pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode, + nfs_req_openctx(req), + req_offset(req), + req->wb_bytes, + IOMODE_RW, + false, + nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + return; + } + } + + /* If no lseg, fall back to write through mds */ + if (pgio->pg_lseg == NULL) + nfs_pageio_reset_write_mds(pgio); +} + +static const struct nfs_pageio_ops filelayout_pg_read_ops = { + .pg_init = filelayout_pg_init_read, + .pg_test = filelayout_pg_test, + .pg_doio = pnfs_generic_pg_readpages, + .pg_cleanup = pnfs_generic_pg_cleanup, +}; + +static const struct nfs_pageio_ops filelayout_pg_write_ops = { + .pg_init = filelayout_pg_init_write, + .pg_test = filelayout_pg_test, + .pg_doio = pnfs_generic_pg_writepages, + .pg_cleanup = pnfs_generic_pg_cleanup, +}; + +static u32 select_bucket_index(struct nfs4_filelayout_segment *fl, u32 j) +{ + if (fl->stripe_type == STRIPE_SPARSE) + return nfs4_fl_calc_ds_index(&fl->generic_hdr, j); + else + return j; +} + +static void +filelayout_mark_request_commit(struct nfs_page *req, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx) + +{ + struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); + u32 i, j; + + if (fl->commit_through_mds) { + nfs_request_add_commit_list(req, cinfo); + } else { + /* Note that we are calling nfs4_fl_calc_j_index on each page + * that ends up being committed to a data server. An attractive + * alternative is to add a field to nfs_write_data and nfs_page + * to store the value calculated in filelayout_write_pagelist + * and just use that here. + */ + j = nfs4_fl_calc_j_index(lseg, req_offset(req)); + i = select_bucket_index(fl, j); + pnfs_layout_mark_request_commit(req, lseg, cinfo, i); + } +} + +static u32 calc_ds_index_from_commit(struct pnfs_layout_segment *lseg, u32 i) +{ + struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); + + if (flseg->stripe_type == STRIPE_SPARSE) + return i; + else + return nfs4_fl_calc_ds_index(lseg, i); +} + +static struct nfs_fh * +select_ds_fh_from_commit(struct pnfs_layout_segment *lseg, u32 i) +{ + struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); + + if (flseg->stripe_type == STRIPE_SPARSE) { + if (flseg->num_fh == 1) + i = 0; + else if (flseg->num_fh == 0) + /* Use the MDS OPEN fh set in nfs_read_rpcsetup */ + return NULL; + } + return flseg->fh_array[i]; +} + +static int filelayout_initiate_commit(struct nfs_commit_data *data, int how) +{ + struct pnfs_layout_segment *lseg = data->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + u32 idx; + struct nfs_fh *fh; + + idx = calc_ds_index_from_commit(lseg, data->ds_commit_index); + ds = nfs4_fl_prepare_ds(lseg, idx); + if (!ds) + goto out_err; + + ds_clnt = nfs4_find_or_create_ds_client(ds->ds_clp, data->inode); + if (IS_ERR(ds_clnt)) + goto out_err; + + dprintk("%s ino %llu, how %d cl_count %d\n", __func__, + data->inode->i_ino, how, refcount_read(&ds->ds_clp->cl_count)); + data->commit_done_cb = filelayout_commit_done_cb; + refcount_inc(&ds->ds_clp->cl_count); + data->ds_clp = ds->ds_clp; + fh = select_ds_fh_from_commit(lseg, data->ds_commit_index); + if (fh) + data->args.fh = fh; + return nfs_initiate_commit(ds_clnt, data, NFS_PROTO(data->inode), + &filelayout_commit_call_ops, how, + RPC_TASK_SOFTCONN, NULL); +out_err: + pnfs_generic_prepare_to_resend_writes(data); + pnfs_generic_commit_release(data); + return -EAGAIN; +} + +static int +filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages, + int how, struct nfs_commit_info *cinfo) +{ + return pnfs_generic_commit_pagelist(inode, mds_pages, how, cinfo, + filelayout_initiate_commit); +} + +static struct nfs4_deviceid_node * +filelayout_alloc_deviceid_node(struct nfs_server *server, + struct pnfs_device *pdev, gfp_t gfp_flags) +{ + struct nfs4_file_layout_dsaddr *dsaddr; + + dsaddr = nfs4_fl_alloc_deviceid_node(server, pdev, gfp_flags); + if (!dsaddr) + return NULL; + return &dsaddr->id_node; +} + +static void +filelayout_free_deviceid_node(struct nfs4_deviceid_node *d) +{ + nfs4_fl_free_deviceid(container_of(d, struct nfs4_file_layout_dsaddr, id_node)); +} + +static struct pnfs_layout_hdr * +filelayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags) +{ + struct nfs4_filelayout *flo; + + flo = kzalloc_obj(*flo, gfp_flags); + if (flo == NULL) + return NULL; + pnfs_init_ds_commit_info(&flo->commit_info); + flo->commit_info.ops = &filelayout_commit_ops; + return &flo->generic_hdr; +} + +static void +filelayout_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + kfree_rcu(FILELAYOUT_FROM_HDR(lo), generic_hdr.plh_rcu); +} + +static struct pnfs_ds_commit_info * +filelayout_get_ds_info(struct inode *inode) +{ + struct pnfs_layout_hdr *layout = NFS_I(inode)->layout; + + if (layout == NULL) + return NULL; + else + return &FILELAYOUT_FROM_HDR(layout)->commit_info; +} + +static void +filelayout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg) +{ + struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); + struct inode *inode = lseg->pls_layout->plh_inode; + struct pnfs_commit_array *array, *new; + unsigned int size = (fl->stripe_type == STRIPE_SPARSE) ? + fl->dsaddr->ds_num : fl->dsaddr->stripe_count; + + new = pnfs_alloc_commit_array(size, nfs_io_gfp_mask()); + if (new) { + spin_lock(&inode->i_lock); + array = pnfs_add_commit_array(fl_cinfo, new, lseg); + spin_unlock(&inode->i_lock); + if (array != new) + pnfs_free_commit_array(new); + } +} + +static void +filelayout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, + struct inode *inode) +{ + spin_lock(&inode->i_lock); + pnfs_generic_ds_cinfo_destroy(fl_cinfo); + spin_unlock(&inode->i_lock); +} + +static const struct pnfs_commit_ops filelayout_commit_ops = { + .setup_ds_info = filelayout_setup_ds_info, + .release_ds_info = filelayout_release_ds_info, + .mark_request_commit = filelayout_mark_request_commit, + .clear_request_commit = pnfs_generic_clear_request_commit, + .scan_commit_lists = pnfs_generic_scan_commit_lists, + .recover_commit_reqs = pnfs_generic_recover_commit_reqs, + .commit_pagelist = filelayout_commit_pagelist, +}; + +static struct pnfs_layoutdriver_type filelayout_type = { + .id = LAYOUT_NFSV4_1_FILES, + .name = "LAYOUT_NFSV4_1_FILES", + .owner = THIS_MODULE, + .flags = PNFS_LAYOUTGET_ON_OPEN, + .max_layoutget_response = 4096, /* 1 page or so... */ + .alloc_layout_hdr = filelayout_alloc_layout_hdr, + .free_layout_hdr = filelayout_free_layout_hdr, + .alloc_lseg = filelayout_alloc_lseg, + .free_lseg = filelayout_free_lseg, + .pg_read_ops = &filelayout_pg_read_ops, + .pg_write_ops = &filelayout_pg_write_ops, + .get_ds_info = &filelayout_get_ds_info, + .read_pagelist = filelayout_read_pagelist, + .write_pagelist = filelayout_write_pagelist, + .alloc_deviceid_node = filelayout_alloc_deviceid_node, + .free_deviceid_node = filelayout_free_deviceid_node, + .sync = pnfs_nfs_generic_sync, +}; + +static int __init nfs4filelayout_init(void) +{ + printk(KERN_INFO "%s: NFSv4 File Layout Driver Registering...\n", + __func__); + return pnfs_register_layoutdriver(&filelayout_type); +} + +static void __exit nfs4filelayout_exit(void) +{ + printk(KERN_INFO "%s: NFSv4 File Layout Driver Unregistering...\n", + __func__); + pnfs_unregister_layoutdriver(&filelayout_type); +} + +MODULE_ALIAS("nfs-layouttype4-1"); + +module_init(nfs4filelayout_init); +module_exit(nfs4filelayout_exit); diff --git a/fs/nfs/filelayout/filelayout.h b/fs/nfs/filelayout/filelayout.h new file mode 100644 index 000000000..c7bb5da93 --- /dev/null +++ b/fs/nfs/filelayout/filelayout.h @@ -0,0 +1,118 @@ +/* + * NFSv4 file layout driver data structures. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#ifndef FS_NFS_NFS4FILELAYOUT_H +#define FS_NFS_NFS4FILELAYOUT_H + +#include "../pnfs.h" + +/* + * Field testing shows we need to support up to 4096 stripe indices. + * We store each index as a u8 (u32 on the wire) to keep the memory footprint + * reasonable. This in turn means we support a maximum of 256 + * RFC 5661 multipath_list4 structures. + */ +#define NFS4_PNFS_MAX_STRIPE_CNT 4096 +#define NFS4_PNFS_MAX_MULTI_CNT 256 /* 256 fit into a u8 stripe_index */ + +enum stripetype4 { + STRIPE_SPARSE = 1, + STRIPE_DENSE = 2 +}; + +struct nfs4_file_layout_dsaddr { + struct nfs4_deviceid_node id_node; + u32 stripe_count; + u8 *stripe_indices; + u32 ds_num; + struct nfs4_pnfs_ds *ds_list[] __counted_by(ds_num); +}; + +struct nfs4_filelayout_segment { + struct pnfs_layout_segment generic_hdr; + u32 stripe_type; + u32 commit_through_mds; + u32 stripe_unit; + u32 first_stripe_index; + u64 pattern_offset; + struct nfs4_deviceid deviceid; + struct nfs4_file_layout_dsaddr *dsaddr; /* Point to GETDEVINFO data */ + unsigned int num_fh; + struct nfs_fh **fh_array; +}; + +struct nfs4_filelayout { + struct pnfs_layout_hdr generic_hdr; + struct pnfs_ds_commit_info commit_info; +}; + +static inline struct nfs4_filelayout * +FILELAYOUT_FROM_HDR(struct pnfs_layout_hdr *lo) +{ + return container_of(lo, struct nfs4_filelayout, generic_hdr); +} + +static inline struct nfs4_filelayout_segment * +FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg) +{ + return container_of(lseg, + struct nfs4_filelayout_segment, + generic_hdr); +} + +static inline struct nfs4_deviceid_node * +FILELAYOUT_DEVID_NODE(struct pnfs_layout_segment *lseg) +{ + return &FILELAYOUT_LSEG(lseg)->dsaddr->id_node; +} + +static inline bool +filelayout_test_devid_invalid(struct nfs4_deviceid_node *node) +{ + return test_bit(NFS_DEVICEID_INVALID, &node->flags); +} + +extern bool +filelayout_test_devid_unavailable(struct nfs4_deviceid_node *node); + +extern struct nfs_fh * +nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j); + +u32 nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset); +u32 nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j); +struct nfs4_pnfs_ds *nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, + u32 ds_idx); + +extern struct nfs4_file_layout_dsaddr * +nfs4_fl_alloc_deviceid_node(struct nfs_server *server, + struct pnfs_device *pdev, gfp_t gfp_flags); +extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); +extern void nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); + +#endif /* FS_NFS_NFS4FILELAYOUT_H */ diff --git a/fs/nfs/filelayout/filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c new file mode 100644 index 000000000..7226989ee --- /dev/null +++ b/fs/nfs/filelayout/filelayoutdev.c @@ -0,0 +1,305 @@ +/* + * Device operations for the pnfs nfs4 file layout driver. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * Garth Goodson + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#include +#include +#include + +#include "../internal.h" +#include "../nfs4session.h" +#include "filelayout.h" +#include "../nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +static unsigned int dataserver_timeo = NFS4_DEF_DS_TIMEO; +static unsigned int dataserver_retrans = NFS4_DEF_DS_RETRANS; + +void +nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr) +{ + struct nfs4_pnfs_ds *ds; + int i; + + nfs4_print_deviceid(&dsaddr->id_node.deviceid); + + for (i = 0; i < dsaddr->ds_num; i++) { + ds = dsaddr->ds_list[i]; + if (ds != NULL) + nfs4_pnfs_ds_put(ds); + } + kfree(dsaddr->stripe_indices); + kfree_rcu(dsaddr, id_node.rcu); +} + +/* Decode opaque device data and return the result */ +struct nfs4_file_layout_dsaddr * +nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, + gfp_t gfp_flags) +{ + int i; + u32 cnt, num; + u8 *indexp; + __be32 *p; + u8 *stripe_indices; + u8 max_stripe_index; + struct nfs4_file_layout_dsaddr *dsaddr = NULL; + struct xdr_stream stream; + struct xdr_buf buf; + struct folio *scratch; + struct list_head dsaddrs; + struct nfs4_pnfs_ds_addr *da; + struct net *net = server->nfs_client->cl_net; + + /* set up xdr stream */ + scratch = folio_alloc(gfp_flags, 0); + if (!scratch) + goto out_err; + + xdr_init_decode_pages(&stream, &buf, pdev->pages, pdev->pglen); + xdr_set_scratch_folio(&stream, scratch); + + /* Get the stripe count (number of stripe index) */ + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err_free_scratch; + + cnt = be32_to_cpup(p); + dprintk("%s stripe count %d\n", __func__, cnt); + if (cnt > NFS4_PNFS_MAX_STRIPE_CNT) { + printk(KERN_WARNING "NFS: %s: stripe count %d greater than " + "supported maximum %d\n", __func__, + cnt, NFS4_PNFS_MAX_STRIPE_CNT); + goto out_err_free_scratch; + } + + /* read stripe indices */ + stripe_indices = kcalloc(cnt, sizeof(u8), gfp_flags); + if (!stripe_indices) + goto out_err_free_scratch; + + p = xdr_inline_decode(&stream, cnt << 2); + if (unlikely(!p)) + goto out_err_free_stripe_indices; + + indexp = &stripe_indices[0]; + max_stripe_index = 0; + for (i = 0; i < cnt; i++) { + *indexp = be32_to_cpup(p++); + max_stripe_index = max(max_stripe_index, *indexp); + indexp++; + } + + /* Check the multipath list count */ + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err_free_stripe_indices; + + num = be32_to_cpup(p); + dprintk("%s ds_num %u\n", __func__, num); + if (num > NFS4_PNFS_MAX_MULTI_CNT) { + printk(KERN_WARNING "NFS: %s: multipath count %d greater than " + "supported maximum %d\n", __func__, + num, NFS4_PNFS_MAX_MULTI_CNT); + goto out_err_free_stripe_indices; + } + + /* validate stripe indices are all < num */ + if (max_stripe_index >= num) { + printk(KERN_WARNING "NFS: %s: stripe index %u >= num ds %u\n", + __func__, max_stripe_index, num); + goto out_err_free_stripe_indices; + } + + dsaddr = kzalloc_flex(*dsaddr, ds_list, num, gfp_flags); + if (!dsaddr) + goto out_err_free_stripe_indices; + + dsaddr->stripe_count = cnt; + dsaddr->stripe_indices = stripe_indices; + stripe_indices = NULL; + dsaddr->ds_num = num; + nfs4_init_deviceid_node(&dsaddr->id_node, server, &pdev->dev_id); + + INIT_LIST_HEAD(&dsaddrs); + + for (i = 0; i < dsaddr->ds_num; i++) { + int j; + u32 mp_count; + + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err_free_deviceid; + + mp_count = be32_to_cpup(p); /* multipath count */ + for (j = 0; j < mp_count; j++) { + da = nfs4_decode_mp_ds_addr(net, &stream, gfp_flags); + if (da) + list_add_tail(&da->da_node, &dsaddrs); + } + if (list_empty(&dsaddrs)) { + dprintk("%s: no suitable DS addresses found\n", + __func__); + goto out_err_free_deviceid; + } + + dsaddr->ds_list[i] = nfs4_pnfs_ds_add(net, &dsaddrs, gfp_flags); + if (!dsaddr->ds_list[i]) + goto out_err_drain_dsaddrs; + trace_fl_getdevinfo(server, &pdev->dev_id, dsaddr->ds_list[i]->ds_remotestr); + + /* If DS was already in cache, free ds addrs */ + while (!list_empty(&dsaddrs)) { + da = list_first_entry(&dsaddrs, + struct nfs4_pnfs_ds_addr, + da_node); + list_del_init(&da->da_node); + kfree(da->da_remotestr); + kfree(da); + } + } + + folio_put(scratch); + return dsaddr; + +out_err_drain_dsaddrs: + while (!list_empty(&dsaddrs)) { + da = list_first_entry(&dsaddrs, struct nfs4_pnfs_ds_addr, + da_node); + list_del_init(&da->da_node); + kfree(da->da_remotestr); + kfree(da); + } +out_err_free_deviceid: + nfs4_fl_free_deviceid(dsaddr); + /* stripe_indicies was part of dsaddr */ + goto out_err_free_scratch; +out_err_free_stripe_indices: + kfree(stripe_indices); +out_err_free_scratch: + folio_put(scratch); +out_err: + dprintk("%s ERROR: returning NULL\n", __func__); + return NULL; +} + +void +nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr) +{ + nfs4_put_deviceid_node(&dsaddr->id_node); +} + +/* + * Want res = (offset - layout->pattern_offset)/ layout->stripe_unit + * Then: ((res + fsi) % dsaddr->stripe_count) + */ +u32 +nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset) +{ + struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); + u64 tmp; + + tmp = offset - flseg->pattern_offset; + do_div(tmp, flseg->stripe_unit); + tmp += flseg->first_stripe_index; + return do_div(tmp, flseg->dsaddr->stripe_count); +} + +u32 +nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j) +{ + return FILELAYOUT_LSEG(lseg)->dsaddr->stripe_indices[j]; +} + +struct nfs_fh * +nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j) +{ + struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg); + u32 i; + + if (flseg->stripe_type == STRIPE_SPARSE) { + if (flseg->num_fh == 1) + i = 0; + else if (flseg->num_fh == 0) + /* Use the MDS OPEN fh set in nfs_read_rpcsetup */ + return NULL; + else + i = nfs4_fl_calc_ds_index(lseg, j); + } else + i = j; + return flseg->fh_array[i]; +} + +/* Upon return, either ds is connected, or ds is NULL */ +struct nfs4_pnfs_ds * +nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx) +{ + struct nfs4_file_layout_dsaddr *dsaddr = FILELAYOUT_LSEG(lseg)->dsaddr; + struct nfs4_pnfs_ds *ds = dsaddr->ds_list[ds_idx]; + struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(lseg); + struct nfs4_pnfs_ds *ret = ds; + struct nfs_server *s = NFS_SERVER(lseg->pls_layout->plh_inode); + int status; + + if (ds == NULL) { + printk(KERN_ERR "NFS: %s: No data server for offset index %d\n", + __func__, ds_idx); + pnfs_generic_mark_devid_invalid(devid); + goto out; + } + smp_rmb(); + if (ds->ds_clp) + goto out_test_devid; + + status = nfs4_pnfs_ds_connect(s, ds, devid, dataserver_timeo, + dataserver_retrans, 4, + s->nfs_client->cl_minorversion); + if (status) { + nfs4_mark_deviceid_unavailable(devid); + ret = NULL; + goto out; + } + +out_test_devid: + if (ret->ds_clp == NULL || + filelayout_test_devid_unavailable(devid)) + ret = NULL; +out: + return ret; +} + +module_param(dataserver_retrans, uint, 0644); +MODULE_PARM_DESC(dataserver_retrans, "The number of times the NFSv4.1 client " + "retries a request before it attempts further " + " recovery action."); +module_param(dataserver_timeo, uint, 0644); +MODULE_PARM_DESC(dataserver_timeo, "The time (in tenths of a second) the " + "NFSv4.1 client waits for a response from a " + " data server before it retries an NFS request."); diff --git a/fs/nfs/flexfilelayout/Makefile b/fs/nfs/flexfilelayout/Makefile new file mode 100644 index 000000000..49f03422b --- /dev/null +++ b/fs/nfs/flexfilelayout/Makefile @@ -0,0 +1,6 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Makefile for the pNFS Flexfile Layout Driver kernel module +# +obj-$(CONFIG_PNFS_FLEXFILE_LAYOUT) += nfs_layout_flexfiles.o +nfs_layout_flexfiles-y := flexfilelayout.o flexfilelayoutdev.o diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c new file mode 100644 index 000000000..c4aa99502 --- /dev/null +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -0,0 +1,3069 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Module for pnfs flexfile layout driver. + * + * Copyright (c) 2014, Primary Data, Inc. All rights reserved. + * + * Tao Peng + */ + +#include +#include +#include +#include +#include +#include + +#include + +#include "flexfilelayout.h" +#include "../nfs4session.h" +#include "../nfs4idmap.h" +#include "../internal.h" +#include "../delegation.h" +#include "../nfs4trace.h" +#include "../iostat.h" +#include "../nfs.h" +#include "../nfs42.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +#define FF_LAYOUT_POLL_RETRY_MAX (15*HZ) +#define FF_LAYOUTRETURN_MAXERR 20 + +enum nfs4_ff_op_type { + NFS4_FF_OP_LAYOUTSTATS, + NFS4_FF_OP_LAYOUTRETURN, +}; + +static unsigned short io_maxretrans; + +static const struct pnfs_commit_ops ff_layout_commit_ops; +static void ff_layout_read_record_layoutstats_done(struct rpc_task *task, + struct nfs_pgio_header *hdr); +static int +ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, + struct nfs42_layoutstat_devinfo *devinfo, + int dev_limit, enum nfs4_ff_op_type type); +static void ff_layout_encode_ff_layoutupdate(struct xdr_stream *xdr, + const struct nfs42_layoutstat_devinfo *devinfo, + struct nfs4_ff_layout_ds_stripe *dss_info); + +static struct pnfs_layout_hdr * +ff_layout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags) +{ + struct nfs4_flexfile_layout *ffl; + + ffl = kzalloc_obj(*ffl, gfp_flags); + if (ffl) { + pnfs_init_ds_commit_info(&ffl->commit_info); + INIT_LIST_HEAD(&ffl->error_list); + INIT_LIST_HEAD(&ffl->mirrors); + ffl->last_report_time = ktime_get(); + ffl->commit_info.ops = &ff_layout_commit_ops; + return &ffl->generic_hdr; + } else + return NULL; +} + +static void +ff_layout_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct nfs4_flexfile_layout *ffl = FF_LAYOUT_FROM_HDR(lo); + struct nfs4_ff_layout_ds_err *err, *n; + + list_for_each_entry_safe(err, n, &ffl->error_list, list) { + list_del(&err->list); + kfree(err); + } + kfree_rcu(ffl, generic_hdr.plh_rcu); +} + +static int decode_pnfs_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS4_STATEID_SIZE); + if (unlikely(p == NULL)) + return -ENOBUFS; + stateid->type = NFS4_PNFS_DS_STATEID_TYPE; + memcpy(stateid->data, p, NFS4_STATEID_SIZE); + dprintk("%s: stateid id= [%x%x%x%x]\n", __func__, + p[0], p[1], p[2], p[3]); + return 0; +} + +static int decode_deviceid(struct xdr_stream *xdr, struct nfs4_deviceid *devid) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS4_DEVICEID4_SIZE); + if (unlikely(!p)) + return -ENOBUFS; + memcpy(devid, p, NFS4_DEVICEID4_SIZE); + nfs4_print_deviceid(devid); + return 0; +} + +static int decode_nfs_fh(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -ENOBUFS; + fh->size = be32_to_cpup(p++); + if (fh->size > NFS_MAXFHSIZE) { + printk(KERN_ERR "NFS flexfiles: Too big fh received %d\n", + fh->size); + return -EOVERFLOW; + } + /* fh.data */ + p = xdr_inline_decode(xdr, fh->size); + if (unlikely(!p)) + return -ENOBUFS; + memcpy(&fh->data, p, fh->size); + dprintk("%s: fh len %d\n", __func__, fh->size); + + return 0; +} + +/* + * Currently only stringified uids and gids are accepted. + * I.e., kerberos is not supported to the DSes, so no pricipals. + * + * That means that one common function will suffice, but when + * principals are added, this should be split to accomodate + * calls to both nfs_map_name_to_uid() and nfs_map_group_to_gid(). + */ +static int +decode_name(struct xdr_stream *xdr, u32 *id) +{ + __be32 *p; + int len; + + /* opaque_length(4)*/ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -ENOBUFS; + len = be32_to_cpup(p++); + if (len < 0) + return -EINVAL; + + dprintk("%s: len %u\n", __func__, len); + + /* opaque body */ + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -ENOBUFS; + + if (!nfs_map_string_to_numeric((char *)p, len, id)) + return -EINVAL; + + return 0; +} + +static struct nfsd_file * +ff_local_open_fh(struct pnfs_layout_segment *lseg, u32 ds_idx, u32 dss_id, + struct nfs_client *clp, const struct cred *cred, + struct nfs_fh *fh, fmode_t mode) +{ +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + struct nfs4_ff_layout_mirror *mirror = FF_LAYOUT_COMP(lseg, ds_idx); + + return nfs_local_open_fh(clp, cred, fh, &mirror->dss[dss_id].nfl, mode); +#else + return NULL; +#endif +} + +static bool ff_dss_match_fh(const struct nfs4_ff_layout_ds_stripe *dss1, + const struct nfs4_ff_layout_ds_stripe *dss2) +{ + int i, j; + + if (dss1->fh_versions_cnt != dss2->fh_versions_cnt) + return false; + + for (i = 0; i < dss1->fh_versions_cnt; i++) { + bool found_fh = false; + for (j = 0; j < dss2->fh_versions_cnt; j++) { + if (nfs_compare_fh(&dss1->fh_versions[i], + &dss2->fh_versions[j]) == 0) { + found_fh = true; + break; + } + } + if (!found_fh) + return false; + } + return true; +} + +static bool ff_mirror_match_fh(const struct nfs4_ff_layout_mirror *m1, + const struct nfs4_ff_layout_mirror *m2) +{ + u32 dss_id; + + if (m1->dss_count != m2->dss_count) + return false; + + for (dss_id = 0; dss_id < m1->dss_count; dss_id++) + if (!ff_dss_match_fh(&m1->dss[dss_id], &m2->dss[dss_id])) + return false; + + return true; +} + +static bool ff_mirror_match_devid(const struct nfs4_ff_layout_mirror *m1, + const struct nfs4_ff_layout_mirror *m2) +{ + u32 dss_id; + + if (m1->dss_count != m2->dss_count) + return false; + + for (dss_id = 0; dss_id < m1->dss_count; dss_id++) + if (memcmp(&m1->dss[dss_id].devid, + &m2->dss[dss_id].devid, + sizeof(m1->dss[dss_id].devid)) != 0) + return false; + + return true; +} + +static struct nfs4_ff_layout_mirror * +ff_layout_add_mirror(struct pnfs_layout_hdr *lo, + struct nfs4_ff_layout_mirror *mirror) +{ + struct nfs4_flexfile_layout *ff_layout = FF_LAYOUT_FROM_HDR(lo); + struct nfs4_ff_layout_mirror *pos; + struct inode *inode = lo->plh_inode; + + spin_lock(&inode->i_lock); + list_for_each_entry(pos, &ff_layout->mirrors, mirrors) { + if (!ff_mirror_match_devid(mirror, pos)) + continue; + if (!ff_mirror_match_fh(mirror, pos)) + continue; + if (refcount_inc_not_zero(&pos->ref)) { + spin_unlock(&inode->i_lock); + return pos; + } + } + list_add(&mirror->mirrors, &ff_layout->mirrors); + mirror->layout = lo; + spin_unlock(&inode->i_lock); + return mirror; +} + +static void +ff_layout_remove_mirror(struct nfs4_ff_layout_mirror *mirror) +{ + struct inode *inode; + if (mirror->layout == NULL) + return; + inode = mirror->layout->plh_inode; + spin_lock(&inode->i_lock); + list_del(&mirror->mirrors); + spin_unlock(&inode->i_lock); + mirror->layout = NULL; +} + +static struct nfs4_ff_layout_mirror *ff_layout_alloc_mirror(u32 dss_count, + gfp_t gfp_flags) +{ + struct nfs4_ff_layout_mirror *mirror; + + mirror = kzalloc_obj(*mirror, gfp_flags); + if (mirror == NULL) + return NULL; + + spin_lock_init(&mirror->lock); + refcount_set(&mirror->ref, 1); + INIT_LIST_HEAD(&mirror->mirrors); + + mirror->dss_count = dss_count; + mirror->dss = + kzalloc_objs(struct nfs4_ff_layout_ds_stripe, dss_count, + gfp_flags); + if (mirror->dss == NULL) { + kfree(mirror); + return NULL; + } + + for (u32 dss_id = 0; dss_id < mirror->dss_count; dss_id++) + nfs_localio_file_init(&mirror->dss[dss_id].nfl); + + return mirror; +} + +static void ff_layout_free_mirror(struct nfs4_ff_layout_mirror *mirror) +{ + const struct cred *cred; + u32 dss_id; + + ff_layout_remove_mirror(mirror); + + for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { + kfree(mirror->dss[dss_id].fh_versions); + cred = rcu_access_pointer(mirror->dss[dss_id].ro_cred); + put_cred(cred); + cred = rcu_access_pointer(mirror->dss[dss_id].rw_cred); + put_cred(cred); + nfs_close_local_fh(&mirror->dss[dss_id].nfl); + nfs4_ff_layout_put_deviceid(mirror->dss[dss_id].mirror_ds); + } + + kfree(mirror->dss); + kfree(mirror); +} + +static void ff_layout_put_mirror(struct nfs4_ff_layout_mirror *mirror) +{ + if (mirror != NULL && refcount_dec_and_test(&mirror->ref)) + ff_layout_free_mirror(mirror); +} + +static void ff_layout_free_mirror_array(struct nfs4_ff_layout_segment *fls) +{ + u32 i; + + for (i = 0; i < fls->mirror_array_cnt; i++) + ff_layout_put_mirror(fls->mirror_array[i]); +} + +static void _ff_layout_free_lseg(struct nfs4_ff_layout_segment *fls) +{ + if (fls) { + ff_layout_free_mirror_array(fls); + kfree(fls); + } +} + +static bool +ff_lseg_match_mirrors(struct pnfs_layout_segment *l1, + struct pnfs_layout_segment *l2) +{ + const struct nfs4_ff_layout_segment *fl1 = FF_LAYOUT_LSEG(l1); + const struct nfs4_ff_layout_segment *fl2 = FF_LAYOUT_LSEG(l2); + u32 i; + + if (fl1->mirror_array_cnt != fl2->mirror_array_cnt) + return false; + for (i = 0; i < fl1->mirror_array_cnt; i++) { + if (fl1->mirror_array[i] != fl2->mirror_array[i]) + return false; + } + return true; +} + +static bool +ff_lseg_range_is_after(const struct pnfs_layout_range *l1, + const struct pnfs_layout_range *l2) +{ + u64 end1, end2; + + if (l1->iomode != l2->iomode) + return l1->iomode != IOMODE_READ; + end1 = pnfs_calc_offset_end(l1->offset, l1->length); + end2 = pnfs_calc_offset_end(l2->offset, l2->length); + if (end1 < l2->offset) + return false; + if (end2 < l1->offset) + return true; + return l2->offset <= l1->offset; +} + +static bool +ff_lseg_merge(struct pnfs_layout_segment *new, + struct pnfs_layout_segment *old) +{ + u64 new_end, old_end; + + if (test_bit(NFS_LSEG_LAYOUTRETURN, &old->pls_flags)) + return false; + if (new->pls_range.iomode != old->pls_range.iomode) + return false; + old_end = pnfs_calc_offset_end(old->pls_range.offset, + old->pls_range.length); + if (old_end < new->pls_range.offset) + return false; + new_end = pnfs_calc_offset_end(new->pls_range.offset, + new->pls_range.length); + if (new_end < old->pls_range.offset) + return false; + if (!ff_lseg_match_mirrors(new, old)) + return false; + + /* Mergeable: copy info from 'old' to 'new' */ + if (new_end < old_end) + new_end = old_end; + if (new->pls_range.offset < old->pls_range.offset) + new->pls_range.offset = old->pls_range.offset; + new->pls_range.length = pnfs_calc_offset_length(new->pls_range.offset, + new_end); + if (test_bit(NFS_LSEG_ROC, &old->pls_flags)) + set_bit(NFS_LSEG_ROC, &new->pls_flags); + return true; +} + +static void +ff_layout_add_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg, + struct list_head *free_me) +{ + pnfs_generic_layout_insert_lseg(lo, lseg, + ff_lseg_range_is_after, + ff_lseg_merge, + free_me); +} + +static u32 ff_mirror_efficiency_sum(const struct nfs4_ff_layout_mirror *mirror) +{ + u32 dss_id, sum = 0; + + for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) + sum += mirror->dss[dss_id].efficiency; + + return sum; +} + +static void ff_layout_sort_mirrors(struct nfs4_ff_layout_segment *fls) +{ + int i, j; + + for (i = 0; i < fls->mirror_array_cnt - 1; i++) { + for (j = i + 1; j < fls->mirror_array_cnt; j++) + if (ff_mirror_efficiency_sum(fls->mirror_array[i]) < + ff_mirror_efficiency_sum(fls->mirror_array[j])) + swap(fls->mirror_array[i], + fls->mirror_array[j]); + } +} + +static struct pnfs_layout_segment * +ff_layout_alloc_lseg(struct pnfs_layout_hdr *lh, + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) +{ + struct pnfs_layout_segment *ret; + struct nfs4_ff_layout_segment *fls = NULL; + struct xdr_stream stream; + struct xdr_buf buf; + struct folio *scratch; + u64 stripe_unit; + u32 mirror_array_cnt; + __be32 *p; + int i, rc; + struct nfs4_ff_layout_ds_stripe *dss_info; + + dprintk("--> %s\n", __func__); + scratch = folio_alloc(gfp_flags, 0); + if (!scratch) + return ERR_PTR(-ENOMEM); + + xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, + lgr->layoutp->len); + xdr_set_scratch_folio(&stream, scratch); + + /* stripe unit and mirror_array_cnt */ + rc = -EIO; + p = xdr_inline_decode(&stream, 8 + 4); + if (!p) + goto out_err_free; + + p = xdr_decode_hyper(p, &stripe_unit); + mirror_array_cnt = be32_to_cpup(p++); + dprintk("%s: stripe_unit=%llu mirror_array_cnt=%u\n", __func__, + stripe_unit, mirror_array_cnt); + + if (mirror_array_cnt > NFS4_FLEXFILE_LAYOUT_MAX_MIRROR_CNT || + mirror_array_cnt == 0) + goto out_err_free; + + rc = -ENOMEM; + fls = kzalloc_flex(*fls, mirror_array, mirror_array_cnt, gfp_flags); + if (!fls) + goto out_err_free; + + fls->mirror_array_cnt = mirror_array_cnt; + fls->stripe_unit = stripe_unit; + + u32 dss_count = 0; + for (i = 0; i < fls->mirror_array_cnt; i++) { + struct nfs4_ff_layout_mirror *mirror; + struct cred *kcred; + const struct cred __rcu *cred; + kuid_t uid; + kgid_t gid; + u32 fh_count, id; + int j, dss_id; + + rc = -EIO; + p = xdr_inline_decode(&stream, 4); + if (!p) + goto out_err_free; + + // Ensure all mirrors have same stripe count. + if (dss_count == 0) + dss_count = be32_to_cpup(p); + else if (dss_count != be32_to_cpup(p)) + goto out_err_free; + + if (dss_count > NFS4_FLEXFILE_LAYOUT_MAX_STRIPE_CNT || + dss_count == 0) + goto out_err_free; + + if (dss_count > 1 && stripe_unit == 0) + goto out_err_free; + + fls->mirror_array[i] = ff_layout_alloc_mirror(dss_count, gfp_flags); + if (fls->mirror_array[i] == NULL) { + rc = -ENOMEM; + goto out_err_free; + } + + for (dss_id = 0; dss_id < dss_count; dss_id++) { + dss_info = &fls->mirror_array[i]->dss[dss_id]; + dss_info->mirror = fls->mirror_array[i]; + + /* deviceid */ + rc = decode_deviceid(&stream, &dss_info->devid); + if (rc) + goto out_err_free; + + /* efficiency */ + rc = -EIO; + p = xdr_inline_decode(&stream, 4); + if (!p) + goto out_err_free; + dss_info->efficiency = be32_to_cpup(p); + + /* stateid */ + rc = decode_pnfs_stateid(&stream, &dss_info->stateid); + if (rc) + goto out_err_free; + + /* fh */ + rc = -EIO; + p = xdr_inline_decode(&stream, 4); + if (!p) + goto out_err_free; + fh_count = be32_to_cpup(p); + if (fh_count == 0) { + rc = -EINVAL; + goto out_err_free; + } + + dss_info->fh_versions = + kzalloc_objs(struct nfs_fh, fh_count, gfp_flags); + if (dss_info->fh_versions == NULL) { + rc = -ENOMEM; + goto out_err_free; + } + + for (j = 0; j < fh_count; j++) { + rc = decode_nfs_fh(&stream, + &dss_info->fh_versions[j]); + if (rc) + goto out_err_free; + } + + dss_info->fh_versions_cnt = fh_count; + + /* user */ + rc = decode_name(&stream, &id); + if (rc) + goto out_err_free; + + uid = make_kuid(&init_user_ns, id); + + /* group */ + rc = decode_name(&stream, &id); + if (rc) + goto out_err_free; + + gid = make_kgid(&init_user_ns, id); + + if (gfp_flags & __GFP_FS) + kcred = prepare_kernel_cred(&init_task); + else { + unsigned int nofs_flags = memalloc_nofs_save(); + + kcred = prepare_kernel_cred(&init_task); + memalloc_nofs_restore(nofs_flags); + } + rc = -ENOMEM; + if (!kcred) + goto out_err_free; + kcred->fsuid = uid; + kcred->fsgid = gid; + cred = RCU_INITIALIZER(kcred); + + if (lgr->range.iomode == IOMODE_READ) + rcu_assign_pointer(dss_info->ro_cred, cred); + else + rcu_assign_pointer(dss_info->rw_cred, cred); + } + + mirror = ff_layout_add_mirror(lh, fls->mirror_array[i]); + if (mirror != fls->mirror_array[i]) { + for (dss_id = 0; dss_id < dss_count; dss_id++) { + dss_info = &fls->mirror_array[i]->dss[dss_id]; + /* swap cred ptrs so free_mirror will clean up old */ + if (lgr->range.iomode == IOMODE_READ) { + cred = xchg(&mirror->dss[dss_id].ro_cred, + dss_info->ro_cred); + rcu_assign_pointer(dss_info->ro_cred, cred); + } else { + cred = xchg(&mirror->dss[dss_id].rw_cred, + dss_info->rw_cred); + rcu_assign_pointer(dss_info->rw_cred, cred); + } + } + ff_layout_free_mirror(fls->mirror_array[i]); + fls->mirror_array[i] = mirror; + } + + dprintk("%s: iomode %s uid %u gid %u\n", __func__, + lgr->range.iomode == IOMODE_READ ? "READ" : "RW", + from_kuid(&init_user_ns, uid), + from_kgid(&init_user_ns, gid)); + } + + p = xdr_inline_decode(&stream, 4); + if (!p) + goto out_sort_mirrors; + fls->flags = be32_to_cpup(p); + if (fls->flags & FF_FLAGS_NO_IO_THRU_MDS) + set_bit(NFS4_FF_HDR_NO_IO_THRU_MDS, + &FF_LAYOUT_FROM_HDR(lh)->flags); + + p = xdr_inline_decode(&stream, 4); + if (!p) + goto out_sort_mirrors; + for (i=0; i < fls->mirror_array_cnt; i++) + fls->mirror_array[i]->report_interval = be32_to_cpup(p); + +out_sort_mirrors: + ff_layout_sort_mirrors(fls); + ret = &fls->generic_hdr; + dprintk("<-- %s (success)\n", __func__); +out_free_page: + folio_put(scratch); + return ret; +out_err_free: + _ff_layout_free_lseg(fls); + ret = ERR_PTR(rc); + dprintk("<-- %s (%d)\n", __func__, rc); + goto out_free_page; +} + +static void +ff_layout_free_lseg(struct pnfs_layout_segment *lseg) +{ + struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg); + + dprintk("--> %s\n", __func__); + + if (lseg->pls_range.iomode == IOMODE_RW) { + struct nfs4_flexfile_layout *ffl; + struct inode *inode; + + ffl = FF_LAYOUT_FROM_HDR(lseg->pls_layout); + inode = ffl->generic_hdr.plh_inode; + spin_lock(&inode->i_lock); + pnfs_generic_ds_cinfo_release_lseg(&ffl->commit_info, lseg); + spin_unlock(&inode->i_lock); + } + _ff_layout_free_lseg(fls); +} + +static u32 calc_commit_idx(struct pnfs_layout_segment *lseg, + u32 mirror_idx, u32 dss_id) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg); + + return (mirror_idx * flseg->mirror_array[0]->dss_count) + dss_id; +} + +static u32 calc_mirror_idx_from_commit(struct pnfs_layout_segment *lseg, + u32 commit_index) +{ + return commit_index / FF_LAYOUT_LSEG(lseg)->mirror_array[0]->dss_count; +} + +static u32 calc_dss_id_from_commit(struct pnfs_layout_segment *lseg, + u32 commit_index) +{ + return commit_index % FF_LAYOUT_LSEG(lseg)->mirror_array[0]->dss_count; +} + +static void +nfs4_ff_start_busy_timer(struct nfs4_ff_busy_timer *timer, ktime_t now) +{ + /* first IO request? */ + if (atomic_inc_return(&timer->n_ops) == 1) { + timer->start_time = now; + } +} + +static ktime_t +nfs4_ff_end_busy_timer(struct nfs4_ff_busy_timer *timer, ktime_t now) +{ + ktime_t start; + + if (atomic_dec_return(&timer->n_ops) < 0) + WARN_ON_ONCE(1); + + start = timer->start_time; + timer->start_time = now; + return ktime_sub(now, start); +} + +static bool +nfs4_ff_layoutstat_start_io(struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + struct nfs4_ff_layoutstat *layoutstat, + ktime_t now) +{ + s64 report_interval = FF_LAYOUTSTATS_REPORT_INTERVAL; + struct nfs4_flexfile_layout *ffl = FF_LAYOUT_FROM_HDR(mirror->layout); + + nfs4_ff_start_busy_timer(&layoutstat->busy_timer, now); + if (!mirror->dss[dss_id].start_time) + mirror->dss[dss_id].start_time = now; + if (mirror->report_interval != 0) + report_interval = (s64)mirror->report_interval * 1000LL; + else if (layoutstats_timer != 0) + report_interval = (s64)layoutstats_timer * 1000LL; + if (ktime_to_ms(ktime_sub(now, ffl->last_report_time)) >= + report_interval) { + ffl->last_report_time = now; + return true; + } + + return false; +} + +static void +nfs4_ff_layout_stat_io_update_requested(struct nfs4_ff_layoutstat *layoutstat, + __u64 requested) +{ + struct nfs4_ff_io_stat *iostat = &layoutstat->io_stat; + + iostat->ops_requested++; + iostat->bytes_requested += requested; +} + +static void +nfs4_ff_layout_stat_io_update_completed(struct nfs4_ff_layoutstat *layoutstat, + __u64 requested, + __u64 completed, + ktime_t time_completed, + ktime_t time_started) +{ + struct nfs4_ff_io_stat *iostat = &layoutstat->io_stat; + ktime_t completion_time = ktime_sub(time_completed, time_started); + ktime_t timer; + + iostat->ops_completed++; + iostat->bytes_completed += completed; + iostat->bytes_not_delivered += requested - completed; + + timer = nfs4_ff_end_busy_timer(&layoutstat->busy_timer, time_completed); + iostat->total_busy_time = + ktime_add(iostat->total_busy_time, timer); + iostat->aggregate_completion_time = + ktime_add(iostat->aggregate_completion_time, + completion_time); +} + +static void +nfs4_ff_layout_stat_io_start_read(struct inode *inode, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + __u64 requested, ktime_t now) +{ + bool report; + + spin_lock(&mirror->lock); + report = nfs4_ff_layoutstat_start_io( + mirror, dss_id, &mirror->dss[dss_id].read_stat, now); + nfs4_ff_layout_stat_io_update_requested( + &mirror->dss[dss_id].read_stat, requested); + set_bit(NFS4_FF_MIRROR_STAT_AVAIL, &mirror->flags); + spin_unlock(&mirror->lock); + + if (report) + pnfs_report_layoutstat(inode, nfs_io_gfp_mask()); +} + +static void +nfs4_ff_layout_stat_io_end_read(struct rpc_task *task, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + __u64 requested, + __u64 completed) +{ + spin_lock(&mirror->lock); + nfs4_ff_layout_stat_io_update_completed(&mirror->dss[dss_id].read_stat, + requested, completed, + ktime_get(), task->tk_start); + set_bit(NFS4_FF_MIRROR_STAT_AVAIL, &mirror->flags); + spin_unlock(&mirror->lock); +} + +static void +nfs4_ff_layout_stat_io_start_write(struct inode *inode, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + __u64 requested, ktime_t now) +{ + bool report; + + spin_lock(&mirror->lock); + report = nfs4_ff_layoutstat_start_io( + mirror, + dss_id, + &mirror->dss[dss_id].write_stat, + now); + nfs4_ff_layout_stat_io_update_requested( + &mirror->dss[dss_id].write_stat, + requested); + set_bit(NFS4_FF_MIRROR_STAT_AVAIL, &mirror->flags); + spin_unlock(&mirror->lock); + + if (report) + pnfs_report_layoutstat(inode, nfs_io_gfp_mask()); +} + +static void +nfs4_ff_layout_stat_io_end_write(struct rpc_task *task, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + __u64 requested, + __u64 completed, + enum nfs3_stable_how committed) +{ + if (committed == NFS_UNSTABLE) + requested = completed = 0; + + spin_lock(&mirror->lock); + nfs4_ff_layout_stat_io_update_completed(&mirror->dss[dss_id].write_stat, + requested, completed, ktime_get(), task->tk_start); + set_bit(NFS4_FF_MIRROR_STAT_AVAIL, &mirror->flags); + spin_unlock(&mirror->lock); +} + +static void +ff_layout_mark_ds_unreachable(struct pnfs_layout_segment *lseg, u32 idx, u32 dss_id) +{ + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx, dss_id); + + if (devid) + nfs4_mark_deviceid_unavailable(devid); +} + +static void +ff_layout_mark_ds_reachable(struct pnfs_layout_segment *lseg, u32 idx, u32 dss_id) +{ + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx, dss_id); + + if (devid) + nfs4_mark_deviceid_available(devid); +} + +static struct nfs4_pnfs_ds * +ff_layout_choose_ds_for_read(struct pnfs_layout_segment *lseg, + u32 start_idx, u32 *best_idx, + u32 offset, u32 *dss_id, + bool check_device) +{ + struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg); + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_pnfs_ds *ds = ERR_PTR(-EAGAIN); + u32 idx; + + /* mirrors are initially sorted by efficiency */ + for (idx = start_idx; idx < fls->mirror_array_cnt; idx++) { + mirror = FF_LAYOUT_COMP(lseg, idx); + *dss_id = nfs4_ff_layout_calc_dss_id( + fls->stripe_unit, + fls->mirror_array[idx]->dss_count, + offset); + ds = nfs4_ff_layout_prepare_ds(lseg, mirror, *dss_id, false); + if (IS_ERR(ds)) + continue; + + if (check_device && + nfs4_test_deviceid_unavailable(&mirror->dss[*dss_id].mirror_ds->id_node)) { + // reinitialize the error state in case if this is the last iteration + ds = ERR_PTR(-EINVAL); + continue; + } + + *best_idx = idx; + break; + } + + return ds; +} + +static struct nfs4_pnfs_ds * +ff_layout_choose_any_ds_for_read(struct pnfs_layout_segment *lseg, + u32 start_idx, u32 *best_idx, + u32 offset, u32 *dss_id) +{ + return ff_layout_choose_ds_for_read(lseg, start_idx, best_idx, + offset, dss_id, false); +} + +static struct nfs4_pnfs_ds * +ff_layout_choose_valid_ds_for_read(struct pnfs_layout_segment *lseg, + u32 start_idx, u32 *best_idx, + u32 offset, u32 *dss_id) +{ + return ff_layout_choose_ds_for_read(lseg, start_idx, best_idx, + offset, dss_id, true); +} + +static struct nfs4_pnfs_ds * +ff_layout_choose_best_ds_for_read(struct pnfs_layout_segment *lseg, + u32 start_idx, u32 *best_idx, + u32 offset, u32 *dss_id) +{ + struct nfs4_pnfs_ds *ds; + + ds = ff_layout_choose_valid_ds_for_read(lseg, start_idx, best_idx, + offset, dss_id); + if (!IS_ERR(ds)) + return ds; + return ff_layout_choose_any_ds_for_read(lseg, start_idx, best_idx, + offset, dss_id); +} + +static struct nfs4_pnfs_ds * +ff_layout_get_ds_for_read(struct nfs_pageio_descriptor *pgio, + u32 *best_idx, + u32 offset, + u32 *dss_id) +{ + struct pnfs_layout_segment *lseg = pgio->pg_lseg; + struct nfs4_pnfs_ds *ds; + + ds = ff_layout_choose_best_ds_for_read(lseg, pgio->pg_mirror_idx, + best_idx, offset, dss_id); + if (!IS_ERR(ds) || !pgio->pg_mirror_idx) + return ds; + return ff_layout_choose_best_ds_for_read(lseg, 0, best_idx, + offset, dss_id); +} + +static void +ff_layout_pg_get_read(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, + bool strict_iomode) +{ + pnfs_put_lseg(pgio->pg_lseg); + pgio->pg_lseg = + pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), + req_offset(req), req->wb_bytes, IOMODE_READ, + strict_iomode, nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + } +} + +static bool +ff_layout_lseg_is_striped(const struct nfs4_ff_layout_segment *fls) +{ + return fls->mirror_array[0]->dss_count > 1; +} + +/* + * ff_layout_pg_test(). Called by nfs_can_coalesce_requests() + * + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t +ff_layout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + unsigned int size; + u64 p_stripe, r_stripe; + u32 stripe_offset; + u64 segment_offset = pgio->pg_lseg->pls_range.offset; + u32 stripe_unit = FF_LAYOUT_LSEG(pgio->pg_lseg)->stripe_unit; + + /* calls nfs_generic_pg_test */ + size = pnfs_generic_pg_test(pgio, prev, req); + if (!size) + return 0; + else if (!ff_layout_lseg_is_striped(FF_LAYOUT_LSEG(pgio->pg_lseg))) + return size; + + /* see if req and prev are in the same stripe */ + if (prev) { + p_stripe = (u64)req_offset(prev) - segment_offset; + r_stripe = (u64)req_offset(req) - segment_offset; + do_div(p_stripe, stripe_unit); + do_div(r_stripe, stripe_unit); + + if (p_stripe != r_stripe) + return 0; + } + + /* calculate remaining bytes in the current stripe */ + div_u64_rem((u64)req_offset(req) - segment_offset, + stripe_unit, + &stripe_offset); + WARN_ON_ONCE(stripe_offset > stripe_unit); + if (stripe_offset >= stripe_unit) + return 0; + return min(stripe_unit - (unsigned int)stripe_offset, size); +} + +static void +ff_layout_pg_init_read(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + struct nfs_pgio_mirror *pgm; + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_pnfs_ds *ds; + u32 ds_idx, dss_id; + + if (NFS_SERVER(pgio->pg_inode)->flags & + (NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR)) + pgio->pg_maxretrans = io_maxretrans; +retry: + pnfs_generic_pg_check_layout(pgio, req); + /* Use full layout for now */ + if (!pgio->pg_lseg) { + ff_layout_pg_get_read(pgio, req, false); + if (!pgio->pg_lseg) + goto out_nolseg; + } + if (ff_layout_avoid_read_on_rw(pgio->pg_lseg)) { + ff_layout_pg_get_read(pgio, req, true); + if (!pgio->pg_lseg) + goto out_nolseg; + } + /* Reset wb_nio, since getting layout segment was successful */ + req->wb_nio = 0; + + ds = ff_layout_get_ds_for_read(pgio, &ds_idx, + req_offset(req), &dss_id); + if (IS_ERR(ds)) { + if (!ff_layout_no_fallback_to_mds(pgio->pg_lseg)) + goto out_mds; + pnfs_generic_pg_cleanup(pgio); + /* Sleep for 1 second before retrying */ + ssleep(1); + goto retry; + } + + mirror = FF_LAYOUT_COMP(pgio->pg_lseg, ds_idx); + pgm = &pgio->pg_mirrors[0]; + pgm->pg_bsize = mirror->dss[dss_id].mirror_ds->ds_versions[0].rsize; + + pgio->pg_mirror_idx = ds_idx; + return; +out_nolseg: + if (pgio->pg_error < 0) { + if (pgio->pg_error != -EAGAIN) + return; + /* Retry getting layout segment if lower layer returned -EAGAIN */ + if (pgio->pg_maxretrans && req->wb_nio++ > pgio->pg_maxretrans) { + if (NFS_SERVER(pgio->pg_inode)->flags & NFS_MOUNT_SOFTERR) + pgio->pg_error = -ETIMEDOUT; + else + pgio->pg_error = -EIO; + return; + } + pgio->pg_error = 0; + /* Sleep for 1 second before retrying */ + ssleep(1); + goto retry; + } +out_mds: + trace_pnfs_mds_fallback_pg_init_read(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_READ, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); + pgio->pg_maxretrans = 0; + nfs_pageio_reset_read_mds(pgio); +} + +static void +ff_layout_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + struct nfs4_ff_layout_mirror *mirror; + struct nfs_pgio_mirror *pgm; + struct nfs4_pnfs_ds *ds; + u32 i, dss_id; + +retry: + pnfs_generic_pg_check_layout(pgio, req); + if (!pgio->pg_lseg) { + pgio->pg_lseg = + pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), + req_offset(req), req->wb_bytes, + IOMODE_RW, false, nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + return; + } + } + /* If no lseg, fall back to write through mds */ + if (pgio->pg_lseg == NULL) + goto out_mds; + + /* Use a direct mapping of ds_idx to pgio mirror_idx */ + if (pgio->pg_mirror_count != FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg)) + goto out_eagain; + + for (i = 0; i < pgio->pg_mirror_count; i++) { + mirror = FF_LAYOUT_COMP(pgio->pg_lseg, i); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(pgio->pg_lseg)->stripe_unit, + mirror->dss_count, + req_offset(req)); + ds = nfs4_ff_layout_prepare_ds(pgio->pg_lseg, mirror, + dss_id, true); + if (IS_ERR(ds)) { + if (!ff_layout_no_fallback_to_mds(pgio->pg_lseg)) + goto out_mds; + pnfs_generic_pg_cleanup(pgio); + /* Sleep for 1 second before retrying */ + ssleep(1); + goto retry; + } + pgm = &pgio->pg_mirrors[i]; + pgm->pg_bsize = mirror->dss[dss_id].mirror_ds->ds_versions[0].wsize; + } + + if (NFS_SERVER(pgio->pg_inode)->flags & + (NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR)) + pgio->pg_maxretrans = io_maxretrans; + return; +out_eagain: + pnfs_generic_pg_cleanup(pgio); + pgio->pg_error = -EAGAIN; + return; +out_mds: + trace_pnfs_mds_fallback_pg_init_write(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_RW, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); + pgio->pg_maxretrans = 0; + nfs_pageio_reset_write_mds(pgio); + pgio->pg_error = -EAGAIN; +} + +static unsigned int +ff_layout_pg_get_mirror_count_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + if (!pgio->pg_lseg) { + pgio->pg_lseg = + pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), + req_offset(req), req->wb_bytes, + IOMODE_RW, false, nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + goto out; + } + } + if (pgio->pg_lseg) + return FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg); + + trace_pnfs_mds_fallback_pg_get_mirror_count(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_RW, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); + if (NFS_I(pgio->pg_inode)->layout && + ff_layout_hdr_no_fallback_to_mds(NFS_I(pgio->pg_inode)->layout)) { + /* + * FF_FLAGS_NO_IO_THRU_MDS: no current lseg but the server's + * policy forbids MDS fallback. Surface -EAGAIN so writeback + * retries rather than silently issuing the WRITE via MDS. + */ + pgio->pg_error = -EAGAIN; + goto out; + } + /* no lseg means that pnfs is not in use, so no mirroring here */ + nfs_pageio_reset_write_mds(pgio); +out: + return 1; +} + +static u32 +ff_layout_pg_set_mirror_write(struct nfs_pageio_descriptor *desc, u32 idx) +{ + u32 old = desc->pg_mirror_idx; + + desc->pg_mirror_idx = idx; + return old; +} + +static struct nfs_pgio_mirror * +ff_layout_pg_get_mirror_write(struct nfs_pageio_descriptor *desc, u32 idx) +{ + return &desc->pg_mirrors[idx]; +} + +static const struct nfs_pageio_ops ff_layout_pg_read_ops = { + .pg_init = ff_layout_pg_init_read, + .pg_test = ff_layout_pg_test, + .pg_doio = pnfs_generic_pg_readpages, + .pg_cleanup = pnfs_generic_pg_cleanup, +}; + +static const struct nfs_pageio_ops ff_layout_pg_write_ops = { + .pg_init = ff_layout_pg_init_write, + .pg_test = ff_layout_pg_test, + .pg_doio = pnfs_generic_pg_writepages, + .pg_get_mirror_count = ff_layout_pg_get_mirror_count_write, + .pg_cleanup = pnfs_generic_pg_cleanup, + .pg_get_mirror = ff_layout_pg_get_mirror_write, + .pg_set_mirror = ff_layout_pg_set_mirror_write, +}; + +static void ff_layout_reset_write(struct nfs_pgio_header *hdr, bool retry_pnfs) +{ + struct rpc_task *task = &hdr->task; + + pnfs_layoutcommit_inode(hdr->inode, false); + + if (retry_pnfs) { + dprintk("%s Reset task %5u for i/o through pNFS " + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, + hdr->task.tk_pid, + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + hdr->completion_ops->reschedule_io(hdr); + return; + } + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + dprintk("%s Reset task %5u for i/o through MDS " + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, + hdr->task.tk_pid, + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + trace_pnfs_mds_fallback_write_done(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_RW, NFS_I(hdr->inode)->layout, + hdr->lseg); + task->tk_status = pnfs_write_done_resend_to_mds(hdr); + } +} + +static void ff_layout_resend_pnfs_read(struct nfs_pgio_header *hdr) +{ + u32 idx = hdr->pgio_mirror_idx + 1; + u32 new_idx = 0; + u32 dss_id = 0; + struct nfs4_pnfs_ds *ds; + + ds = ff_layout_choose_any_ds_for_read(hdr->lseg, idx, &new_idx, + hdr->args.offset, &dss_id); + if (IS_ERR(ds)) + pnfs_error_mark_layout_for_return(hdr->inode, hdr->lseg); + else + ff_layout_send_layouterror(hdr->lseg); + pnfs_read_resend_pnfs(hdr, new_idx); +} + +static void ff_layout_reset_read(struct nfs_pgio_header *hdr) +{ + struct rpc_task *task = &hdr->task; + + pnfs_layoutcommit_inode(hdr->inode, false); + pnfs_error_mark_layout_for_return(hdr->inode, hdr->lseg); + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + dprintk("%s Reset task %5u for i/o through MDS " + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, + hdr->task.tk_pid, + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + trace_pnfs_mds_fallback_read_done(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_READ, NFS_I(hdr->inode)->layout, + hdr->lseg); + task->tk_status = pnfs_read_done_resend_to_mds(hdr); + } +} + +static int ff_layout_async_handle_error_v4(struct rpc_task *task, + u32 op_status, + struct nfs4_state *state, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, + u32 idx, u32 dss_id) +{ + struct pnfs_layout_hdr *lo = lseg->pls_layout; + struct inode *inode = lo->plh_inode; + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx, dss_id); + struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table; + + switch (op_status) { + case NFS4_OK: + case NFS4ERR_NXIO: + break; + case NFSERR_PERM: + if (!task->tk_xprt) + break; + xprt_force_disconnect(task->tk_xprt); + goto out_retry; + case NFS4ERR_BADSESSION: + case NFS4ERR_BADSLOT: + case NFS4ERR_BAD_HIGH_SLOT: + case NFS4ERR_DEADSESSION: + case NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case NFS4ERR_SEQ_FALSE_RETRY: + case NFS4ERR_SEQ_MISORDERED: + dprintk("%s ERROR %d, Reset session. Exchangeid " + "flags 0x%x\n", __func__, task->tk_status, + clp->cl_exchange_flags); + nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); + goto out_retry; + case NFS4ERR_DELAY: + nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); + fallthrough; + case NFS4ERR_GRACE: + rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX); + goto out_retry; + case NFS4ERR_RETRY_UNCACHED_REP: + goto out_retry; + /* Invalidate Layout errors */ + case NFS4ERR_PNFS_NO_LAYOUT: + case NFS4ERR_STALE: + case NFS4ERR_BADHANDLE: + case NFS4ERR_ISDIR: + case NFS4ERR_FHEXPIRED: + case NFS4ERR_WRONG_TYPE: + dprintk("%s Invalid layout error %d\n", __func__, + task->tk_status); + /* + * Destroy layout so new i/o will get a new layout. + * Layout will not be destroyed until all current lseg + * references are put. Mark layout as invalid to resend failed + * i/o and all i/o waiting on the slot table to the MDS until + * layout is destroyed and a new valid layout is obtained. + */ + pnfs_destroy_layout(NFS_I(inode)); + rpc_wake_up(&tbl->slot_tbl_waitq); + goto reset; + default: + break; + } + + switch (task->tk_status) { + /* RPC connection errors */ + case -ENETDOWN: + case -ENETUNREACH: + if (test_bit(NFS_CS_NETUNREACH_FATAL, &clp->cl_flags)) + return -NFS4ERR_FATAL_IOERROR; + fallthrough; + case -ECONNREFUSED: + case -EHOSTDOWN: + case -EHOSTUNREACH: + case -EIO: + case -ETIMEDOUT: + case -EPIPE: + case -EPROTO: + case -ENODEV: + dprintk("%s DS connection error %d\n", __func__, + task->tk_status); + nfs4_delete_deviceid(devid->ld, devid->nfs_client, + &devid->deviceid); + rpc_wake_up(&tbl->slot_tbl_waitq); + break; + default: + break; + } + + if (ff_layout_avoid_mds_available_ds(lseg)) + return -NFS4ERR_RESET_TO_PNFS; +reset: + dprintk("%s Retry through MDS. Error %d\n", __func__, + task->tk_status); + return -NFS4ERR_RESET_TO_MDS; + +out_retry: + task->tk_status = 0; + return -EAGAIN; +} + +/* Retry all errors through either pNFS or MDS except for -EJUKEBOX */ +static int ff_layout_async_handle_error_v3(struct rpc_task *task, + u32 op_status, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, + u32 idx, u32 dss_id) +{ + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx, dss_id); + + switch (op_status) { + case NFS_OK: + case NFSERR_NXIO: + break; + case NFSERR_PERM: + if (!task->tk_xprt) + break; + xprt_force_disconnect(task->tk_xprt); + goto out_retry; + case NFSERR_ACCES: + case NFSERR_BADHANDLE: + case NFSERR_FBIG: + case NFSERR_IO: + case NFSERR_NOSPC: + case NFSERR_ROFS: + case NFSERR_STALE: + goto out_reset_to_pnfs; + case NFSERR_JUKEBOX: + nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); + goto out_retry; + default: + break; + } + + switch (task->tk_status) { + /* File access problems. Don't mark the device as unavailable */ + case -EACCES: + case -ESTALE: + case -EISDIR: + case -EBADHANDLE: + case -ELOOP: + case -ENOSPC: + break; + case -EJUKEBOX: + nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); + goto out_retry; + case -ENETDOWN: + case -ENETUNREACH: + if (test_bit(NFS_CS_NETUNREACH_FATAL, &clp->cl_flags)) + return -NFS4ERR_FATAL_IOERROR; + fallthrough; + default: + dprintk("%s DS connection error %d\n", __func__, + task->tk_status); + nfs4_delete_deviceid(devid->ld, devid->nfs_client, + &devid->deviceid); + } +out_reset_to_pnfs: + /* FIXME: Need to prevent infinite looping here. */ + return -NFS4ERR_RESET_TO_PNFS; +out_retry: + task->tk_status = 0; + rpc_restart_call_prepare(task); + rpc_delay(task, NFS_JUKEBOX_RETRY_TIME); + return -EAGAIN; +} + +static int ff_layout_async_handle_error(struct rpc_task *task, + u32 op_status, + struct nfs4_state *state, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, + u32 idx, u32 dss_id) +{ + int vers = clp->cl_nfs_mod->rpc_vers->number; + + if (task->tk_status >= 0) { + ff_layout_mark_ds_reachable(lseg, idx, dss_id); + return 0; + } + + /* Handle the case of an invalid layout segment */ + if (!pnfs_is_valid_lseg(lseg)) + return -NFS4ERR_RESET_TO_PNFS; + + switch (vers) { + case 3: + return ff_layout_async_handle_error_v3(task, op_status, clp, + lseg, idx, dss_id); + case 4: + return ff_layout_async_handle_error_v4(task, op_status, state, + clp, lseg, idx, dss_id); + default: + /* should never happen */ + WARN_ON_ONCE(1); + return 0; + } +} + +static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg, + u32 idx, u32 dss_id, u64 offset, u64 length, + u32 *op_status, int opnum, int error) +{ + struct nfs4_ff_layout_mirror *mirror; + u32 status = *op_status; + int err; + + if (status == 0) { + switch (error) { + case -ETIMEDOUT: + case -EPFNOSUPPORT: + case -EPROTONOSUPPORT: + case -EOPNOTSUPP: + case -EINVAL: + case -ECONNREFUSED: + case -ECONNRESET: + case -EHOSTDOWN: + case -EHOSTUNREACH: + case -ENETDOWN: + case -ENETUNREACH: + case -EADDRINUSE: + case -ENOBUFS: + case -EPIPE: + case -EPERM: + case -EPROTO: + case -ENODEV: + *op_status = status = NFS4ERR_NXIO; + break; + case -EACCES: + *op_status = status = NFS4ERR_ACCESS; + break; + default: + return; + } + } + + mirror = FF_LAYOUT_COMP(lseg, idx); + err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout), + mirror, dss_id, offset, length, status, opnum, + nfs_io_gfp_mask()); + + switch (status) { + case NFS4ERR_DELAY: + case NFS4ERR_GRACE: + case NFS4ERR_PERM: + break; + case NFS4ERR_NXIO: + ff_layout_mark_ds_unreachable(lseg, idx, dss_id); + /* + * Don't return the layout if this is a read and we still + * have layouts to try + */ + if (opnum == OP_READ) + break; + fallthrough; + default: + pnfs_error_mark_layout_for_return(lseg->pls_layout->plh_inode, + lseg); + } + + dprintk("%s: err %d op %d status %u\n", __func__, err, opnum, status); +} + +/* NFS_PROTO call done callback routines */ +static int ff_layout_read_done_cb(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(hdr->lseg); + u32 dss_id = nfs4_ff_layout_calc_dss_id( + flseg->stripe_unit, + flseg->mirror_array[hdr->pgio_mirror_idx]->dss_count, + hdr->args.offset); + int err; + + if (task->tk_status < 0) { + ff_layout_io_track_ds_error(hdr->lseg, + hdr->pgio_mirror_idx, dss_id, + hdr->args.offset, hdr->args.count, + &hdr->res.op_status, OP_READ, + task->tk_status); + trace_ff_layout_read_error(hdr, task->tk_status); + } + + err = ff_layout_async_handle_error(task, hdr->res.op_status, + hdr->args.context->state, + hdr->ds_clp, hdr->lseg, + hdr->pgio_mirror_idx, + dss_id); + + trace_nfs4_pnfs_read(hdr, err); + clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); + clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); + switch (err) { + case -NFS4ERR_RESET_TO_PNFS: + set_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); + return task->tk_status; + case -NFS4ERR_RESET_TO_MDS: + set_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); + return task->tk_status; + case -EAGAIN: + goto out_eagain; + case -NFS4ERR_FATAL_IOERROR: + task->tk_status = -EIO; + return 0; + } + + return 0; +out_eagain: + rpc_restart_call_prepare(task); + return -EAGAIN; +} + +static bool +ff_layout_need_layoutcommit(struct pnfs_layout_segment *lseg) +{ + return !(FF_LAYOUT_LSEG(lseg)->flags & FF_FLAGS_NO_LAYOUTCOMMIT); +} + +/* + * We reference the rpc_cred of the first WRITE that triggers the need for + * a LAYOUTCOMMIT, and use it to send the layoutcommit compound. + * rfc5661 is not clear about which credential should be used. + * + * Flexlayout client should treat DS replied FILE_SYNC as DATA_SYNC, so + * to follow http://www.rfc-editor.org/errata_search.php?rfc=5661&eid=2751 + * we always send layoutcommit after DS writes. + */ +static void +ff_layout_set_layoutcommit(struct inode *inode, + struct pnfs_layout_segment *lseg, + loff_t end_offset) +{ + if (!ff_layout_need_layoutcommit(lseg)) + return; + + pnfs_set_layoutcommit(inode, lseg, end_offset); + dprintk("%s inode %llu pls_end_pos %llu\n", __func__, inode->i_ino, + (unsigned long long) NFS_I(inode)->layout->plh_lwb); +} + +static void ff_layout_read_record_layoutstats_start(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_mirror *mirror; + u32 dss_id; + + if (test_and_set_bit(NFS_IOHDR_STAT, &hdr->flags)) + return; + + mirror = FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(hdr->lseg)->stripe_unit, + mirror->dss_count, + hdr->args.offset); + + nfs4_ff_layout_stat_io_start_read( + hdr->inode, + mirror, + dss_id, + hdr->args.count, + task->tk_start); +} + +static void ff_layout_read_record_layoutstats_done(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_mirror *mirror; + u32 dss_id; + + if (!test_and_clear_bit(NFS_IOHDR_STAT, &hdr->flags)) + return; + + mirror = FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(hdr->lseg)->stripe_unit, + mirror->dss_count, + hdr->args.offset); + + nfs4_ff_layout_stat_io_end_read( + task, + mirror, + dss_id, + hdr->args.count, + hdr->res.count); + set_bit(NFS_LSEG_LAYOUTRETURN, &hdr->lseg->pls_flags); +} + +static int ff_layout_read_prepare_common(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) { + rpc_exit(task, -EIO); + return -EIO; + } + + if (!pnfs_is_valid_lseg(hdr->lseg)) { + rpc_exit(task, -EAGAIN); + return -EAGAIN; + } + + ff_layout_read_record_layoutstats_start(task, hdr); + return 0; +} + +/* + * Call ops for the async read/write cases + * In the case of dense layouts, the offset needs to be reset to its + * original value. + */ +static void ff_layout_read_prepare_v3(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (ff_layout_read_prepare_common(task, hdr)) + return; + + rpc_call_start(task); +} + +static void ff_layout_read_prepare_v4(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (nfs4_setup_sequence(hdr->ds_clp, + &hdr->args.seq_args, + &hdr->res.seq_res, + task)) + return; + + ff_layout_read_prepare_common(task, hdr); +} + +static void ff_layout_read_call_done(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && + task->tk_status == 0) { + nfs4_sequence_done(task, &hdr->res.seq_res); + return; + } + + /* Note this may cause RPC to be resent */ + hdr->mds_ops->rpc_call_done(task, hdr); +} + +static void ff_layout_read_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + ff_layout_read_record_layoutstats_done(task, hdr); + rpc_count_iostats_metrics(task, + &NFS_CLIENT(hdr->inode)->cl_metrics[NFSPROC4_CLNT_READ]); +} + +static void ff_layout_read_release(void *data) +{ + struct nfs_pgio_header *hdr = data; + + ff_layout_read_record_layoutstats_done(&hdr->task, hdr); + if (test_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags)) + ff_layout_resend_pnfs_read(hdr); + else if (test_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags)) + ff_layout_reset_read(hdr); + pnfs_generic_rw_release(data); +} + + +static int ff_layout_write_done_cb(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(hdr->lseg); + u32 dss_id = nfs4_ff_layout_calc_dss_id( + flseg->stripe_unit, + flseg->mirror_array[hdr->pgio_mirror_idx]->dss_count, + hdr->args.offset); + loff_t end_offs = 0; + int err; + + if (task->tk_status < 0) { + ff_layout_io_track_ds_error(hdr->lseg, + hdr->pgio_mirror_idx, dss_id, + hdr->args.offset, hdr->args.count, + &hdr->res.op_status, OP_WRITE, + task->tk_status); + trace_ff_layout_write_error(hdr, task->tk_status); + } + + err = ff_layout_async_handle_error(task, hdr->res.op_status, + hdr->args.context->state, + hdr->ds_clp, hdr->lseg, + hdr->pgio_mirror_idx, + dss_id); + + trace_nfs4_pnfs_write(hdr, err); + clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); + clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); + switch (err) { + case -NFS4ERR_RESET_TO_PNFS: + set_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); + return task->tk_status; + case -NFS4ERR_RESET_TO_MDS: + set_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); + return task->tk_status; + case -EAGAIN: + return -EAGAIN; + case -NFS4ERR_FATAL_IOERROR: + task->tk_status = -EIO; + return 0; + } + + if (hdr->res.verf->committed == NFS_FILE_SYNC || + hdr->res.verf->committed == NFS_DATA_SYNC) + end_offs = hdr->mds_offset + (loff_t)hdr->res.count; + + /* Note: if the write is unstable, don't set end_offs until commit */ + ff_layout_set_layoutcommit(hdr->inode, hdr->lseg, end_offs); + + /* zero out fattr since we don't care DS attr at all */ + hdr->fattr.valid = 0; + if (task->tk_status >= 0) + nfs_writeback_update_inode(hdr); + + return 0; +} + +static int ff_layout_commit_done_cb(struct rpc_task *task, + struct nfs_commit_data *data) +{ + int err; + u32 idx = calc_mirror_idx_from_commit(data->lseg, data->ds_commit_index); + u32 dss_id = calc_dss_id_from_commit(data->lseg, data->ds_commit_index); + + if (task->tk_status < 0) { + ff_layout_io_track_ds_error(data->lseg, idx, dss_id, + data->args.offset, data->args.count, + &data->res.op_status, OP_COMMIT, + task->tk_status); + trace_ff_layout_commit_error(data, task->tk_status); + } + + err = ff_layout_async_handle_error(task, data->res.op_status, + NULL, data->ds_clp, data->lseg, idx, + dss_id); + + trace_nfs4_pnfs_commit_ds(data, err); + switch (err) { + case -NFS4ERR_RESET_TO_PNFS: + pnfs_generic_prepare_to_resend_writes(data); + return -EAGAIN; + case -NFS4ERR_RESET_TO_MDS: + pnfs_generic_prepare_to_resend_writes(data); + return -EAGAIN; + case -EAGAIN: + rpc_restart_call_prepare(task); + return -EAGAIN; + case -NFS4ERR_FATAL_IOERROR: + task->tk_status = -EIO; + return 0; + } + + ff_layout_set_layoutcommit(data->inode, data->lseg, data->lwb); + return 0; +} + +static void ff_layout_write_record_layoutstats_start(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_mirror *mirror; + u32 dss_id; + + if (test_and_set_bit(NFS_IOHDR_STAT, &hdr->flags)) + return; + + mirror = FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(hdr->lseg)->stripe_unit, + mirror->dss_count, + hdr->args.offset); + + nfs4_ff_layout_stat_io_start_write( + hdr->inode, + mirror, + dss_id, + hdr->args.count, + task->tk_start); +} + +static void ff_layout_write_record_layoutstats_done(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs4_ff_layout_mirror *mirror; + u32 dss_id; + + if (!test_and_clear_bit(NFS_IOHDR_STAT, &hdr->flags)) + return; + + mirror = FF_LAYOUT_COMP(hdr->lseg, hdr->pgio_mirror_idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(hdr->lseg)->stripe_unit, + mirror->dss_count, + hdr->args.offset); + + nfs4_ff_layout_stat_io_end_write( + task, + mirror, + dss_id, + hdr->args.count, + hdr->res.count, + hdr->res.verf->committed); + set_bit(NFS_LSEG_LAYOUTRETURN, &hdr->lseg->pls_flags); +} + +static int ff_layout_write_prepare_common(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) { + rpc_exit(task, -EIO); + return -EIO; + } + + if (!pnfs_is_valid_lseg(hdr->lseg)) { + rpc_exit(task, -EAGAIN); + return -EAGAIN; + } + + ff_layout_write_record_layoutstats_start(task, hdr); + return 0; +} + +static void ff_layout_write_prepare_v3(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (ff_layout_write_prepare_common(task, hdr)) + return; + + rpc_call_start(task); +} + +static void ff_layout_write_prepare_v4(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (nfs4_setup_sequence(hdr->ds_clp, + &hdr->args.seq_args, + &hdr->res.seq_res, + task)) + return; + + ff_layout_write_prepare_common(task, hdr); +} + +static void ff_layout_write_call_done(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags) && + task->tk_status == 0) { + nfs4_sequence_done(task, &hdr->res.seq_res); + return; + } + + /* Note this may cause RPC to be resent */ + hdr->mds_ops->rpc_call_done(task, hdr); +} + +static void ff_layout_write_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_pgio_header *hdr = data; + + ff_layout_write_record_layoutstats_done(task, hdr); + rpc_count_iostats_metrics(task, + &NFS_CLIENT(hdr->inode)->cl_metrics[NFSPROC4_CLNT_WRITE]); +} + +static void ff_layout_write_release(void *data) +{ + struct nfs_pgio_header *hdr = data; + + ff_layout_write_record_layoutstats_done(&hdr->task, hdr); + if (test_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags)) { + ff_layout_send_layouterror(hdr->lseg); + ff_layout_reset_write(hdr, true); + } else if (test_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags)) + ff_layout_reset_write(hdr, false); + pnfs_generic_rw_release(data); +} + +static void ff_layout_commit_record_layoutstats_start(struct rpc_task *task, + struct nfs_commit_data *cdata) +{ + u32 idx, dss_id; + + if (test_and_set_bit(NFS_IOHDR_STAT, &cdata->flags)) + return; + + idx = calc_mirror_idx_from_commit(cdata->lseg, cdata->ds_commit_index); + dss_id = calc_dss_id_from_commit(cdata->lseg, cdata->ds_commit_index); + nfs4_ff_layout_stat_io_start_write(cdata->inode, + FF_LAYOUT_COMP(cdata->lseg, idx), + dss_id, + 0, task->tk_start); +} + +static void ff_layout_commit_record_layoutstats_done(struct rpc_task *task, + struct nfs_commit_data *cdata) +{ + struct nfs_page *req; + __u64 count = 0; + u32 idx, dss_id; + + if (!test_and_clear_bit(NFS_IOHDR_STAT, &cdata->flags)) + return; + + if (task->tk_status == 0) { + list_for_each_entry(req, &cdata->pages, wb_list) + count += req->wb_bytes; + } + + idx = calc_mirror_idx_from_commit(cdata->lseg, cdata->ds_commit_index); + dss_id = calc_dss_id_from_commit(cdata->lseg, cdata->ds_commit_index); + nfs4_ff_layout_stat_io_end_write(task, + FF_LAYOUT_COMP(cdata->lseg, idx), + dss_id, + count, count, NFS_FILE_SYNC); + set_bit(NFS_LSEG_LAYOUTRETURN, &cdata->lseg->pls_flags); +} + +static int ff_layout_commit_prepare_common(struct rpc_task *task, + struct nfs_commit_data *cdata) +{ + if (!pnfs_is_valid_lseg(cdata->lseg)) { + rpc_exit(task, -EAGAIN); + return -EAGAIN; + } + + ff_layout_commit_record_layoutstats_start(task, cdata); + return 0; +} + +static void ff_layout_commit_prepare_v3(struct rpc_task *task, void *data) +{ + if (ff_layout_commit_prepare_common(task, data)) + return; + + rpc_call_start(task); +} + +static void ff_layout_commit_prepare_v4(struct rpc_task *task, void *data) +{ + struct nfs_commit_data *wdata = data; + + if (nfs4_setup_sequence(wdata->ds_clp, + &wdata->args.seq_args, + &wdata->res.seq_res, + task)) + return; + ff_layout_commit_prepare_common(task, data); +} + +static void ff_layout_commit_done(struct rpc_task *task, void *data) +{ + pnfs_generic_write_commit_done(task, data); +} + +static void ff_layout_commit_count_stats(struct rpc_task *task, void *data) +{ + struct nfs_commit_data *cdata = data; + + ff_layout_commit_record_layoutstats_done(task, cdata); + rpc_count_iostats_metrics(task, + &NFS_CLIENT(cdata->inode)->cl_metrics[NFSPROC4_CLNT_COMMIT]); +} + +static void ff_layout_commit_release(void *data) +{ + struct nfs_commit_data *cdata = data; + + ff_layout_commit_record_layoutstats_done(&cdata->task, cdata); + pnfs_generic_commit_release(data); +} + +static const struct rpc_call_ops ff_layout_read_call_ops_v3 = { + .rpc_call_prepare = ff_layout_read_prepare_v3, + .rpc_call_done = ff_layout_read_call_done, + .rpc_count_stats = ff_layout_read_count_stats, + .rpc_release = ff_layout_read_release, +}; + +static const struct rpc_call_ops ff_layout_read_call_ops_v4 = { + .rpc_call_prepare = ff_layout_read_prepare_v4, + .rpc_call_done = ff_layout_read_call_done, + .rpc_count_stats = ff_layout_read_count_stats, + .rpc_release = ff_layout_read_release, +}; + +static const struct rpc_call_ops ff_layout_write_call_ops_v3 = { + .rpc_call_prepare = ff_layout_write_prepare_v3, + .rpc_call_done = ff_layout_write_call_done, + .rpc_count_stats = ff_layout_write_count_stats, + .rpc_release = ff_layout_write_release, +}; + +static const struct rpc_call_ops ff_layout_write_call_ops_v4 = { + .rpc_call_prepare = ff_layout_write_prepare_v4, + .rpc_call_done = ff_layout_write_call_done, + .rpc_count_stats = ff_layout_write_count_stats, + .rpc_release = ff_layout_write_release, +}; + +static const struct rpc_call_ops ff_layout_commit_call_ops_v3 = { + .rpc_call_prepare = ff_layout_commit_prepare_v3, + .rpc_call_done = ff_layout_commit_done, + .rpc_count_stats = ff_layout_commit_count_stats, + .rpc_release = ff_layout_commit_release, +}; + +static const struct rpc_call_ops ff_layout_commit_call_ops_v4 = { + .rpc_call_prepare = ff_layout_commit_prepare_v4, + .rpc_call_done = ff_layout_commit_done, + .rpc_count_stats = ff_layout_commit_count_stats, + .rpc_release = ff_layout_commit_release, +}; + +static enum pnfs_try_status +ff_layout_read_pagelist(struct nfs_pgio_header *hdr) +{ + struct pnfs_layout_segment *lseg = hdr->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + struct nfsd_file *localio; + struct nfs4_ff_layout_mirror *mirror; + const struct cred *ds_cred; + loff_t offset = hdr->args.offset; + u32 idx = hdr->pgio_mirror_idx; + int vers; + struct nfs_fh *fh; + u32 dss_id; + bool ds_fatal_error = false; + + dprintk("--> %s ino %llu pgbase %u req %zu@%llu\n", + __func__, hdr->inode->i_ino, + hdr->args.pgbase, (size_t)hdr->args.count, offset); + + mirror = FF_LAYOUT_COMP(lseg, idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(lseg)->stripe_unit, + mirror->dss_count, + offset); + ds = nfs4_ff_layout_prepare_ds(lseg, mirror, dss_id, false); + if (IS_ERR(ds)) { + ds_fatal_error = nfs_error_is_fatal(PTR_ERR(ds)); + goto out_failed; + } + + ds_clnt = nfs4_ff_find_or_create_ds_client(mirror, ds->ds_clp, + hdr->inode, dss_id); + if (IS_ERR(ds_clnt)) + goto out_failed; + + ds_cred = ff_layout_get_ds_cred(mirror, &lseg->pls_range, hdr->cred, dss_id); + if (!ds_cred) + goto out_failed; + + vers = nfs4_ff_layout_ds_version(mirror, dss_id); + + dprintk("%s USE DS: %s cl_count %d vers %d\n", __func__, + ds->ds_remotestr, refcount_read(&ds->ds_clp->cl_count), vers); + + hdr->pgio_done_cb = ff_layout_read_done_cb; + refcount_inc(&ds->ds_clp->cl_count); + hdr->ds_clp = ds->ds_clp; + fh = nfs4_ff_layout_select_ds_fh(mirror, dss_id); + if (fh) + hdr->args.fh = fh; + + nfs4_ff_layout_select_ds_stateid(mirror, dss_id, &hdr->args.stateid); + + /* + * Note that if we ever decide to split across DSes, + * then we may need to handle dense-like offsets. + */ + hdr->args.offset = offset; + hdr->mds_offset = offset; + + /* Start IO accounting for local read */ + localio = ff_local_open_fh(lseg, idx, dss_id, ds->ds_clp, ds_cred, fh, + FMODE_READ); + if (localio) { + hdr->task.tk_start = ktime_get(); + ff_layout_read_record_layoutstats_start(&hdr->task, hdr); + } + + /* Perform an asynchronous read to ds */ + nfs_initiate_pgio(ds_clnt, hdr, ds_cred, ds->ds_clp->rpc_ops, + vers == 3 ? &ff_layout_read_call_ops_v3 : + &ff_layout_read_call_ops_v4, + 0, RPC_TASK_SOFTCONN, localio); + put_cred(ds_cred); + return PNFS_ATTEMPTED; + +out_failed: + if (ff_layout_avoid_mds_available_ds(lseg) && !ds_fatal_error) + return PNFS_TRY_AGAIN; + if (ff_layout_no_fallback_to_mds(lseg)) { + /* + * FF_FLAGS_NO_IO_THRU_MDS: force fresh LAYOUTGET, + * never fall through to MDS I/O. + */ + pnfs_error_mark_layout_for_return(hdr->inode, lseg); + return PNFS_TRY_AGAIN; + } + trace_pnfs_mds_fallback_read_pagelist(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_READ, NFS_I(hdr->inode)->layout, lseg); + return PNFS_NOT_ATTEMPTED; +} + +/* Perform async writes. */ +static enum pnfs_try_status +ff_layout_write_pagelist(struct nfs_pgio_header *hdr, int sync) +{ + struct pnfs_layout_segment *lseg = hdr->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + struct nfsd_file *localio; + struct nfs4_ff_layout_mirror *mirror; + const struct cred *ds_cred; + loff_t offset = hdr->args.offset; + int vers; + struct nfs_fh *fh; + u32 idx = hdr->pgio_mirror_idx; + u32 dss_id; + bool ds_fatal_error = false; + + mirror = FF_LAYOUT_COMP(lseg, idx); + dss_id = nfs4_ff_layout_calc_dss_id( + FF_LAYOUT_LSEG(lseg)->stripe_unit, + mirror->dss_count, + offset); + ds = nfs4_ff_layout_prepare_ds(lseg, mirror, dss_id, true); + if (IS_ERR(ds)) { + ds_fatal_error = nfs_error_is_fatal(PTR_ERR(ds)); + goto out_failed; + } + + ds_clnt = nfs4_ff_find_or_create_ds_client(mirror, ds->ds_clp, + hdr->inode, dss_id); + if (IS_ERR(ds_clnt)) + goto out_failed; + + ds_cred = ff_layout_get_ds_cred(mirror, &lseg->pls_range, hdr->cred, dss_id); + if (!ds_cred) + goto out_failed; + + vers = nfs4_ff_layout_ds_version(mirror, dss_id); + + dprintk("%s ino %llu sync %d req %zu@%llu DS: %s cl_count %d vers %d\n", + __func__, hdr->inode->i_ino, sync, (size_t) hdr->args.count, + offset, ds->ds_remotestr, refcount_read(&ds->ds_clp->cl_count), + vers); + + hdr->pgio_done_cb = ff_layout_write_done_cb; + refcount_inc(&ds->ds_clp->cl_count); + hdr->ds_clp = ds->ds_clp; + hdr->ds_commit_idx = calc_commit_idx(lseg, idx, dss_id); + fh = nfs4_ff_layout_select_ds_fh(mirror, dss_id); + if (fh) + hdr->args.fh = fh; + + nfs4_ff_layout_select_ds_stateid(mirror, dss_id, &hdr->args.stateid); + + /* + * Note that if we ever decide to split across DSes, + * then we may need to handle dense-like offsets. + */ + hdr->args.offset = offset; + + /* Start IO accounting for local write */ + localio = ff_local_open_fh(lseg, idx, dss_id, ds->ds_clp, ds_cred, fh, + FMODE_READ|FMODE_WRITE); + if (localio) { + hdr->task.tk_start = ktime_get(); + ff_layout_write_record_layoutstats_start(&hdr->task, hdr); + } + + /* Perform an asynchronous write */ + nfs_initiate_pgio(ds_clnt, hdr, ds_cred, ds->ds_clp->rpc_ops, + vers == 3 ? &ff_layout_write_call_ops_v3 : + &ff_layout_write_call_ops_v4, + sync, RPC_TASK_SOFTCONN, localio); + put_cred(ds_cred); + return PNFS_ATTEMPTED; + +out_failed: + if (ff_layout_avoid_mds_available_ds(lseg) && !ds_fatal_error) + return PNFS_TRY_AGAIN; + if (ff_layout_no_fallback_to_mds(lseg)) { + /* + * FF_FLAGS_NO_IO_THRU_MDS: force fresh LAYOUTGET, + * never fall through to MDS I/O. + */ + pnfs_error_mark_layout_for_return(hdr->inode, lseg); + return PNFS_TRY_AGAIN; + } + trace_pnfs_mds_fallback_write_pagelist(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_RW, NFS_I(hdr->inode)->layout, lseg); + return PNFS_NOT_ATTEMPTED; +} + +static struct nfs_fh * +select_ds_fh_from_commit(struct pnfs_layout_segment *lseg, u32 i, u32 dss_id) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg); + + /* FIXME: Assume that there is only one NFS version available + * for the DS. + */ + return &flseg->mirror_array[i]->dss[dss_id].fh_versions[0]; +} + +static int ff_layout_initiate_commit(struct nfs_commit_data *data, int how) +{ + struct pnfs_layout_segment *lseg = data->lseg; + struct nfs4_pnfs_ds *ds; + struct rpc_clnt *ds_clnt; + struct nfsd_file *localio; + struct nfs4_ff_layout_mirror *mirror; + const struct cred *ds_cred; + u32 idx, dss_id; + int vers, ret; + struct nfs_fh *fh; + + if (!lseg || !(pnfs_is_valid_lseg(lseg) || + test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags))) + goto out_err; + + idx = calc_mirror_idx_from_commit(lseg, data->ds_commit_index); + mirror = FF_LAYOUT_COMP(lseg, idx); + dss_id = calc_dss_id_from_commit(lseg, data->ds_commit_index); + ds = nfs4_ff_layout_prepare_ds(lseg, mirror, dss_id, true); + if (IS_ERR(ds)) + goto out_err; + + ds_clnt = nfs4_ff_find_or_create_ds_client(mirror, ds->ds_clp, + data->inode, dss_id); + if (IS_ERR(ds_clnt)) + goto out_err; + + ds_cred = ff_layout_get_ds_cred(mirror, &lseg->pls_range, data->cred, dss_id); + if (!ds_cred) + goto out_err; + + vers = nfs4_ff_layout_ds_version(mirror, dss_id); + + dprintk("%s ino %llu, how %d cl_count %d vers %d\n", __func__, + data->inode->i_ino, how, refcount_read(&ds->ds_clp->cl_count), + vers); + data->commit_done_cb = ff_layout_commit_done_cb; + data->cred = ds_cred; + refcount_inc(&ds->ds_clp->cl_count); + data->ds_clp = ds->ds_clp; + fh = select_ds_fh_from_commit(lseg, idx, dss_id); + if (fh) + data->args.fh = fh; + + /* Start IO accounting for local commit */ + localio = ff_local_open_fh(lseg, idx, dss_id, ds->ds_clp, ds_cred, fh, + FMODE_READ|FMODE_WRITE); + if (localio) { + data->task.tk_start = ktime_get(); + ff_layout_commit_record_layoutstats_start(&data->task, data); + } + + ret = nfs_initiate_commit(ds_clnt, data, ds->ds_clp->rpc_ops, + vers == 3 ? &ff_layout_commit_call_ops_v3 : + &ff_layout_commit_call_ops_v4, + how, RPC_TASK_SOFTCONN, localio); + put_cred(ds_cred); + return ret; +out_err: + pnfs_generic_prepare_to_resend_writes(data); + pnfs_generic_commit_release(data); + return -EAGAIN; +} + +static int +ff_layout_commit_pagelist(struct inode *inode, struct list_head *mds_pages, + int how, struct nfs_commit_info *cinfo) +{ + return pnfs_generic_commit_pagelist(inode, mds_pages, how, cinfo, + ff_layout_initiate_commit); +} + +static bool ff_layout_match_rw(const struct rpc_task *task, + const struct nfs_pgio_header *hdr, + const struct pnfs_layout_segment *lseg) +{ + return hdr->lseg == lseg; +} + +static bool ff_layout_match_commit(const struct rpc_task *task, + const struct nfs_commit_data *cdata, + const struct pnfs_layout_segment *lseg) +{ + return cdata->lseg == lseg; +} + +static bool ff_layout_match_io(const struct rpc_task *task, const void *data) +{ + const struct rpc_call_ops *ops = task->tk_ops; + + if (ops == &ff_layout_read_call_ops_v3 || + ops == &ff_layout_read_call_ops_v4 || + ops == &ff_layout_write_call_ops_v3 || + ops == &ff_layout_write_call_ops_v4) + return ff_layout_match_rw(task, task->tk_calldata, data); + if (ops == &ff_layout_commit_call_ops_v3 || + ops == &ff_layout_commit_call_ops_v4) + return ff_layout_match_commit(task, task->tk_calldata, data); + return false; +} + +static void ff_layout_cancel_io(struct pnfs_layout_segment *lseg) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg); + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_ff_layout_ds *mirror_ds; + struct nfs4_pnfs_ds *ds; + struct nfs_client *ds_clp; + struct rpc_clnt *clnt; + u32 idx, dss_id; + + for (idx = 0; idx < flseg->mirror_array_cnt; idx++) { + mirror = flseg->mirror_array[idx]; + for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { + mirror_ds = mirror->dss[dss_id].mirror_ds; + if (IS_ERR_OR_NULL(mirror_ds)) + continue; + ds = mirror->dss[dss_id].mirror_ds->ds; + if (!ds) + continue; + ds_clp = ds->ds_clp; + if (!ds_clp) + continue; + clnt = ds_clp->cl_rpcclient; + if (!clnt) + continue; + if (!rpc_cancel_tasks(clnt, -EAGAIN, + ff_layout_match_io, lseg)) + continue; + rpc_clnt_disconnect(clnt); + } + } +} + +static struct pnfs_ds_commit_info * +ff_layout_get_ds_info(struct inode *inode) +{ + struct pnfs_layout_hdr *layout = NFS_I(inode)->layout; + + if (layout == NULL) + return NULL; + + return &FF_LAYOUT_FROM_HDR(layout)->commit_info; +} + +static void +ff_layout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg) +{ + struct nfs4_ff_layout_segment *flseg = FF_LAYOUT_LSEG(lseg); + struct inode *inode = lseg->pls_layout->plh_inode; + struct pnfs_commit_array *array, *new; + u32 size = flseg->mirror_array_cnt * flseg->mirror_array[0]->dss_count; + + new = pnfs_alloc_commit_array(size, + nfs_io_gfp_mask()); + if (new) { + spin_lock(&inode->i_lock); + array = pnfs_add_commit_array(fl_cinfo, new, lseg); + spin_unlock(&inode->i_lock); + if (array != new) + pnfs_free_commit_array(new); + } +} + +static void +ff_layout_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, + struct inode *inode) +{ + spin_lock(&inode->i_lock); + pnfs_generic_ds_cinfo_destroy(fl_cinfo); + spin_unlock(&inode->i_lock); +} + +static void +ff_layout_free_deviceid_node(struct nfs4_deviceid_node *d) +{ + nfs4_ff_layout_free_deviceid(container_of(d, struct nfs4_ff_layout_ds, + id_node)); +} + +static int ff_layout_encode_ioerr(struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args, + const struct nfs4_flexfile_layoutreturn_args *ff_args) +{ + __be32 *start; + + start = xdr_reserve_space(xdr, 4); + if (unlikely(!start)) + return -E2BIG; + + *start = cpu_to_be32(ff_args->num_errors); + /* This assume we always return _ALL_ layouts */ + return ff_layout_encode_ds_ioerr(xdr, &ff_args->errors); +} + +static void +ff_layout_encode_ff_iostat_head(struct xdr_stream *xdr, + const nfs4_stateid *stateid, + const struct nfs42_layoutstat_devinfo *devinfo) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 8 + 8); + p = xdr_encode_hyper(p, devinfo->offset); + p = xdr_encode_hyper(p, devinfo->length); + encode_opaque_fixed(xdr, stateid->data, NFS4_STATEID_SIZE); + p = xdr_reserve_space(xdr, 4*8); + p = xdr_encode_hyper(p, devinfo->read_count); + p = xdr_encode_hyper(p, devinfo->read_bytes); + p = xdr_encode_hyper(p, devinfo->write_count); + p = xdr_encode_hyper(p, devinfo->write_bytes); + encode_opaque_fixed(xdr, devinfo->dev_id.data, NFS4_DEVICEID4_SIZE); +} + +static void +ff_layout_encode_ff_iostat(struct xdr_stream *xdr, + const nfs4_stateid *stateid, + const struct nfs42_layoutstat_devinfo *devinfo) +{ + ff_layout_encode_ff_iostat_head(xdr, stateid, devinfo); + ff_layout_encode_ff_layoutupdate(xdr, devinfo, + devinfo->ld_private.data); +} + +/* report nothing for now */ +static void ff_layout_encode_iostats_array(struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args, + struct nfs4_flexfile_layoutreturn_args *ff_args) +{ + __be32 *p; + int i; + + p = xdr_reserve_space(xdr, 4); + *p = cpu_to_be32(ff_args->num_dev); + for (i = 0; i < ff_args->num_dev; i++) + ff_layout_encode_ff_iostat(xdr, + &args->layout->plh_stateid, + &ff_args->devinfo[i]); +} + +static void +ff_layout_free_iostats_array(struct nfs42_layoutstat_devinfo *devinfo, + unsigned int num_entries) +{ + unsigned int i; + + for (i = 0; i < num_entries; i++) { + if (!devinfo[i].ld_private.ops) + continue; + if (!devinfo[i].ld_private.ops->free) + continue; + devinfo[i].ld_private.ops->free(&devinfo[i].ld_private); + } +} + +static struct nfs4_deviceid_node * +ff_layout_alloc_deviceid_node(struct nfs_server *server, + struct pnfs_device *pdev, gfp_t gfp_flags) +{ + struct nfs4_ff_layout_ds *dsaddr; + + dsaddr = nfs4_ff_alloc_deviceid_node(server, pdev, gfp_flags); + if (!dsaddr) + return NULL; + return &dsaddr->id_node; +} + +static void +ff_layout_encode_layoutreturn(struct xdr_stream *xdr, + const void *voidargs, + const struct nfs4_xdr_opaque_data *ff_opaque) +{ + const struct nfs4_layoutreturn_args *args = voidargs; + struct nfs4_flexfile_layoutreturn_args *ff_args = ff_opaque->data; + struct xdr_buf tmp_buf = { + .head = { + [0] = { + .iov_base = page_address(ff_args->pages[0]), + }, + }, + .buflen = PAGE_SIZE, + }; + struct xdr_stream tmp_xdr; + __be32 *start; + + dprintk("%s: Begin\n", __func__); + + xdr_init_encode(&tmp_xdr, &tmp_buf, NULL, NULL); + + ff_layout_encode_ioerr(&tmp_xdr, args, ff_args); + ff_layout_encode_iostats_array(&tmp_xdr, args, ff_args); + + start = xdr_reserve_space(xdr, 4); + *start = cpu_to_be32(tmp_buf.len); + xdr_write_pages(xdr, ff_args->pages, 0, tmp_buf.len); + + dprintk("%s: Return\n", __func__); +} + +static void +ff_layout_free_layoutreturn(struct nfs4_xdr_opaque_data *args) +{ + struct nfs4_flexfile_layoutreturn_args *ff_args; + + if (!args->data) + return; + ff_args = args->data; + args->data = NULL; + + ff_layout_free_ds_ioerr(&ff_args->errors); + ff_layout_free_iostats_array(ff_args->devinfo, ff_args->num_dev); + + put_page(ff_args->pages[0]); + kfree(ff_args); +} + +static const struct nfs4_xdr_opaque_ops layoutreturn_ops = { + .encode = ff_layout_encode_layoutreturn, + .free = ff_layout_free_layoutreturn, +}; + +static int +ff_layout_prepare_layoutreturn(struct nfs4_layoutreturn_args *args) +{ + struct nfs4_flexfile_layoutreturn_args *ff_args; + struct nfs4_flexfile_layout *ff_layout = FF_LAYOUT_FROM_HDR(args->layout); + + ff_args = kmalloc_obj(*ff_args, nfs_io_gfp_mask()); + if (!ff_args) + goto out_nomem; + ff_args->pages[0] = alloc_page(nfs_io_gfp_mask()); + if (!ff_args->pages[0]) + goto out_nomem_free; + + INIT_LIST_HEAD(&ff_args->errors); + ff_args->num_errors = ff_layout_fetch_ds_ioerr(args->layout, + &args->range, &ff_args->errors, + FF_LAYOUTRETURN_MAXERR); + + spin_lock(&args->inode->i_lock); + ff_args->num_dev = ff_layout_mirror_prepare_stats( + &ff_layout->generic_hdr, &ff_args->devinfo[0], + ARRAY_SIZE(ff_args->devinfo), NFS4_FF_OP_LAYOUTRETURN); + spin_unlock(&args->inode->i_lock); + + args->ld_private->ops = &layoutreturn_ops; + args->ld_private->data = ff_args; + return 0; +out_nomem_free: + kfree(ff_args); +out_nomem: + return -ENOMEM; +} + +#ifdef CONFIG_NFS_V4_2 +void +ff_layout_send_layouterror(struct pnfs_layout_segment *lseg) +{ + struct pnfs_layout_hdr *lo = lseg->pls_layout; + struct nfs42_layout_error *errors; + LIST_HEAD(head); + + if (!nfs_server_capable(lo->plh_inode, NFS_CAP_LAYOUTERROR)) + return; + ff_layout_fetch_ds_ioerr(lo, &lseg->pls_range, &head, -1); + if (list_empty(&head)) + return; + + errors = kmalloc_objs(*errors, NFS42_LAYOUTERROR_MAX, nfs_io_gfp_mask()); + if (errors != NULL) { + const struct nfs4_ff_layout_ds_err *pos; + size_t n = 0; + + list_for_each_entry(pos, &head, list) { + errors[n].offset = pos->offset; + errors[n].length = pos->length; + nfs4_stateid_copy(&errors[n].stateid, &pos->stateid); + errors[n].errors[0].dev_id = pos->deviceid; + errors[n].errors[0].status = pos->status; + errors[n].errors[0].opnum = pos->opnum; + n++; + if (!list_is_last(&pos->list, &head) && + n < NFS42_LAYOUTERROR_MAX) + continue; + if (nfs42_proc_layouterror(lseg, errors, n) < 0) + break; + n = 0; + } + kfree(errors); + } + ff_layout_free_ds_ioerr(&head); +} +#else +void +ff_layout_send_layouterror(struct pnfs_layout_segment *lseg) +{ +} +#endif + +static int +ff_layout_ntop4(const struct sockaddr *sap, char *buf, const size_t buflen) +{ + const struct sockaddr_in *sin = (struct sockaddr_in *)sap; + + return snprintf(buf, buflen, "%pI4", &sin->sin_addr); +} + +static size_t +ff_layout_ntop6_noscopeid(const struct sockaddr *sap, char *buf, + const int buflen) +{ + const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap; + const struct in6_addr *addr = &sin6->sin6_addr; + + /* + * RFC 4291, Section 2.2.2 + * + * Shorthanded ANY address + */ + if (ipv6_addr_any(addr)) + return snprintf(buf, buflen, "::"); + + /* + * RFC 4291, Section 2.2.2 + * + * Shorthanded loopback address + */ + if (ipv6_addr_loopback(addr)) + return snprintf(buf, buflen, "::1"); + + /* + * RFC 4291, Section 2.2.3 + * + * Special presentation address format for mapped v4 + * addresses. + */ + if (ipv6_addr_v4mapped(addr)) + return snprintf(buf, buflen, "::ffff:%pI4", + &addr->s6_addr32[3]); + + /* + * RFC 4291, Section 2.2.1 + */ + return snprintf(buf, buflen, "%pI6c", addr); +} + +/* Derived from rpc_sockaddr2uaddr */ +static void +ff_layout_encode_netaddr(struct xdr_stream *xdr, struct nfs4_pnfs_ds_addr *da) +{ + struct sockaddr *sap = (struct sockaddr *)&da->da_addr; + char portbuf[RPCBIND_MAXUADDRPLEN]; + char addrbuf[RPCBIND_MAXUADDRLEN]; + unsigned short port; + int len, netid_len; + __be32 *p; + + switch (sap->sa_family) { + case AF_INET: + if (ff_layout_ntop4(sap, addrbuf, sizeof(addrbuf)) == 0) + return; + port = ntohs(((struct sockaddr_in *)sap)->sin_port); + break; + case AF_INET6: + if (ff_layout_ntop6_noscopeid(sap, addrbuf, sizeof(addrbuf)) == 0) + return; + port = ntohs(((struct sockaddr_in6 *)sap)->sin6_port); + break; + default: + WARN_ON_ONCE(1); + return; + } + + snprintf(portbuf, sizeof(portbuf), ".%u.%u", port >> 8, port & 0xff); + len = strlcat(addrbuf, portbuf, sizeof(addrbuf)); + + netid_len = strlen(da->da_netid); + p = xdr_reserve_space(xdr, 4 + netid_len); + xdr_encode_opaque(p, da->da_netid, netid_len); + + p = xdr_reserve_space(xdr, 4 + len); + xdr_encode_opaque(p, addrbuf, len); +} + +static void +ff_layout_encode_nfstime(struct xdr_stream *xdr, + ktime_t t) +{ + struct timespec64 ts; + __be32 *p; + + p = xdr_reserve_space(xdr, 12); + ts = ktime_to_timespec64(t); + p = xdr_encode_hyper(p, ts.tv_sec); + *p++ = cpu_to_be32(ts.tv_nsec); +} + +static void +ff_layout_encode_io_latency(struct xdr_stream *xdr, + struct nfs4_ff_io_stat *stat) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 5 * 8); + p = xdr_encode_hyper(p, stat->ops_requested); + p = xdr_encode_hyper(p, stat->bytes_requested); + p = xdr_encode_hyper(p, stat->ops_completed); + p = xdr_encode_hyper(p, stat->bytes_completed); + p = xdr_encode_hyper(p, stat->bytes_not_delivered); + ff_layout_encode_nfstime(xdr, stat->total_busy_time); + ff_layout_encode_nfstime(xdr, stat->aggregate_completion_time); +} + +static void +ff_layout_encode_ff_layoutupdate(struct xdr_stream *xdr, + const struct nfs42_layoutstat_devinfo *devinfo, + struct nfs4_ff_layout_ds_stripe *dss_info) +{ + struct nfs4_pnfs_ds_addr *da; + struct nfs4_pnfs_ds *ds = dss_info->mirror_ds->ds; + struct nfs_fh *fh = &dss_info->fh_versions[0]; + __be32 *p; + + da = list_first_entry(&ds->ds_addrs, struct nfs4_pnfs_ds_addr, da_node); + dprintk("%s: DS %s: encoding address %s\n", + __func__, ds->ds_remotestr, da->da_remotestr); + /* netaddr4 */ + ff_layout_encode_netaddr(xdr, da); + /* nfs_fh4 */ + p = xdr_reserve_space(xdr, 4 + fh->size); + xdr_encode_opaque(p, fh->data, fh->size); + /* ff_io_latency4 read */ + spin_lock(&dss_info->mirror->lock); + ff_layout_encode_io_latency(xdr, + &dss_info->read_stat.io_stat); + /* ff_io_latency4 write */ + ff_layout_encode_io_latency(xdr, + &dss_info->write_stat.io_stat); + spin_unlock(&dss_info->mirror->lock); + /* nfstime4 */ + ff_layout_encode_nfstime(xdr, + ktime_sub(ktime_get(), + dss_info->start_time)); + /* bool */ + p = xdr_reserve_space(xdr, 4); + *p = cpu_to_be32(false); +} + +static void +ff_layout_encode_layoutstats(struct xdr_stream *xdr, const void *args, + const struct nfs4_xdr_opaque_data *opaque) +{ + struct nfs42_layoutstat_devinfo *devinfo = container_of(opaque, + struct nfs42_layoutstat_devinfo, ld_private); + __be32 *start; + + /* layoutupdate length */ + start = xdr_reserve_space(xdr, 4); + ff_layout_encode_ff_layoutupdate(xdr, devinfo, opaque->data); + + *start = cpu_to_be32((xdr->p - start - 1) * 4); +} + +static void +ff_layout_free_layoutstats(struct nfs4_xdr_opaque_data *opaque) +{ + struct nfs4_ff_layout_ds_stripe *dss_info = opaque->data; + struct nfs4_ff_layout_mirror *mirror = dss_info->mirror; + + ff_layout_put_mirror(mirror); +} + +static const struct nfs4_xdr_opaque_ops layoutstat_ops = { + .encode = ff_layout_encode_layoutstats, + .free = ff_layout_free_layoutstats, +}; + +static int +ff_layout_mirror_prepare_stats(struct pnfs_layout_hdr *lo, + struct nfs42_layoutstat_devinfo *devinfo, + int dev_limit, enum nfs4_ff_op_type type) +{ + struct nfs4_flexfile_layout *ff_layout = FF_LAYOUT_FROM_HDR(lo); + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_ff_layout_ds_stripe *dss_info; + struct nfs4_deviceid_node *dev; + int i = 0, dss_id; + + list_for_each_entry(mirror, &ff_layout->mirrors, mirrors) { + for (dss_id = 0; dss_id < mirror->dss_count; ++dss_id) { + dss_info = &mirror->dss[dss_id]; + if (i >= dev_limit) + break; + if (IS_ERR_OR_NULL(dss_info->mirror_ds)) + continue; + if (!test_and_clear_bit(NFS4_FF_MIRROR_STAT_AVAIL, + &mirror->flags) && + type != NFS4_FF_OP_LAYOUTRETURN) + continue; + /* mirror refcount put in cleanup_layoutstats */ + if (!refcount_inc_not_zero(&mirror->ref)) + continue; + dev = &dss_info->mirror_ds->id_node; + memcpy(&devinfo->dev_id, + &dev->deviceid, + NFS4_DEVICEID4_SIZE); + devinfo->offset = 0; + devinfo->length = NFS4_MAX_UINT64; + spin_lock(&mirror->lock); + devinfo->read_count = + dss_info->read_stat.io_stat.ops_completed; + devinfo->read_bytes = + dss_info->read_stat.io_stat.bytes_completed; + devinfo->write_count = + dss_info->write_stat.io_stat.ops_completed; + devinfo->write_bytes = + dss_info->write_stat.io_stat.bytes_completed; + spin_unlock(&mirror->lock); + devinfo->layout_type = LAYOUT_FLEX_FILES; + devinfo->ld_private.ops = &layoutstat_ops; + devinfo->ld_private.data = &mirror->dss[dss_id]; + + devinfo++; + i++; + } + } + return i; +} + +static int ff_layout_prepare_layoutstats(struct nfs42_layoutstat_args *args) +{ + struct pnfs_layout_hdr *lo; + struct nfs4_flexfile_layout *ff_layout; + const int dev_count = PNFS_LAYOUTSTATS_MAXDEV; + + /* For now, send at most PNFS_LAYOUTSTATS_MAXDEV statistics */ + args->devinfo = kmalloc_objs(*args->devinfo, dev_count, + nfs_io_gfp_mask()); + if (!args->devinfo) + return -ENOMEM; + + spin_lock(&args->inode->i_lock); + lo = NFS_I(args->inode)->layout; + if (lo && pnfs_layout_is_valid(lo)) { + ff_layout = FF_LAYOUT_FROM_HDR(lo); + args->num_dev = ff_layout_mirror_prepare_stats( + &ff_layout->generic_hdr, &args->devinfo[0], dev_count, + NFS4_FF_OP_LAYOUTSTATS); + } else + args->num_dev = 0; + spin_unlock(&args->inode->i_lock); + if (!args->num_dev) { + kfree(args->devinfo); + args->devinfo = NULL; + return -ENOENT; + } + + return 0; +} + +static int +ff_layout_set_layoutdriver(struct nfs_server *server, + const struct nfs_fh *dummy) +{ +#if IS_ENABLED(CONFIG_NFS_V4_2) + server->caps |= NFS_CAP_LAYOUTSTATS | NFS_CAP_REBOOT_LAYOUTRETURN; +#endif + return 0; +} + +static const struct pnfs_commit_ops ff_layout_commit_ops = { + .setup_ds_info = ff_layout_setup_ds_info, + .release_ds_info = ff_layout_release_ds_info, + .mark_request_commit = pnfs_layout_mark_request_commit, + .clear_request_commit = pnfs_generic_clear_request_commit, + .scan_commit_lists = pnfs_generic_scan_commit_lists, + .recover_commit_reqs = pnfs_generic_recover_commit_reqs, + .commit_pagelist = ff_layout_commit_pagelist, +}; + +static struct pnfs_layoutdriver_type flexfilelayout_type = { + .id = LAYOUT_FLEX_FILES, + .name = "LAYOUT_FLEX_FILES", + .owner = THIS_MODULE, + .flags = PNFS_LAYOUTGET_ON_OPEN, + .max_layoutget_response = 4096, /* 1 page or so... */ + .set_layoutdriver = ff_layout_set_layoutdriver, + .alloc_layout_hdr = ff_layout_alloc_layout_hdr, + .free_layout_hdr = ff_layout_free_layout_hdr, + .alloc_lseg = ff_layout_alloc_lseg, + .free_lseg = ff_layout_free_lseg, + .add_lseg = ff_layout_add_lseg, + .pg_read_ops = &ff_layout_pg_read_ops, + .pg_write_ops = &ff_layout_pg_write_ops, + .get_ds_info = ff_layout_get_ds_info, + .free_deviceid_node = ff_layout_free_deviceid_node, + .read_pagelist = ff_layout_read_pagelist, + .write_pagelist = ff_layout_write_pagelist, + .alloc_deviceid_node = ff_layout_alloc_deviceid_node, + .prepare_layoutreturn = ff_layout_prepare_layoutreturn, + .sync = pnfs_nfs_generic_sync, + .prepare_layoutstats = ff_layout_prepare_layoutstats, + .cancel_io = ff_layout_cancel_io, +}; + +static int __init nfs4flexfilelayout_init(void) +{ + printk(KERN_INFO "%s: NFSv4 Flexfile Layout Driver Registering...\n", + __func__); + return pnfs_register_layoutdriver(&flexfilelayout_type); +} + +static void __exit nfs4flexfilelayout_exit(void) +{ + printk(KERN_INFO "%s: NFSv4 Flexfile Layout Driver Unregistering...\n", + __func__); + pnfs_unregister_layoutdriver(&flexfilelayout_type); +} + +MODULE_ALIAS("nfs-layouttype4-4"); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("The NFSv4 flexfile layout driver"); + +module_init(nfs4flexfilelayout_init); +module_exit(nfs4flexfilelayout_exit); + +module_param(io_maxretrans, ushort, 0644); +MODULE_PARM_DESC(io_maxretrans, "The number of times the NFSv4.1 client " + "retries an I/O request before returning an error. "); diff --git a/fs/nfs/flexfilelayout/flexfilelayout.h b/fs/nfs/flexfilelayout/flexfilelayout.h new file mode 100644 index 000000000..a5bd00f69 --- /dev/null +++ b/fs/nfs/flexfilelayout/flexfilelayout.h @@ -0,0 +1,269 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * NFSv4 flexfile layout driver data structures. + * + * Copyright (c) 2014, Primary Data, Inc. All rights reserved. + * + * Tao Peng + */ + +#ifndef FS_NFS_NFS4FLEXFILELAYOUT_H +#define FS_NFS_NFS4FLEXFILELAYOUT_H + +#define FF_FLAGS_NO_LAYOUTCOMMIT 1 +#define FF_FLAGS_NO_IO_THRU_MDS 2 +#define FF_FLAGS_NO_READ_IO 4 + +#include +#include "../pnfs.h" + +/* XXX: Let's filter out insanely large mirror count for now to avoid oom + * due to network error etc. */ +#define NFS4_FLEXFILE_LAYOUT_MAX_MIRROR_CNT 4096 + +#define NFS4_FLEXFILE_LAYOUT_MAX_STRIPE_CNT 4096 + +/* LAYOUTSTATS report interval in ms */ +#define FF_LAYOUTSTATS_REPORT_INTERVAL (60000L) +#define FF_LAYOUTSTATS_MAXDEV 4 + +struct nfs4_ff_ds_version { + u32 version; + u32 minor_version; + u32 rsize; + u32 wsize; + bool tightly_coupled; +}; + +/* chained in global deviceid hlist */ +struct nfs4_ff_layout_ds { + struct nfs4_deviceid_node id_node; + u32 ds_versions_cnt; + struct nfs4_ff_ds_version *ds_versions; + struct nfs4_pnfs_ds *ds; +}; + +struct nfs4_ff_layout_ds_err { + struct list_head list; /* linked in mirror error_list */ + u64 offset; + u64 length; + int status; + enum nfs_opnum4 opnum; + nfs4_stateid stateid; + struct nfs4_deviceid deviceid; +}; + +struct nfs4_ff_io_stat { + __u64 ops_requested; + __u64 bytes_requested; + __u64 ops_completed; + __u64 bytes_completed; + __u64 bytes_not_delivered; + ktime_t total_busy_time; + ktime_t aggregate_completion_time; +}; + +struct nfs4_ff_busy_timer { + ktime_t start_time; + atomic_t n_ops; +}; + +struct nfs4_ff_layoutstat { + struct nfs4_ff_io_stat io_stat; + struct nfs4_ff_busy_timer busy_timer; +}; + +struct nfs4_ff_layout_mirror; + +struct nfs4_ff_layout_ds_stripe { + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_deviceid devid; + u32 efficiency; + struct nfs4_ff_layout_ds *mirror_ds; + u32 fh_versions_cnt; + struct nfs_fh *fh_versions; + nfs4_stateid stateid; + const struct cred __rcu *ro_cred; + const struct cred __rcu *rw_cred; + struct nfs_file_localio nfl; + struct nfs4_ff_layoutstat read_stat; + struct nfs4_ff_layoutstat write_stat; + ktime_t start_time; +}; + +struct nfs4_ff_layout_mirror { + struct pnfs_layout_hdr *layout; + struct list_head mirrors; + u32 dss_count; + struct nfs4_ff_layout_ds_stripe *dss; + refcount_t ref; + spinlock_t lock; + unsigned long flags; + u32 report_interval; +}; + +#define NFS4_FF_MIRROR_STAT_AVAIL (0) + +struct nfs4_ff_layout_segment { + struct pnfs_layout_segment generic_hdr; + u64 stripe_unit; + u32 flags; + u32 mirror_array_cnt; + struct nfs4_ff_layout_mirror *mirror_array[] __counted_by(mirror_array_cnt); +}; + +/* nfs4_flexfile_layout::flags bit indices */ +#define NFS4_FF_HDR_NO_IO_THRU_MDS 0 /* any lseg has had FF_FLAGS_NO_IO_THRU_MDS */ + +struct nfs4_flexfile_layout { + struct pnfs_layout_hdr generic_hdr; + struct pnfs_ds_commit_info commit_info; + struct list_head mirrors; + struct list_head error_list; /* nfs4_ff_layout_ds_err */ + ktime_t last_report_time; /* Layoutstat report times */ + unsigned long flags; +}; + +struct nfs4_flexfile_layoutreturn_args { + struct list_head errors; + struct nfs42_layoutstat_devinfo devinfo[FF_LAYOUTSTATS_MAXDEV]; + unsigned int num_errors; + unsigned int num_dev; + struct page *pages[1]; +}; + +static inline struct nfs4_flexfile_layout * +FF_LAYOUT_FROM_HDR(struct pnfs_layout_hdr *lo) +{ + return container_of(lo, struct nfs4_flexfile_layout, generic_hdr); +} + +static inline struct nfs4_ff_layout_segment * +FF_LAYOUT_LSEG(struct pnfs_layout_segment *lseg) +{ + return container_of(lseg, + struct nfs4_ff_layout_segment, + generic_hdr); +} + +static inline struct nfs4_ff_layout_ds * +FF_LAYOUT_MIRROR_DS(struct nfs4_deviceid_node *node) +{ + return container_of(node, struct nfs4_ff_layout_ds, id_node); +} + +static inline struct nfs4_ff_layout_mirror * +FF_LAYOUT_COMP(struct pnfs_layout_segment *lseg, u32 idx) +{ + struct nfs4_ff_layout_segment *fls = FF_LAYOUT_LSEG(lseg); + + if (idx < fls->mirror_array_cnt) + return fls->mirror_array[idx]; + return NULL; +} + +static inline struct nfs4_deviceid_node * +FF_LAYOUT_DEVID_NODE(struct pnfs_layout_segment *lseg, u32 idx, u32 dss_id) +{ + struct nfs4_ff_layout_mirror *mirror = FF_LAYOUT_COMP(lseg, idx); + + if (mirror != NULL) { + struct nfs4_ff_layout_ds *mirror_ds = mirror->dss[dss_id].mirror_ds; + + if (!IS_ERR_OR_NULL(mirror_ds)) + return &mirror_ds->id_node; + } + return NULL; +} + +static inline u32 +FF_LAYOUT_MIRROR_COUNT(struct pnfs_layout_segment *lseg) +{ + return FF_LAYOUT_LSEG(lseg)->mirror_array_cnt; +} + +static inline bool +ff_layout_no_fallback_to_mds(struct pnfs_layout_segment *lseg) +{ + return FF_LAYOUT_LSEG(lseg)->flags & FF_FLAGS_NO_IO_THRU_MDS; +} + +/* + * Sticky hdr-level mirror of FF_FLAGS_NO_IO_THRU_MDS so callers that have + * no current lseg (e.g. between LAYOUTRETURN and the next LAYOUTGET) can + * still honor the no-MDS-fallback policy. + */ +static inline bool +ff_layout_hdr_no_fallback_to_mds(struct pnfs_layout_hdr *lo) +{ + return test_bit(NFS4_FF_HDR_NO_IO_THRU_MDS, + &FF_LAYOUT_FROM_HDR(lo)->flags); +} + +static inline bool +ff_layout_no_read_on_rw(struct pnfs_layout_segment *lseg) +{ + return FF_LAYOUT_LSEG(lseg)->flags & FF_FLAGS_NO_READ_IO; +} + +static inline int +nfs4_ff_layout_ds_version(const struct nfs4_ff_layout_mirror *mirror, u32 dss_id) +{ + return mirror->dss[dss_id].mirror_ds->ds_versions[0].version; +} + +static inline u32 +nfs4_ff_layout_calc_dss_id(const u64 stripe_unit, const u32 dss_count, const loff_t offset) +{ + u64 tmp = offset; + + if (dss_count == 1 || stripe_unit == 0) + return 0; + + do_div(tmp, stripe_unit); + + return do_div(tmp, dss_count); +} + +struct nfs4_ff_layout_ds * +nfs4_ff_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, + gfp_t gfp_flags); +void nfs4_ff_layout_put_deviceid(struct nfs4_ff_layout_ds *mirror_ds); +void nfs4_ff_layout_free_deviceid(struct nfs4_ff_layout_ds *mirror_ds); +int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, u64 offset, u64 length, int status, + enum nfs_opnum4 opnum, gfp_t gfp_flags); +void ff_layout_send_layouterror(struct pnfs_layout_segment *lseg); +int ff_layout_encode_ds_ioerr(struct xdr_stream *xdr, const struct list_head *head); +void ff_layout_free_ds_ioerr(struct list_head *head); +unsigned int ff_layout_fetch_ds_ioerr(struct pnfs_layout_hdr *lo, + const struct pnfs_layout_range *range, + struct list_head *head, + unsigned int maxnum); +struct nfs_fh * +nfs4_ff_layout_select_ds_fh(struct nfs4_ff_layout_mirror *mirror, u32 dss_id); +void +nfs4_ff_layout_select_ds_stateid(const struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + nfs4_stateid *stateid); + +struct nfs4_pnfs_ds * +nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + bool fail_return); + +struct rpc_clnt * +nfs4_ff_find_or_create_ds_client(struct nfs4_ff_layout_mirror *mirror, + struct nfs_client *ds_clp, + struct inode *inode, + u32 dss_id); +const struct cred *ff_layout_get_ds_cred(struct nfs4_ff_layout_mirror *mirror, + const struct pnfs_layout_range *range, + const struct cred *mdscred, + u32 dss_id); +bool ff_layout_avoid_mds_available_ds(struct pnfs_layout_segment *lseg); +bool ff_layout_avoid_read_on_rw(struct pnfs_layout_segment *lseg); + +#endif /* FS_NFS_NFS4FLEXFILELAYOUT_H */ diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c new file mode 100644 index 000000000..1109462a9 --- /dev/null +++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c @@ -0,0 +1,644 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Device operations for the pnfs nfs4 file layout driver. + * + * Copyright (c) 2014, Primary Data, Inc. All rights reserved. + * + * Tao Peng + */ + +#include +#include +#include +#include + +#include "../internal.h" +#include "../nfs4session.h" +#include "flexfilelayout.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +static unsigned int dataserver_timeo = NFS_DEF_TCP_TIMEO; +static unsigned int dataserver_retrans; + +static bool ff_layout_has_available_ds(struct pnfs_layout_segment *lseg); + +void nfs4_ff_layout_put_deviceid(struct nfs4_ff_layout_ds *mirror_ds) +{ + if (!IS_ERR_OR_NULL(mirror_ds)) + nfs4_put_deviceid_node(&mirror_ds->id_node); +} + +void nfs4_ff_layout_free_deviceid(struct nfs4_ff_layout_ds *mirror_ds) +{ + nfs4_print_deviceid(&mirror_ds->id_node.deviceid); + nfs4_pnfs_ds_put(mirror_ds->ds); + kfree(mirror_ds->ds_versions); + kfree_rcu(mirror_ds, id_node.rcu); +} + +/* Decode opaque device data and construct new_ds using it */ +struct nfs4_ff_layout_ds * +nfs4_ff_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, + gfp_t gfp_flags) +{ + struct xdr_stream stream; + struct xdr_buf buf; + struct folio *scratch; + struct list_head dsaddrs; + struct nfs4_pnfs_ds_addr *da; + struct nfs4_ff_layout_ds *new_ds = NULL; + struct nfs4_ff_ds_version *ds_versions = NULL; + struct net *net = server->nfs_client->cl_net; + u32 mp_count; + u32 version_count; + __be32 *p; + int i, ret = -ENOMEM; + + /* set up xdr stream */ + scratch = folio_alloc(gfp_flags, 0); + if (!scratch) + goto out_err; + + new_ds = kzalloc_obj(struct nfs4_ff_layout_ds, gfp_flags); + if (!new_ds) + goto out_scratch; + + nfs4_init_deviceid_node(&new_ds->id_node, + server, + &pdev->dev_id); + INIT_LIST_HEAD(&dsaddrs); + + xdr_init_decode_pages(&stream, &buf, pdev->pages, pdev->pglen); + xdr_set_scratch_folio(&stream, scratch); + + /* multipath count */ + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err_drain_dsaddrs; + mp_count = be32_to_cpup(p); + dprintk("%s: multipath ds count %d\n", __func__, mp_count); + + for (i = 0; i < mp_count; i++) { + /* multipath ds */ + da = nfs4_decode_mp_ds_addr(net, &stream, gfp_flags); + if (da) + list_add_tail(&da->da_node, &dsaddrs); + } + if (list_empty(&dsaddrs)) { + dprintk("%s: no suitable DS addresses found\n", + __func__); + ret = -ENOMEDIUM; + goto out_err_drain_dsaddrs; + } + + /* version count */ + p = xdr_inline_decode(&stream, 4); + if (unlikely(!p)) + goto out_err_drain_dsaddrs; + version_count = be32_to_cpup(p); + + if (version_count == 0) { + ret = -EINVAL; + goto out_err_drain_dsaddrs; + } + dprintk("%s: version count %d\n", __func__, version_count); + + ds_versions = kzalloc_objs(struct nfs4_ff_ds_version, version_count, + gfp_flags); + if (!ds_versions) + goto out_err_drain_dsaddrs; + + for (i = 0; i < version_count; i++) { + /* 20 = version(4) + minor_version(4) + rsize(4) + wsize(4) + + * tightly_coupled(4) */ + p = xdr_inline_decode(&stream, 20); + if (unlikely(!p)) + goto out_err_drain_dsaddrs; + ds_versions[i].version = be32_to_cpup(p++); + ds_versions[i].minor_version = be32_to_cpup(p++); + ds_versions[i].rsize = nfs_io_size(be32_to_cpup(p++), + server->nfs_client->cl_proto); + ds_versions[i].wsize = nfs_io_size(be32_to_cpup(p++), + server->nfs_client->cl_proto); + ds_versions[i].tightly_coupled = be32_to_cpup(p); + + if (ds_versions[i].rsize > NFS_MAX_FILE_IO_SIZE) + ds_versions[i].rsize = NFS_MAX_FILE_IO_SIZE; + if (ds_versions[i].wsize > NFS_MAX_FILE_IO_SIZE) + ds_versions[i].wsize = NFS_MAX_FILE_IO_SIZE; + + /* + * check for valid major/minor combination. + * currently we support dataserver which talk: + * v3, v4.0, v4.1, v4.2 + */ + if (!((ds_versions[i].version == 3 && ds_versions[i].minor_version == 0) || + (ds_versions[i].version == 4 && ds_versions[i].minor_version < 3))) { + dprintk("%s: [%d] unsupported ds version %d-%d\n", __func__, + i, ds_versions[i].version, + ds_versions[i].minor_version); + ret = -EPROTONOSUPPORT; + goto out_err_drain_dsaddrs; + } + + dprintk("%s: [%d] vers %u minor_ver %u rsize %u wsize %u coupled %d\n", + __func__, i, ds_versions[i].version, + ds_versions[i].minor_version, + ds_versions[i].rsize, + ds_versions[i].wsize, + ds_versions[i].tightly_coupled); + } + + new_ds->ds_versions = ds_versions; + new_ds->ds_versions_cnt = version_count; + + new_ds->ds = nfs4_pnfs_ds_add(net, &dsaddrs, gfp_flags); + if (!new_ds->ds) + goto out_err_drain_dsaddrs; + + /* If DS was already in cache, free ds addrs */ + while (!list_empty(&dsaddrs)) { + da = list_first_entry(&dsaddrs, + struct nfs4_pnfs_ds_addr, + da_node); + list_del_init(&da->da_node); + kfree(da->da_remotestr); + kfree(da); + } + + folio_put(scratch); + return new_ds; + +out_err_drain_dsaddrs: + while (!list_empty(&dsaddrs)) { + da = list_first_entry(&dsaddrs, struct nfs4_pnfs_ds_addr, + da_node); + list_del_init(&da->da_node); + kfree(da->da_remotestr); + kfree(da); + } + + kfree(ds_versions); +out_scratch: + folio_put(scratch); +out_err: + kfree(new_ds); + + dprintk("%s ERROR: returning %d\n", __func__, ret); + return NULL; +} + +static void extend_ds_error(struct nfs4_ff_layout_ds_err *err, + u64 offset, u64 length) +{ + u64 end; + + end = max_t(u64, pnfs_end_offset(err->offset, err->length), + pnfs_end_offset(offset, length)); + err->offset = min_t(u64, err->offset, offset); + err->length = end - err->offset; +} + +static int +ff_ds_error_match(const struct nfs4_ff_layout_ds_err *e1, + const struct nfs4_ff_layout_ds_err *e2) +{ + int ret; + + if (e1->opnum != e2->opnum) + return e1->opnum < e2->opnum ? -1 : 1; + if (e1->status != e2->status) + return e1->status < e2->status ? -1 : 1; + ret = memcmp(e1->stateid.data, e2->stateid.data, + sizeof(e1->stateid.data)); + if (ret != 0) + return ret; + ret = memcmp(&e1->deviceid, &e2->deviceid, sizeof(e1->deviceid)); + if (ret != 0) + return ret; + if (pnfs_end_offset(e1->offset, e1->length) < e2->offset) + return -1; + if (e1->offset > pnfs_end_offset(e2->offset, e2->length)) + return 1; + /* If ranges overlap or are contiguous, they are the same */ + return 0; +} + +static void +ff_layout_add_ds_error_locked(struct nfs4_flexfile_layout *flo, + struct nfs4_ff_layout_ds_err *dserr) +{ + struct nfs4_ff_layout_ds_err *err, *tmp; + struct list_head *head = &flo->error_list; + int match; + + /* Do insertion sort w/ merges */ + list_for_each_entry_safe(err, tmp, &flo->error_list, list) { + match = ff_ds_error_match(err, dserr); + if (match < 0) + continue; + if (match > 0) { + /* Add entry "dserr" _before_ entry "err" */ + head = &err->list; + break; + } + /* Entries match, so merge "err" into "dserr" */ + extend_ds_error(dserr, err->offset, err->length); + list_replace(&err->list, &dserr->list); + kfree(err); + return; + } + + list_add_tail(&dserr->list, head); +} + +int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, u64 offset, u64 length, int status, + enum nfs_opnum4 opnum, gfp_t gfp_flags) +{ + struct nfs4_ff_layout_ds_err *dserr; + + if (status == 0) + return 0; + + if (IS_ERR_OR_NULL(mirror->dss[dss_id].mirror_ds)) + return -EINVAL; + + dserr = kmalloc_obj(*dserr, gfp_flags); + if (!dserr) + return -ENOMEM; + + INIT_LIST_HEAD(&dserr->list); + dserr->offset = offset; + dserr->length = length; + dserr->status = status; + dserr->opnum = opnum; + nfs4_stateid_copy(&dserr->stateid, &mirror->dss[dss_id].stateid); + memcpy(&dserr->deviceid, &mirror->dss[dss_id].mirror_ds->id_node.deviceid, + NFS4_DEVICEID4_SIZE); + + spin_lock(&flo->generic_hdr.plh_inode->i_lock); + ff_layout_add_ds_error_locked(flo, dserr); + spin_unlock(&flo->generic_hdr.plh_inode->i_lock); + return 0; +} + +static const struct cred * +ff_layout_get_mirror_cred(struct nfs4_ff_layout_mirror *mirror, u32 iomode, u32 dss_id) +{ + const struct cred *cred, __rcu **pcred; + + if (iomode == IOMODE_READ) + pcred = &mirror->dss[dss_id].ro_cred; + else + pcred = &mirror->dss[dss_id].rw_cred; + + rcu_read_lock(); + do { + cred = rcu_dereference(*pcred); + if (!cred) + break; + + cred = get_cred_rcu(cred); + } while(!cred); + rcu_read_unlock(); + return cred; +} + +struct nfs_fh * +nfs4_ff_layout_select_ds_fh(struct nfs4_ff_layout_mirror *mirror, u32 dss_id) +{ + /* FIXME: For now assume there is only 1 version available for the DS */ + return &mirror->dss[dss_id].fh_versions[0]; +} + +void +nfs4_ff_layout_select_ds_stateid(const struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + nfs4_stateid *stateid) +{ + if (nfs4_ff_layout_ds_version(mirror, dss_id) == 4) + nfs4_stateid_copy(stateid, &mirror->dss[dss_id].stateid); +} + +static bool +ff_layout_init_mirror_ds(struct pnfs_layout_hdr *lo, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id) +{ + if (mirror == NULL) + goto outerr; + if (mirror->dss[dss_id].mirror_ds == NULL) { + struct nfs4_deviceid_node *node; + struct nfs4_ff_layout_ds *mirror_ds = ERR_PTR(-ENODEV); + + node = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode), + &mirror->dss[dss_id].devid, lo->plh_lc_cred, + GFP_KERNEL); + if (node) + mirror_ds = FF_LAYOUT_MIRROR_DS(node); + + /* check for race with another call to this function */ + if (cmpxchg(&mirror->dss[dss_id].mirror_ds, NULL, mirror_ds) && + mirror_ds != ERR_PTR(-ENODEV)) + nfs4_put_deviceid_node(node); + } + + if (IS_ERR(mirror->dss[dss_id].mirror_ds)) + goto outerr; + + return true; +outerr: + return false; +} + +/** + * nfs4_ff_layout_prepare_ds - prepare a DS connection for an RPC call + * @lseg: the layout segment we're operating on + * @mirror: layout mirror describing the DS to use + * @dss_id: DS stripe id to select stripe to use + * @fail_return: return layout on connect failure? + * + * Try to prepare a DS connection to accept an RPC call. This involves + * selecting a mirror to use and connecting the client to it if it's not + * already connected. + * + * Since we only need a single functioning mirror to satisfy a read, we don't + * want to return the layout if there is one. For writes though, any down + * mirror should result in a LAYOUTRETURN. @fail_return is how we distinguish + * between the two cases. + * + * Returns a pointer to a connected DS object on success or NULL on failure. + */ +struct nfs4_pnfs_ds * +nfs4_ff_layout_prepare_ds(struct pnfs_layout_segment *lseg, + struct nfs4_ff_layout_mirror *mirror, + u32 dss_id, + bool fail_return) +{ + struct nfs4_pnfs_ds *ds; + struct inode *ino = lseg->pls_layout->plh_inode; + struct nfs_server *s = NFS_SERVER(ino); + unsigned int max_payload; + int status = -EAGAIN; + + if (!ff_layout_init_mirror_ds(lseg->pls_layout, mirror, dss_id)) + goto noconnect; + + ds = mirror->dss[dss_id].mirror_ds->ds; + if (READ_ONCE(ds->ds_clp)) + goto out; + /* matching smp_wmb() in _nfs4_pnfs_v3/4_ds_connect */ + smp_rmb(); + + /* FIXME: For now we assume the server sent only one version of NFS + * to use for the DS. + */ + status = nfs4_pnfs_ds_connect(s, ds, &mirror->dss[dss_id].mirror_ds->id_node, + dataserver_timeo, dataserver_retrans, + mirror->dss[dss_id].mirror_ds->ds_versions[0].version, + mirror->dss[dss_id].mirror_ds->ds_versions[0].minor_version); + + /* connect success, check rsize/wsize limit */ + if (!status) { + /* + * ds_clp is put in destroy_ds(). + * keep ds_clp even if DS is local, so that if local IO cannot + * proceed somehow, we can fall back to NFS whenever we want. + */ + nfs_local_probe_async(ds->ds_clp); + max_payload = + nfs_block_size(rpc_max_payload(ds->ds_clp->cl_rpcclient), + NULL); + if (mirror->dss[dss_id].mirror_ds->ds_versions[0].rsize > max_payload) + mirror->dss[dss_id].mirror_ds->ds_versions[0].rsize = max_payload; + if (mirror->dss[dss_id].mirror_ds->ds_versions[0].wsize > max_payload) + mirror->dss[dss_id].mirror_ds->ds_versions[0].wsize = max_payload; + goto out; + } +noconnect: + ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout), + mirror, dss_id, lseg->pls_range.offset, + lseg->pls_range.length, NFS4ERR_NXIO, + OP_ILLEGAL, GFP_NOIO); + ff_layout_send_layouterror(lseg); + if (fail_return || !ff_layout_has_available_ds(lseg)) + pnfs_error_mark_layout_for_return(ino, lseg); + ds = ERR_PTR(status); +out: + return ds; +} + +const struct cred * +ff_layout_get_ds_cred(struct nfs4_ff_layout_mirror *mirror, + const struct pnfs_layout_range *range, + const struct cred *mdscred, + u32 dss_id) +{ + const struct cred *cred; + + if (mirror && !mirror->dss[dss_id].mirror_ds->ds_versions[0].tightly_coupled) { + cred = ff_layout_get_mirror_cred(mirror, range->iomode, dss_id); + if (!cred) + cred = get_cred(mdscred); + } else { + cred = get_cred(mdscred); + } + return cred; +} + +/** + * nfs4_ff_find_or_create_ds_client - Find or create a DS rpc client + * @mirror: pointer to the mirror + * @ds_clp: nfs_client for the DS + * @inode: pointer to inode + * @dss_id: DS stripe id + * + * Find or create a DS rpc client with th MDS server rpc client auth flavor + * in the nfs_client cl_ds_clients list. + */ +struct rpc_clnt * +nfs4_ff_find_or_create_ds_client(struct nfs4_ff_layout_mirror *mirror, + struct nfs_client *ds_clp, struct inode *inode, + u32 dss_id) +{ + switch (mirror->dss[dss_id].mirror_ds->ds_versions[0].version) { + case 3: + /* For NFSv3 DS, flavor is set when creating DS connections */ + return ds_clp->cl_rpcclient; + case 4: + return nfs4_find_or_create_ds_client(ds_clp, inode); + default: + BUG(); + } +} + +void ff_layout_free_ds_ioerr(struct list_head *head) +{ + struct nfs4_ff_layout_ds_err *err; + + while (!list_empty(head)) { + err = list_first_entry(head, + struct nfs4_ff_layout_ds_err, + list); + list_del(&err->list); + kfree(err); + } +} + +/* called with inode i_lock held */ +int ff_layout_encode_ds_ioerr(struct xdr_stream *xdr, const struct list_head *head) +{ + struct nfs4_ff_layout_ds_err *err; + __be32 *p; + + list_for_each_entry(err, head, list) { + /* offset(8) + length(8) + stateid(NFS4_STATEID_SIZE) + * + array length + deviceid(NFS4_DEVICEID4_SIZE) + * + status(4) + opnum(4) + */ + p = xdr_reserve_space(xdr, + 28 + NFS4_STATEID_SIZE + NFS4_DEVICEID4_SIZE); + if (unlikely(!p)) + return -ENOBUFS; + p = xdr_encode_hyper(p, err->offset); + p = xdr_encode_hyper(p, err->length); + p = xdr_encode_opaque_fixed(p, &err->stateid, + NFS4_STATEID_SIZE); + /* Encode 1 error */ + *p++ = cpu_to_be32(1); + p = xdr_encode_opaque_fixed(p, &err->deviceid, + NFS4_DEVICEID4_SIZE); + *p++ = cpu_to_be32(err->status); + *p++ = cpu_to_be32(err->opnum); + dprintk("%s: offset %llu length %llu status %d op %d\n", + __func__, err->offset, err->length, err->status, + err->opnum); + } + + return 0; +} + +static +unsigned int do_layout_fetch_ds_ioerr(struct pnfs_layout_hdr *lo, + const struct pnfs_layout_range *range, + struct list_head *head, + unsigned int maxnum) +{ + struct nfs4_flexfile_layout *flo = FF_LAYOUT_FROM_HDR(lo); + struct inode *inode = lo->plh_inode; + struct nfs4_ff_layout_ds_err *err, *n; + unsigned int ret = 0; + + spin_lock(&inode->i_lock); + list_for_each_entry_safe(err, n, &flo->error_list, list) { + if (!pnfs_is_range_intersecting(err->offset, + pnfs_end_offset(err->offset, err->length), + range->offset, + pnfs_end_offset(range->offset, range->length))) + continue; + if (!maxnum) + break; + list_move(&err->list, head); + maxnum--; + ret++; + } + spin_unlock(&inode->i_lock); + return ret; +} + +unsigned int ff_layout_fetch_ds_ioerr(struct pnfs_layout_hdr *lo, + const struct pnfs_layout_range *range, + struct list_head *head, + unsigned int maxnum) +{ + unsigned int ret; + + ret = do_layout_fetch_ds_ioerr(lo, range, head, maxnum); + /* If we're over the max, discard all remaining entries */ + if (ret == maxnum) { + LIST_HEAD(discard); + do_layout_fetch_ds_ioerr(lo, range, &discard, -1); + ff_layout_free_ds_ioerr(&discard); + } + return ret; +} + +static bool ff_read_layout_has_available_ds(struct pnfs_layout_segment *lseg) +{ + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_deviceid_node *devid; + u32 idx, dss_id; + + for (idx = 0; idx < FF_LAYOUT_MIRROR_COUNT(lseg); idx++) { + mirror = FF_LAYOUT_COMP(lseg, idx); + if (!mirror) + continue; + for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { + if (!mirror->dss[dss_id].mirror_ds) + return true; + if (IS_ERR(mirror->dss[dss_id].mirror_ds)) + continue; + devid = &mirror->dss[dss_id].mirror_ds->id_node; + if (!nfs4_test_deviceid_unavailable(devid)) + return true; + } + } + + return false; +} + +static bool ff_rw_layout_has_available_ds(struct pnfs_layout_segment *lseg) +{ + struct nfs4_ff_layout_mirror *mirror; + struct nfs4_deviceid_node *devid; + u32 idx, dss_id; + + for (idx = 0; idx < FF_LAYOUT_MIRROR_COUNT(lseg); idx++) { + mirror = FF_LAYOUT_COMP(lseg, idx); + if (!mirror) + return false; + for (dss_id = 0; dss_id < mirror->dss_count; dss_id++) { + if (IS_ERR(mirror->dss[dss_id].mirror_ds)) + return false; + if (!mirror->dss[dss_id].mirror_ds) + continue; + devid = &mirror->dss[dss_id].mirror_ds->id_node; + if (nfs4_test_deviceid_unavailable(devid)) + return false; + } + } + + return FF_LAYOUT_MIRROR_COUNT(lseg) != 0; +} + +static bool ff_layout_has_available_ds(struct pnfs_layout_segment *lseg) +{ + if (lseg->pls_range.iomode == IOMODE_READ) + return ff_read_layout_has_available_ds(lseg); + /* Note: RW layout needs all mirrors available */ + return ff_rw_layout_has_available_ds(lseg); +} + +bool ff_layout_avoid_mds_available_ds(struct pnfs_layout_segment *lseg) +{ + return ff_layout_no_fallback_to_mds(lseg) || + ff_layout_has_available_ds(lseg); +} + +bool ff_layout_avoid_read_on_rw(struct pnfs_layout_segment *lseg) +{ + return lseg->pls_range.iomode == IOMODE_RW && + ff_layout_no_read_on_rw(lseg); +} + +module_param(dataserver_retrans, uint, 0644); +MODULE_PARM_DESC(dataserver_retrans, "The number of times the NFSv4.1 client " + "retries a request before it attempts further " + " recovery action."); +module_param(dataserver_timeo, uint, 0644); +MODULE_PARM_DESC(dataserver_timeo, "The time (in tenths of a second) the " + "NFSv4.1 client waits for a response from a " + " data server before it retries an NFS request."); diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c new file mode 100644 index 000000000..1967de7d1 --- /dev/null +++ b/fs/nfs/fs_context.c @@ -0,0 +1,1793 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/fs_context.c + * + * Copyright (C) 1992 Rick Sladkey + * Conversion to new mount api Copyright (C) David Howells + * + * NFS mount handling. + * + * Split from fs/nfs/super.c by David Howells + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "nfs.h" +#include "internal.h" + +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_MOUNT + +#if IS_ENABLED(CONFIG_NFS_V3) +#define NFS_DEFAULT_VERSION 3 +#else +#define NFS_DEFAULT_VERSION 2 +#endif + +#define NFS_MAX_CONNECTIONS 16 + +enum nfs_param { + Opt_ac, + Opt_acdirmax, + Opt_acdirmin, + Opt_acl, + Opt_acregmax, + Opt_acregmin, + Opt_actimeo, + Opt_addr, + Opt_bg, + Opt_bsize, + Opt_clientaddr, + Opt_cto, + Opt_alignwrite, + Opt_fatal_neterrors, + Opt_fg, + Opt_fscache, + Opt_fscache_flag, + Opt_hard, + Opt_intr, + Opt_local_lock, + Opt_lock, + Opt_lookupcache, + Opt_migration, + Opt_minorversion, + Opt_mountaddr, + Opt_mounthost, + Opt_mountport, + Opt_mountproto, + Opt_mountvers, + Opt_namelen, + Opt_nconnect, + Opt_max_connect, + Opt_port, + Opt_posix, + Opt_proto, + Opt_rdirplus, + Opt_rdirplus_none, + Opt_rdirplus_force, + Opt_rdma, + Opt_resvport, + Opt_retrans, + Opt_retry, + Opt_rsize, + Opt_sec, + Opt_sharecache, + Opt_sloppy, + Opt_soft, + Opt_softerr, + Opt_softreval, + Opt_source, + Opt_tcp, + Opt_timeo, + Opt_trunkdiscovery, + Opt_udp, + Opt_v, + Opt_vers, + Opt_wsize, + Opt_write, + Opt_xprtsec, + Opt_cert_serial, + Opt_privkey_serial, +}; + +enum { + Opt_fatal_neterrors_default, + Opt_fatal_neterrors_enetunreach, + Opt_fatal_neterrors_none, +}; + +static const struct constant_table nfs_param_enums_fatal_neterrors[] = { + { "default", Opt_fatal_neterrors_default }, + { "ENETDOWN:ENETUNREACH", Opt_fatal_neterrors_enetunreach }, + { "ENETUNREACH:ENETDOWN", Opt_fatal_neterrors_enetunreach }, + { "none", Opt_fatal_neterrors_none }, + {} +}; + +enum { + Opt_local_lock_all, + Opt_local_lock_flock, + Opt_local_lock_none, + Opt_local_lock_posix, +}; + +static const struct constant_table nfs_param_enums_local_lock[] = { + { "all", Opt_local_lock_all }, + { "flock", Opt_local_lock_flock }, + { "posix", Opt_local_lock_posix }, + { "none", Opt_local_lock_none }, + {} +}; + +enum { + Opt_lookupcache_all, + Opt_lookupcache_none, + Opt_lookupcache_positive, +}; + +static const struct constant_table nfs_param_enums_lookupcache[] = { + { "all", Opt_lookupcache_all }, + { "none", Opt_lookupcache_none }, + { "pos", Opt_lookupcache_positive }, + { "positive", Opt_lookupcache_positive }, + {} +}; + +enum { + Opt_write_lazy, + Opt_write_eager, + Opt_write_wait, +}; + +static const struct constant_table nfs_param_enums_write[] = { + { "lazy", Opt_write_lazy }, + { "eager", Opt_write_eager }, + { "wait", Opt_write_wait }, + {} +}; + +static const struct fs_parameter_spec nfs_fs_parameters[] = { + fsparam_flag_no("ac", Opt_ac), + fsparam_u32 ("acdirmax", Opt_acdirmax), + fsparam_u32 ("acdirmin", Opt_acdirmin), + fsparam_flag_no("acl", Opt_acl), + fsparam_u32 ("acregmax", Opt_acregmax), + fsparam_u32 ("acregmin", Opt_acregmin), + fsparam_u32 ("actimeo", Opt_actimeo), + fsparam_string("addr", Opt_addr), + fsparam_flag ("bg", Opt_bg), + fsparam_u32 ("bsize", Opt_bsize), + fsparam_string("clientaddr", Opt_clientaddr), + fsparam_flag_no("cto", Opt_cto), + fsparam_flag_no("alignwrite", Opt_alignwrite), + fsparam_enum("fatal_neterrors", Opt_fatal_neterrors, + nfs_param_enums_fatal_neterrors), + fsparam_flag ("fg", Opt_fg), + fsparam_flag_no("fsc", Opt_fscache_flag), + fsparam_string("fsc", Opt_fscache), + fsparam_flag ("hard", Opt_hard), + __fsparam(NULL, "intr", Opt_intr, + fs_param_neg_with_no|fs_param_deprecated, NULL), + fsparam_enum ("local_lock", Opt_local_lock, nfs_param_enums_local_lock), + fsparam_flag_no("lock", Opt_lock), + fsparam_enum ("lookupcache", Opt_lookupcache, nfs_param_enums_lookupcache), + fsparam_flag_no("migration", Opt_migration), + fsparam_u32 ("minorversion", Opt_minorversion), + fsparam_string("mountaddr", Opt_mountaddr), + fsparam_string("mounthost", Opt_mounthost), + fsparam_u32 ("mountport", Opt_mountport), + fsparam_string("mountproto", Opt_mountproto), + fsparam_u32 ("mountvers", Opt_mountvers), + fsparam_u32 ("namlen", Opt_namelen), + fsparam_u32 ("nconnect", Opt_nconnect), + fsparam_u32 ("max_connect", Opt_max_connect), + fsparam_string("nfsvers", Opt_vers), + fsparam_u32 ("port", Opt_port), + fsparam_flag_no("posix", Opt_posix), + fsparam_string("proto", Opt_proto), + fsparam_flag_no("rdirplus", Opt_rdirplus), // rdirplus|nordirplus + fsparam_string("rdirplus", Opt_rdirplus), // rdirplus=... + fsparam_flag ("rdma", Opt_rdma), + fsparam_flag_no("resvport", Opt_resvport), + fsparam_u32 ("retrans", Opt_retrans), + fsparam_string("retry", Opt_retry), + fsparam_u32 ("rsize", Opt_rsize), + fsparam_string("sec", Opt_sec), + fsparam_flag_no("sharecache", Opt_sharecache), + fsparam_flag ("sloppy", Opt_sloppy), + fsparam_flag ("soft", Opt_soft), + fsparam_flag ("softerr", Opt_softerr), + fsparam_flag ("softreval", Opt_softreval), + fsparam_string("source", Opt_source), + fsparam_flag ("tcp", Opt_tcp), + fsparam_u32 ("timeo", Opt_timeo), + fsparam_flag_no("trunkdiscovery", Opt_trunkdiscovery), + fsparam_flag ("udp", Opt_udp), + fsparam_flag ("v2", Opt_v), + fsparam_flag ("v3", Opt_v), + fsparam_flag ("v4", Opt_v), + fsparam_flag ("v4.0", Opt_v), + fsparam_flag ("v4.1", Opt_v), + fsparam_flag ("v4.2", Opt_v), + fsparam_string("vers", Opt_vers), + fsparam_enum ("write", Opt_write, nfs_param_enums_write), + fsparam_u32 ("wsize", Opt_wsize), + fsparam_string("xprtsec", Opt_xprtsec), + fsparam_s32("cert_serial", Opt_cert_serial), + fsparam_s32("privkey_serial", Opt_privkey_serial), + {} +}; + +enum { + Opt_vers_2, + Opt_vers_3, + Opt_vers_4, + Opt_vers_4_0, + Opt_vers_4_1, + Opt_vers_4_2, +}; + +static const struct constant_table nfs_vers_tokens[] = { + { "2", Opt_vers_2 }, + { "3", Opt_vers_3 }, + { "4", Opt_vers_4 }, + { "4.0", Opt_vers_4_0 }, + { "4.1", Opt_vers_4_1 }, + { "4.2", Opt_vers_4_2 }, + {} +}; + +enum { + Opt_xprt_rdma, + Opt_xprt_rdma6, + Opt_xprt_tcp, + Opt_xprt_tcp6, + Opt_xprt_udp, + Opt_xprt_udp6, + nr__Opt_xprt +}; + +static const struct constant_table nfs_xprt_protocol_tokens[] = { + { "rdma", Opt_xprt_rdma }, + { "rdma6", Opt_xprt_rdma6 }, + { "tcp", Opt_xprt_tcp }, + { "tcp6", Opt_xprt_tcp6 }, + { "udp", Opt_xprt_udp }, + { "udp6", Opt_xprt_udp6 }, + {} +}; + +enum { + Opt_sec_krb5, + Opt_sec_krb5i, + Opt_sec_krb5p, + Opt_sec_lkey, + Opt_sec_lkeyi, + Opt_sec_lkeyp, + Opt_sec_none, + Opt_sec_spkm, + Opt_sec_spkmi, + Opt_sec_spkmp, + Opt_sec_sys, + nr__Opt_sec +}; + +static const struct constant_table nfs_secflavor_tokens[] = { + { "krb5", Opt_sec_krb5 }, + { "krb5i", Opt_sec_krb5i }, + { "krb5p", Opt_sec_krb5p }, + { "lkey", Opt_sec_lkey }, + { "lkeyi", Opt_sec_lkeyi }, + { "lkeyp", Opt_sec_lkeyp }, + { "none", Opt_sec_none }, + { "null", Opt_sec_none }, + { "spkm3", Opt_sec_spkm }, + { "spkm3i", Opt_sec_spkmi }, + { "spkm3p", Opt_sec_spkmp }, + { "sys", Opt_sec_sys }, + {} +}; + +enum { + Opt_xprtsec_none, + Opt_xprtsec_tls, + Opt_xprtsec_mtls, + nr__Opt_xprtsec +}; + +static const struct constant_table nfs_xprtsec_policies[] = { + { "none", Opt_xprtsec_none }, + { "tls", Opt_xprtsec_tls }, + { "mtls", Opt_xprtsec_mtls }, + {} +}; + +static const struct constant_table nfs_rdirplus_tokens[] = { + { "none", Opt_rdirplus_none }, + { "force", Opt_rdirplus_force }, + {} +}; + +/* + * Sanity-check a server address provided by the mount command. + * + * Address family must be initialized, and address must not be + * the ANY address for that family. + */ +static int nfs_verify_server_address(struct sockaddr_storage *addr) +{ + switch (addr->ss_family) { + case AF_INET: { + struct sockaddr_in *sa = (struct sockaddr_in *)addr; + return sa->sin_addr.s_addr != htonl(INADDR_ANY); + } + case AF_INET6: { + struct in6_addr *sa = &((struct sockaddr_in6 *)addr)->sin6_addr; + return !ipv6_addr_any(sa); + } + } + + return 0; +} + +#ifdef CONFIG_NFS_DISABLE_UDP_SUPPORT +static bool nfs_server_transport_udp_invalid(const struct nfs_fs_context *ctx) +{ + return true; +} +#else +static bool nfs_server_transport_udp_invalid(const struct nfs_fs_context *ctx) +{ + if (ctx->version == 4) + return true; + return false; +} +#endif + +/* + * Sanity check the NFS transport protocol. + */ +static int nfs_validate_transport_protocol(struct fs_context *fc, + struct nfs_fs_context *ctx) +{ + switch (ctx->nfs_server.protocol) { + case XPRT_TRANSPORT_UDP: + if (nfs_server_transport_udp_invalid(ctx)) + goto out_invalid_transport_udp; + break; + case XPRT_TRANSPORT_TCP: + case XPRT_TRANSPORT_RDMA: + break; + default: + ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP; + } + + if (ctx->xprtsec.policy != RPC_XPRTSEC_NONE) + switch (ctx->nfs_server.protocol) { + case XPRT_TRANSPORT_TCP: + ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP_TLS; + break; + default: + goto out_invalid_xprtsec_policy; + } + + return 0; +out_invalid_transport_udp: + return nfs_invalf(fc, "NFS: Unsupported transport protocol udp"); +out_invalid_xprtsec_policy: + return nfs_invalf(fc, "NFS: Transport does not support xprtsec"); +} + +/* + * For text based NFSv2/v3 mounts, the mount protocol transport default + * settings should depend upon the specified NFS transport. + */ +static void nfs_set_mount_transport_protocol(struct nfs_fs_context *ctx) +{ + if (ctx->mount_server.protocol == XPRT_TRANSPORT_UDP || + ctx->mount_server.protocol == XPRT_TRANSPORT_TCP) + return; + switch (ctx->nfs_server.protocol) { + case XPRT_TRANSPORT_UDP: + ctx->mount_server.protocol = XPRT_TRANSPORT_UDP; + break; + case XPRT_TRANSPORT_TCP: + case XPRT_TRANSPORT_RDMA: + ctx->mount_server.protocol = XPRT_TRANSPORT_TCP; + } +} + +/* + * Add 'flavor' to 'auth_info' if not already present. + * Returns true if 'flavor' ends up in the list, false otherwise + */ +static int nfs_auth_info_add(struct fs_context *fc, + struct nfs_auth_info *auth_info, + rpc_authflavor_t flavor) +{ + unsigned int i; + unsigned int max_flavor_len = ARRAY_SIZE(auth_info->flavors); + + /* make sure this flavor isn't already in the list */ + for (i = 0; i < auth_info->flavor_len; i++) { + if (flavor == auth_info->flavors[i]) + return 0; + } + + if (auth_info->flavor_len + 1 >= max_flavor_len) + return nfs_invalf(fc, "NFS: too many sec= flavors"); + + auth_info->flavors[auth_info->flavor_len++] = flavor; + return 0; +} + +/* + * Parse the value of the 'sec=' option. + */ +static int nfs_parse_security_flavors(struct fs_context *fc, + struct fs_parameter *param) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + rpc_authflavor_t pseudoflavor; + char *string = param->string, *p; + int ret; + + trace_nfs_mount_assign(param->key, string); + + while ((p = strsep(&string, ":")) != NULL) { + if (!*p) + continue; + switch (lookup_constant(nfs_secflavor_tokens, p, -1)) { + case Opt_sec_none: + pseudoflavor = RPC_AUTH_NULL; + break; + case Opt_sec_sys: + pseudoflavor = RPC_AUTH_UNIX; + break; + case Opt_sec_krb5: + pseudoflavor = RPC_AUTH_GSS_KRB5; + break; + case Opt_sec_krb5i: + pseudoflavor = RPC_AUTH_GSS_KRB5I; + break; + case Opt_sec_krb5p: + pseudoflavor = RPC_AUTH_GSS_KRB5P; + break; + case Opt_sec_lkey: + pseudoflavor = RPC_AUTH_GSS_LKEY; + break; + case Opt_sec_lkeyi: + pseudoflavor = RPC_AUTH_GSS_LKEYI; + break; + case Opt_sec_lkeyp: + pseudoflavor = RPC_AUTH_GSS_LKEYP; + break; + case Opt_sec_spkm: + pseudoflavor = RPC_AUTH_GSS_SPKM; + break; + case Opt_sec_spkmi: + pseudoflavor = RPC_AUTH_GSS_SPKMI; + break; + case Opt_sec_spkmp: + pseudoflavor = RPC_AUTH_GSS_SPKMP; + break; + default: + return nfs_invalf(fc, "NFS: sec=%s option not recognized", p); + } + + ret = nfs_auth_info_add(fc, &ctx->auth_info, pseudoflavor); + if (ret < 0) + return ret; + } + + return 0; +} + +static int nfs_parse_xprtsec_policy(struct fs_context *fc, + struct fs_parameter *param) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + + trace_nfs_mount_assign(param->key, param->string); + + switch (lookup_constant(nfs_xprtsec_policies, param->string, -1)) { + case Opt_xprtsec_none: + ctx->xprtsec.policy = RPC_XPRTSEC_NONE; + break; + case Opt_xprtsec_tls: + ctx->xprtsec.policy = RPC_XPRTSEC_TLS_ANON; + break; + case Opt_xprtsec_mtls: + ctx->xprtsec.policy = RPC_XPRTSEC_TLS_X509; + break; + default: + return nfs_invalf(fc, "NFS: Unrecognized transport security policy"); + } + return 0; +} + +static int nfs_parse_version_string(struct fs_context *fc, + const char *string) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + + ctx->flags &= ~NFS_MOUNT_VER3; + switch (lookup_constant(nfs_vers_tokens, string, -1)) { + case Opt_vers_2: + ctx->version = 2; + break; + case Opt_vers_3: + ctx->flags |= NFS_MOUNT_VER3; + ctx->version = 3; + break; + case Opt_vers_4: + /* Backward compatibility option. In future, + * the mount program should always supply + * a NFSv4 minor version number. + */ + ctx->version = 4; + break; + case Opt_vers_4_0: + ctx->version = 4; + ctx->minorversion = 0; + break; + case Opt_vers_4_1: + ctx->version = 4; + ctx->minorversion = 1; + break; + case Opt_vers_4_2: + ctx->version = 4; + ctx->minorversion = 2; + break; + default: + return nfs_invalf(fc, "NFS: Unsupported NFS version"); + } + return 0; +} + +#ifdef CONFIG_KEYS +static int nfs_tls_key_verify(key_serial_t key_id) +{ + struct key *key = key_lookup(key_id); + int error = 0; + + if (IS_ERR(key)) { + pr_err("key id %08x not found\n", key_id); + return PTR_ERR(key); + } + if (test_bit(KEY_FLAG_REVOKED, &key->flags) || + test_bit(KEY_FLAG_INVALIDATED, &key->flags)) { + pr_err("key id %08x revoked\n", key_id); + error = -EKEYREVOKED; + } + + key_put(key); + return error; +} +#else +static inline int nfs_tls_key_verify(key_serial_t key_id) +{ + return -ENOENT; +} +#endif /* CONFIG_KEYS */ + +/* + * Parse a single mount parameter. + */ +static int nfs_fs_context_parse_param(struct fs_context *fc, + struct fs_parameter *param) +{ + struct fs_parse_result result; + struct nfs_fs_context *ctx = nfs_fc2context(fc); + unsigned short protofamily, mountfamily; + unsigned int len; + int ret, opt; + + trace_nfs_mount_option(param); + + opt = fs_parse(fc, nfs_fs_parameters, param, &result); + if (opt < 0) + return (opt == -ENOPARAM && ctx->sloppy) ? 1 : opt; + + if (fc->security) + ctx->has_sec_mnt_opts = 1; + + switch (opt) { + case Opt_source: + if (fc->source) + return nfs_invalf(fc, "NFS: Multiple sources not supported"); + fc->source = param->string; + param->string = NULL; + break; + + /* + * boolean options: foo/nofoo + */ + case Opt_soft: + ctx->flags |= NFS_MOUNT_SOFT; + ctx->flags &= ~NFS_MOUNT_SOFTERR; + break; + case Opt_softerr: + ctx->flags |= NFS_MOUNT_SOFTERR | NFS_MOUNT_SOFTREVAL; + ctx->flags &= ~NFS_MOUNT_SOFT; + break; + case Opt_hard: + ctx->flags &= ~(NFS_MOUNT_SOFT | + NFS_MOUNT_SOFTERR | + NFS_MOUNT_SOFTREVAL); + break; + case Opt_softreval: + if (result.negated) + ctx->flags &= ~NFS_MOUNT_SOFTREVAL; + else + ctx->flags |= NFS_MOUNT_SOFTREVAL; + break; + case Opt_posix: + if (result.negated) + ctx->flags &= ~NFS_MOUNT_POSIX; + else + ctx->flags |= NFS_MOUNT_POSIX; + break; + case Opt_cto: + if (result.negated) + ctx->flags |= NFS_MOUNT_NOCTO; + else + ctx->flags &= ~NFS_MOUNT_NOCTO; + break; + case Opt_trunkdiscovery: + if (result.negated) + ctx->flags &= ~NFS_MOUNT_TRUNK_DISCOVERY; + else + ctx->flags |= NFS_MOUNT_TRUNK_DISCOVERY; + break; + case Opt_alignwrite: + if (result.negated) + ctx->flags |= NFS_MOUNT_NO_ALIGNWRITE; + else + ctx->flags &= ~NFS_MOUNT_NO_ALIGNWRITE; + break; + case Opt_ac: + if (result.negated) + ctx->flags |= NFS_MOUNT_NOAC; + else + ctx->flags &= ~NFS_MOUNT_NOAC; + break; + case Opt_lock: + if (result.negated) { + ctx->lock_status = NFS_LOCK_NOLOCK; + ctx->flags |= NFS_MOUNT_NONLM; + ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } else { + ctx->lock_status = NFS_LOCK_LOCK; + ctx->flags &= ~NFS_MOUNT_NONLM; + ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } + break; + case Opt_udp: + ctx->flags &= ~NFS_MOUNT_TCP; + ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP; + break; + case Opt_tcp: + case Opt_rdma: + ctx->flags |= NFS_MOUNT_TCP; /* for side protocols */ + ret = xprt_find_transport_ident(param->key); + if (ret < 0) + goto out_bad_transport; + ctx->nfs_server.protocol = ret; + break; + case Opt_acl: + if (result.negated) + ctx->flags |= NFS_MOUNT_NOACL; + else + ctx->flags &= ~NFS_MOUNT_NOACL; + break; + case Opt_rdirplus: + if (result.negated) { + ctx->flags &= ~NFS_MOUNT_FORCE_RDIRPLUS; + ctx->flags |= NFS_MOUNT_NORDIRPLUS; + } else if (!param->string) { + ctx->flags &= ~(NFS_MOUNT_NORDIRPLUS | NFS_MOUNT_FORCE_RDIRPLUS); + } else { + switch (lookup_constant(nfs_rdirplus_tokens, param->string, -1)) { + case Opt_rdirplus_none: + ctx->flags &= ~NFS_MOUNT_FORCE_RDIRPLUS; + ctx->flags |= NFS_MOUNT_NORDIRPLUS; + break; + case Opt_rdirplus_force: + ctx->flags &= ~NFS_MOUNT_NORDIRPLUS; + ctx->flags |= NFS_MOUNT_FORCE_RDIRPLUS; + break; + default: + goto out_invalid_value; + } + } + break; + case Opt_sharecache: + if (result.negated) + ctx->flags |= NFS_MOUNT_UNSHARED; + else + ctx->flags &= ~NFS_MOUNT_UNSHARED; + break; + case Opt_resvport: + if (result.negated) + ctx->flags |= NFS_MOUNT_NORESVPORT; + else + ctx->flags &= ~NFS_MOUNT_NORESVPORT; + break; + case Opt_fscache_flag: + if (result.negated) + ctx->options &= ~NFS_OPTION_FSCACHE; + else + ctx->options |= NFS_OPTION_FSCACHE; + kfree(ctx->fscache_uniq); + ctx->fscache_uniq = NULL; + break; + case Opt_fscache: + trace_nfs_mount_assign(param->key, param->string); + ctx->options |= NFS_OPTION_FSCACHE; + kfree(ctx->fscache_uniq); + ctx->fscache_uniq = param->string; + param->string = NULL; + break; + case Opt_migration: + if (result.negated) + ctx->options &= ~NFS_OPTION_MIGRATION; + else + ctx->options |= NFS_OPTION_MIGRATION; + break; + + /* + * options that take numeric values + */ + case Opt_port: + if (result.uint_32 > USHRT_MAX) + goto out_of_bounds; + ctx->nfs_server.port = result.uint_32; + break; + case Opt_rsize: + ctx->rsize = result.uint_32; + break; + case Opt_wsize: + ctx->wsize = result.uint_32; + break; + case Opt_bsize: + ctx->bsize = result.uint_32; + break; + case Opt_timeo: + if (result.uint_32 < 1 || result.uint_32 > INT_MAX) + goto out_of_bounds; + ctx->timeo = result.uint_32; + break; + case Opt_retrans: + if (result.uint_32 > INT_MAX) + goto out_of_bounds; + ctx->retrans = result.uint_32; + break; + case Opt_acregmin: + ctx->acregmin = result.uint_32; + break; + case Opt_acregmax: + ctx->acregmax = result.uint_32; + break; + case Opt_acdirmin: + ctx->acdirmin = result.uint_32; + break; + case Opt_acdirmax: + ctx->acdirmax = result.uint_32; + break; + case Opt_actimeo: + ctx->acregmin = result.uint_32; + ctx->acregmax = result.uint_32; + ctx->acdirmin = result.uint_32; + ctx->acdirmax = result.uint_32; + break; + case Opt_namelen: + ctx->namlen = result.uint_32; + break; + case Opt_mountport: + if (result.uint_32 > USHRT_MAX) + goto out_of_bounds; + ctx->mount_server.port = result.uint_32; + break; + case Opt_mountvers: + if (result.uint_32 < NFS_MNT_VERSION || + result.uint_32 > NFS_MNT3_VERSION) + goto out_of_bounds; + ctx->mount_server.version = result.uint_32; + break; + case Opt_minorversion: + if (result.uint_32 < NFS4_MIN_MINOR_VERSION || + result.uint_32 > NFS4_MAX_MINOR_VERSION) + goto out_of_bounds; + ctx->minorversion = result.uint_32; + break; + + /* + * options that take text values + */ + case Opt_v: + ret = nfs_parse_version_string(fc, param->key + 1); + if (ret < 0) + return ret; + break; + case Opt_vers: + if (!param->string) + goto out_invalid_value; + trace_nfs_mount_assign(param->key, param->string); + ret = nfs_parse_version_string(fc, param->string); + if (ret < 0) + return ret; + break; + case Opt_sec: + ret = nfs_parse_security_flavors(fc, param); + if (ret < 0) + return ret; + break; + case Opt_xprtsec: + ret = nfs_parse_xprtsec_policy(fc, param); + if (ret < 0) + return ret; + break; + case Opt_cert_serial: + ret = nfs_tls_key_verify(result.int_32); + if (ret < 0) + return ret; + ctx->xprtsec.cert_serial = result.int_32; + break; + case Opt_privkey_serial: + ret = nfs_tls_key_verify(result.int_32); + if (ret < 0) + return ret; + ctx->xprtsec.privkey_serial = result.int_32; + break; + + case Opt_proto: + if (!param->string) + goto out_invalid_value; + trace_nfs_mount_assign(param->key, param->string); + protofamily = AF_INET; + switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) { + case Opt_xprt_udp6: + protofamily = AF_INET6; + fallthrough; + case Opt_xprt_udp: + ctx->flags &= ~NFS_MOUNT_TCP; + ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP; + break; + case Opt_xprt_tcp6: + protofamily = AF_INET6; + fallthrough; + case Opt_xprt_tcp: + ctx->flags |= NFS_MOUNT_TCP; + ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP; + break; + case Opt_xprt_rdma6: + protofamily = AF_INET6; + fallthrough; + case Opt_xprt_rdma: + /* vector side protocols to TCP */ + ctx->flags |= NFS_MOUNT_TCP; + ret = xprt_find_transport_ident(param->string); + if (ret < 0) + goto out_bad_transport; + ctx->nfs_server.protocol = ret; + break; + default: + goto out_bad_transport; + } + + ctx->protofamily = protofamily; + break; + + case Opt_mountproto: + if (!param->string) + goto out_invalid_value; + trace_nfs_mount_assign(param->key, param->string); + mountfamily = AF_INET; + switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) { + case Opt_xprt_udp6: + mountfamily = AF_INET6; + fallthrough; + case Opt_xprt_udp: + ctx->mount_server.protocol = XPRT_TRANSPORT_UDP; + break; + case Opt_xprt_tcp6: + mountfamily = AF_INET6; + fallthrough; + case Opt_xprt_tcp: + ctx->mount_server.protocol = XPRT_TRANSPORT_TCP; + break; + case Opt_xprt_rdma: /* not used for side protocols */ + default: + goto out_bad_transport; + } + ctx->mountfamily = mountfamily; + break; + + case Opt_addr: + trace_nfs_mount_assign(param->key, param->string); + len = rpc_pton(fc->net_ns, param->string, param->size, + &ctx->nfs_server.address, + sizeof(ctx->nfs_server._address)); + if (len == 0) + goto out_invalid_address; + ctx->nfs_server.addrlen = len; + break; + case Opt_clientaddr: + trace_nfs_mount_assign(param->key, param->string); + kfree(ctx->client_address); + ctx->client_address = param->string; + param->string = NULL; + break; + case Opt_mounthost: + trace_nfs_mount_assign(param->key, param->string); + kfree(ctx->mount_server.hostname); + ctx->mount_server.hostname = param->string; + param->string = NULL; + break; + case Opt_mountaddr: + trace_nfs_mount_assign(param->key, param->string); + len = rpc_pton(fc->net_ns, param->string, param->size, + &ctx->mount_server.address, + sizeof(ctx->mount_server._address)); + if (len == 0) + goto out_invalid_address; + ctx->mount_server.addrlen = len; + break; + case Opt_nconnect: + trace_nfs_mount_assign(param->key, param->string); + if (result.uint_32 < 1 || result.uint_32 > NFS_MAX_CONNECTIONS) + goto out_of_bounds; + ctx->nfs_server.nconnect = result.uint_32; + break; + case Opt_max_connect: + trace_nfs_mount_assign(param->key, param->string); + if (result.uint_32 < 1 || result.uint_32 > NFS_MAX_TRANSPORTS) + goto out_of_bounds; + ctx->nfs_server.max_connect = result.uint_32; + break; + case Opt_fatal_neterrors: + trace_nfs_mount_assign(param->key, param->string); + switch (result.uint_32) { + case Opt_fatal_neterrors_default: + if (fc->net_ns != &init_net) + ctx->flags |= NFS_MOUNT_NETUNREACH_FATAL; + else + ctx->flags &= ~NFS_MOUNT_NETUNREACH_FATAL; + break; + case Opt_fatal_neterrors_enetunreach: + ctx->flags |= NFS_MOUNT_NETUNREACH_FATAL; + break; + case Opt_fatal_neterrors_none: + ctx->flags &= ~NFS_MOUNT_NETUNREACH_FATAL; + break; + default: + goto out_invalid_value; + } + break; + case Opt_lookupcache: + trace_nfs_mount_assign(param->key, param->string); + switch (result.uint_32) { + case Opt_lookupcache_all: + ctx->flags &= ~(NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE); + break; + case Opt_lookupcache_positive: + ctx->flags &= ~NFS_MOUNT_LOOKUP_CACHE_NONE; + ctx->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG; + break; + case Opt_lookupcache_none: + ctx->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE; + break; + default: + goto out_invalid_value; + } + break; + case Opt_local_lock: + trace_nfs_mount_assign(param->key, param->string); + switch (result.uint_32) { + case Opt_local_lock_all: + ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | + NFS_MOUNT_LOCAL_FCNTL); + break; + case Opt_local_lock_flock: + ctx->flags |= NFS_MOUNT_LOCAL_FLOCK; + break; + case Opt_local_lock_posix: + ctx->flags |= NFS_MOUNT_LOCAL_FCNTL; + break; + case Opt_local_lock_none: + ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | + NFS_MOUNT_LOCAL_FCNTL); + break; + default: + goto out_invalid_value; + } + break; + case Opt_write: + trace_nfs_mount_assign(param->key, param->string); + switch (result.uint_32) { + case Opt_write_lazy: + ctx->flags &= + ~(NFS_MOUNT_WRITE_EAGER | NFS_MOUNT_WRITE_WAIT); + break; + case Opt_write_eager: + ctx->flags |= NFS_MOUNT_WRITE_EAGER; + ctx->flags &= ~NFS_MOUNT_WRITE_WAIT; + break; + case Opt_write_wait: + ctx->flags |= + NFS_MOUNT_WRITE_EAGER | NFS_MOUNT_WRITE_WAIT; + break; + default: + goto out_invalid_value; + } + break; + + /* + * Special options + */ + case Opt_sloppy: + ctx->sloppy = true; + break; + } + + return 0; + +out_invalid_value: + return nfs_invalf(fc, "NFS: Bad mount option value specified"); +out_invalid_address: + return nfs_invalf(fc, "NFS: Bad IP address specified"); +out_of_bounds: + return nfs_invalf(fc, "NFS: Value for '%s' out of range", param->key); +out_bad_transport: + return nfs_invalf(fc, "NFS: Unrecognized transport protocol"); +} + +/* + * Split fc->source into "hostname:export_path". + * + * The leftmost colon demarks the split between the server's hostname + * and the export path. If the hostname starts with a left square + * bracket, then it may contain colons. + * + * Note: caller frees hostname and export path, even on error. + */ +static int nfs_parse_source(struct fs_context *fc, + size_t maxnamlen, size_t maxpathlen) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + const char *dev_name = fc->source; + size_t len; + const char *end; + + if (unlikely(!dev_name || !*dev_name)) + return -EINVAL; + + /* Is the host name protected with square brakcets? */ + if (*dev_name == '[') { + end = strchr(++dev_name, ']'); + if (end == NULL || end[1] != ':') + goto out_bad_devname; + + len = end - dev_name; + end++; + } else { + const char *comma; + + end = strchr(dev_name, ':'); + if (end == NULL) + goto out_bad_devname; + len = end - dev_name; + + /* kill possible hostname list: not supported */ + comma = memchr(dev_name, ',', len); + if (comma) + len = comma - dev_name; + } + + if (len > maxnamlen) + goto out_hostname; + + kfree(ctx->nfs_server.hostname); + + /* N.B. caller will free nfs_server.hostname in all cases */ + ctx->nfs_server.hostname = kmemdup_nul(dev_name, len, GFP_KERNEL); + if (!ctx->nfs_server.hostname) + goto out_nomem; + len = strlen(++end); + if (len > maxpathlen) + goto out_path; + ctx->nfs_server.export_path = kmemdup_nul(end, len, GFP_KERNEL); + if (!ctx->nfs_server.export_path) + goto out_nomem; + + trace_nfs_mount_path(ctx->nfs_server.export_path); + return 0; + +out_bad_devname: + return nfs_invalf(fc, "NFS: device name not in host:path format"); +out_nomem: + nfs_errorf(fc, "NFS: not enough memory to parse device name"); + return -ENOMEM; +out_hostname: + nfs_errorf(fc, "NFS: server hostname too long"); + return -ENAMETOOLONG; +out_path: + nfs_errorf(fc, "NFS: export pathname too long"); + return -ENAMETOOLONG; +} + +static inline bool is_remount_fc(struct fs_context *fc) +{ + return fc->root != NULL; +} + +/* + * Parse monolithic NFS2/NFS3 mount data + * - fills in the mount root filehandle + * + * For option strings, user space handles the following behaviors: + * + * + DNS: mapping server host name to IP address ("addr=" option) + * + * + failure mode: how to behave if a mount request can't be handled + * immediately ("fg/bg" option) + * + * + retry: how often to retry a mount request ("retry=" option) + * + * + breaking back: trying proto=udp after proto=tcp, v2 after v3, + * mountproto=tcp after mountproto=udp, and so on + */ +static int nfs23_parse_monolithic(struct fs_context *fc, + struct nfs_mount_data *data) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_fh *mntfh = ctx->mntfh; + struct sockaddr_storage *sap = &ctx->nfs_server._address; + int extra_flags = NFS_MOUNT_LEGACY_INTERFACE; + int ret; + + if (data == NULL) + goto out_no_data; + + ctx->version = NFS_DEFAULT_VERSION; + switch (data->version) { + case 1: + data->namlen = 0; + fallthrough; + case 2: + data->bsize = 0; + fallthrough; + case 3: + if (data->flags & NFS_MOUNT_VER3) + goto out_no_v3; + data->root.size = NFS2_FHSIZE; + memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE); + /* Turn off security negotiation */ + extra_flags |= NFS_MOUNT_SECFLAVOUR; + fallthrough; + case 4: + if (data->flags & NFS_MOUNT_SECFLAVOUR) + goto out_no_sec; + fallthrough; + case 5: + memset(data->context, 0, sizeof(data->context)); + fallthrough; + case 6: + if (data->flags & NFS_MOUNT_VER3) { + if (data->root.size > NFS3_FHSIZE || data->root.size == 0) + goto out_invalid_fh; + mntfh->size = data->root.size; + ctx->version = 3; + } else { + mntfh->size = NFS2_FHSIZE; + ctx->version = 2; + } + + + memcpy(mntfh->data, data->root.data, mntfh->size); + if (mntfh->size < sizeof(mntfh->data)) + memset(mntfh->data + mntfh->size, 0, + sizeof(mntfh->data) - mntfh->size); + + /* + * for proto == XPRT_TRANSPORT_UDP, which is what uses + * to_exponential, implying shift: limit the shift value + * to BITS_PER_LONG (majortimeo is unsigned long) + */ + if (!(data->flags & NFS_MOUNT_TCP)) /* this will be UDP */ + if (data->retrans >= 64) /* shift value is too large */ + goto out_invalid_data; + + /* + * Translate to nfs_fs_context, which nfs_fill_super + * can deal with. + */ + ctx->flags = data->flags & NFS_MOUNT_FLAGMASK; + ctx->flags |= extra_flags; + ctx->rsize = data->rsize; + ctx->wsize = data->wsize; + ctx->timeo = data->timeo; + ctx->retrans = data->retrans; + ctx->acregmin = data->acregmin; + ctx->acregmax = data->acregmax; + ctx->acdirmin = data->acdirmin; + ctx->acdirmax = data->acdirmax; + ctx->need_mount = false; + + if (!is_remount_fc(fc)) { + memcpy(sap, &data->addr, sizeof(data->addr)); + ctx->nfs_server.addrlen = sizeof(data->addr); + ctx->nfs_server.port = ntohs(data->addr.sin_port); + } + + if (sap->ss_family != AF_INET || + !nfs_verify_server_address(sap)) + goto out_no_address; + + if (!(data->flags & NFS_MOUNT_TCP)) + ctx->nfs_server.protocol = XPRT_TRANSPORT_UDP; + /* N.B. caller will free nfs_server.hostname in all cases */ + ctx->nfs_server.hostname = kstrdup(data->hostname, GFP_KERNEL); + if (!ctx->nfs_server.hostname) + goto out_nomem; + + ctx->namlen = data->namlen; + ctx->bsize = data->bsize; + + if (data->flags & NFS_MOUNT_SECFLAVOUR) + ctx->selected_flavor = data->pseudoflavor; + else + ctx->selected_flavor = RPC_AUTH_UNIX; + + if (!(data->flags & NFS_MOUNT_NONLM)) + ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK| + NFS_MOUNT_LOCAL_FCNTL); + else + ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK| + NFS_MOUNT_LOCAL_FCNTL); + + /* + * The legacy version 6 binary mount data from userspace has a + * field used only to transport selinux information into the + * kernel. To continue to support that functionality we + * have a touch of selinux knowledge here in the NFS code. The + * userspace code converted context=blah to just blah so we are + * converting back to the full string selinux understands. + */ + if (data->context[0]){ +#ifdef CONFIG_SECURITY_SELINUX + int ret; + + data->context[NFS_MAX_CONTEXT_LEN] = '\0'; + ret = vfs_parse_fs_string(fc, "context", data->context); + if (ret < 0) + return ret; +#else + return -EINVAL; +#endif + } + + break; + default: + goto generic; + } + + ret = nfs_validate_transport_protocol(fc, ctx); + if (ret) + return ret; + + ctx->skip_reconfig_option_check = true; + return 0; + +generic: + return generic_parse_monolithic(fc, data); + +out_no_data: + if (is_remount_fc(fc)) { + ctx->skip_reconfig_option_check = true; + return 0; + } + return nfs_invalf(fc, "NFS: mount program didn't pass any mount data"); + +out_no_v3: + return nfs_invalf(fc, "NFS: nfs_mount_data version does not support v3"); + +out_no_sec: + return nfs_invalf(fc, "NFS: nfs_mount_data version supports only AUTH_SYS"); + +out_nomem: + return -ENOMEM; + +out_no_address: + return nfs_invalf(fc, "NFS: mount program didn't pass remote address"); + +out_invalid_fh: + return nfs_invalf(fc, "NFS: invalid root filehandle"); + +out_invalid_data: + return nfs_invalf(fc, "NFS: invalid binary mount data"); +} + +#if IS_ENABLED(CONFIG_NFS_V4) +struct compat_nfs_string { + compat_uint_t len; + compat_uptr_t data; +}; + +static inline void compat_nfs_string(struct nfs_string *dst, + struct compat_nfs_string *src) +{ + dst->data = compat_ptr(src->data); + dst->len = src->len; +} + +struct compat_nfs4_mount_data_v1 { + compat_int_t version; + compat_int_t flags; + compat_int_t rsize; + compat_int_t wsize; + compat_int_t timeo; + compat_int_t retrans; + compat_int_t acregmin; + compat_int_t acregmax; + compat_int_t acdirmin; + compat_int_t acdirmax; + struct compat_nfs_string client_addr; + struct compat_nfs_string mnt_path; + struct compat_nfs_string hostname; + compat_uint_t host_addrlen; + compat_uptr_t host_addr; + compat_int_t proto; + compat_int_t auth_flavourlen; + compat_uptr_t auth_flavours; +}; + +static void nfs4_compat_mount_data_conv(struct nfs4_mount_data *data) +{ + struct compat_nfs4_mount_data_v1 *compat = + (struct compat_nfs4_mount_data_v1 *)data; + + /* copy the fields backwards */ + data->auth_flavours = compat_ptr(compat->auth_flavours); + data->auth_flavourlen = compat->auth_flavourlen; + data->proto = compat->proto; + data->host_addr = compat_ptr(compat->host_addr); + data->host_addrlen = compat->host_addrlen; + compat_nfs_string(&data->hostname, &compat->hostname); + compat_nfs_string(&data->mnt_path, &compat->mnt_path); + compat_nfs_string(&data->client_addr, &compat->client_addr); + data->acdirmax = compat->acdirmax; + data->acdirmin = compat->acdirmin; + data->acregmax = compat->acregmax; + data->acregmin = compat->acregmin; + data->retrans = compat->retrans; + data->timeo = compat->timeo; + data->wsize = compat->wsize; + data->rsize = compat->rsize; + data->flags = compat->flags; + data->version = compat->version; +} + +/* + * Validate NFSv4 mount options + */ +static int nfs4_parse_monolithic(struct fs_context *fc, + struct nfs4_mount_data *data) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct sockaddr_storage *sap = &ctx->nfs_server._address; + int ret; + char *c; + + if (!data) { + if (is_remount_fc(fc)) + goto done; + return nfs_invalf(fc, + "NFS4: mount program didn't pass any mount data"); + } + + ctx->version = 4; + + if (data->version != 1) + return generic_parse_monolithic(fc, data); + + if (in_compat_syscall()) + nfs4_compat_mount_data_conv(data); + + if (data->host_addrlen > sizeof(ctx->nfs_server.address)) + goto out_no_address; + if (data->host_addrlen == 0) + goto out_no_address; + ctx->nfs_server.addrlen = data->host_addrlen; + if (copy_from_user(sap, data->host_addr, data->host_addrlen)) + return -EFAULT; + if (!nfs_verify_server_address(sap)) + goto out_no_address; + ctx->nfs_server.port = ntohs(((struct sockaddr_in *)sap)->sin_port); + + if (data->auth_flavourlen) { + rpc_authflavor_t pseudoflavor; + + if (data->auth_flavourlen > 1) + goto out_inval_auth; + if (copy_from_user(&pseudoflavor, data->auth_flavours, + sizeof(pseudoflavor))) + return -EFAULT; + ctx->selected_flavor = pseudoflavor; + } else { + ctx->selected_flavor = RPC_AUTH_UNIX; + } + + c = strndup_user(data->hostname.data, NFS4_MAXNAMLEN); + if (IS_ERR(c)) + return PTR_ERR(c); + ctx->nfs_server.hostname = c; + + c = strndup_user(data->mnt_path.data, NFS4_MAXPATHLEN); + if (IS_ERR(c)) + return PTR_ERR(c); + ctx->nfs_server.export_path = c; + trace_nfs_mount_path(c); + + c = strndup_user(data->client_addr.data, 16); + if (IS_ERR(c)) + return PTR_ERR(c); + ctx->client_address = c; + + /* + * Translate to nfs_fs_context, which nfs_fill_super + * can deal with. + */ + + ctx->flags = data->flags & NFS4_MOUNT_FLAGMASK; + ctx->rsize = data->rsize; + ctx->wsize = data->wsize; + ctx->timeo = data->timeo; + ctx->retrans = data->retrans; + ctx->acregmin = data->acregmin; + ctx->acregmax = data->acregmax; + ctx->acdirmin = data->acdirmin; + ctx->acdirmax = data->acdirmax; + ctx->nfs_server.protocol = data->proto; + ret = nfs_validate_transport_protocol(fc, ctx); + if (ret) + return ret; +done: + ctx->skip_reconfig_option_check = true; + return 0; + +out_inval_auth: + return nfs_invalf(fc, "NFS4: Invalid number of RPC auth flavours %d", + data->auth_flavourlen); + +out_no_address: + return nfs_invalf(fc, "NFS4: mount program didn't pass remote address"); +} +#endif + +/* + * Parse a monolithic block of data from sys_mount(). + */ +static int nfs_fs_context_parse_monolithic(struct fs_context *fc, + void *data) +{ + if (fc->fs_type == &nfs_fs_type) + return nfs23_parse_monolithic(fc, data); + +#if IS_ENABLED(CONFIG_NFS_V4) + if (fc->fs_type == &nfs4_fs_type) + return nfs4_parse_monolithic(fc, data); +#endif + + return nfs_invalf(fc, "NFS: Unsupported monolithic data version"); +} + +/* + * Validate the preparsed information in the config. + */ +static int nfs_fs_context_validate(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_subversion *nfs_mod; + struct sockaddr_storage *sap = &ctx->nfs_server._address; + int max_namelen = PAGE_SIZE; + int max_pathlen = NFS_MAXPATHLEN; + int port = 0; + int ret; + + if (!fc->source) + goto out_no_device_name; + + /* Check for sanity first. */ + if (ctx->minorversion && ctx->version != 4) + goto out_minorversion_mismatch; + + if (ctx->options & NFS_OPTION_MIGRATION && + (ctx->version != 4 || ctx->minorversion != 0)) + goto out_migration_misuse; + + /* Verify that any proto=/mountproto= options match the address + * families in the addr=/mountaddr= options. + */ + if (ctx->protofamily != AF_UNSPEC && + ctx->protofamily != ctx->nfs_server.address.sa_family) + goto out_proto_mismatch; + + if (ctx->mountfamily != AF_UNSPEC) { + if (ctx->mount_server.addrlen) { + if (ctx->mountfamily != ctx->mount_server.address.sa_family) + goto out_mountproto_mismatch; + } else { + if (ctx->mountfamily != ctx->nfs_server.address.sa_family) + goto out_mountproto_mismatch; + } + } + + if (!nfs_verify_server_address(sap)) + goto out_no_address; + + ret = nfs_validate_transport_protocol(fc, ctx); + if (ret) + return ret; + + if (ctx->version == 4) { + if (IS_ENABLED(CONFIG_NFS_V4)) { + if (ctx->nfs_server.protocol == XPRT_TRANSPORT_RDMA) + port = NFS_RDMA_PORT; + else + port = NFS_PORT; + max_namelen = NFS4_MAXNAMLEN; + max_pathlen = NFS4_MAXPATHLEN; + ctx->flags &= ~(NFS_MOUNT_NONLM | NFS_MOUNT_NOACL | + NFS_MOUNT_VER3 | NFS_MOUNT_LOCAL_FLOCK | + NFS_MOUNT_LOCAL_FCNTL); + } else { + goto out_v4_not_compiled; + } + } else { + nfs_set_mount_transport_protocol(ctx); + if (ctx->nfs_server.protocol == XPRT_TRANSPORT_RDMA) + port = NFS_RDMA_PORT; + } + + nfs_set_port(sap, &ctx->nfs_server.port, port); + + ret = nfs_parse_source(fc, max_namelen, max_pathlen); + if (ret < 0) + return ret; + + /* Load the NFS protocol module if we haven't done so yet */ + if (!ctx->nfs_mod) { + nfs_mod = find_nfs_version(ctx->version); + if (IS_ERR(nfs_mod)) { + ret = PTR_ERR(nfs_mod); + goto out_version_unavailable; + } + ctx->nfs_mod = nfs_mod; + } + + /* Ensure the filesystem context has the correct fs_type */ + if (fc->fs_type != ctx->nfs_mod->nfs_fs) { + module_put(fc->fs_type->owner); + __module_get(ctx->nfs_mod->nfs_fs->owner); + fc->fs_type = ctx->nfs_mod->nfs_fs; + } + return 0; + +out_no_device_name: + return nfs_invalf(fc, "NFS: Device name not specified"); +out_v4_not_compiled: + nfs_errorf(fc, "NFS: NFSv4 is not compiled into kernel"); + return -EPROTONOSUPPORT; +out_no_address: + return nfs_invalf(fc, "NFS: mount program didn't pass remote address"); +out_mountproto_mismatch: + return nfs_invalf(fc, "NFS: Mount server address does not match mountproto= option"); +out_proto_mismatch: + return nfs_invalf(fc, "NFS: Server address does not match proto= option"); +out_minorversion_mismatch: + return nfs_invalf(fc, "NFS: Mount option vers=%u does not support minorversion=%u", + ctx->version, ctx->minorversion); +out_migration_misuse: + return nfs_invalf(fc, "NFS: 'Migration' not supported for this NFS version"); +out_version_unavailable: + nfs_errorf(fc, "NFS: Version unavailable"); + return ret; +} + +/* + * Create an NFS superblock by the appropriate method. + */ +static int nfs_get_tree(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + int err = nfs_fs_context_validate(fc); + + if (err) + return err; + if (!ctx->internal) + return ctx->nfs_mod->rpc_ops->try_get_tree(fc); + else + return nfs_get_tree_common(fc); +} + +/* + * Handle duplication of a configuration. The caller copied *src into *sc, but + * it can't deal with resource pointers in the filesystem context, so we have + * to do that. We need to clear pointers, copy data or get extra refs as + * appropriate. + */ +static int nfs_fs_context_dup(struct fs_context *fc, struct fs_context *src_fc) +{ + struct nfs_fs_context *src = nfs_fc2context(src_fc), *ctx; + + ctx = kmemdup(src, sizeof(struct nfs_fs_context), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->mntfh = nfs_alloc_fhandle(); + if (!ctx->mntfh) { + kfree(ctx); + return -ENOMEM; + } + nfs_copy_fh(ctx->mntfh, src->mntfh); + + get_nfs_version(ctx->nfs_mod); + ctx->client_address = NULL; + ctx->mount_server.hostname = NULL; + ctx->nfs_server.export_path = NULL; + ctx->nfs_server.hostname = NULL; + ctx->fscache_uniq = NULL; + ctx->clone_data.fattr = NULL; + fc->fs_private = ctx; + return 0; +} + +static void nfs_fs_context_free(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + + if (ctx) { + if (ctx->server) + nfs_free_server(ctx->server); + if (ctx->nfs_mod) + put_nfs_version(ctx->nfs_mod); + kfree(ctx->client_address); + kfree(ctx->mount_server.hostname); + kfree(ctx->nfs_server.export_path); + kfree(ctx->nfs_server.hostname); + kfree(ctx->fscache_uniq); + nfs_free_fhandle(ctx->mntfh); + nfs_free_fattr(ctx->clone_data.fattr); + kfree(ctx); + } +} + +static const struct fs_context_operations nfs_fs_context_ops = { + .free = nfs_fs_context_free, + .dup = nfs_fs_context_dup, + .parse_param = nfs_fs_context_parse_param, + .parse_monolithic = nfs_fs_context_parse_monolithic, + .get_tree = nfs_get_tree, + .reconfigure = nfs_reconfigure, +}; + +/* + * Prepare superblock configuration. We use the namespaces attached to the + * context. This may be the current process's namespaces, or it may be a + * container's namespaces. + */ +static int nfs_init_fs_context(struct fs_context *fc) +{ + struct nfs_fs_context *ctx; + + ctx = kzalloc_obj(struct nfs_fs_context); + if (unlikely(!ctx)) + return -ENOMEM; + + ctx->mntfh = nfs_alloc_fhandle(); + if (unlikely(!ctx->mntfh)) { + kfree(ctx); + return -ENOMEM; + } + + ctx->protofamily = AF_UNSPEC; + ctx->mountfamily = AF_UNSPEC; + ctx->mount_server.port = NFS_UNSPEC_PORT; + + if (fc->root) { + /* reconfigure, start with the current config */ + struct nfs_server *nfss = fc->root->d_sb->s_fs_info; + struct net *net = nfss->nfs_client->cl_net; + + ctx->flags = nfss->flags; + ctx->rsize = nfss->rsize; + ctx->wsize = nfss->wsize; + ctx->retrans = nfss->client->cl_timeout->to_retries; + ctx->selected_flavor = nfss->client->cl_auth->au_flavor; + ctx->acregmin = nfss->acregmin / HZ; + ctx->acregmax = nfss->acregmax / HZ; + ctx->acdirmin = nfss->acdirmin / HZ; + ctx->acdirmax = nfss->acdirmax / HZ; + ctx->timeo = 10U * nfss->client->cl_timeout->to_initval / HZ; + ctx->nfs_server.port = nfss->port; + ctx->nfs_server.addrlen = nfss->nfs_client->cl_addrlen; + ctx->version = nfss->nfs_client->rpc_ops->version; + ctx->minorversion = nfss->nfs_client->cl_minorversion; + + memcpy(&ctx->nfs_server._address, &nfss->nfs_client->cl_addr, + ctx->nfs_server.addrlen); + + if (fc->net_ns != net) { + put_net(fc->net_ns); + fc->net_ns = get_net(net); + } + + ctx->nfs_mod = nfss->nfs_client->cl_nfs_mod; + get_nfs_version(ctx->nfs_mod); + } else { + /* defaults */ + ctx->timeo = NFS_UNSPEC_TIMEO; + ctx->retrans = NFS_UNSPEC_RETRANS; + ctx->acregmin = NFS_DEF_ACREGMIN; + ctx->acregmax = NFS_DEF_ACREGMAX; + ctx->acdirmin = NFS_DEF_ACDIRMIN; + ctx->acdirmax = NFS_DEF_ACDIRMAX; + ctx->nfs_server.port = NFS_UNSPEC_PORT; + ctx->nfs_server.protocol = XPRT_TRANSPORT_TCP; + ctx->selected_flavor = RPC_AUTH_MAXFLAVOR; + ctx->minorversion = 0; + ctx->need_mount = true; + ctx->xprtsec.policy = RPC_XPRTSEC_NONE; + ctx->xprtsec.cert_serial = TLS_NO_CERT; + ctx->xprtsec.privkey_serial = TLS_NO_PRIVKEY; + + if (fc->net_ns != &init_net) + ctx->flags |= NFS_MOUNT_NETUNREACH_FATAL; + + fc->s_iflags |= SB_I_STABLE_WRITES; + } + fc->fs_private = ctx; + fc->ops = &nfs_fs_context_ops; + return 0; +} + +struct file_system_type nfs_fs_type = { + .owner = THIS_MODULE, + .name = "nfs", + .init_fs_context = nfs_init_fs_context, + .parameters = nfs_fs_parameters, + .kill_sb = nfs_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE | + FS_BINARY_MOUNTDATA | + FS_USERNS_DELEGATABLE, +}; +MODULE_ALIAS_FS("nfs"); +EXPORT_SYMBOL_GPL(nfs_fs_type); + +#if IS_ENABLED(CONFIG_NFS_V4) +struct file_system_type nfs4_fs_type = { + .owner = THIS_MODULE, + .name = "nfs4", + .init_fs_context = nfs_init_fs_context, + .parameters = nfs_fs_parameters, + .kill_sb = nfs_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE | + FS_BINARY_MOUNTDATA | + FS_USERNS_DELEGATABLE, +}; +MODULE_ALIAS_FS("nfs4"); +MODULE_ALIAS("nfs4"); +EXPORT_SYMBOL_GPL(nfs4_fs_type); +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c new file mode 100644 index 000000000..9b7fdad4a --- /dev/null +++ b/fs/nfs/fscache.c @@ -0,0 +1,387 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* NFS filesystem cache interface + * + * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "iostat.h" +#include "fscache.h" +#include "nfstrace.h" + +#define NFS_MAX_KEY_LEN 1000 + +static bool nfs_append_int(char *key, int *_len, unsigned long long x) +{ + if (*_len > NFS_MAX_KEY_LEN) + return false; + if (x == 0) + key[(*_len)++] = ','; + else + *_len += sprintf(key + *_len, ",%llx", x); + return true; +} + +/* + * Get the per-client index cookie for an NFS client if the appropriate mount + * flag was set + * - We always try and get an index cookie for the client, but get filehandle + * cookies on a per-superblock basis, depending on the mount flags + */ +static bool nfs_fscache_get_client_key(struct nfs_client *clp, + char *key, int *_len) +{ + const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) &clp->cl_addr; + const struct sockaddr_in *sin = (struct sockaddr_in *) &clp->cl_addr; + + *_len += snprintf(key + *_len, NFS_MAX_KEY_LEN - *_len, + ",%u.%u,%x", + clp->rpc_ops->version, + clp->cl_minorversion, + clp->cl_addr.ss_family); + + switch (clp->cl_addr.ss_family) { + case AF_INET: + if (!nfs_append_int(key, _len, sin->sin_port) || + !nfs_append_int(key, _len, sin->sin_addr.s_addr)) + return false; + return true; + + case AF_INET6: + if (!nfs_append_int(key, _len, sin6->sin6_port) || + !nfs_append_int(key, _len, sin6->sin6_addr.s6_addr32[0]) || + !nfs_append_int(key, _len, sin6->sin6_addr.s6_addr32[1]) || + !nfs_append_int(key, _len, sin6->sin6_addr.s6_addr32[2]) || + !nfs_append_int(key, _len, sin6->sin6_addr.s6_addr32[3])) + return false; + return true; + + default: + printk(KERN_WARNING "NFS: Unknown network family '%d'\n", + clp->cl_addr.ss_family); + return false; + } +} + +/* + * Get the cache cookie for an NFS superblock. + * + * The default uniquifier is just an empty string, but it may be overridden + * either by the 'fsc=xxx' option to mount, or by inheriting it from the parent + * superblock across an automount point of some nature. + */ +int nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int ulen) +{ + struct fscache_volume *vcookie; + struct nfs_server *nfss = NFS_SB(sb); + unsigned int len = 3; + char *key; + + if (uniq) { + nfss->fscache_uniq = kmemdup_nul(uniq, ulen, GFP_KERNEL); + if (!nfss->fscache_uniq) + return -ENOMEM; + } + + key = kmalloc(NFS_MAX_KEY_LEN + 24, GFP_KERNEL); + if (!key) + return -ENOMEM; + + memcpy(key, "nfs", 3); + if (!nfs_fscache_get_client_key(nfss->nfs_client, key, &len) || + !nfs_append_int(key, &len, nfss->fsid.major) || + !nfs_append_int(key, &len, nfss->fsid.minor) || + !nfs_append_int(key, &len, sb->s_flags & NFS_SB_MASK) || + !nfs_append_int(key, &len, nfss->flags) || + !nfs_append_int(key, &len, nfss->rsize) || + !nfs_append_int(key, &len, nfss->wsize) || + !nfs_append_int(key, &len, nfss->acregmin) || + !nfs_append_int(key, &len, nfss->acregmax) || + !nfs_append_int(key, &len, nfss->acdirmin) || + !nfs_append_int(key, &len, nfss->acdirmax) || + !nfs_append_int(key, &len, nfss->client->cl_auth->au_flavor)) + goto out; + + if (ulen > 0) { + if (ulen > NFS_MAX_KEY_LEN - len) + goto out; + key[len++] = ','; + memcpy(key + len, uniq, ulen); + len += ulen; + } + key[len] = 0; + + /* create a cache index for looking up filehandles */ + vcookie = fscache_acquire_volume(key, + NULL, /* preferred_cache */ + NULL, 0 /* coherency_data */); + if (IS_ERR(vcookie)) { + if (vcookie != ERR_PTR(-EBUSY)) { + kfree(key); + return PTR_ERR(vcookie); + } + pr_err("NFS: Cache volume key already in use (%s)\n", key); + vcookie = NULL; + } + nfss->fscache = vcookie; + +out: + kfree(key); + return 0; +} + +/* + * release a per-superblock cookie + */ +void nfs_fscache_release_super_cookie(struct super_block *sb) +{ + struct nfs_server *nfss = NFS_SB(sb); + + fscache_relinquish_volume(nfss->fscache, NULL, false); + nfss->fscache = NULL; + kfree(nfss->fscache_uniq); +} + +/* + * Initialise the per-inode cache cookie pointer for an NFS inode. + */ +void nfs_fscache_init_inode(struct inode *inode) +{ + struct nfs_fscache_inode_auxdata auxdata; + struct nfs_server *nfss = NFS_SERVER(inode); + struct nfs_inode *nfsi = NFS_I(inode); + + netfs_inode(inode)->cache = NULL; + if (!(nfss->fscache && S_ISREG(inode->i_mode))) + return; + + nfs_fscache_update_auxdata(&auxdata, inode); + + netfs_inode(inode)->cache = fscache_acquire_cookie( + nfss->fscache, + 0, + nfsi->fh.data, /* index_key */ + nfsi->fh.size, + &auxdata, /* aux_data */ + sizeof(auxdata), + i_size_read(inode)); + + if (netfs_inode(inode)->cache) + mapping_set_release_always(inode->i_mapping); +} + +/* + * Release a per-inode cookie. + */ +void nfs_fscache_clear_inode(struct inode *inode) +{ + fscache_relinquish_cookie(netfs_i_cookie(netfs_inode(inode)), false); + netfs_inode(inode)->cache = NULL; +} + +/* + * Enable or disable caching for a file that is being opened as appropriate. + * The cookie is allocated when the inode is initialised, but is not enabled at + * that time. Enablement is deferred to file-open time to avoid stat() and + * access() thrashing the cache. + * + * For now, with NFS, only regular files that are open read-only will be able + * to use the cache. + * + * We enable the cache for an inode if we open it read-only and it isn't + * currently open for writing. We disable the cache if the inode is open + * write-only. + * + * The caller uses the file struct to pin i_writecount on the inode before + * calling us when a file is opened for writing, so we can make use of that. + * + * Note that this may be invoked multiple times in parallel by parallel + * nfs_open() functions. + */ +void nfs_fscache_open_file(struct inode *inode, struct file *filp) +{ + struct nfs_fscache_inode_auxdata auxdata; + struct fscache_cookie *cookie = netfs_i_cookie(netfs_inode(inode)); + bool open_for_write = inode_is_open_for_write(inode); + + if (!fscache_cookie_valid(cookie)) + return; + + fscache_use_cookie(cookie, open_for_write); + if (open_for_write) { + nfs_fscache_update_auxdata(&auxdata, inode); + fscache_invalidate(cookie, &auxdata, i_size_read(inode), + FSCACHE_INVAL_DIO_WRITE); + } +} +EXPORT_SYMBOL_GPL(nfs_fscache_open_file); + +void nfs_fscache_release_file(struct inode *inode, struct file *filp) +{ + struct nfs_fscache_inode_auxdata auxdata; + struct fscache_cookie *cookie = netfs_i_cookie(netfs_inode(inode)); + loff_t i_size = i_size_read(inode); + + nfs_fscache_update_auxdata(&auxdata, inode); + fscache_unuse_cookie(cookie, &auxdata, &i_size); +} + +int nfs_netfs_read_folio(struct file *file, struct folio *folio) +{ + if (!netfs_inode(folio_inode(folio))->cache) + return -ENOBUFS; + + return netfs_read_folio(file, folio); +} + +int nfs_netfs_readahead(struct readahead_control *ractl) +{ + struct inode *inode = ractl->mapping->host; + + if (!netfs_inode(inode)->cache) + return -ENOBUFS; + + netfs_readahead(ractl); + return 0; +} + +static atomic_t nfs_netfs_debug_id; +static int nfs_netfs_init_request(struct netfs_io_request *rreq, struct file *file) +{ + if (!file) { + if (WARN_ON_ONCE(rreq->origin != NETFS_PGPRIV2_COPY_TO_CACHE)) + return -EIO; + return 0; + } + + rreq->netfs_priv = get_nfs_open_context(nfs_file_open_context(file)); + rreq->debug_id = atomic_inc_return(&nfs_netfs_debug_id); + /* [DEPRECATED] Use PG_private_2 to mark folio being written to the cache. */ + __set_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags); + rreq->io_streams[0].sreq_max_len = NFS_SB(rreq->inode->i_sb)->rsize; + + return 0; +} + +static void nfs_netfs_free_request(struct netfs_io_request *rreq) +{ + if (rreq->netfs_priv) + put_nfs_open_context(rreq->netfs_priv); +} + +static struct nfs_netfs_io_data *nfs_netfs_alloc(struct netfs_io_subrequest *sreq) +{ + struct nfs_netfs_io_data *netfs; + + netfs = kzalloc_obj(*netfs, GFP_KERNEL_ACCOUNT); + if (!netfs) + return NULL; + netfs->sreq = sreq; + refcount_set(&netfs->refcount, 1); + return netfs; +} + +static void nfs_netfs_issue_read(struct netfs_io_subrequest *sreq) +{ + struct nfs_netfs_io_data *netfs; + struct nfs_pageio_descriptor pgio; + struct inode *inode = sreq->rreq->inode; + struct nfs_open_context *ctx = sreq->rreq->netfs_priv; + struct page *page; + unsigned long idx; + pgoff_t start, last; + int err; + + start = (sreq->start + sreq->transferred) >> PAGE_SHIFT; + last = ((sreq->start + sreq->len - sreq->transferred - 1) >> PAGE_SHIFT); + + nfs_pageio_init_read(&pgio, inode, false, + &nfs_async_read_completion_ops); + + netfs = nfs_netfs_alloc(sreq); + if (!netfs) { + sreq->error = -ENOMEM; + return netfs_read_subreq_terminated(sreq); + } + + pgio.pg_netfs = netfs; /* used in completion */ + + xa_for_each_range(&sreq->rreq->mapping->i_pages, idx, page, start, last) { + /* nfs_read_add_folio() may schedule() due to pNFS layout and other RPCs */ + err = nfs_read_add_folio(&pgio, ctx, page_folio(page)); + if (err < 0) { + netfs->error = err; + goto out; + } + } +out: + nfs_pageio_complete_read(&pgio); + nfs_netfs_put(netfs); +} + +void nfs_netfs_initiate_read(struct nfs_pgio_header *hdr) +{ + struct nfs_netfs_io_data *netfs = hdr->netfs; + + if (!netfs) + return; + + nfs_netfs_get(netfs); +} + +int nfs_netfs_folio_unlock(struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + + /* + * If fscache is enabled, netfs will unlock pages. + */ + if (netfs_inode(inode)->cache) + return 0; + + return 1; +} + +void nfs_netfs_read_completion(struct nfs_pgio_header *hdr) +{ + struct nfs_netfs_io_data *netfs = hdr->netfs; + struct netfs_io_subrequest *sreq; + + if (!netfs) + return; + + sreq = netfs->sreq; + if (test_bit(NFS_IOHDR_EOF, &hdr->flags) && + sreq->rreq->origin != NETFS_UNBUFFERED_READ && + sreq->rreq->origin != NETFS_DIO_READ) + __set_bit(NETFS_SREQ_CLEAR_TAIL, &sreq->flags); + + if (hdr->error) + netfs->error = hdr->error; + else + atomic64_add(hdr->res.count, &netfs->transferred); + + nfs_netfs_put(netfs); + hdr->netfs = NULL; +} + +const struct netfs_request_ops nfs_netfs_ops = { + .init_request = nfs_netfs_init_request, + .free_request = nfs_netfs_free_request, + .issue_read = nfs_netfs_issue_read, +}; diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h new file mode 100644 index 000000000..9d86868f4 --- /dev/null +++ b/fs/nfs/fscache.h @@ -0,0 +1,202 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* NFS filesystem cache interface definitions + * + * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#ifndef _NFS_FSCACHE_H +#define _NFS_FSCACHE_H + +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_NFS_FSCACHE + +/* + * Definition of the auxiliary data attached to NFS inode storage objects + * within the cache. + * + * The contents of this struct are recorded in the on-disk local cache in the + * auxiliary data attached to the data storage object backing an inode. This + * permits coherency to be managed when a new inode binds to an already extant + * cache object. + */ +struct nfs_fscache_inode_auxdata { + s64 mtime_sec; + s64 mtime_nsec; + s64 ctime_sec; + s64 ctime_nsec; + u64 change_attr; +}; + +struct nfs_netfs_io_data { + /* + * NFS may split a netfs_io_subrequest into multiple RPCs, each + * with their own read completion. In netfs, we can only call + * netfs_subreq_terminated() once for each subrequest. Use the + * refcount here to double as a marker of the last RPC completion, + * and only call netfs via netfs_subreq_terminated() once. + */ + refcount_t refcount; + struct netfs_io_subrequest *sreq; + + /* + * Final disposition of the netfs_io_subrequest, sent in + * netfs_subreq_terminated() + */ + atomic64_t transferred; + int error; +}; + +static inline void nfs_netfs_get(struct nfs_netfs_io_data *netfs) +{ + refcount_inc(&netfs->refcount); +} + +static inline void nfs_netfs_put(struct nfs_netfs_io_data *netfs) +{ + /* Only the last RPC completion should call netfs_subreq_terminated() */ + if (!refcount_dec_and_test(&netfs->refcount)) + return; + + /* + * The NFS pageio interface may read a complete page, even when netfs + * only asked for a partial page. Specifically, this may be seen when + * one thread is truncating a file while another one is reading the last + * page of the file. + * Correct the final length here to be no larger than the netfs subrequest + * length, and thus avoid netfs's "Subreq overread" warning message. + */ + netfs->sreq->transferred = min_t(s64, netfs->sreq->len, + atomic64_read(&netfs->transferred)); + netfs->sreq->error = netfs->error; + netfs_read_subreq_terminated(netfs->sreq); + kfree(netfs); +} +static inline void nfs_netfs_inode_init(struct nfs_inode *nfsi) +{ + netfs_inode_init(&nfsi->netfs, &nfs_netfs_ops, false); +} +extern void nfs_netfs_initiate_read(struct nfs_pgio_header *hdr); +extern void nfs_netfs_read_completion(struct nfs_pgio_header *hdr); +extern int nfs_netfs_folio_unlock(struct folio *folio); + +/* + * fscache.c + */ +extern int nfs_fscache_get_super_cookie(struct super_block *, const char *, int); +extern void nfs_fscache_release_super_cookie(struct super_block *); + +extern void nfs_fscache_init_inode(struct inode *); +extern void nfs_fscache_clear_inode(struct inode *); +extern void nfs_fscache_open_file(struct inode *, struct file *); +extern void nfs_fscache_release_file(struct inode *, struct file *); +extern int nfs_netfs_readahead(struct readahead_control *ractl); +extern int nfs_netfs_read_folio(struct file *file, struct folio *folio); + +static inline bool nfs_fscache_release_folio(struct folio *folio, gfp_t gfp) +{ + if (folio_test_private_2(folio)) { /* [DEPRECATED] */ + if (current_is_kswapd() || !(gfp & __GFP_FS)) + return false; + folio_wait_private_2(folio); + } + fscache_note_page_release(netfs_i_cookie(netfs_inode(folio->mapping->host))); + return true; +} + +static inline void nfs_fscache_update_auxdata(struct nfs_fscache_inode_auxdata *auxdata, + struct inode *inode) +{ + memset(auxdata, 0, sizeof(*auxdata)); + auxdata->mtime_sec = inode_get_mtime(inode).tv_sec; + auxdata->mtime_nsec = inode_get_mtime(inode).tv_nsec; + auxdata->ctime_sec = inode_get_ctime(inode).tv_sec; + auxdata->ctime_nsec = inode_get_ctime(inode).tv_nsec; + + if (NFS_SERVER(inode)->nfs_client->rpc_ops->version == 4) + auxdata->change_attr = inode_peek_iversion_raw(inode); +} + +/* + * Invalidate the contents of fscache for this inode. This will not sleep. + */ +static inline void nfs_fscache_invalidate(struct inode *inode, int flags) +{ + struct nfs_fscache_inode_auxdata auxdata; + struct fscache_cookie *cookie = netfs_i_cookie(&NFS_I(inode)->netfs); + + nfs_fscache_update_auxdata(&auxdata, inode); + fscache_invalidate(cookie, &auxdata, i_size_read(inode), flags); +} + +/* + * indicate the client caching state as readable text + */ +static inline const char *nfs_server_fscache_state(struct nfs_server *server) +{ + if (server->fscache) + return "yes"; + return "no "; +} + +static inline void nfs_netfs_set_pgio_header(struct nfs_pgio_header *hdr, + struct nfs_pageio_descriptor *desc) +{ + hdr->netfs = desc->pg_netfs; +} +static inline void nfs_netfs_set_pageio_descriptor(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + desc->pg_netfs = hdr->netfs; +} +static inline void nfs_netfs_reset_pageio_descriptor(struct nfs_pageio_descriptor *desc) +{ + desc->pg_netfs = NULL; +} +#else /* CONFIG_NFS_FSCACHE */ +static inline void nfs_netfs_inode_init(struct nfs_inode *nfsi) {} +static inline void nfs_netfs_initiate_read(struct nfs_pgio_header *hdr) {} +static inline void nfs_netfs_read_completion(struct nfs_pgio_header *hdr) {} +static inline int nfs_netfs_folio_unlock(struct folio *folio) +{ + return 1; +} +static inline void nfs_fscache_release_super_cookie(struct super_block *sb) {} + +static inline void nfs_fscache_init_inode(struct inode *inode) {} +static inline void nfs_fscache_clear_inode(struct inode *inode) {} +static inline void nfs_fscache_open_file(struct inode *inode, + struct file *filp) {} +static inline void nfs_fscache_release_file(struct inode *inode, struct file *file) {} +static inline int nfs_netfs_readahead(struct readahead_control *ractl) +{ + return -ENOBUFS; +} +static inline int nfs_netfs_read_folio(struct file *file, struct folio *folio) +{ + return -ENOBUFS; +} + +static inline bool nfs_fscache_release_folio(struct folio *folio, gfp_t gfp) +{ + return true; /* may release folio */ +} +static inline void nfs_fscache_invalidate(struct inode *inode, int flags) {} + +static inline const char *nfs_server_fscache_state(struct nfs_server *server) +{ + return "no "; +} +static inline void nfs_netfs_set_pgio_header(struct nfs_pgio_header *hdr, + struct nfs_pageio_descriptor *desc) {} +static inline void nfs_netfs_set_pageio_descriptor(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) {} +static inline void nfs_netfs_reset_pageio_descriptor(struct nfs_pageio_descriptor *desc) {} +#endif /* CONFIG_NFS_FSCACHE */ +#endif /* _NFS_FSCACHE_H */ diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c new file mode 100644 index 000000000..ff7424bc4 --- /dev/null +++ b/fs/nfs/getroot.c @@ -0,0 +1,133 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* getroot.c: get the root dentry for an NFS mount + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +/* + * get a root dentry from the root filehandle + */ +int nfs_get_root(struct super_block *s, struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_server *server = NFS_SB(s), *clone_server; + struct nfs_fsinfo fsinfo; + struct dentry *root; + struct inode *inode; + char *name; + int error = -ENOMEM; + unsigned long kflags = 0, kflags_out = 0; + + name = kstrdup(fc->source, GFP_KERNEL); + if (!name) + goto out; + + /* get the actual root for this mount */ + fsinfo.fattr = nfs_alloc_fattr_with_label(server); + if (fsinfo.fattr == NULL) + goto out_name; + + error = server->nfs_client->rpc_ops->getroot(server, ctx->mntfh, &fsinfo); + if (error < 0) { + dprintk("nfs_get_root: getattr error = %d\n", -error); + nfs_errorf(fc, "NFS: Couldn't getattr on root"); + goto out_fattr; + } + + inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr); + if (IS_ERR(inode)) { + dprintk("nfs_get_root: get root inode failed\n"); + error = PTR_ERR(inode); + nfs_errorf(fc, "NFS: Couldn't get root inode"); + goto out_fattr; + } + + /* root dentries normally start off anonymous and get spliced in later + * if the dentry tree reaches them; however if the dentry already + * exists, we'll pick it up at this point and use it as the root + */ + root = d_obtain_root(inode); + if (IS_ERR(root)) { + dprintk("nfs_get_root: get root dentry failed\n"); + error = PTR_ERR(root); + nfs_errorf(fc, "NFS: Couldn't get root dentry"); + goto out_fattr; + } + + security_d_instantiate(root, inode); + spin_lock(&root->d_lock); + if (IS_ROOT(root) && !root->d_fsdata && + !(root->d_flags & DCACHE_NFSFS_RENAMED)) { + root->d_fsdata = name; + name = NULL; + } + spin_unlock(&root->d_lock); + if (!s->s_root) + s->s_root = dget(root); + fc->root = root; + if (server->caps & NFS_CAP_SECURITY_LABEL) + kflags |= SECURITY_LSM_NATIVE_LABELS; + if (ctx->clone_data.sb) { + if (d_inode(fc->root)->i_fop != &nfs_dir_operations) { + error = -ESTALE; + goto error_splat_root; + } + /* clone lsm security options from the parent to the new sb */ + error = security_sb_clone_mnt_opts(ctx->clone_data.sb, + s, kflags, &kflags_out); + if (error) + goto error_splat_root; + clone_server = NFS_SB(ctx->clone_data.sb); + server->has_sec_mnt_opts = clone_server->has_sec_mnt_opts; + } else { + error = security_sb_set_mnt_opts(s, fc->security, + kflags, &kflags_out); + } + if (error) + goto error_splat_root; + if (server->caps & NFS_CAP_SECURITY_LABEL && + !(kflags_out & SECURITY_LSM_NATIVE_LABELS)) + server->caps &= ~NFS_CAP_SECURITY_LABEL; + + nfs_setsecurity(inode, fsinfo.fattr); + error = 0; + +out_fattr: + nfs_free_fattr(fsinfo.fattr); +out_name: + kfree(name); +out: + return error; +error_splat_root: + dput(fc->root); + fc->root = NULL; + goto out_fattr; +} diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c new file mode 100644 index 000000000..5bcd4027d --- /dev/null +++ b/fs/nfs/inode.c @@ -0,0 +1,2793 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/inode.c + * + * Copyright (C) 1992 Rick Sladkey + * + * nfs inode and superblock handling functions + * + * Modularised by Alan Cox , while hacking some + * experimental NFS changes. Modularisation taken straight from SYS5 fs. + * + * Change to nfs_read_super() to permit NFS mounts to multi-homed hosts. + * J.S.Peatfield@damtp.cam.ac.uk + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "iostat.h" +#include "internal.h" +#include "fscache.h" +#include "pnfs.h" +#include "nfs.h" +#include "netns.h" +#include "sysfs.h" + +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static bool enable_ino64; + +static int param_set_enable_ino64(const char *val, const struct kernel_param *kp) +{ + pr_notice("enable_ino64 is deprecated and has no effect\n"); + return 0; +} + +static const struct kernel_param_ops param_ops_enable_ino64 = { + .set = param_set_enable_ino64, + .get = param_get_bool, +}; + +static int nfs_update_inode(struct inode *, struct nfs_fattr *); + +static struct kmem_cache * nfs_inode_cachep; + +int nfs_wait_bit_killable(struct wait_bit_key *key, int mode) +{ + if (unlikely(nfs_current_task_exiting())) + return -EINTR; + schedule(); + if (signal_pending_state(mode, current)) + return -ERESTARTSYS; + return 0; +} +EXPORT_SYMBOL_GPL(nfs_wait_bit_killable); + +int nfs_drop_inode(struct inode *inode) +{ + return NFS_STALE(inode) || inode_generic_drop(inode); +} +EXPORT_SYMBOL_GPL(nfs_drop_inode); + +void nfs_clear_inode(struct inode *inode) +{ + /* + * The following should never happen... + */ + WARN_ON_ONCE(nfs_have_writebacks(inode)); + WARN_ON_ONCE(!list_empty(&NFS_I(inode)->open_files)); + nfs_zap_acl_cache(inode); + nfs_access_zap_cache(inode); + nfs_fscache_clear_inode(inode); +} +EXPORT_SYMBOL_GPL(nfs_clear_inode); + +void nfs_evict_inode(struct inode *inode) +{ + truncate_inode_pages_final(&inode->i_data); + clear_inode(inode); + nfs_clear_inode(inode); +} + +int nfs_sync_inode(struct inode *inode) +{ + inode_dio_wait(inode); + return nfs_wb_all(inode); +} +EXPORT_SYMBOL_GPL(nfs_sync_inode); + +/** + * nfs_sync_mapping - helper to flush all mmapped dirty data to disk + * @mapping: pointer to struct address_space + */ +int nfs_sync_mapping(struct address_space *mapping) +{ + int ret = 0; + + if (mapping->nrpages != 0) { + unmap_mapping_range(mapping, 0, 0, 0); + ret = nfs_wb_all(mapping->host); + } + return ret; +} + +static int nfs_attribute_timeout(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + return !time_in_range_open(jiffies, nfsi->read_cache_jiffies, nfsi->read_cache_jiffies + nfsi->attrtimeo); +} + +static bool nfs_check_cache_flags_invalid(struct inode *inode, + unsigned long flags) +{ + unsigned long cache_validity = READ_ONCE(NFS_I(inode)->cache_validity); + + return (cache_validity & flags) != 0; +} + +bool nfs_check_cache_invalid(struct inode *inode, unsigned long flags) +{ + if (nfs_check_cache_flags_invalid(inode, flags)) + return true; + return nfs_attribute_cache_expired(inode); +} +EXPORT_SYMBOL_GPL(nfs_check_cache_invalid); + +#ifdef CONFIG_NFS_V4_2 +static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi) +{ + return nfsi->xattr_cache != NULL; +} +#else +static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi) +{ + return false; +} +#endif + +void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + if (nfs_have_delegated_attributes(inode)) { + if (!(flags & NFS_INO_REVAL_FORCED)) + flags &= ~(NFS_INO_INVALID_MODE | + NFS_INO_INVALID_OTHER | + NFS_INO_INVALID_BTIME | + NFS_INO_INVALID_XATTR); + flags &= ~(NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE); + } + + if (!nfs_has_xattr_cache(nfsi)) + flags &= ~NFS_INO_INVALID_XATTR; + if (flags & NFS_INO_INVALID_DATA) + nfs_fscache_invalidate(inode, 0); + flags &= ~NFS_INO_REVAL_FORCED; + + flags |= nfsi->cache_validity; + if (inode->i_mapping->nrpages == 0) + flags &= ~NFS_INO_INVALID_DATA; + + /* pairs with nfs_clear_invalid_mapping()'s smp_load_acquire() */ + smp_store_release(&nfsi->cache_validity, flags); + + if (inode->i_mapping->nrpages == 0 || + nfsi->cache_validity & NFS_INO_INVALID_DATA) { + nfs_ooo_clear(nfsi); + } + trace_nfs_set_cache_invalid(inode, 0); +} +EXPORT_SYMBOL_GPL(nfs_set_cache_invalid); + +/* + * Invalidate the local caches + */ +static void nfs_zap_caches_locked(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int mode = inode->i_mode; + + nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE); + + nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); + nfsi->attrtimeo_timestamp = jiffies; + + if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR | + NFS_INO_INVALID_DATA | + NFS_INO_INVALID_ACCESS | + NFS_INO_INVALID_ACL | + NFS_INO_INVALID_XATTR); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR | + NFS_INO_INVALID_ACCESS | + NFS_INO_INVALID_ACL | + NFS_INO_INVALID_XATTR); + nfs_zap_label_cache_locked(nfsi); +} + +void nfs_zap_caches(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_zap_caches_locked(inode); + spin_unlock(&inode->i_lock); +} + +void nfs_zap_mapping(struct inode *inode, struct address_space *mapping) +{ + if (mapping->nrpages != 0) { + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); + spin_unlock(&inode->i_lock); + } +} + +void nfs_zap_acl_cache(struct inode *inode) +{ + void (*clear_acl_cache)(struct inode *); + + clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache; + if (clear_acl_cache != NULL) + clear_acl_cache(inode); + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_ACL; + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_zap_acl_cache); + +void nfs_invalidate_atime(struct inode *inode) +{ + if (nfs_have_delegated_atime(inode)) + return; + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_invalidate_atime); + +/* + * Invalidate, but do not unhash, the inode. + * NB: must be called with inode->i_lock held! + */ +static void nfs_set_inode_stale_locked(struct inode *inode) +{ + set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); + nfs_zap_caches_locked(inode); + trace_nfs_set_inode_stale(inode); +} + +void nfs_set_inode_stale(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_set_inode_stale_locked(inode); + spin_unlock(&inode->i_lock); +} + +struct nfs_find_desc { + struct nfs_fh *fh; + struct nfs_fattr *fattr; +}; + +/* + * For re-exported directories (also seen in NFSv2) + * we are forced to allow 2 different inodes to have the same + * i_ino. + */ +static int +nfs_find_actor(struct inode *inode, void *opaque) +{ + struct nfs_find_desc *desc = opaque; + struct nfs_fh *fh = desc->fh; + struct nfs_fattr *fattr = desc->fattr; + + if (inode->i_ino != fattr->fileid) + return 0; + if (inode_wrong_type(inode, fattr->mode)) + return 0; + if (nfs_compare_fh(NFS_FH(inode), fh)) + return 0; + if (is_bad_inode(inode) || NFS_STALE(inode)) + return 0; + return 1; +} + +static int +nfs_init_locked(struct inode *inode, void *opaque) +{ + struct nfs_find_desc *desc = opaque; + struct nfs_fattr *fattr = desc->fattr; + + inode->i_ino = fattr->fileid; + inode->i_mode = fattr->mode; + nfs_copy_fh(NFS_FH(inode), desc->fh); + return 0; +} + +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +static void nfs_clear_label_invalid(struct inode *inode) +{ + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_LABEL; + spin_unlock(&inode->i_lock); +} + +void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr) +{ + int error; + + if (fattr->label == NULL) + return; + + if ((fattr->valid & NFS_ATTR_FATTR_V4_SECURITY_LABEL) && inode->i_security) { + error = security_inode_notifysecctx(inode, fattr->label->label, + fattr->label->len); + if (error) + printk(KERN_ERR "%s() %s %d " + "security_inode_notifysecctx() %d\n", + __func__, + (char *)fattr->label->label, + fattr->label->len, error); + nfs_clear_label_invalid(inode); + } +} + +struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) +{ + struct nfs4_label *label; + + if (!(server->caps & NFS_CAP_SECURITY_LABEL)) + return NULL; + + label = kzalloc_obj(struct nfs4_label, flags); + if (label == NULL) + return ERR_PTR(-ENOMEM); + + label->label = kzalloc(NFS4_MAXLABELLEN, flags); + if (label->label == NULL) { + kfree(label); + return ERR_PTR(-ENOMEM); + } + label->len = NFS4_MAXLABELLEN; + + return label; +} +EXPORT_SYMBOL_GPL(nfs4_label_alloc); +#else +void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr) +{ +} +#endif +EXPORT_SYMBOL_GPL(nfs_setsecurity); + +/* Search for inode identified by fh, fileid and i_mode in inode cache. */ +struct inode * +nfs_ilookup(struct super_block *sb, struct nfs_fattr *fattr, struct nfs_fh *fh) +{ + struct nfs_find_desc desc = { + .fh = fh, + .fattr = fattr, + }; + struct inode *inode; + u64 hash; + + if (!(fattr->valid & NFS_ATTR_FATTR_FILEID) || + !(fattr->valid & NFS_ATTR_FATTR_TYPE)) + return NULL; + + hash = fattr->fileid; + inode = ilookup5(sb, hash, nfs_find_actor, &desc); + + dprintk("%s: returning %p\n", __func__, inode); + return inode; +} + +static void nfs_inode_init_regular(struct nfs_inode *nfsi) +{ + atomic_long_set(&nfsi->nrequests, 0); + atomic_long_set(&nfsi->redirtied_pages, 0); + INIT_LIST_HEAD(&nfsi->commit_info.list); + atomic_long_set(&nfsi->commit_info.ncommit, 0); + atomic_set(&nfsi->commit_info.rpcs_out, 0); + mutex_init(&nfsi->commit_mutex); +} + +static void nfs_inode_init_dir(struct nfs_inode *nfsi) +{ + nfsi->cache_change_attribute = 0; + memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); + init_rwsem(&nfsi->rmdir_sem); +} + +/* + * This is our front-end to iget that looks up inodes by file handle + * instead of inode number. + */ +struct inode * +nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) +{ + struct nfs_find_desc desc = { + .fh = fh, + .fattr = fattr + }; + struct inode *inode = ERR_PTR(-ENOENT); + u64 fattr_supported = NFS_SB(sb)->fattr_valid; + u64 hash; + + nfs_attr_check_mountpoint(sb, fattr); + + if (nfs_attr_use_mounted_on_fileid(fattr)) + fattr->fileid = fattr->mounted_on_fileid; + else if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0) + goto out_no_inode; + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) == 0) + goto out_no_inode; + + hash = fattr->fileid; + + inode = iget5_locked(sb, hash, nfs_find_actor, nfs_init_locked, &desc); + if (inode == NULL) { + inode = ERR_PTR(-ENOMEM); + goto out_no_inode; + } + + if (inode_state_read_once(inode) & I_NEW) { + struct nfs_inode *nfsi = NFS_I(inode); + unsigned long now = jiffies; + + /* We can't support update_atime(), since the server will reset it */ + inode->i_flags |= S_NOATIME|S_NOCMTIME; + inode->i_mode = fattr->mode; + nfsi->cache_validity = 0; + if ((fattr->valid & NFS_ATTR_FATTR_MODE) == 0 + && (fattr_supported & NFS_ATTR_FATTR_MODE)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MODE); + /* Why so? Because we want revalidate for devices/FIFOs, and + * that's precisely what we have in nfs_file_inode_operations. + */ + inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops; + if (S_ISREG(inode->i_mode)) { + inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops; + inode->i_data.a_ops = &nfs_file_aops; + nfs_inode_init_regular(nfsi); + mapping_set_large_folios(inode->i_mapping); + } else if (S_ISDIR(inode->i_mode)) { + inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops; + inode->i_fop = &nfs_dir_operations; + inode->i_data.a_ops = &nfs_dir_aops; + nfs_inode_init_dir(nfsi); + /* Deal with crossing mountpoints */ + if (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT || + fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) { + if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) + inode->i_op = &nfs_referral_inode_operations; + else + inode->i_op = &nfs_mountpoint_inode_operations; + inode->i_fop = NULL; + inode->i_flags |= S_AUTOMOUNT; + } + } else if (S_ISLNK(inode->i_mode)) { + inode->i_op = &nfs_symlink_inode_operations; + inode_nohighmem(inode); + } else + init_special_inode(inode, inode->i_mode, fattr->rdev); + + inode_set_atime(inode, 0, 0); + inode_set_mtime(inode, 0, 0); + inode_set_ctime(inode, 0, 0); + memset(&nfsi->btime, 0, sizeof(nfsi->btime)); + inode_set_iversion_raw(inode, 0); + inode->i_size = 0; + clear_nlink(inode); + inode->i_uid = make_kuid(&init_user_ns, -2); + inode->i_gid = make_kgid(&init_user_ns, -2); + inode->i_blocks = 0; + nfsi->write_io = 0; + nfsi->read_io = 0; + + nfsi->read_cache_jiffies = fattr->time_start; + nfsi->attr_gencount = fattr->gencount; + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + inode_set_atime_to_ts(inode, fattr->atime); + else if (fattr_supported & NFS_ATTR_FATTR_ATIME) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); + if (fattr->valid & NFS_ATTR_FATTR_MTIME) + inode_set_mtime_to_ts(inode, fattr->mtime); + else if (fattr_supported & NFS_ATTR_FATTR_MTIME) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode_set_ctime_to_ts(inode, fattr->ctime); + else if (fattr_supported & NFS_ATTR_FATTR_CTIME) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CTIME); + if (fattr->valid & NFS_ATTR_FATTR_BTIME) + nfsi->btime = fattr->btime; + else if (fattr_supported & NFS_ATTR_FATTR_BTIME) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BTIME); + if (fattr->valid & NFS_ATTR_FATTR_CHANGE) + inode_set_iversion_raw(inode, fattr->change_attr); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE); + if (fattr->valid & NFS_ATTR_FATTR_SIZE) + inode->i_size = nfs_size_to_loff_t(fattr->size); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_SIZE); + if (fattr->valid & NFS_ATTR_FATTR_NLINK) + set_nlink(inode, fattr->nlink); + else if (fattr_supported & NFS_ATTR_FATTR_NLINK) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_NLINK); + else + set_nlink(inode, 1); + if (fattr->valid & NFS_ATTR_FATTR_OWNER) + inode->i_uid = fattr->uid; + else if (fattr_supported & NFS_ATTR_FATTR_OWNER) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); + if (fattr->valid & NFS_ATTR_FATTR_GROUP) + inode->i_gid = fattr->gid; + else if (fattr_supported & NFS_ATTR_FATTR_GROUP) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); + if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) + inode->i_blocks = fattr->du.nfs2.blocks; + else if (fattr_supported & NFS_ATTR_FATTR_BLOCKS_USED && + fattr->size != 0) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { + /* + * report the blocks in 512byte units + */ + inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); + } else if (fattr_supported & NFS_ATTR_FATTR_SPACE_USED && + fattr->size != 0) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + + nfs_setsecurity(inode, fattr); + + nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); + nfsi->attrtimeo_timestamp = now; + nfsi->access_cache = RB_ROOT; + + nfs_fscache_init_inode(inode); + + unlock_new_inode(inode); + } else { + int err = nfs_refresh_inode(inode, fattr); + if (err < 0) { + iput(inode); + inode = ERR_PTR(err); + goto out_no_inode; + } + } + dprintk("NFS: nfs_fhget(%s/%Lu fh_crc=0x%08x ct=%d)\n", + inode->i_sb->s_id, + (unsigned long long)inode->i_ino, + nfs_display_fhandle_hash(fh), + icount_read_once(inode)); + +out: + return inode; + +out_no_inode: + dprintk("nfs_fhget: iget failed with error %ld\n", PTR_ERR(inode)); + goto out; +} +EXPORT_SYMBOL_GPL(nfs_fhget); + +static void +nfs_fattr_fixup_delegated(struct inode *inode, struct nfs_fattr *fattr) +{ + unsigned long cache_validity = NFS_I(inode)->cache_validity; + + if (nfs_have_delegated_mtime(inode)) { + if (!(cache_validity & NFS_INO_INVALID_CTIME)) + fattr->valid &= ~(NFS_ATTR_FATTR_PRECTIME | + NFS_ATTR_FATTR_CTIME); + + if (!(cache_validity & NFS_INO_INVALID_MTIME)) + fattr->valid &= ~(NFS_ATTR_FATTR_PREMTIME | + NFS_ATTR_FATTR_MTIME); + + if (!(cache_validity & NFS_INO_INVALID_ATIME)) + fattr->valid &= ~NFS_ATTR_FATTR_ATIME; + } else if (nfs_have_delegated_atime(inode)) { + if (!(cache_validity & NFS_INO_INVALID_ATIME)) + fattr->valid &= ~NFS_ATTR_FATTR_ATIME; + } +} + +static void nfs_set_timestamps_to_ts(struct inode *inode, struct iattr *attr) +{ + unsigned int cache_flags = 0; + + if (attr->ia_valid & ATTR_MTIME_SET) { + struct timespec64 ctime = inode_get_ctime(inode); + struct timespec64 mtime = inode_get_mtime(inode); + struct timespec64 now; + bool updated = false; + + now = inode_set_ctime_current(inode); + if (!timespec64_equal(&now, &ctime)) + updated = true; + + inode_set_mtime_to_ts(inode, attr->ia_mtime); + if (!timespec64_equal(&now, &mtime)) + updated = true; + + inode_maybe_inc_iversion(inode, updated); + cache_flags |= NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME; + } + if (attr->ia_valid & ATTR_ATIME_SET) { + inode_set_atime_to_ts(inode, attr->ia_atime); + cache_flags |= NFS_INO_INVALID_ATIME; + } + NFS_I(inode)->cache_validity &= ~cache_flags; +} + +static void nfs_update_atime(struct inode *inode) +{ + inode_update_time(inode, FS_UPD_ATIME, 0); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_ATIME; +} + +static void nfs_update_mtime(struct inode *inode) +{ + inode_update_time(inode, FS_UPD_CMTIME, 0); + NFS_I(inode)->cache_validity &= + ~(NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME); +} + +void nfs_update_delegated_atime(struct inode *inode) +{ + spin_lock(&inode->i_lock); + if (nfs_have_delegated_atime(inode)) + nfs_update_atime(inode); + spin_unlock(&inode->i_lock); +} + +void nfs_update_delegated_mtime_locked(struct inode *inode) +{ + if (nfs_have_delegated_mtime(inode) || + nfs_have_directory_delegation(inode)) + nfs_update_mtime(inode); +} + +void nfs_update_delegated_mtime(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_update_delegated_mtime_locked(inode); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_update_delegated_mtime); + +#define NFS_VALID_ATTRS (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_SIZE|ATTR_ATIME|ATTR_ATIME_SET|ATTR_MTIME|ATTR_MTIME_SET|ATTR_FILE|ATTR_OPEN) + +int +nfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, + struct iattr *attr) +{ + struct inode *inode = d_inode(dentry); + struct nfs_fattr *fattr; + loff_t oldsize; + int error = 0; + kuid_t task_uid = current_fsuid(); + kuid_t owner_uid = inode->i_uid; + + nfs_inc_stats(inode, NFSIOS_VFSSETATTR); + + /* skip mode change if it's just for clearing setuid/setgid */ + if (attr->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID)) + attr->ia_valid &= ~ATTR_MODE; + + if (S_ISREG(inode->i_mode)) + nfs_file_block_o_direct(NFS_I(inode)); + + oldsize = i_size_read(inode); + if (attr->ia_valid & ATTR_SIZE) { + BUG_ON(!S_ISREG(inode->i_mode)); + + error = inode_newsize_ok(inode, attr->ia_size); + if (error) + return error; + + if (attr->ia_size == oldsize) + attr->ia_valid &= ~ATTR_SIZE; + } + + if (nfs_have_delegated_mtime(inode) && attr->ia_valid & ATTR_MTIME) { + spin_lock(&inode->i_lock); + if (attr->ia_valid & ATTR_MTIME_SET) { + if (uid_eq(task_uid, owner_uid)) { + nfs_set_timestamps_to_ts(inode, attr); + attr->ia_valid &= ~(ATTR_MTIME|ATTR_MTIME_SET| + ATTR_ATIME|ATTR_ATIME_SET); + } + } else { + if (attr->ia_valid & ATTR_MTIME) + nfs_update_mtime(inode); + if (attr->ia_valid & ATTR_ATIME) + nfs_update_atime(inode); + attr->ia_valid &= ~(ATTR_MTIME|ATTR_ATIME); + } + spin_unlock(&inode->i_lock); + } else if (nfs_have_delegated_atime(inode) && + attr->ia_valid & ATTR_ATIME && + !(attr->ia_valid & ATTR_MTIME)) { + if (!(attr->ia_valid & ATTR_ATIME_SET)) { + nfs_update_delegated_atime(inode); + attr->ia_valid &= ~ATTR_ATIME; + } + } + + /* Optimization: if the end result is no change, don't RPC */ + if (((attr->ia_valid & NFS_VALID_ATTRS) & ~(ATTR_FILE|ATTR_OPEN)) == 0) + return 0; + + trace_nfs_setattr_enter(inode); + + /* Write all dirty data */ + if (S_ISREG(inode->i_mode)) + nfs_sync_inode(inode); + + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fattr == NULL) { + error = -ENOMEM; + goto out; + } + + error = NFS_PROTO(inode)->setattr(dentry, fattr, attr); + if (error == 0) { + if (attr->ia_valid & ATTR_SIZE) + nfs_truncate_last_folio(inode->i_mapping, oldsize, + attr->ia_size); + error = nfs_refresh_inode(inode, fattr); + } + nfs_free_fattr(fattr); +out: + trace_nfs_setattr_exit(inode, error); + return error; +} +EXPORT_SYMBOL_GPL(nfs_setattr); + +/** + * nfs_vmtruncate - unmap mappings "freed" by truncate() syscall + * @inode: inode of the file used + * @offset: file offset to start truncating + * + * This is a copy of the common vmtruncate, but with the locking + * corrected to take into account the fact that NFS requires + * inode->i_size to be updated under the inode->i_lock. + * Note: must be called with inode->i_lock held! + */ +static int nfs_vmtruncate(struct inode * inode, loff_t offset) +{ + int err; + + err = inode_newsize_ok(inode, offset); + if (err) + goto out; + + trace_nfs_size_truncate(inode, offset); + i_size_write(inode, offset); + /* Optimisation */ + if (offset == 0) { + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA; + nfs_ooo_clear(NFS_I(inode)); + } + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; + + spin_unlock(&inode->i_lock); + truncate_pagecache(inode, offset); + nfs_update_delegated_mtime_locked(inode); + spin_lock(&inode->i_lock); +out: + return err; +} + +/** + * nfs_setattr_update_inode - Update inode metadata after a setattr call. + * @inode: pointer to struct inode + * @attr: pointer to struct iattr + * @fattr: pointer to struct nfs_fattr + * + * Note: we do this in the *proc.c in order to ensure that + * it works for things like exclusive creates too. + */ +void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, + struct nfs_fattr *fattr) +{ + /* Barrier: bump the attribute generation count. */ + nfs_fattr_set_barrier(fattr); + + spin_lock(&inode->i_lock); + NFS_I(inode)->attr_gencount = fattr->gencount; + if ((attr->ia_valid & ATTR_SIZE) != 0) { + if (!nfs_have_delegated_mtime(inode)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC); + nfs_vmtruncate(inode, attr->ia_size); + } + if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) { + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_CTIME; + if ((attr->ia_valid & ATTR_KILL_SUID) != 0 && + inode->i_mode & S_ISUID) + inode->i_mode &= ~S_ISUID; + if (setattr_should_drop_sgid(&nop_mnt_idmap, inode)) + inode->i_mode &= ~S_ISGID; + if ((attr->ia_valid & ATTR_MODE) != 0) { + int mode = attr->ia_mode & S_IALLUGO; + mode |= inode->i_mode & ~S_IALLUGO; + inode->i_mode = mode; + } + if ((attr->ia_valid & ATTR_UID) != 0) + inode->i_uid = attr->ia_uid; + if ((attr->ia_valid & ATTR_GID) != 0) + inode->i_gid = attr->ia_gid; + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode_set_ctime_to_ts(inode, fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL); + } + if (attr->ia_valid & (ATTR_ATIME_SET|ATTR_ATIME)) { + NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_ATIME + | NFS_INO_INVALID_CTIME); + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + inode_set_atime_to_ts(inode, fattr->atime); + else if (attr->ia_valid & ATTR_ATIME_SET) + inode_set_atime_to_ts(inode, attr->ia_atime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); + + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode_set_ctime_to_ts(inode, fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); + } + if (attr->ia_valid & (ATTR_MTIME_SET|ATTR_MTIME)) { + NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_CTIME); + if (fattr->valid & NFS_ATTR_FATTR_MTIME) + inode_set_mtime_to_ts(inode, fattr->mtime); + else if (attr->ia_valid & ATTR_MTIME_SET) + inode_set_mtime_to_ts(inode, attr->ia_mtime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); + + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode_set_ctime_to_ts(inode, fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); + } + if (fattr->valid) + nfs_update_inode(inode, fattr); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_setattr_update_inode); + +/* + * Don't request help from readdirplus if the file is being written to, + * or if attribute caching is turned off + */ +static bool nfs_getattr_readdirplus_enable(const struct inode *inode) +{ + return nfs_server_capable(inode, NFS_CAP_READDIRPLUS) && + !nfs_have_writebacks(inode) && NFS_MAXATTRTIMEO(inode) > 5 * HZ; +} + +static void nfs_readdirplus_parent_cache_miss(struct dentry *dentry) +{ + if (!IS_ROOT(dentry)) { + struct dentry *parent = dget_parent(dentry); + nfs_readdir_record_entry_cache_miss(d_inode(parent)); + dput(parent); + } +} + +static void nfs_readdirplus_parent_cache_hit(struct dentry *dentry) +{ + if (!IS_ROOT(dentry)) { + struct dentry *parent = dget_parent(dentry); + nfs_readdir_record_entry_cache_hit(d_inode(parent)); + dput(parent); + } +} + +static u32 nfs_get_valid_attrmask(struct inode *inode) +{ + u64 fattr_valid = NFS_SERVER(inode)->fattr_valid; + unsigned long cache_validity = READ_ONCE(NFS_I(inode)->cache_validity); + u32 reply_mask = STATX_INO | STATX_TYPE; + + if (!(cache_validity & NFS_INO_INVALID_ATIME)) + reply_mask |= STATX_ATIME; + if (!(cache_validity & NFS_INO_INVALID_CTIME)) + reply_mask |= STATX_CTIME; + if (!(cache_validity & NFS_INO_INVALID_MTIME)) + reply_mask |= STATX_MTIME; + if (!(cache_validity & NFS_INO_INVALID_SIZE)) + reply_mask |= STATX_SIZE; + if (!(cache_validity & NFS_INO_INVALID_NLINK)) + reply_mask |= STATX_NLINK; + if (!(cache_validity & NFS_INO_INVALID_MODE)) + reply_mask |= STATX_MODE; + if (!(cache_validity & NFS_INO_INVALID_OTHER)) + reply_mask |= STATX_UID | STATX_GID; + if (!(cache_validity & NFS_INO_INVALID_BLOCKS)) + reply_mask |= STATX_BLOCKS; + if (!(cache_validity & NFS_INO_INVALID_BTIME) && + (fattr_valid & NFS_ATTR_FATTR_BTIME)) + reply_mask |= STATX_BTIME; + if (!(cache_validity & NFS_INO_INVALID_CHANGE)) + reply_mask |= STATX_CHANGE_COOKIE; + return reply_mask; +} + +int nfs_getattr(struct mnt_idmap *idmap, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) +{ + struct inode *inode = d_inode(path->dentry); + struct nfs_server *server = NFS_SERVER(inode); + u64 fattr_valid = server->fattr_valid; + unsigned long cache_validity; + int err = 0; + bool force_sync = query_flags & AT_STATX_FORCE_SYNC; + bool do_update = false; + bool readdirplus_enabled = nfs_getattr_readdirplus_enable(inode); + + trace_nfs_getattr_enter(inode); + + request_mask &= STATX_TYPE | STATX_MODE | STATX_NLINK | STATX_UID | + STATX_GID | STATX_ATIME | STATX_MTIME | STATX_CTIME | + STATX_INO | STATX_SIZE | STATX_BLOCKS | STATX_BTIME | + STATX_CHANGE_COOKIE; + + if (!(fattr_valid & NFS_ATTR_FATTR_BTIME)) + request_mask &= ~STATX_BTIME; + + if ((query_flags & AT_STATX_DONT_SYNC) && !force_sync) { + if (readdirplus_enabled) + nfs_readdirplus_parent_cache_hit(path->dentry); + goto out_no_revalidate; + } + + /* Flush out writes to the server in order to update c/mtime/version. */ + if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_CHANGE_COOKIE)) && + S_ISREG(inode->i_mode)) { + if (nfs_have_delegated_mtime(inode)) + filemap_fdatawrite(inode->i_mapping); + else + filemap_write_and_wait(inode->i_mapping); + } + + /* + * We may force a getattr if the user cares about atime. + * + * Note that we only have to check the vfsmount flags here: + * - NFS always sets S_NOATIME by so checking it would give a + * bogus result + * - NFS never sets SB_NOATIME or SB_NODIRATIME so there is + * no point in checking those. + */ + if ((path->mnt->mnt_flags & MNT_NOATIME) || + ((path->mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode))) + request_mask &= ~STATX_ATIME; + + /* Is the user requesting attributes that might need revalidation? */ + if (!(request_mask & (STATX_MODE|STATX_NLINK|STATX_ATIME|STATX_CTIME| + STATX_MTIME|STATX_UID|STATX_GID| + STATX_SIZE|STATX_BLOCKS|STATX_BTIME| + STATX_CHANGE_COOKIE))) + goto out_no_revalidate; + + /* Check whether the cached attributes are stale */ + do_update |= force_sync || nfs_attribute_cache_expired(inode); + cache_validity = READ_ONCE(NFS_I(inode)->cache_validity); + do_update |= cache_validity & NFS_INO_INVALID_CHANGE; + if (request_mask & STATX_ATIME) + do_update |= cache_validity & NFS_INO_INVALID_ATIME; + if (request_mask & STATX_CTIME) + do_update |= cache_validity & NFS_INO_INVALID_CTIME; + if (request_mask & STATX_MTIME) + do_update |= cache_validity & NFS_INO_INVALID_MTIME; + if (request_mask & STATX_SIZE) + do_update |= cache_validity & NFS_INO_INVALID_SIZE; + if (request_mask & STATX_NLINK) + do_update |= cache_validity & NFS_INO_INVALID_NLINK; + if (request_mask & STATX_MODE) + do_update |= cache_validity & NFS_INO_INVALID_MODE; + if (request_mask & (STATX_UID | STATX_GID)) + do_update |= cache_validity & NFS_INO_INVALID_OTHER; + if (request_mask & STATX_BLOCKS) + do_update |= cache_validity & NFS_INO_INVALID_BLOCKS; + if (request_mask & STATX_BTIME) + do_update |= cache_validity & NFS_INO_INVALID_BTIME; + + if (do_update) { + if (readdirplus_enabled) + nfs_readdirplus_parent_cache_miss(path->dentry); + err = __nfs_revalidate_inode(server, inode); + if (err) + goto out; + } else if (readdirplus_enabled) + nfs_readdirplus_parent_cache_hit(path->dentry); +out_no_revalidate: + /* Only return attributes that were revalidated. */ + stat->result_mask = nfs_get_valid_attrmask(inode) | request_mask; + + generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat); + stat->change_cookie = inode_peek_iversion_raw(inode); + stat->attributes_mask |= STATX_ATTR_CHANGE_MONOTONIC; + if (server->change_attr_type != NFS4_CHANGE_TYPE_IS_UNDEFINED) + stat->attributes |= STATX_ATTR_CHANGE_MONOTONIC; + if (S_ISDIR(inode->i_mode)) + stat->blksize = NFS_SERVER(inode)->dtsize; + stat->btime = NFS_I(inode)->btime; + + /* Special handling for STATX_DIOALIGN and STATX_DIO_READ_ALIGN + * - NFS doesn't have DIO alignment constraints, avoid getting + * these DIO attrs from remote and just respond with most + * accommodating limits (so client will issue supported DIO). + * - this is unintuitive, but the most coarse-grained + * dio_offset_align is the most accommodating. + */ + if ((request_mask & (STATX_DIOALIGN | STATX_DIO_READ_ALIGN)) && + S_ISREG(inode->i_mode)) { + stat->result_mask |= STATX_DIOALIGN | STATX_DIO_READ_ALIGN; + stat->dio_mem_align = 4; /* 4-byte alignment */ + stat->dio_offset_align = PAGE_SIZE; + stat->dio_read_offset_align = stat->dio_offset_align; + } +out: + trace_nfs_getattr_exit(inode, err); + return err; +} +EXPORT_SYMBOL_GPL(nfs_getattr); + +int nfs_fileattr_get(struct dentry *dentry, struct file_kattr *fa) +{ + struct inode *inode = d_inode(dentry); + + if (nfs_server_capable(inode, NFS_CAP_CASE_INSENSITIVE)) { + fa->fsx_xflags |= FS_XFLAG_CASEFOLD; + fa->flags |= FS_CASEFOLD_FL; + } + if (nfs_server_capable(inode, NFS_CAP_CASE_NONPRESERVING)) + fa->fsx_xflags |= FS_XFLAG_CASENONPRESERVING; + return 0; +} +EXPORT_SYMBOL_GPL(nfs_fileattr_get); + +static void nfs_init_lock_context(struct nfs_lock_context *l_ctx) +{ + refcount_set(&l_ctx->count, 1); + l_ctx->lockowner = current->files; + INIT_LIST_HEAD(&l_ctx->list); + atomic_set(&l_ctx->io_count, 0); +} + +static struct nfs_lock_context *__nfs_find_lock_context(struct nfs_open_context *ctx) +{ + struct nfs_lock_context *pos; + + list_for_each_entry_rcu(pos, &ctx->lock_context.list, list) { + if (pos->lockowner != current->files) + continue; + if (refcount_inc_not_zero(&pos->count)) + return pos; + } + return NULL; +} + +struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx) +{ + struct nfs_lock_context *res, *new = NULL; + struct inode *inode = d_inode(ctx->dentry); + + rcu_read_lock(); + res = __nfs_find_lock_context(ctx); + rcu_read_unlock(); + if (res == NULL) { + new = kmalloc_obj(*new, GFP_KERNEL_ACCOUNT); + if (new == NULL) + return ERR_PTR(-ENOMEM); + nfs_init_lock_context(new); + spin_lock(&inode->i_lock); + res = __nfs_find_lock_context(ctx); + if (res == NULL) { + new->open_context = get_nfs_open_context(ctx); + if (new->open_context) { + list_add_tail_rcu(&new->list, + &ctx->lock_context.list); + res = new; + new = NULL; + } else + res = ERR_PTR(-EBADF); + } + spin_unlock(&inode->i_lock); + kfree(new); + } + return res; +} +EXPORT_SYMBOL_GPL(nfs_get_lock_context); + +void nfs_put_lock_context(struct nfs_lock_context *l_ctx) +{ + struct nfs_open_context *ctx = l_ctx->open_context; + struct inode *inode = d_inode(ctx->dentry); + + if (!refcount_dec_and_lock(&l_ctx->count, &inode->i_lock)) + return; + list_del_rcu(&l_ctx->list); + spin_unlock(&inode->i_lock); + put_nfs_open_context(ctx); + kfree_rcu(l_ctx, rcu_head); +} +EXPORT_SYMBOL_GPL(nfs_put_lock_context); + +/** + * nfs_close_context - Common close_context() routine NFSv2/v3 + * @ctx: pointer to context + * @is_sync: is this a synchronous close + * + * Ensure that the attributes are up to date if we're mounted + * with close-to-open semantics and we have cached data that will + * need to be revalidated on open. + */ +void nfs_close_context(struct nfs_open_context *ctx, int is_sync) +{ + struct nfs_inode *nfsi; + struct inode *inode; + + if (!(ctx->mode & FMODE_WRITE)) + return; + if (!is_sync) + return; + inode = d_inode(ctx->dentry); + if (nfs_have_read_or_write_delegation(inode)) + return; + nfsi = NFS_I(inode); + if (inode->i_mapping->nrpages == 0) + return; + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + return; + if (!list_empty(&nfsi->open_files)) + return; + if (NFS_SERVER(inode)->flags & NFS_MOUNT_NOCTO) + return; + nfs_revalidate_inode(inode, + NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE); +} +EXPORT_SYMBOL_GPL(nfs_close_context); + +struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, + fmode_t f_mode, + struct file *filp) +{ + struct nfs_open_context *ctx; + + ctx = kmalloc_obj(*ctx, GFP_KERNEL_ACCOUNT); + if (!ctx) + return ERR_PTR(-ENOMEM); + nfs_sb_active(dentry->d_sb); + ctx->dentry = dget(dentry); + if (filp) + ctx->cred = get_cred(filp->f_cred); + else + ctx->cred = get_current_cred(); + rcu_assign_pointer(ctx->ll_cred, NULL); + ctx->state = NULL; + ctx->mode = f_mode; + ctx->flags = 0; + ctx->error = 0; + ctx->flock_owner = (fl_owner_t)filp; + nfs_init_lock_context(&ctx->lock_context); + ctx->lock_context.open_context = ctx; + INIT_LIST_HEAD(&ctx->list); + ctx->mdsthreshold = NULL; + nfs_localio_file_init(&ctx->nfl); + + return ctx; +} +EXPORT_SYMBOL_GPL(alloc_nfs_open_context); + +struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx) +{ + if (ctx != NULL && refcount_inc_not_zero(&ctx->lock_context.count)) + return ctx; + return NULL; +} +EXPORT_SYMBOL_GPL(get_nfs_open_context); + +static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync) +{ + struct inode *inode = d_inode(ctx->dentry); + struct super_block *sb = ctx->dentry->d_sb; + + if (!refcount_dec_and_test(&ctx->lock_context.count)) + return; + if (!list_empty(&ctx->list)) { + spin_lock(&inode->i_lock); + list_del_rcu(&ctx->list); + spin_unlock(&inode->i_lock); + } + if (inode != NULL) + NFS_PROTO(inode)->close_context(ctx, is_sync); + put_cred(ctx->cred); + dput(ctx->dentry); + nfs_sb_deactive(sb); + put_rpccred(rcu_dereference_protected(ctx->ll_cred, 1)); + kfree(ctx->mdsthreshold); + nfs_close_local_fh(&ctx->nfl); + kfree_rcu(ctx, rcu_head); +} + +void put_nfs_open_context(struct nfs_open_context *ctx) +{ + __put_nfs_open_context(ctx, 0); +} +EXPORT_SYMBOL_GPL(put_nfs_open_context); + +static void put_nfs_open_context_sync(struct nfs_open_context *ctx) +{ + __put_nfs_open_context(ctx, 1); +} + +/* + * Ensure that mmap has a recent RPC credential for use when writing out + * shared pages + */ +void nfs_inode_attach_open_context(struct nfs_open_context *ctx) +{ + struct inode *inode = d_inode(ctx->dentry); + struct nfs_inode *nfsi = NFS_I(inode); + + spin_lock(&inode->i_lock); + if (list_empty(&nfsi->open_files) && + nfs_ooo_test(nfsi)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); + list_add_tail_rcu(&ctx->list, &nfsi->open_files); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_inode_attach_open_context); + +void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx) +{ + filp->private_data = get_nfs_open_context(ctx); + set_bit(NFS_CONTEXT_FILE_OPEN, &ctx->flags); + if (list_empty(&ctx->list)) + nfs_inode_attach_open_context(ctx); +} +EXPORT_SYMBOL_GPL(nfs_file_set_open_context); + +/* + * Given an inode, search for an open context with the desired characteristics + */ +struct nfs_open_context *nfs_find_open_context(struct inode *inode, const struct cred *cred, fmode_t mode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_open_context *pos, *ctx = NULL; + + rcu_read_lock(); + list_for_each_entry_rcu(pos, &nfsi->open_files, list) { + if (cred != NULL && cred_fscmp(pos->cred, cred) != 0) + continue; + if ((pos->mode & (FMODE_READ|FMODE_WRITE)) != mode) + continue; + if (!test_bit(NFS_CONTEXT_FILE_OPEN, &pos->flags)) + continue; + ctx = get_nfs_open_context(pos); + if (ctx) + break; + } + rcu_read_unlock(); + return ctx; +} + +void nfs_file_clear_open_context(struct file *filp) +{ + struct nfs_open_context *ctx = nfs_file_open_context(filp); + + if (ctx) { + struct inode *inode = d_inode(ctx->dentry); + + clear_bit(NFS_CONTEXT_FILE_OPEN, &ctx->flags); + /* + * We fatal error on write before. Try to writeback + * every page again. + */ + if (ctx->error < 0) + invalidate_inode_pages2(inode->i_mapping); + filp->private_data = NULL; + put_nfs_open_context_sync(ctx); + } +} + +/* + * These allocate and release file read/write context information. + */ +int nfs_open(struct inode *inode, struct file *filp) +{ + struct nfs_open_context *ctx; + + ctx = alloc_nfs_open_context(file_dentry(filp), + flags_to_mode(filp->f_flags), filp); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + nfs_file_set_open_context(filp, ctx); + put_nfs_open_context(ctx); + nfs_fscache_open_file(inode, filp); + return 0; +} + +/* + * This function is called whenever some part of NFS notices that + * the cached attributes have to be refreshed. + */ +int +__nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) +{ + int status = -ESTALE; + struct nfs_fattr *fattr = NULL; + struct nfs_inode *nfsi = NFS_I(inode); + + dfprintk(PAGECACHE, "NFS: revalidating (%s/%Lu)\n", + inode->i_sb->s_id, (unsigned long long)inode->i_ino); + + trace_nfs_revalidate_inode_enter(inode); + + if (is_bad_inode(inode)) + goto out; + if (NFS_STALE(inode)) + goto out; + + /* pNFS: Attributes aren't updated until we layoutcommit */ + if (S_ISREG(inode->i_mode)) { + status = pnfs_sync_inode(inode, false); + if (status) + goto out; + } else if (nfs_have_directory_delegation(inode) && + !(NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATTR)) { + status = 0; + goto out; + } + + status = -ENOMEM; + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fattr == NULL) + goto out; + + nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); + + status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, inode); + if (status != 0) { + dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) getattr failed, error=%d\n", + inode->i_sb->s_id, + (unsigned long long)inode->i_ino, status); + switch (status) { + case -ETIMEDOUT: + /* A soft timeout occurred. Use cached information? */ + if (server->flags & NFS_MOUNT_SOFTREVAL) + status = 0; + break; + case -ESTALE: + if (!S_ISDIR(inode->i_mode)) + nfs_set_inode_stale(inode); + else + nfs_zap_caches(inode); + } + goto out; + } + + status = nfs_refresh_inode(inode, fattr); + if (status) { + dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) refresh failed, error=%d\n", + inode->i_sb->s_id, + (unsigned long long)inode->i_ino, status); + goto out; + } + + if (nfsi->cache_validity & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); + + nfs_setsecurity(inode, fattr); + + dfprintk(PAGECACHE, "NFS: (%s/%Lu) revalidation complete\n", + inode->i_sb->s_id, + (unsigned long long)inode->i_ino); + +out: + nfs_free_fattr(fattr); + trace_nfs_revalidate_inode_exit(inode, status); + return status; +} + +int nfs_attribute_cache_expired(struct inode *inode) +{ + if (nfs_have_delegated_attributes(inode)) + return 0; + return nfs_attribute_timeout(inode); +} + +/** + * nfs_revalidate_inode - Revalidate the inode attributes + * @inode: pointer to inode struct + * @flags: cache flags to check + * + * Updates inode attribute information by retrieving the data from the server. + */ +int nfs_revalidate_inode(struct inode *inode, unsigned long flags) +{ + if (!nfs_check_cache_invalid(inode, flags)) + return NFS_STALE(inode) ? -ESTALE : 0; + return __nfs_revalidate_inode(NFS_SERVER(inode), inode); +} +EXPORT_SYMBOL_GPL(nfs_revalidate_inode); + +static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping) +{ + int ret; + + nfs_fscache_invalidate(inode, 0); + if (mapping->nrpages != 0) { + if (S_ISREG(inode->i_mode)) { + ret = nfs_sync_mapping(mapping); + if (ret < 0) + return ret; + } + ret = invalidate_inode_pages2(mapping); + if (ret < 0) + return ret; + } + nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); + + dfprintk(PAGECACHE, "NFS: (%s/%Lu) data cache invalidated\n", + inode->i_sb->s_id, + (unsigned long long)inode->i_ino); + return 0; +} + +/** + * nfs_clear_invalid_mapping - Conditionally clear a mapping + * @mapping: pointer to mapping + * + * If the NFS_INO_INVALID_DATA inode flag is set, clear the mapping. + */ +int nfs_clear_invalid_mapping(struct address_space *mapping) +{ + struct inode *inode = mapping->host; + struct nfs_inode *nfsi = NFS_I(inode); + unsigned long *bitlock = &nfsi->flags; + int ret = 0; + + /* + * We must clear NFS_INO_INVALID_DATA first to ensure that + * invalidations that come in while we're shooting down the mappings + * are respected. But, that leaves a race window where one revalidator + * can clear the flag, and then another checks it before the mapping + * gets invalidated. Fix that by serializing access to this part of + * the function. + * + * At the same time, we need to allow other tasks to see whether we + * might be in the middle of invalidating the pages, so we only set + * the bit lock here if it looks like we're going to be doing that. + */ + for (;;) { + ret = wait_on_bit_action(bitlock, NFS_INO_INVALIDATING, + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + if (ret) + goto out; + smp_rmb(); /* pairs with smp_wmb() below */ + if (test_bit(NFS_INO_INVALIDATING, bitlock)) + continue; + /* pairs with nfs_set_cache_invalid()'s smp_store_release() */ + if (!(smp_load_acquire(&nfsi->cache_validity) & NFS_INO_INVALID_DATA)) + goto out; + /* Slow-path that double-checks with spinlock held */ + spin_lock(&inode->i_lock); + if (test_bit(NFS_INO_INVALIDATING, bitlock)) { + spin_unlock(&inode->i_lock); + continue; + } + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + break; + spin_unlock(&inode->i_lock); + goto out; + } + + set_bit(NFS_INO_INVALIDATING, bitlock); + smp_wmb(); + nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; + nfs_ooo_clear(nfsi); + spin_unlock(&inode->i_lock); + trace_nfs_invalidate_mapping_enter(inode); + ret = nfs_invalidate_mapping(inode, mapping); + trace_nfs_invalidate_mapping_exit(inode, ret); + + clear_bit_unlock(NFS_INO_INVALIDATING, bitlock); + smp_mb__after_atomic(); + wake_up_bit(bitlock, NFS_INO_INVALIDATING); +out: + return ret; +} + +bool nfs_mapping_need_revalidate_inode(struct inode *inode) +{ + return nfs_check_cache_invalid(inode, NFS_INO_INVALID_CHANGE) || + NFS_STALE(inode); +} + +int nfs_revalidate_mapping_rcu(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + unsigned long *bitlock = &nfsi->flags; + int ret = 0; + + if (IS_SWAPFILE(inode)) + goto out; + if (nfs_mapping_need_revalidate_inode(inode)) { + ret = -ECHILD; + goto out; + } + spin_lock(&inode->i_lock); + if (test_bit(NFS_INO_INVALIDATING, bitlock) || + (nfsi->cache_validity & NFS_INO_INVALID_DATA)) + ret = -ECHILD; + spin_unlock(&inode->i_lock); +out: + return ret; +} + +/** + * nfs_revalidate_mapping - Revalidate the pagecache + * @inode: pointer to host inode + * @mapping: pointer to mapping + */ +int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) +{ + /* swapfiles are not supposed to be shared. */ + if (IS_SWAPFILE(inode)) + return 0; + + if (nfs_mapping_need_revalidate_inode(inode)) { + int ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + return ret; + } + + return nfs_clear_invalid_mapping(mapping); +} + +static bool nfs_file_has_writers(struct nfs_inode *nfsi) +{ + struct inode *inode = &nfsi->vfs_inode; + + if (!S_ISREG(inode->i_mode)) + return false; + if (list_empty(&nfsi->open_files)) + return false; + return inode_is_open_for_write(inode); +} + +static bool nfs_file_has_buffered_writers(struct nfs_inode *nfsi) +{ + return nfs_file_has_writers(nfsi) && nfs_file_io_is_buffered(nfsi); +} + +static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + struct timespec64 ts; + + if ((fattr->valid & NFS_ATTR_FATTR_PRECHANGE) + && (fattr->valid & NFS_ATTR_FATTR_CHANGE) + && inode_eq_iversion_raw(inode, fattr->pre_change_attr)) { + inode_set_iversion_raw(inode, fattr->change_attr); + if (S_ISDIR(inode->i_mode)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); + else if (nfs_server_capable(inode, NFS_CAP_XATTR)) + nfs_set_cache_invalid(inode, NFS_INO_INVALID_XATTR); + } + /* If we have atomic WCC data, we may update some attributes */ + ts = inode_get_ctime(inode); + if ((fattr->valid & NFS_ATTR_FATTR_PRECTIME) + && (fattr->valid & NFS_ATTR_FATTR_CTIME) + && timespec64_equal(&ts, &fattr->pre_ctime)) { + inode_set_ctime_to_ts(inode, fattr->ctime); + } + + ts = inode_get_mtime(inode); + if ((fattr->valid & NFS_ATTR_FATTR_PREMTIME) + && (fattr->valid & NFS_ATTR_FATTR_MTIME) + && timespec64_equal(&ts, &fattr->pre_mtime)) { + inode_set_mtime_to_ts(inode, fattr->mtime); + } + if ((fattr->valid & NFS_ATTR_FATTR_PRESIZE) + && (fattr->valid & NFS_ATTR_FATTR_SIZE) + && i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) + && !nfs_have_writebacks(inode)) { + trace_nfs_size_wcc(inode, fattr->size); + i_size_write(inode, nfs_size_to_loff_t(fattr->size)); + } +} + +/** + * nfs_check_inode_attributes - verify consistency of the inode attribute cache + * @inode: pointer to inode + * @fattr: updated attributes + * + * Verifies the attribute cache. If we have just changed the attributes, + * so that fattr carries weak cache consistency data, then it may + * also update the ctime/mtime/change_attribute. + */ +static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fattr) +{ + struct nfs_inode *nfsi = NFS_I(inode); + loff_t cur_size, new_isize; + unsigned long invalid = 0; + struct timespec64 ts; + + if (nfs_have_delegated_attributes(inode)) + return 0; + + if (!(fattr->valid & NFS_ATTR_FATTR_FILEID)) { + /* Only a mounted-on-fileid? Just exit */ + if (fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) + return 0; + /* Has the inode gone and changed behind our back? */ + } else if (inode->i_ino != fattr->fileid) { + /* Is this perhaps the mounted-on fileid? */ + if ((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) && + inode->i_ino == fattr->mounted_on_fileid) + return 0; + return -ESTALE; + } + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) && inode_wrong_type(inode, fattr->mode)) + return -ESTALE; + + + if (!nfs_file_has_buffered_writers(nfsi)) { + /* Verify a few of the more important attributes */ + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && !inode_eq_iversion_raw(inode, fattr->change_attr)) + invalid |= NFS_INO_INVALID_CHANGE; + + ts = inode_get_mtime(inode); + if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec64_equal(&ts, &fattr->mtime)) + invalid |= NFS_INO_INVALID_MTIME; + + ts = inode_get_ctime(inode); + if ((fattr->valid & NFS_ATTR_FATTR_CTIME) && !timespec64_equal(&ts, &fattr->ctime)) + invalid |= NFS_INO_INVALID_CTIME; + + if (fattr->valid & NFS_ATTR_FATTR_SIZE) { + cur_size = i_size_read(inode); + new_isize = nfs_size_to_loff_t(fattr->size); + if (cur_size != new_isize) + invalid |= NFS_INO_INVALID_SIZE; + } + } + + /* Have any file permissions changed? */ + if ((fattr->valid & NFS_ATTR_FATTR_MODE) && (inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) + invalid |= NFS_INO_INVALID_MODE; + if ((fattr->valid & NFS_ATTR_FATTR_OWNER) && !uid_eq(inode->i_uid, fattr->uid)) + invalid |= NFS_INO_INVALID_OTHER; + if ((fattr->valid & NFS_ATTR_FATTR_GROUP) && !gid_eq(inode->i_gid, fattr->gid)) + invalid |= NFS_INO_INVALID_OTHER; + + /* Has the link count changed? */ + if ((fattr->valid & NFS_ATTR_FATTR_NLINK) && inode->i_nlink != fattr->nlink) + invalid |= NFS_INO_INVALID_NLINK; + + ts = inode_get_atime(inode); + if ((fattr->valid & NFS_ATTR_FATTR_ATIME) && !timespec64_equal(&ts, &fattr->atime)) + invalid |= NFS_INO_INVALID_ATIME; + + if (invalid != 0) + nfs_set_cache_invalid(inode, invalid); + + nfsi->read_cache_jiffies = fattr->time_start; + return 0; +} + +static atomic_long_t nfs_attr_generation_counter; + +static unsigned long nfs_read_attr_generation_counter(void) +{ + return atomic_long_read(&nfs_attr_generation_counter); +} + +unsigned long nfs_inc_attr_generation_counter(void) +{ + return atomic_long_inc_return(&nfs_attr_generation_counter); +} +EXPORT_SYMBOL_GPL(nfs_inc_attr_generation_counter); + +void nfs_fattr_init(struct nfs_fattr *fattr) +{ + fattr->valid = 0; + fattr->time_start = jiffies; + fattr->gencount = nfs_inc_attr_generation_counter(); + fattr->owner_name = NULL; + fattr->group_name = NULL; + fattr->mdsthreshold = NULL; +} +EXPORT_SYMBOL_GPL(nfs_fattr_init); + +/** + * nfs_fattr_set_barrier + * @fattr: attributes + * + * Used to set a barrier after an attribute was updated. This + * barrier ensures that older attributes from RPC calls that may + * have raced with our update cannot clobber these new values. + * Note that you are still responsible for ensuring that other + * operations which change the attribute on the server do not + * collide. + */ +void nfs_fattr_set_barrier(struct nfs_fattr *fattr) +{ + fattr->gencount = nfs_inc_attr_generation_counter(); +} + +struct nfs_fattr *nfs_alloc_fattr(void) +{ + struct nfs_fattr *fattr; + + fattr = kmalloc_obj(*fattr); + if (fattr != NULL) { + nfs_fattr_init(fattr); + fattr->label = NULL; + } + return fattr; +} +EXPORT_SYMBOL_GPL(nfs_alloc_fattr); + +struct nfs_fattr *nfs_alloc_fattr_with_label(struct nfs_server *server) +{ + struct nfs_fattr *fattr = nfs_alloc_fattr(); + + if (!fattr) + return NULL; + + fattr->label = nfs4_label_alloc(server, GFP_KERNEL); + if (IS_ERR(fattr->label)) { + kfree(fattr); + return NULL; + } + + return fattr; +} +EXPORT_SYMBOL_GPL(nfs_alloc_fattr_with_label); + +struct nfs_fh *nfs_alloc_fhandle(void) +{ + struct nfs_fh *fh; + + fh = kmalloc_obj(struct nfs_fh); + if (fh != NULL) + fh->size = 0; + return fh; +} +EXPORT_SYMBOL_GPL(nfs_alloc_fhandle); + +#ifdef NFS_DEBUG +/* + * _nfs_display_fhandle_hash - calculate the crc32 hash for the filehandle + * in the same way that wireshark does + * + * @fh: file handle + * + * For debugging only. + */ +u32 _nfs_display_fhandle_hash(const struct nfs_fh *fh) +{ + /* wireshark uses 32-bit AUTODIN crc and does a bitwise + * not on the result */ + return nfs_fhandle_hash(fh); +} +EXPORT_SYMBOL_GPL(_nfs_display_fhandle_hash); + +/* + * _nfs_display_fhandle - display an NFS file handle on the console + * + * @fh: file handle to display + * @caption: display caption + * + * For debugging only. + */ +void _nfs_display_fhandle(const struct nfs_fh *fh, const char *caption) +{ + unsigned short i; + + if (fh == NULL || fh->size == 0) { + printk(KERN_DEFAULT "%s at %p is empty\n", caption, fh); + return; + } + + printk(KERN_DEFAULT "%s at %p is %u bytes, crc: 0x%08x:\n", + caption, fh, fh->size, _nfs_display_fhandle_hash(fh)); + for (i = 0; i < fh->size; i += 16) { + __be32 *pos = (__be32 *)&fh->data[i]; + + switch ((fh->size - i - 1) >> 2) { + case 0: + printk(KERN_DEFAULT " %08x\n", + be32_to_cpup(pos)); + break; + case 1: + printk(KERN_DEFAULT " %08x %08x\n", + be32_to_cpup(pos), be32_to_cpup(pos + 1)); + break; + case 2: + printk(KERN_DEFAULT " %08x %08x %08x\n", + be32_to_cpup(pos), be32_to_cpup(pos + 1), + be32_to_cpup(pos + 2)); + break; + default: + printk(KERN_DEFAULT " %08x %08x %08x %08x\n", + be32_to_cpup(pos), be32_to_cpup(pos + 1), + be32_to_cpup(pos + 2), be32_to_cpup(pos + 3)); + } + } +} +EXPORT_SYMBOL_GPL(_nfs_display_fhandle); +#endif + +/** + * nfs_inode_attrs_cmp_generic - compare attributes + * @fattr: attributes + * @inode: pointer to inode + * + * Attempt to divine whether or not an RPC call reply carrying stale + * attributes got scheduled after another call carrying updated ones. + * Note also the check for wraparound of 'attr_gencount' + * + * The function returns '1' if it thinks the attributes in @fattr are + * more recent than the ones cached in @inode. Otherwise it returns + * the value '0'. + */ +static int nfs_inode_attrs_cmp_generic(const struct nfs_fattr *fattr, + const struct inode *inode) +{ + unsigned long attr_gencount = NFS_I(inode)->attr_gencount; + + return (long)(fattr->gencount - attr_gencount) > 0 || + (long)(attr_gencount - nfs_read_attr_generation_counter()) > 0; +} + +/** + * nfs_inode_attrs_cmp_monotonic - compare attributes + * @fattr: attributes + * @inode: pointer to inode + * + * Attempt to divine whether or not an RPC call reply carrying stale + * attributes got scheduled after another call carrying updated ones. + * + * We assume that the server observes monotonic semantics for + * the change attribute, so a larger value means that the attributes in + * @fattr are more recent, in which case the function returns the + * value '1'. + * A return value of '0' indicates no measurable change + * A return value of '-1' means that the attributes in @inode are + * more recent. + */ +static int nfs_inode_attrs_cmp_monotonic(const struct nfs_fattr *fattr, + const struct inode *inode) +{ + s64 diff = fattr->change_attr - inode_peek_iversion_raw(inode); + if (diff > 0) + return 1; + return diff == 0 ? 0 : -1; +} + +/** + * nfs_inode_attrs_cmp_strict_monotonic - compare attributes + * @fattr: attributes + * @inode: pointer to inode + * + * Attempt to divine whether or not an RPC call reply carrying stale + * attributes got scheduled after another call carrying updated ones. + * + * We assume that the server observes strictly monotonic semantics for + * the change attribute, so a larger value means that the attributes in + * @fattr are more recent, in which case the function returns the + * value '1'. + * A return value of '-1' means that the attributes in @inode are + * more recent or unchanged. + */ +static int nfs_inode_attrs_cmp_strict_monotonic(const struct nfs_fattr *fattr, + const struct inode *inode) +{ + return nfs_inode_attrs_cmp_monotonic(fattr, inode) > 0 ? 1 : -1; +} + +/** + * nfs_inode_attrs_cmp - compare attributes + * @fattr: attributes + * @inode: pointer to inode + * + * This function returns '1' if it thinks the attributes in @fattr are + * more recent than the ones cached in @inode. It returns '-1' if + * the attributes in @inode are more recent than the ones in @fattr, + * and it returns 0 if not sure. + */ +static int nfs_inode_attrs_cmp(const struct nfs_fattr *fattr, + const struct inode *inode) +{ + if (nfs_inode_attrs_cmp_generic(fattr, inode) > 0) + return 1; + switch (NFS_SERVER(inode)->change_attr_type) { + case NFS4_CHANGE_TYPE_IS_UNDEFINED: + break; + case NFS4_CHANGE_TYPE_IS_TIME_METADATA: + if (!(fattr->valid & NFS_ATTR_FATTR_CHANGE)) + break; + return nfs_inode_attrs_cmp_monotonic(fattr, inode); + default: + if (!(fattr->valid & NFS_ATTR_FATTR_CHANGE)) + break; + return nfs_inode_attrs_cmp_strict_monotonic(fattr, inode); + } + return 0; +} + +/** + * nfs_inode_finish_partial_attr_update - complete a previous inode update + * @fattr: attributes + * @inode: pointer to inode + * + * Returns '1' if the last attribute update left the inode cached + * attributes in a partially unrevalidated state, and @fattr + * matches the change attribute of that partial update. + * Otherwise returns '0'. + */ +static int nfs_inode_finish_partial_attr_update(const struct nfs_fattr *fattr, + const struct inode *inode) +{ + const unsigned long check_valid = + NFS_INO_INVALID_ATIME | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_MTIME | NFS_INO_INVALID_SIZE | + NFS_INO_INVALID_BLOCKS | NFS_INO_INVALID_OTHER | + NFS_INO_INVALID_NLINK | NFS_INO_INVALID_BTIME; + unsigned long cache_validity = NFS_I(inode)->cache_validity; + enum nfs4_change_attr_type ctype = NFS_SERVER(inode)->change_attr_type; + + if (ctype != NFS4_CHANGE_TYPE_IS_UNDEFINED && + !(cache_validity & NFS_INO_INVALID_CHANGE) && + (cache_validity & check_valid) != 0 && + (fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && + nfs_inode_attrs_cmp_monotonic(fattr, inode) == 0) + return 1; + return 0; +} + +static void nfs_ooo_merge(struct nfs_inode *nfsi, + u64 start, u64 end) +{ + int i, cnt; + + if (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER) + /* No point merging anything */ + return; + + if (!nfsi->ooo) { + nfsi->ooo = kmalloc_obj(*nfsi->ooo, GFP_ATOMIC); + if (!nfsi->ooo) { + nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER; + return; + } + nfsi->ooo->cnt = 0; + } + + /* add this range, merging if possible */ + cnt = nfsi->ooo->cnt; + for (i = 0; i < cnt; i++) { + if (end == nfsi->ooo->gap[i].start) + end = nfsi->ooo->gap[i].end; + else if (start == nfsi->ooo->gap[i].end) + start = nfsi->ooo->gap[i].start; + else + continue; + /* Remove 'i' from table and loop to insert the new range */ + cnt -= 1; + nfsi->ooo->gap[i] = nfsi->ooo->gap[cnt]; + i = -1; + } + if (start != end) { + if (cnt >= ARRAY_SIZE(nfsi->ooo->gap)) { + nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER; + kfree(nfsi->ooo); + nfsi->ooo = NULL; + return; + } + nfsi->ooo->gap[cnt].start = start; + nfsi->ooo->gap[cnt].end = end; + cnt += 1; + } + nfsi->ooo->cnt = cnt; +} + +static void nfs_ooo_record(struct nfs_inode *nfsi, + struct nfs_fattr *fattr) +{ + /* This reply was out-of-order, so record in the + * pre/post change id, possibly cancelling + * gaps created when iversion was jumpped forward. + */ + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) && + (fattr->valid & NFS_ATTR_FATTR_PRECHANGE)) + nfs_ooo_merge(nfsi, + fattr->change_attr, + fattr->pre_change_attr); +} + +static int nfs_refresh_inode_locked(struct inode *inode, + struct nfs_fattr *fattr) +{ + int attr_cmp = nfs_inode_attrs_cmp(fattr, inode); + int ret = 0; + + trace_nfs_refresh_inode_enter(inode); + + if (attr_cmp > 0 || nfs_inode_finish_partial_attr_update(fattr, inode)) + ret = nfs_update_inode(inode, fattr); + else { + nfs_ooo_record(NFS_I(inode), fattr); + + if (attr_cmp == 0) + ret = nfs_check_inode_attributes(inode, fattr); + } + + trace_nfs_refresh_inode_exit(inode, ret); + return ret; +} + +/** + * nfs_refresh_inode - try to update the inode attribute cache + * @inode: pointer to inode + * @fattr: updated attributes + * + * Check that an RPC call that returned attributes has not overlapped with + * other recent updates of the inode metadata, then decide whether it is + * safe to do a full update of the inode attributes, or whether just to + * call nfs_check_inode_attributes. + */ +int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + int status; + + if ((fattr->valid & NFS_ATTR_FATTR) == 0) + return 0; + spin_lock(&inode->i_lock); + status = nfs_refresh_inode_locked(inode, fattr); + spin_unlock(&inode->i_lock); + + return status; +} +EXPORT_SYMBOL_GPL(nfs_refresh_inode); + +static int nfs_post_op_update_inode_locked(struct inode *inode, + struct nfs_fattr *fattr, unsigned int invalid) +{ + if (S_ISDIR(inode->i_mode)) + invalid |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, invalid); + if ((fattr->valid & NFS_ATTR_FATTR) == 0) + return 0; + return nfs_refresh_inode_locked(inode, fattr); +} + +/** + * nfs_post_op_update_inode - try to update the inode attribute cache + * @inode: pointer to inode + * @fattr: updated attributes + * + * After an operation that has changed the inode metadata, mark the + * attribute cache as being invalid, then try to update it. + * + * NB: if the server didn't return any post op attributes, this + * function will force the retrieval of attributes before the next + * NFS request. Thus it should be used only for operations that + * are expected to change one or more attributes, to avoid + * unnecessary NFS requests and trips through nfs_update_inode(). + */ +int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + int status; + + spin_lock(&inode->i_lock); + nfs_fattr_set_barrier(fattr); + status = nfs_post_op_update_inode_locked(inode, fattr, + NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_REVAL_FORCED); + spin_unlock(&inode->i_lock); + + return status; +} +EXPORT_SYMBOL_GPL(nfs_post_op_update_inode); + +/** + * nfs_post_op_update_inode_force_wcc_locked - update the inode attribute cache + * @inode: pointer to inode + * @fattr: updated attributes + * + * After an operation that has changed the inode metadata, mark the + * attribute cache as being invalid, then try to update it. Fake up + * weak cache consistency data, if none exist. + * + * This function is mainly designed to be used by the ->write_done() functions. + */ +int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fattr *fattr) +{ + int attr_cmp = nfs_inode_attrs_cmp(fattr, inode); + int status; + + /* Don't do a WCC update if these attributes are already stale */ + if (attr_cmp < 0) + return 0; + if ((fattr->valid & NFS_ATTR_FATTR) == 0 || !attr_cmp) { + /* Record the pre/post change info before clearing PRECHANGE */ + nfs_ooo_record(NFS_I(inode), fattr); + fattr->valid &= ~(NFS_ATTR_FATTR_PRECHANGE + | NFS_ATTR_FATTR_PRESIZE + | NFS_ATTR_FATTR_PREMTIME + | NFS_ATTR_FATTR_PRECTIME); + goto out_noforce; + } + if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRECHANGE) == 0) { + fattr->pre_change_attr = inode_peek_iversion_raw(inode); + fattr->valid |= NFS_ATTR_FATTR_PRECHANGE; + } + if ((fattr->valid & NFS_ATTR_FATTR_CTIME) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRECTIME) == 0) { + fattr->pre_ctime = inode_get_ctime(inode); + fattr->valid |= NFS_ATTR_FATTR_PRECTIME; + } + if ((fattr->valid & NFS_ATTR_FATTR_MTIME) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PREMTIME) == 0) { + fattr->pre_mtime = inode_get_mtime(inode); + fattr->valid |= NFS_ATTR_FATTR_PREMTIME; + } + if ((fattr->valid & NFS_ATTR_FATTR_SIZE) != 0 && + (fattr->valid & NFS_ATTR_FATTR_PRESIZE) == 0) { + fattr->pre_size = i_size_read(inode); + fattr->valid |= NFS_ATTR_FATTR_PRESIZE; + } +out_noforce: + status = nfs_post_op_update_inode_locked(inode, fattr, + NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_BLOCKS); + return status; +} + +/** + * nfs_post_op_update_inode_force_wcc - try to update the inode attribute cache + * @inode: pointer to inode + * @fattr: updated attributes + * + * After an operation that has changed the inode metadata, mark the + * attribute cache as being invalid, then try to update it. Fake up + * weak cache consistency data, if none exist. + * + * This function is mainly designed to be used by the ->write_done() functions. + */ +int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fattr) +{ + int status; + + spin_lock(&inode->i_lock); + nfs_fattr_set_barrier(fattr); + status = nfs_post_op_update_inode_force_wcc_locked(inode, fattr); + spin_unlock(&inode->i_lock); + return status; +} +EXPORT_SYMBOL_GPL(nfs_post_op_update_inode_force_wcc); + + +/* + * Many nfs protocol calls return the new file attributes after + * an operation. Here we update the inode to reflect the state + * of the server's inode. + * + * This is a bit tricky because we have to make sure all dirty pages + * have been sent off to the server before calling invalidate_inode_pages. + * To make sure no other process adds more write requests while we try + * our best to flush them, we make them sleep during the attribute refresh. + * + * A very similar scenario holds for the dir cache. + */ +static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_inode *nfsi = NFS_I(inode); + loff_t cur_isize, new_isize; + u64 fattr_supported = server->fattr_valid; + unsigned long invalid = 0; + unsigned long now = jiffies; + unsigned long save_cache_validity; + bool have_writers = nfs_file_has_buffered_writers(nfsi); + bool cache_revalidated = true; + bool attr_changed = false; + bool have_delegation; + + dfprintk(VFS, "NFS: %s(%s/%llu fh_crc=0x%08x ct=%d info=0x%llx)\n", + __func__, inode->i_sb->s_id, inode->i_ino, + nfs_display_fhandle_hash(NFS_FH(inode)), + icount_read_once(inode), fattr->valid); + + if (!(fattr->valid & NFS_ATTR_FATTR_FILEID)) { + /* Only a mounted-on-fileid? Just exit */ + if (fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) + return 0; + /* Has the inode gone and changed behind our back? */ + } else if (inode->i_ino != fattr->fileid) { + /* Is this perhaps the mounted-on fileid? */ + if ((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) && + inode->i_ino == fattr->mounted_on_fileid) + return 0; + printk(KERN_ERR "NFS: server %s error: fileid changed\n" + "fsid %s: expected fileid 0x%Lx, got 0x%Lx\n", + NFS_SERVER(inode)->nfs_client->cl_hostname, + inode->i_sb->s_id, (long long)inode->i_ino, + (long long)fattr->fileid); + goto out_err; + } + + /* + * Make sure the inode's type hasn't changed. + */ + if ((fattr->valid & NFS_ATTR_FATTR_TYPE) && inode_wrong_type(inode, fattr->mode)) { + /* + * Big trouble! The inode has become a different object. + */ + printk(KERN_DEBUG "NFS: %s: inode %llu mode changed, %07o to %07o\n", + __func__, inode->i_ino, inode->i_mode, fattr->mode); + goto out_err; + } + + /* Update the fsid? */ + if (S_ISDIR(inode->i_mode) && (fattr->valid & NFS_ATTR_FATTR_FSID) && + !nfs_fsid_equal(&server->fsid, &fattr->fsid) && + !IS_AUTOMOUNT(inode)) + server->fsid = fattr->fsid; + + /* Save the delegation state before clearing cache_validity */ + have_delegation = nfs_have_delegated_attributes(inode); + + /* + * Update the read time so we don't revalidate too often. + */ + nfsi->read_cache_jiffies = fattr->time_start; + + /* Fix up any delegated attributes in the struct nfs_fattr */ + nfs_fattr_fixup_delegated(inode, fattr); + + save_cache_validity = nfsi->cache_validity; + nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR + | NFS_INO_INVALID_ATIME + | NFS_INO_REVAL_FORCED + | NFS_INO_INVALID_BLOCKS); + + /* Do atomic weak cache consistency updates */ + nfs_wcc_update_inode(inode, fattr); + + if (pnfs_layoutcommit_outstanding(inode)) { + nfsi->cache_validity |= + save_cache_validity & + (NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_MTIME | NFS_INO_INVALID_SIZE | + NFS_INO_INVALID_BLOCKS); + cache_revalidated = false; + } + + /* More cache consistency checks */ + if (fattr->valid & NFS_ATTR_FATTR_CHANGE) { + if (!have_writers && nfsi->ooo && nfsi->ooo->cnt == 1 && + nfsi->ooo->gap[0].end == inode_peek_iversion_raw(inode)) { + /* There is one remaining gap that hasn't been + * merged into iversion - do that now. + */ + inode_set_iversion_raw(inode, nfsi->ooo->gap[0].start); + kfree(nfsi->ooo); + nfsi->ooo = NULL; + } + if (!inode_eq_iversion_raw(inode, fattr->change_attr)) { + /* Could it be a race with writeback? */ + if (!(have_writers || have_delegation)) { + invalid |= NFS_INO_INVALID_DATA + | NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_XATTR; + /* Force revalidate of all attributes */ + save_cache_validity |= NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_SIZE + | NFS_INO_INVALID_BLOCKS + | NFS_INO_INVALID_NLINK + | NFS_INO_INVALID_MODE + | NFS_INO_INVALID_OTHER + | NFS_INO_INVALID_BTIME; + if (S_ISDIR(inode->i_mode)) + nfs_force_lookup_revalidate(inode); + attr_changed = true; + dprintk("NFS: change_attr change on server for file %s/%llu\n", + inode->i_sb->s_id, + inode->i_ino); + } else if (!have_delegation) { + nfs_ooo_record(nfsi, fattr); + nfs_ooo_merge(nfsi, inode_peek_iversion_raw(inode), + fattr->change_attr); + } + inode_set_iversion_raw(inode, fattr->change_attr); + } + } else { + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_CHANGE; + if (!have_delegation || + (nfsi->cache_validity & NFS_INO_INVALID_CHANGE) != 0) + cache_revalidated = false; + } + + if (fattr->valid & NFS_ATTR_FATTR_MTIME) + inode_set_mtime_to_ts(inode, fattr->mtime); + else if (fattr_supported & NFS_ATTR_FATTR_MTIME) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_MTIME; + + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode_set_ctime_to_ts(inode, fattr->ctime); + else if (fattr_supported & NFS_ATTR_FATTR_CTIME) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_CTIME; + + if (fattr->valid & NFS_ATTR_FATTR_BTIME) + nfsi->btime = fattr->btime; + else if (fattr_supported & NFS_ATTR_FATTR_BTIME) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_BTIME; + + /* Check if our cached file size is stale */ + if (fattr->valid & NFS_ATTR_FATTR_SIZE) { + new_isize = nfs_size_to_loff_t(fattr->size); + cur_isize = i_size_read(inode); + if (new_isize != cur_isize && !have_delegation) { + /* Do we perhaps have any outstanding writes, or has + * the file grown beyond our last write? */ + if (!nfs_have_writebacks(inode) || new_isize > cur_isize) { + trace_nfs_size_update(inode, new_isize); + i_size_write(inode, new_isize); + if (!have_writers) + invalid |= NFS_INO_INVALID_DATA; + } + } + if (new_isize == 0 && + !(fattr->valid & (NFS_ATTR_FATTR_SPACE_USED | + NFS_ATTR_FATTR_BLOCKS_USED))) { + fattr->du.nfs3.used = 0; + fattr->valid |= NFS_ATTR_FATTR_SPACE_USED; + } + } else + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_SIZE; + + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + inode_set_atime_to_ts(inode, fattr->atime); + else if (fattr_supported & NFS_ATTR_FATTR_ATIME) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_ATIME; + + if (fattr->valid & NFS_ATTR_FATTR_MODE) { + if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) { + umode_t newmode = inode->i_mode & S_IFMT; + newmode |= fattr->mode & S_IALLUGO; + inode->i_mode = newmode; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL; + } + } else if (fattr_supported & NFS_ATTR_FATTR_MODE) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_MODE; + + if (fattr->valid & NFS_ATTR_FATTR_OWNER) { + if (!uid_eq(inode->i_uid, fattr->uid)) { + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL; + inode->i_uid = fattr->uid; + } + } else if (fattr_supported & NFS_ATTR_FATTR_OWNER) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_OTHER; + + if (fattr->valid & NFS_ATTR_FATTR_GROUP) { + if (!gid_eq(inode->i_gid, fattr->gid)) { + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL; + inode->i_gid = fattr->gid; + } + } else if (fattr_supported & NFS_ATTR_FATTR_GROUP) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_OTHER; + + if (fattr->valid & NFS_ATTR_FATTR_NLINK) { + if (inode->i_nlink != fattr->nlink) + set_nlink(inode, fattr->nlink); + } else if (fattr_supported & NFS_ATTR_FATTR_NLINK) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_NLINK; + + if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { + /* + * report the blocks in 512byte units + */ + inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); + } else if (fattr_supported & NFS_ATTR_FATTR_SPACE_USED) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_BLOCKS; + + if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) + inode->i_blocks = fattr->du.nfs2.blocks; + else if (fattr_supported & NFS_ATTR_FATTR_BLOCKS_USED) + nfsi->cache_validity |= + save_cache_validity & NFS_INO_INVALID_BLOCKS; + + /* Update attrtimeo value if we're out of the unstable period */ + if (attr_changed) { + nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE); + nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); + nfsi->attrtimeo_timestamp = now; + /* Set barrier to be more recent than all outstanding updates */ + nfsi->attr_gencount = nfs_inc_attr_generation_counter(); + } else { + if (cache_revalidated) { + if (!time_in_range_open(now, nfsi->attrtimeo_timestamp, + nfsi->attrtimeo_timestamp + nfsi->attrtimeo)) { + nfsi->attrtimeo <<= 1; + if (nfsi->attrtimeo > NFS_MAXATTRTIMEO(inode)) + nfsi->attrtimeo = NFS_MAXATTRTIMEO(inode); + } + nfsi->attrtimeo_timestamp = now; + } + /* Set the barrier to be more recent than this fattr */ + if ((long)(fattr->gencount - nfsi->attr_gencount) > 0) + nfsi->attr_gencount = fattr->gencount; + } + + /* Don't invalidate the data if we were to blame */ + if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) + || S_ISLNK(inode->i_mode))) + invalid &= ~NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, invalid); + + return 0; + out_err: + /* + * No need to worry about unhashing the dentry, as the + * lookup validation will know that the inode is bad. + * (But we fall through to invalidate the caches.) + */ + nfs_set_inode_stale_locked(inode); + return -ESTALE; +} + +struct inode *nfs_alloc_inode(struct super_block *sb) +{ + struct nfs_inode *nfsi; + nfsi = alloc_inode_sb(sb, nfs_inode_cachep, GFP_KERNEL); + if (!nfsi) + return NULL; + nfsi->flags = 0UL; + nfsi->cache_validity = 0UL; + nfsi->ooo = NULL; +#if IS_ENABLED(CONFIG_NFS_V4) + nfsi->nfs4_acl = NULL; +#endif /* CONFIG_NFS_V4 */ +#ifdef CONFIG_NFS_V4_2 + nfsi->xattr_cache = NULL; +#endif + nfs_netfs_inode_init(nfsi); + + return &nfsi->vfs_inode; +} +EXPORT_SYMBOL_GPL(nfs_alloc_inode); + +void nfs_free_inode(struct inode *inode) +{ + kfree(NFS_I(inode)->ooo); + kmem_cache_free(nfs_inode_cachep, NFS_I(inode)); +} +EXPORT_SYMBOL_GPL(nfs_free_inode); + +static inline void nfs4_init_once(struct nfs_inode *nfsi) +{ +#if IS_ENABLED(CONFIG_NFS_V4) + INIT_LIST_HEAD(&nfsi->open_states); + nfsi->delegation = NULL; + init_rwsem(&nfsi->rwsem); + nfsi->layout = NULL; +#endif +} + +static void init_once(void *foo) +{ + struct nfs_inode *nfsi = foo; + + inode_init_once(&nfsi->vfs_inode); + INIT_LIST_HEAD(&nfsi->open_files); + INIT_LIST_HEAD(&nfsi->access_cache_entry_lru); + INIT_LIST_HEAD(&nfsi->access_cache_inode_lru); + nfs4_init_once(nfsi); +} + +static int __init nfs_init_inodecache(void) +{ + nfs_inode_cachep = kmem_cache_create("nfs_inode_cache", + sizeof(struct nfs_inode), + 0, (SLAB_RECLAIM_ACCOUNT| + SLAB_ACCOUNT), + init_once); + if (nfs_inode_cachep == NULL) + return -ENOMEM; + + return 0; +} + +static void nfs_destroy_inodecache(void) +{ + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); + kmem_cache_destroy(nfs_inode_cachep); +} + +struct workqueue_struct *nfslocaliod_workqueue; +struct workqueue_struct *nfsiod_workqueue; +EXPORT_SYMBOL_GPL(nfsiod_workqueue); + +/* + * Destroy the nfsiod workqueues + */ +static void nfsiod_stop(void) +{ + struct workqueue_struct *wq; + + wq = nfsiod_workqueue; + if (wq != NULL) { + nfsiod_workqueue = NULL; + destroy_workqueue(wq); + } +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + wq = nfslocaliod_workqueue; + if (wq != NULL) { + nfslocaliod_workqueue = NULL; + destroy_workqueue(wq); + } +#endif /* CONFIG_NFS_LOCALIO */ +} + +/* + * Start the nfsiod workqueues + */ +static int nfsiod_start(void) +{ + dprintk("RPC: creating workqueue nfsiod\n"); + nfsiod_workqueue = alloc_workqueue("nfsiod", WQ_MEM_RECLAIM | WQ_UNBOUND, 0); + if (nfsiod_workqueue == NULL) + return -ENOMEM; +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + /* + * localio writes need to use a normal (non-memreclaim) workqueue. + * When we start getting low on space, XFS goes and calls flush_work() on + * a non-memreclaim work queue, which causes a priority inversion problem. + */ + dprintk("RPC: creating workqueue nfslocaliod\n"); + nfslocaliod_workqueue = alloc_workqueue("nfslocaliod", WQ_UNBOUND, 0); + if (unlikely(nfslocaliod_workqueue == NULL)) { + nfsiod_stop(); + return -ENOMEM; + } +#endif /* CONFIG_NFS_LOCALIO */ + return 0; +} + +unsigned int nfs_net_id; +EXPORT_SYMBOL_GPL(nfs_net_id); + +static int nfs_net_init(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + int err; + + nfs_clients_init(net); + + if (!rpc_proc_register(net, &nn->rpcstats)) { + err = -ENOMEM; + goto err_proc_rpc; + } + + err = nfs_fs_proc_net_init(net); + if (err) + goto err_proc_nfs; + + return 0; + +err_proc_nfs: + rpc_proc_unregister(net, "nfs"); +err_proc_rpc: + nfs_clients_exit(net); + return err; +} + +static void nfs_net_exit(struct net *net) +{ + rpc_proc_unregister(net, "nfs"); + nfs_fs_proc_net_exit(net); + nfs_clients_exit(net); +} + +static struct pernet_operations nfs_net_ops = { + .init = nfs_net_init, + .exit = nfs_net_exit, + .id = &nfs_net_id, + .size = sizeof(struct nfs_net), +}; + +#ifdef CONFIG_KEYS +static struct key *nfs_keyring; + +static int __init nfs_init_keyring(void) +{ + nfs_keyring = keyring_alloc(".nfs", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, + current_cred(), + (KEY_POS_ALL & ~KEY_POS_SETATTR) | + (KEY_USR_ALL & ~KEY_USR_SETATTR), + KEY_ALLOC_NOT_IN_QUOTA, NULL, NULL); + return PTR_ERR_OR_ZERO(nfs_keyring); +} + +static void nfs_exit_keyring(void) +{ + key_put(nfs_keyring); +} +#else +static inline int nfs_init_keyring(void) +{ + return 0; +} + +static inline void nfs_exit_keyring(void) +{ +} +#endif /* CONFIG_KEYS */ + +/* + * Initialize NFS + */ +static int __init init_nfs_fs(void) +{ + int err; + + err = nfs_init_keyring(); + if (err) + return err; + + err = nfs_sysfs_init(); + if (err < 0) + goto out10; + + err = register_pernet_subsys(&nfs_net_ops); + if (err < 0) + goto out9; + + err = nfsiod_start(); + if (err) + goto out7; + + err = nfs_fs_proc_init(); + if (err) + goto out6; + + err = nfs_init_nfspagecache(); + if (err) + goto out5; + + err = nfs_init_inodecache(); + if (err) + goto out4; + + err = nfs_init_readpagecache(); + if (err) + goto out3; + + err = nfs_init_writepagecache(); + if (err) + goto out2; + + err = nfs_init_directcache(); + if (err) + goto out1; + + err = register_nfs_fs(); + if (err) + goto out0; + + return 0; +out0: + nfs_destroy_directcache(); +out1: + nfs_destroy_writepagecache(); +out2: + nfs_destroy_readpagecache(); +out3: + nfs_destroy_inodecache(); +out4: + nfs_destroy_nfspagecache(); +out5: + nfs_fs_proc_exit(); +out6: + nfsiod_stop(); +out7: + unregister_pernet_subsys(&nfs_net_ops); +out9: + nfs_sysfs_exit(); +out10: + nfs_exit_keyring(); + return err; +} + +static void __exit exit_nfs_fs(void) +{ + nfs_destroy_directcache(); + nfs_destroy_writepagecache(); + nfs_destroy_readpagecache(); + nfs_destroy_inodecache(); + nfs_destroy_nfspagecache(); + unregister_pernet_subsys(&nfs_net_ops); + unregister_nfs_fs(); + nfs_fs_proc_exit(); + nfsiod_stop(); + nfs_sysfs_exit(); + nfs_exit_keyring(); +} + +/* Not quite true; I just maintain it */ +MODULE_AUTHOR("Olaf Kirch "); +MODULE_DESCRIPTION("NFS client support"); +MODULE_LICENSE("GPL"); +module_param_cb(enable_ino64, ¶m_ops_enable_ino64, &enable_ino64, 0644); + +module_init(init_nfs_fs) +module_exit(exit_nfs_fs) diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h new file mode 100644 index 000000000..7f96a258a --- /dev/null +++ b/fs/nfs/internal.h @@ -0,0 +1,1012 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * NFS internal definitions + */ + +#include "nfs4_fs.h" +#include +#include +#include +#include +#include +#include +#include +#include + +#define NFS_SB_MASK (SB_RDONLY|SB_NOSUID|SB_NODEV|SB_NOEXEC|SB_SYNCHRONOUS) + +extern const struct export_operations nfs_export_ops; + +struct nfs_string; +struct nfs_pageio_descriptor; + +static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct nfs_fattr *fattr) +{ + if (!nfs_fsid_equal(&NFS_SB(parent)->fsid, &fattr->fsid)) + fattr->valid |= NFS_ATTR_FATTR_MOUNTPOINT; +} + +static inline int nfs_attr_use_mounted_on_fileid(struct nfs_fattr *fattr) +{ + if (((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) == 0) || + (((fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) && + ((fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) == 0))) + return 0; + return 1; +} + +static inline bool nfs_lookup_is_soft_revalidate(const struct dentry *dentry) +{ + if (!(NFS_SB(dentry->d_sb)->flags & NFS_MOUNT_SOFTREVAL)) + return false; + if (!d_is_positive(dentry) || !NFS_FH(d_inode(dentry))->size) + return false; + return true; +} + +static inline fmode_t flags_to_mode(int flags) +{ + fmode_t res = (__force fmode_t)flags & FMODE_EXEC; + if ((flags & O_ACCMODE) != O_WRONLY) + res |= FMODE_READ; + if ((flags & O_ACCMODE) != O_RDONLY) + res |= FMODE_WRITE; + return res; +} + +/* + * Note: RFC 1813 doesn't limit the number of auth flavors that + * a server can return, so make something up. + */ +#define NFS_MAX_SECFLAVORS (12) + +/* + * Value used if the user did not specify a port value. + */ +#define NFS_UNSPEC_PORT (-1) + +#define NFS_UNSPEC_RETRANS (UINT_MAX) +#define NFS_UNSPEC_TIMEO (UINT_MAX) + +struct nfs_client_initdata { + unsigned long init_flags; + const char *hostname; /* Hostname of the server */ + const struct sockaddr_storage *addr; /* Address of the server */ + const char *nodename; /* Hostname of the client */ + const char *ip_addr; /* IP address of the client */ + size_t addrlen; + struct nfs_subversion *nfs_mod; + int proto; + u32 minorversion; + unsigned int nconnect; + unsigned int max_connect; + struct net *net; + const struct rpc_timeout *timeparms; + const struct cred *cred; + struct xprtsec_parms xprtsec; + unsigned long connect_timeout; + unsigned long reconnect_timeout; +}; + +/* + * In-kernel mount arguments + */ +struct nfs_fs_context { + bool internal; + bool skip_reconfig_option_check; + bool need_mount; + bool sloppy; + unsigned int flags; /* NFS{,4}_MOUNT_* flags */ + unsigned int rsize, wsize; + unsigned int timeo, retrans; + unsigned int acregmin, acregmax; + unsigned int acdirmin, acdirmax; + unsigned int namlen; + unsigned int options; + unsigned int bsize; + struct nfs_auth_info auth_info; + rpc_authflavor_t selected_flavor; + struct xprtsec_parms xprtsec; + char *client_address; + unsigned int version; + unsigned int minorversion; + char *fscache_uniq; + unsigned short protofamily; + unsigned short mountfamily; + bool has_sec_mnt_opts; + int lock_status; + + struct { + union { + struct sockaddr address; + struct sockaddr_storage _address; + }; + size_t addrlen; + char *hostname; + u32 version; + int port; + unsigned short protocol; + } mount_server; + + struct { + union { + struct sockaddr address; + struct sockaddr_storage _address; + }; + size_t addrlen; + char *hostname; + char *export_path; + int port; + unsigned short protocol; + unsigned short nconnect; + unsigned short max_connect; + unsigned short export_path_len; + } nfs_server; + + struct nfs_fh *mntfh; + struct nfs_server *server; + struct nfs_subversion *nfs_mod; + + /* Information for a cloned mount. */ + struct nfs_clone_mount { + struct super_block *sb; + struct dentry *dentry; + struct nfs_fattr *fattr; + } clone_data; +}; + +enum nfs_lock_status { + NFS_LOCK_NOT_SET = 0, + NFS_LOCK_LOCK = 1, + NFS_LOCK_NOLOCK = 2, +}; + +#define nfs_errorf(fc, fmt, ...) ((fc)->log.log ? \ + errorf(fc, fmt, ## __VA_ARGS__) : \ + ({ dprintk(fmt "\n", ## __VA_ARGS__); })) + +#define nfs_ferrorf(fc, fac, fmt, ...) ((fc)->log.log ? \ + errorf(fc, fmt, ## __VA_ARGS__) : \ + ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); })) + +#define nfs_invalf(fc, fmt, ...) ((fc)->log.log ? \ + invalf(fc, fmt, ## __VA_ARGS__) : \ + ({ dprintk(fmt "\n", ## __VA_ARGS__); -EINVAL; })) + +#define nfs_finvalf(fc, fac, fmt, ...) ((fc)->log.log ? \ + invalf(fc, fmt, ## __VA_ARGS__) : \ + ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); -EINVAL; })) + +#define nfs_warnf(fc, fmt, ...) ((fc)->log.log ? \ + warnf(fc, fmt, ## __VA_ARGS__) : \ + ({ dprintk(fmt "\n", ## __VA_ARGS__); })) + +#define nfs_fwarnf(fc, fac, fmt, ...) ((fc)->log.log ? \ + warnf(fc, fmt, ## __VA_ARGS__) : \ + ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); })) + +static inline struct nfs_fs_context *nfs_fc2context(const struct fs_context *fc) +{ + return fc->fs_private; +} + +/* mount_clnt.c */ +struct nfs_mount_request { + struct sockaddr_storage *sap; + size_t salen; + char *hostname; + char *dirpath; + u32 version; + unsigned short protocol; + struct nfs_fh *fh; + int noresvport; + unsigned int *auth_flav_len; + rpc_authflavor_t *auth_flavs; + struct net *net; +}; + +extern int nfs_mount(struct nfs_mount_request *info, int timeo, int retrans); + +/* client.c */ +extern const struct rpc_program nfs_program; +extern void nfs_clients_init(struct net *net); +extern void nfs_clients_exit(struct net *net); +extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *); +int nfs_create_rpc_client(struct nfs_client *, const struct nfs_client_initdata *, rpc_authflavor_t); +struct nfs_client *nfs_get_client(const struct nfs_client_initdata *); +int nfs_probe_server(struct nfs_server *, struct nfs_fh *); +void nfs_server_insert_lists(struct nfs_server *); +void nfs_server_remove_lists(struct nfs_server *); +void nfs_init_timeout_values(struct rpc_timeout *to, int proto, int timeo, int retrans); +int nfs_init_server_rpcclient(struct nfs_server *, const struct rpc_timeout *t, + rpc_authflavor_t); +struct nfs_server *nfs_alloc_server(void); +void nfs_server_copy_userdata(struct nfs_server *, struct nfs_server *); + +extern void nfs_put_client(struct nfs_client *); +extern void nfs_free_client(struct nfs_client *); +extern struct nfs_client *nfs4_find_client_ident(struct net *, int); +extern struct nfs_client * +nfs4_find_client_sessionid(struct net *, const struct sockaddr *, + struct nfs4_sessionid *, u32); +extern struct nfs_server *nfs_create_server(struct fs_context *); +extern void nfs_server_set_init_caps(struct nfs_server *); +extern struct nfs_server *nfs4_create_server(struct fs_context *); +extern struct nfs_server *nfs4_create_referral_server(struct fs_context *); +extern int nfs4_update_server(struct nfs_server *server, const char *hostname, + struct sockaddr_storage *sap, size_t salen, + struct net *net); +extern void nfs_free_server(struct nfs_server *server); +extern struct nfs_server *nfs_clone_server(struct nfs_server *, + struct nfs_fh *, + struct nfs_fattr *, + rpc_authflavor_t); +extern bool nfs_client_init_is_complete(const struct nfs_client *clp); +extern int nfs_client_init_status(const struct nfs_client *clp); +extern int nfs_wait_client_init_complete(const struct nfs_client *clp); +extern void nfs_mark_client_ready(struct nfs_client *clp, int state); +extern struct nfs_client *nfs4_set_ds_client(struct nfs_server *mds_srv, + const struct sockaddr_storage *ds_addr, + int ds_addrlen, int ds_proto, + unsigned int ds_timeo, + unsigned int ds_retrans, + u32 minor_version); +extern struct rpc_clnt *nfs4_find_or_create_ds_client(struct nfs_client *, + struct inode *); +extern void nfs4_session_limit_rwsize(struct nfs_server *server); +extern void nfs4_session_limit_xasize(struct nfs_server *server); +extern struct nfs_client *nfs3_set_ds_client(struct nfs_server *mds_srv, + const struct sockaddr_storage *ds_addr, int ds_addrlen, + int ds_proto, unsigned int ds_timeo, + unsigned int ds_retrans); +#ifdef CONFIG_PROC_FS +extern int __init nfs_fs_proc_init(void); +extern void nfs_fs_proc_exit(void); +extern int nfs_fs_proc_net_init(struct net *net); +extern void nfs_fs_proc_net_exit(struct net *net); +#else +static inline int nfs_fs_proc_net_init(struct net *net) +{ + return 0; +} +static inline void nfs_fs_proc_net_exit(struct net *net) +{ +} +static inline int nfs_fs_proc_init(void) +{ + return 0; +} +static inline void nfs_fs_proc_exit(void) +{ +} +#endif + +/* callback_xdr.c */ +extern const struct svc_version nfs4_callback_version1; +extern const struct svc_version nfs4_callback_version4; + +/* fs_context.c */ +extern struct file_system_type nfs_fs_type; + +/* pagelist.c */ +extern int __init nfs_init_nfspagecache(void); +extern void nfs_destroy_nfspagecache(void); +extern int __init nfs_init_readpagecache(void); +extern void nfs_destroy_readpagecache(void); +extern int __init nfs_init_writepagecache(void); +extern void nfs_destroy_writepagecache(void); + +extern int __init nfs_init_directcache(void); +extern void nfs_destroy_directcache(void); +extern void nfs_pgheader_init(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr, + void (*release)(struct nfs_pgio_header *hdr)); +void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos); +int nfs_iocounter_wait(struct nfs_lock_context *l_ctx); + +extern const struct nfs_pageio_ops nfs_pgio_rw_ops; +struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *); +void nfs_pgio_header_free(struct nfs_pgio_header *); +int nfs_generic_pgio(struct nfs_pageio_descriptor *, struct nfs_pgio_header *); +int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr, + const struct cred *cred, const struct nfs_rpc_ops *rpc_ops, + const struct rpc_call_ops *call_ops, int how, int flags, + struct nfsd_file *localio); +void nfs_free_request(struct nfs_page *req); +struct nfs_pgio_mirror * +nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc); + +static inline bool nfs_match_open_context(const struct nfs_open_context *ctx1, + const struct nfs_open_context *ctx2) +{ + return cred_fscmp(ctx1->cred, ctx2->cred) == 0 && ctx1->state == ctx2->state; +} + +/* nfs2xdr.c */ +extern const struct rpc_procinfo nfs_procedures[]; +extern int nfs2_decode_dirent(struct xdr_stream *, + struct nfs_entry *, bool); + +/* nfs3xdr.c */ +extern const struct rpc_procinfo nfs3_procedures[]; +extern int nfs3_decode_dirent(struct xdr_stream *, + struct nfs_entry *, bool); + +/* nfs4xdr.c */ +#if IS_ENABLED(CONFIG_NFS_V4) +extern int nfs4_decode_dirent(struct xdr_stream *, + struct nfs_entry *, bool); +extern const u32 nfs41_maxread_overhead; +extern const u32 nfs41_maxwrite_overhead; +extern const u32 nfs41_maxgetdevinfo_overhead; + +/* nfs4proc.c */ +extern const struct rpc_procinfo nfs4_procedures[]; +#endif /* CONFIG_NFS_V4 */ + +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +extern struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags); +static inline struct nfs4_label * +nfs4_label_copy(struct nfs4_label *dst, struct nfs4_label *src) +{ + if (!dst || !src) + return NULL; + + if (src->len > NFS4_MAXLABELLEN) + return NULL; + + dst->lfs = src->lfs; + dst->pi = src->pi; + dst->len = src->len; + memcpy(dst->label, src->label, src->len); + + return dst; +} + +static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) +{ + if (nfs_server_capable(&nfsi->vfs_inode, NFS_CAP_SECURITY_LABEL)) + nfsi->cache_validity |= NFS_INO_INVALID_LABEL; +} +#else +static inline struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { return NULL; } +static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) +{ +} +static inline struct nfs4_label * +nfs4_label_copy(struct nfs4_label *dst, struct nfs4_label *src) +{ + return NULL; +} +#endif /* CONFIG_NFS_V4_SECURITY_LABEL */ + +/* proc.c */ +void nfs_close_context(struct nfs_open_context *ctx, int is_sync); +extern struct nfs_client *nfs_init_client(struct nfs_client *clp, + const struct nfs_client_initdata *); + +/* dir.c */ +extern void nfs_readdir_record_entry_cache_hit(struct inode *dir); +extern void nfs_readdir_record_entry_cache_miss(struct inode *dir); +extern unsigned long nfs_access_cache_count(struct shrinker *shrink, + struct shrink_control *sc); +extern unsigned long nfs_access_cache_scan(struct shrinker *shrink, + struct shrink_control *sc); +struct dentry *nfs_lookup(struct inode *, struct dentry *, unsigned int); +void nfs_d_prune_case_insensitive_aliases(struct inode *inode); +int nfs_create(struct mnt_idmap *, struct inode *, struct dentry *, + umode_t); +struct dentry *nfs_mkdir(struct mnt_idmap *, struct inode *, struct dentry *, + umode_t); +int nfs_rmdir(struct inode *, struct dentry *); +int nfs_unlink(struct inode *, struct dentry *); +int nfs_symlink(struct mnt_idmap *, struct inode *, struct dentry *, + const char *); +int nfs_link(struct dentry *, struct inode *, struct dentry *); +int nfs_mknod(struct mnt_idmap *, struct inode *, struct dentry *, umode_t, + dev_t); +int nfs_rename(struct mnt_idmap *, struct inode *, struct dentry *, + struct inode *, struct dentry *, unsigned int); + +#ifdef CONFIG_NFS_V4_2 +static inline __u32 nfs_access_xattr_mask(const struct nfs_server *server) +{ + if (!(server->caps & NFS_CAP_XATTR)) + return 0; + return NFS4_ACCESS_XAREAD | NFS4_ACCESS_XAWRITE | NFS4_ACCESS_XALIST; +} +#else +static inline __u32 nfs_access_xattr_mask(const struct nfs_server *server) +{ + return 0; +} +#endif + +/* file.c */ +int nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync); +loff_t nfs_file_llseek(struct file *, loff_t, int); +ssize_t nfs_file_read(struct kiocb *, struct iov_iter *); +ssize_t nfs_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, + size_t len, unsigned int flags); +int nfs_file_mmap_prepare(struct vm_area_desc *); +ssize_t nfs_file_write(struct kiocb *, struct iov_iter *); +int nfs_file_release(struct inode *, struct file *); +int nfs_lock(struct file *, int, struct file_lock *); +int nfs_flock(struct file *, int, struct file_lock *); +int nfs_check_flags(int); +void nfs_truncate_last_folio(struct address_space *mapping, loff_t from, + loff_t to); + +/* inode.c */ +extern struct workqueue_struct *nfsiod_workqueue; +extern struct workqueue_struct *nfslocaliod_workqueue; +extern struct inode *nfs_alloc_inode(struct super_block *sb); +extern void nfs_free_inode(struct inode *); +extern int nfs_write_inode(struct inode *, struct writeback_control *); +extern int nfs_drop_inode(struct inode *); +extern void nfs_clear_inode(struct inode *); +extern void nfs_evict_inode(struct inode *); +extern void nfs_zap_acl_cache(struct inode *inode); +extern void nfs_set_cache_invalid(struct inode *inode, unsigned long flags); +extern bool nfs_check_cache_invalid(struct inode *, unsigned long); +extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode); + +struct file_kattr; +int nfs_fileattr_get(struct dentry *dentry, struct file_kattr *fa); + +#if IS_ENABLED(CONFIG_NFS_LOCALIO) +/* localio.c */ +struct nfs_local_dio { + u32 mem_align; + u32 offset_align; + loff_t middle_offset; + loff_t end_offset; + ssize_t start_len; /* Length for misaligned first extent */ + ssize_t middle_len; /* Length for DIO-aligned middle extent */ + ssize_t end_len; /* Length for misaligned last extent */ +}; + +extern void nfs_local_probe_async(struct nfs_client *); +extern void nfs_local_probe_async_work(struct work_struct *); +extern struct nfsd_file *nfs_local_open_fh(struct nfs_client *, + const struct cred *, + struct nfs_fh *, + struct nfs_file_localio *, + const fmode_t); +extern int nfs_local_doio(struct nfs_client *, + struct nfsd_file *, + struct nfs_pgio_header *, + const struct rpc_call_ops *); +extern int nfs_local_commit(struct nfsd_file *, + struct nfs_commit_data *, + const struct rpc_call_ops *, int); +extern bool nfs_server_is_local(const struct nfs_client *clp); + +#else /* CONFIG_NFS_LOCALIO */ +static inline void nfs_local_probe(struct nfs_client *clp) {} +static inline void nfs_local_probe_async(struct nfs_client *clp) {} +static inline struct nfsd_file * +nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, + struct nfs_fh *fh, struct nfs_file_localio *nfl, + const fmode_t mode) +{ + return NULL; +} +static inline int nfs_local_doio(struct nfs_client *clp, + struct nfsd_file *localio, + struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops) +{ + return -EINVAL; +} +static inline int nfs_local_commit(struct nfsd_file *localio, + struct nfs_commit_data *data, + const struct rpc_call_ops *call_ops, int how) +{ + return -EINVAL; +} +static inline bool nfs_server_is_local(const struct nfs_client *clp) +{ + return false; +} +#endif /* CONFIG_NFS_LOCALIO */ + +/* super.c */ +extern const struct super_operations nfs_sops; +bool nfs_auth_info_match(const struct nfs_auth_info *, rpc_authflavor_t); +int nfs_try_get_tree(struct fs_context *); +int nfs_get_tree_common(struct fs_context *); +void nfs_kill_super(struct super_block *); + +extern int __init register_nfs_fs(void); +extern void __exit unregister_nfs_fs(void); +extern bool nfs_sb_active(struct super_block *sb); +extern void nfs_sb_deactive(struct super_block *sb); +extern int nfs_client_for_each_server(struct nfs_client *clp, + int (*fn)(struct nfs_server *, void *), + void *data); +#ifdef CONFIG_NFS_FSCACHE +extern const struct netfs_request_ops nfs_netfs_ops; +#endif + +/* io.c */ +extern __must_check int nfs_start_io_read(struct inode *inode); +extern void nfs_end_io_read(struct inode *inode); +extern __must_check int nfs_start_io_write(struct inode *inode); +extern void nfs_end_io_write(struct inode *inode); +extern __must_check int nfs_start_io_direct(struct inode *inode); +extern __must_check int nfs_start_io_direct_nowait(struct inode *inode); +extern void nfs_end_io_direct(struct inode *inode); + +static inline bool nfs_file_io_is_buffered(struct nfs_inode *nfsi) +{ + return test_bit(NFS_INO_ODIRECT, &nfsi->flags) == 0; +} + +/* Must be called with exclusively locked inode->i_rwsem */ +static inline void nfs_file_block_o_direct(struct nfs_inode *nfsi) +{ + if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) { + clear_bit(NFS_INO_ODIRECT, &nfsi->flags); + inode_dio_wait(&nfsi->vfs_inode); + } +} + + +/* namespace.c */ +#define NFS_PATH_CANONICAL 1 +extern char *nfs_path(char **p, struct dentry *dentry, + char *buffer, ssize_t buflen, unsigned flags); +extern struct vfsmount *nfs_d_automount(struct path *path); +int nfs_submount(struct fs_context *, struct nfs_server *); +int nfs_do_submount(struct fs_context *); + +/* getroot.c */ +extern int nfs_get_root(struct super_block *s, struct fs_context *fc); +#if IS_ENABLED(CONFIG_NFS_V4) +extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool); +#endif + +struct nfs_pgio_completion_ops; +/* read.c */ +extern const struct nfs_pgio_completion_ops nfs_async_read_completion_ops; +extern void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, + struct inode *inode, bool force_mds, + const struct nfs_pgio_completion_ops *compl_ops); +extern bool nfs_read_alloc_scratch(struct nfs_pgio_header *hdr, size_t size); +extern int nfs_read_add_folio(struct nfs_pageio_descriptor *pgio, + struct nfs_open_context *ctx, + struct folio *folio); +extern void nfs_pageio_complete_read(struct nfs_pageio_descriptor *pgio); +extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio); + +/* super.c */ +void nfs_umount_begin(struct super_block *); +int nfs_statfs(struct dentry *, struct kstatfs *); +int nfs_show_options(struct seq_file *, struct dentry *); +int nfs_show_devname(struct seq_file *, struct dentry *); +int nfs_show_path(struct seq_file *, struct dentry *); +int nfs_show_stats(struct seq_file *, struct dentry *); +int nfs_reconfigure(struct fs_context *); + +/* write.c */ +extern void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, + struct inode *inode, int ioflags, bool force_mds, + const struct nfs_pgio_completion_ops *compl_ops); +extern void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio); +extern void nfs_commit_free(struct nfs_commit_data *p); +extern void nfs_commit_prepare(struct rpc_task *task, void *calldata); +extern int nfs_initiate_commit(struct rpc_clnt *clnt, + struct nfs_commit_data *data, + const struct nfs_rpc_ops *nfs_ops, + const struct rpc_call_ops *call_ops, + int how, int flags, + struct nfsd_file *localio); +extern void nfs_init_commit(struct nfs_commit_data *data, + struct list_head *head, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo); +int nfs_scan_commit_list(struct list_head *src, struct list_head *dst, + struct nfs_commit_info *cinfo, int max); +unsigned long nfs_reqs_to_commit(struct nfs_commit_info *); +int nfs_scan_commit(struct inode *inode, struct list_head *dst, + struct nfs_commit_info *cinfo); +void nfs_mark_request_commit(struct nfs_page *req, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx); +int nfs_write_need_commit(struct nfs_pgio_header *); +void nfs_writeback_update_inode(struct nfs_pgio_header *hdr); +int nfs_generic_commit_list(struct inode *inode, struct list_head *head, + int how, struct nfs_commit_info *cinfo); +void nfs_retry_commit(struct list_head *page_list, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx); +void nfs_commitdata_release(struct nfs_commit_data *data); +void nfs_request_add_commit_list(struct nfs_page *req, + struct nfs_commit_info *cinfo); +void nfs_request_add_commit_list_locked(struct nfs_page *req, + struct list_head *dst, + struct nfs_commit_info *cinfo); +void nfs_request_remove_commit_list(struct nfs_page *req, + struct nfs_commit_info *cinfo); +void nfs_init_cinfo(struct nfs_commit_info *cinfo, + struct inode *inode, + struct nfs_direct_req *dreq); +int nfs_key_timeout_notify(struct file *filp, struct inode *inode); +bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode); +void nfs_pageio_stop_mirroring(struct nfs_pageio_descriptor *pgio); + +int nfs_filemap_write_and_wait_range(struct address_space *mapping, + loff_t lstart, loff_t lend); + +#ifdef CONFIG_NFS_V4 +static inline void +pnfs_bucket_clear_pnfs_ds_commit_verifiers(struct pnfs_commit_bucket *buckets, + unsigned int nbuckets) +{ + unsigned int i; + + for (i = 0; i < nbuckets; i++) + buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW; +} +static inline +void nfs_clear_pnfs_ds_commit_verifiers(struct pnfs_ds_commit_info *cinfo) +{ + struct pnfs_commit_array *array; + + rcu_read_lock(); + list_for_each_entry_rcu(array, &cinfo->commits, cinfo_list) + pnfs_bucket_clear_pnfs_ds_commit_verifiers(array->buckets, + array->nbuckets); + rcu_read_unlock(); +} +#else /* CONFIG_NFS_V4 */ +static inline +void nfs_clear_pnfs_ds_commit_verifiers(struct pnfs_ds_commit_info *cinfo) +{ +} +#endif /* CONFIG_NFS_V4 */ + +#ifdef CONFIG_MIGRATION +int nfs_migrate_folio(struct address_space *, struct folio *dst, + struct folio *src, enum migrate_mode); +#else +#define nfs_migrate_folio NULL +#endif + +static inline int +nfs_write_verifier_cmp(const struct nfs_write_verifier *v1, + const struct nfs_write_verifier *v2) +{ + return memcmp(v1->data, v2->data, sizeof(v1->data)); +} + +static inline bool +nfs_write_match_verf(const struct nfs_writeverf *verf, + struct nfs_page *req) +{ + return verf->committed > NFS_UNSTABLE && + !nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier); +} + +static inline gfp_t nfs_io_gfp_mask(void) +{ + gfp_t ret = current_gfp_context(GFP_KERNEL); + + /* For workers __GFP_NORETRY only with __GFP_IO or __GFP_FS */ + if ((current->flags & PF_WQ_WORKER) && ret == GFP_KERNEL) + ret |= __GFP_NORETRY | __GFP_NOWARN; + return ret; +} + +/* + * Special version of should_remove_suid() that ignores capabilities. + */ +static inline int nfs_should_remove_suid(const struct inode *inode) +{ + umode_t mode = inode->i_mode; + int kill = 0; + + /* suid always must be killed */ + if (unlikely(mode & S_ISUID)) + kill = ATTR_KILL_SUID; + + /* + * sgid without any exec bits is just a mandatory locking mark; leave + * it alone. If some exec bits are set, it's a real sgid; kill it. + */ + if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) + kill |= ATTR_KILL_SGID; + + if (unlikely(kill && S_ISREG(mode))) + return kill; + + return 0; +} + +/* unlink.c */ +extern struct rpc_task * +nfs_async_rename(struct inode *old_dir, struct inode *new_dir, + struct dentry *old_dentry, struct dentry *new_dentry, + void (*complete)(struct rpc_task *, struct nfs_renamedata *)); +extern int nfs_sillyrename(struct inode *dir, struct dentry *dentry); + +/* direct.c */ +void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo, + struct nfs_direct_req *dreq); +extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq, loff_t offset); + +/* nfs4proc.c */ +extern struct nfs_client *nfs4_init_client(struct nfs_client *clp, + const struct nfs_client_initdata *); +extern int nfs41_walk_client_list(struct nfs_client *clp, + struct nfs_client **result, + const struct cred *cred); +extern void nfs4_test_session_trunk(struct rpc_clnt *clnt, + struct rpc_xprt *xprt, + void *data); + +static inline struct inode *nfs_igrab_and_active(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + + if (sb && nfs_sb_active(sb)) { + if (igrab(inode)) + return inode; + nfs_sb_deactive(sb); + } + return NULL; +} + +static inline void nfs_iput_and_deactive(struct inode *inode) +{ + if (inode != NULL) { + struct super_block *sb = inode->i_sb; + + iput(inode); + nfs_sb_deactive(sb); + } +} + +/* + * Determine the device name as a string + */ +static inline char *nfs_devname(struct dentry *dentry, + char *buffer, ssize_t buflen) +{ + char *dummy; + return nfs_path(&dummy, dentry, buffer, buflen, NFS_PATH_CANONICAL); +} + +/* + * Determine the actual block size (and log2 thereof) + */ +static inline +unsigned long nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp) +{ + /* make sure blocksize is a power of two */ + if ((bsize & (bsize - 1)) || nrbitsp) { + unsigned char nrbits; + + for (nrbits = 31; nrbits && !(bsize & (1UL << nrbits)); nrbits--) + ; + bsize = 1UL << nrbits; + if (nrbitsp) + *nrbitsp = nrbits; + } + + return bsize; +} + +/* + * Calculate the number of 512byte blocks used. + */ +static inline blkcnt_t nfs_calc_block_size(u64 tsize) +{ + blkcnt_t used = (tsize + 511) >> 9; + return (used > ULONG_MAX) ? ULONG_MAX : used; +} + +/* + * Compute and set NFS server blocksize + */ +static inline +unsigned long nfs_block_size(unsigned long bsize, unsigned char *nrbitsp) +{ + if (bsize < NFS_MIN_FILE_IO_SIZE) + bsize = NFS_DEF_FILE_IO_SIZE; + else if (bsize >= NFS_MAX_FILE_IO_SIZE) + bsize = NFS_MAX_FILE_IO_SIZE; + + return nfs_block_bits(bsize, nrbitsp); +} + +/* + * Compute and set NFS server rsize / wsize + */ +static inline +unsigned long nfs_io_size(unsigned long iosize, enum xprt_transports proto) +{ + if (iosize < NFS_MIN_FILE_IO_SIZE) + iosize = NFS_DEF_FILE_IO_SIZE; + else if (iosize >= NFS_MAX_FILE_IO_SIZE) + iosize = NFS_MAX_FILE_IO_SIZE; + + if (proto == XPRT_TRANSPORT_UDP || iosize < PAGE_SIZE) + return nfs_block_bits(iosize, NULL); + return iosize & PAGE_MASK; +} + +/* + * Determine the maximum file size for a superblock + */ +static inline +void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize) +{ + sb->s_maxbytes = (loff_t)maxfilesize; + if (sb->s_maxbytes > MAX_LFS_FILESIZE || sb->s_maxbytes <= 0) + sb->s_maxbytes = MAX_LFS_FILESIZE; +} + +/* + * Record the request's range as unstable (an extra writeback period) and + * mark its inode as dirty. + */ +static inline void nfs_folio_mark_unstable(struct nfs_page *req, + struct nfs_commit_info *cinfo) +{ + struct folio *folio = nfs_page_to_folio(req); + + if (folio && !cinfo->dreq) { + struct inode *inode = folio->mapping->host; + long nr = DIV_ROUND_UP(req->wb_bytes, PAGE_SIZE); + + /* This range is really still in write-back - just that the + * writeback is happening on the server now. + */ + node_stat_mod_folio(folio, NR_WRITEBACK, nr); + bdi_wb_stat_mod(inode, WB_WRITEBACK, nr); + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + } +} + +/* + * Determine the number of bytes of data the page contains + */ +static inline size_t nfs_folio_length(struct folio *folio) +{ + loff_t i_size = i_size_read(folio->mapping->host); + + if (i_size > 0) { + pgoff_t index = folio->index >> folio_order(folio); + pgoff_t end_index = (i_size - 1) >> folio_shift(folio); + if (index < end_index) + return folio_size(folio); + if (index == end_index) + return offset_in_folio(folio, i_size - 1) + 1; + } + return 0; +} + +/* + * Convert a umode to a dirent->d_type + */ +static inline +unsigned char nfs_umode_to_dtype(umode_t mode) +{ + return (mode >> 12) & 15; +} + +/* + * Determine the number of pages in an array of length 'len' and + * with a base offset of 'base' + */ +static inline unsigned int nfs_page_array_len(unsigned int base, size_t len) +{ + return ((unsigned long)len + (unsigned long)base + PAGE_SIZE - 1) >> + PAGE_SHIFT; +} + +/* + * Convert a struct timespec64 into a 64-bit change attribute + * + * This does approximately the same thing as timespec64_to_ns(), + * but for calculation efficiency, we multiply the seconds by + * 1024*1024*1024. + */ +static inline +u64 nfs_timespec_to_change_attr(const struct timespec64 *ts) +{ + return ((u64)ts->tv_sec << 30) + ts->tv_nsec; +} + +static inline u32 nfs_stateid_hash(const nfs4_stateid *stateid) +{ + return ~crc32_le(0xFFFFFFFF, &stateid->other[0], + NFS4_STATEID_OTHER_SIZE); +} + +static inline bool nfs_current_task_exiting(void) +{ + return (current->flags & PF_EXITING) != 0; +} + +static inline bool nfs_error_is_fatal(int err) +{ + switch (err) { + case -ERESTARTSYS: + case -EINTR: + case -EACCES: + case -EDQUOT: + case -EFBIG: + case -EIO: + case -ENOSPC: + case -EROFS: + case -ESTALE: + case -E2BIG: + case -ENOMEM: + case -ETIMEDOUT: + return true; + default: + return false; + } +} + +static inline bool nfs_error_is_fatal_on_server(int err) +{ + switch (err) { + case 0: + case -ERESTARTSYS: + case -EINTR: + case -ENOMEM: + return false; + } + return nfs_error_is_fatal(err); +} + +/* + * Select between a default port value and a user-specified port value. + * If a zero value is set, then autobind will be used. + */ +static inline void nfs_set_port(struct sockaddr_storage *sap, int *port, + const unsigned short default_port) +{ + if (*port == NFS_UNSPEC_PORT) + *port = default_port; + + rpc_set_port((struct sockaddr *)sap, *port); +} + +struct nfs_direct_req { + struct kref kref; /* release manager */ + + /* I/O parameters */ + struct nfs_open_context *ctx; /* file open context info */ + struct nfs_lock_context *l_ctx; /* Lock context info */ + struct kiocb * iocb; /* controlling i/o request */ + struct inode * inode; /* target file of i/o */ + + /* completion state */ + atomic_t io_count; /* i/os we're waiting for */ + spinlock_t lock; /* protect completion state */ + + loff_t io_start; /* Start offset for I/O */ + ssize_t count, /* bytes actually processed */ + max_count, /* max expected count */ + error; /* any reported error */ + struct completion completion; /* wait for i/o completion */ + + /* commit state */ + struct nfs_mds_commit_info mds_cinfo; /* Storage for cinfo */ + struct pnfs_ds_commit_info ds_cinfo; /* Storage for cinfo */ + struct work_struct work; + int flags; + /* for write */ +#define NFS_ODIRECT_DO_COMMIT (1) /* an unstable reply was received */ +#define NFS_ODIRECT_RESCHED_WRITES (2) /* write verification failed */ + /* for read */ +#define NFS_ODIRECT_SHOULD_DIRTY (3) /* dirty user-space page after read */ +#define NFS_ODIRECT_DONE INT_MAX /* write verification failed */ +}; diff --git a/fs/nfs/io.c b/fs/nfs/io.c new file mode 100644 index 000000000..2faf2003f --- /dev/null +++ b/fs/nfs/io.c @@ -0,0 +1,204 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2016 Trond Myklebust + * + * I/O and data path helper functionality. + */ + +#include +#include +#include +#include +#include +#include + +#include "internal.h" + +/** + * nfs_start_io_read - declare the file is being used for buffered reads + * @inode: file inode + * + * Declare that a buffered read operation is about to start, and ensure + * that we block all direct I/O. + * On exit, the function ensures that the NFS_INO_ODIRECT flag is unset, + * and holds a shared lock on inode->i_rwsem to ensure that the flag + * cannot be changed. + * In practice, this means that buffered read operations are allowed to + * execute in parallel, thanks to the shared lock, whereas direct I/O + * operations need to wait to grab an exclusive lock in order to set + * NFS_INO_ODIRECT. + * Note that buffered writes and truncates both take a write lock on + * inode->i_rwsem, meaning that those are serialised w.r.t. the reads. + */ +int +nfs_start_io_read(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int err; + + /* Be an optimist! */ + err = down_read_killable(&inode->i_rwsem); + if (err) + return err; + if (test_bit(NFS_INO_ODIRECT, &nfsi->flags) == 0) + return 0; + up_read(&inode->i_rwsem); + + /* Slow path.... */ + err = down_write_killable(&inode->i_rwsem); + if (err) + return err; + nfs_file_block_o_direct(nfsi); + downgrade_write(&inode->i_rwsem); + + return 0; +} + +/** + * nfs_end_io_read - declare that the buffered read operation is done + * @inode: file inode + * + * Declare that a buffered read operation is done, and release the shared + * lock on inode->i_rwsem. + */ +void +nfs_end_io_read(struct inode *inode) +{ + up_read(&inode->i_rwsem); +} + +/** + * nfs_start_io_write - declare the file is being used for buffered writes + * @inode: file inode + * + * Declare that a buffered read operation is about to start, and ensure + * that we block all direct I/O. + */ +int +nfs_start_io_write(struct inode *inode) +{ + int err; + + err = down_write_killable(&inode->i_rwsem); + if (!err) + nfs_file_block_o_direct(NFS_I(inode)); + return err; +} +EXPORT_SYMBOL_GPL(nfs_start_io_write); + +/** + * nfs_end_io_write - declare that the buffered write operation is done + * @inode: file inode + * + * Declare that a buffered write operation is done, and release the + * lock on inode->i_rwsem. + */ +void +nfs_end_io_write(struct inode *inode) +{ + up_write(&inode->i_rwsem); +} +EXPORT_SYMBOL_GPL(nfs_end_io_write); + +/* Call with exclusively locked inode->i_rwsem */ +static void nfs_block_buffered(struct nfs_inode *nfsi, struct inode *inode) +{ + if (!test_bit(NFS_INO_ODIRECT, &nfsi->flags)) { + set_bit(NFS_INO_ODIRECT, &nfsi->flags); + nfs_sync_mapping(inode->i_mapping); + } +} + +static int nfs_block_buffered_nowait(struct nfs_inode *nfsi, struct inode *inode) +{ + if (!test_bit(NFS_INO_ODIRECT, &nfsi->flags)) { + if (inode->i_mapping->nrpages != 0) + return 1; + set_bit(NFS_INO_ODIRECT, &nfsi->flags); + } + return 0; +} + +/** + * nfs_start_io_direct - declare the file is being used for direct i/o + * @inode: file inode + * + * Declare that a direct I/O operation is about to start, and ensure + * that we block all buffered I/O. + * On exit, the function ensures that the NFS_INO_ODIRECT flag is set, + * and holds a shared lock on inode->i_rwsem to ensure that the flag + * cannot be changed. + * In practice, this means that direct I/O operations are allowed to + * execute in parallel, thanks to the shared lock, whereas buffered I/O + * operations need to wait to grab an exclusive lock in order to clear + * NFS_INO_ODIRECT. + * Note that buffered writes and truncates both take a write lock on + * inode->i_rwsem, meaning that those are serialised w.r.t. O_DIRECT. + */ +int +nfs_start_io_direct(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int err; + + /* Be an optimist! */ + err = down_read_killable(&inode->i_rwsem); + if (err) + return err; + if (test_bit(NFS_INO_ODIRECT, &nfsi->flags) != 0) + return 0; + up_read(&inode->i_rwsem); + + /* Slow path.... */ + err = down_write_killable(&inode->i_rwsem); + if (err) + return err; + nfs_block_buffered(nfsi, inode); + downgrade_write(&inode->i_rwsem); + + return 0; +} + +/** + * nfs_start_io_direct_nowait - non-blocking variant of nfs_start_io_direct() + * @inode: file inode + * + * Try to declare that a direct I/O operation is about to start without + * blocking. + * Ensure all buffered I/O is blocked. + * If this could not be done without blocking then returns -EAGAIN. + */ +int +nfs_start_io_direct_nowait(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + if (!down_read_trylock(&inode->i_rwsem)) + return -EAGAIN; + if (test_bit(NFS_INO_ODIRECT, &nfsi->flags)) + return 0; + up_read(&inode->i_rwsem); + + /* Slow path: try to flip NFS_INO_ODIRECT without blocking. */ + if (!down_write_trylock(&inode->i_rwsem)) + return -EAGAIN; + if (nfs_block_buffered_nowait(nfsi, inode)) { + up_write(&inode->i_rwsem); + return -EAGAIN; + } + downgrade_write(&inode->i_rwsem); + return 0; +} + +/** + * nfs_end_io_direct - declare that the direct i/o operation is done + * @inode: file inode + * + * Declare that a direct I/O operation is done, and release the shared + * lock on inode->i_rwsem. + */ +void +nfs_end_io_direct(struct inode *inode) +{ + up_read(&inode->i_rwsem); +} diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h new file mode 100644 index 000000000..49862c95b --- /dev/null +++ b/fs/nfs/iostat.h @@ -0,0 +1,61 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * linux/fs/nfs/iostat.h + * + * Declarations for NFS client per-mount statistics + * + * Copyright (C) 2005, 2006 Chuck Lever + * + */ + +#ifndef _NFS_IOSTAT +#define _NFS_IOSTAT + +#include +#include +#include + +struct nfs_iostats { + unsigned long long bytes[__NFSIOS_BYTESMAX]; + unsigned long events[__NFSIOS_COUNTSMAX]; +} ____cacheline_aligned; + +static inline void nfs_inc_server_stats(const struct nfs_server *server, + enum nfs_stat_eventcounters stat) +{ + this_cpu_inc(server->io_stats->events[stat]); +} + +static inline void nfs_inc_stats(const struct inode *inode, + enum nfs_stat_eventcounters stat) +{ + nfs_inc_server_stats(NFS_SERVER(inode), stat); +} + +static inline void nfs_add_server_stats(const struct nfs_server *server, + enum nfs_stat_bytecounters stat, + long addend) +{ + this_cpu_add(server->io_stats->bytes[stat], addend); +} + +static inline void nfs_add_stats(const struct inode *inode, + enum nfs_stat_bytecounters stat, + long addend) +{ + nfs_add_server_stats(NFS_SERVER(inode), stat, addend); +} + +/* + * This specialized allocator has to be a macro for its allocations to be + * accounted separately (to have a separate alloc_tag). + */ +#define nfs_alloc_iostats() alloc_percpu(struct nfs_iostats) + +static inline void nfs_free_iostats(struct nfs_iostats __percpu *stats) +{ + if (stats != NULL) + free_percpu(stats); +} + +#endif /* _NFS_IOSTAT */ diff --git a/fs/nfs/localio.c b/fs/nfs/localio.c new file mode 100644 index 000000000..e55c5977f --- /dev/null +++ b/fs/nfs/localio.c @@ -0,0 +1,1120 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * NFS client support for local clients to bypass network stack + * + * Copyright (C) 2014 Weston Andros Adamson + * Copyright (C) 2019 Trond Myklebust + * Copyright (C) 2024 Mike Snitzer + * Copyright (C) 2024 NeilBrown + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" +#include "pnfs.h" +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +#define NFSLOCAL_MAX_IOS 3 + +struct nfs_local_kiocb { + struct kiocb kiocb; + struct bio_vec *bvec; + struct nfs_pgio_header *hdr; + struct work_struct work; + void (*aio_complete_work)(struct work_struct *); + struct nfsd_file *localio; + /* Begin mostly DIO-specific members */ + size_t end_len; + short int end_iter_index; + atomic_t n_iters; + struct iov_iter iters[NFSLOCAL_MAX_IOS]; + bool iter_is_dio_aligned[NFSLOCAL_MAX_IOS]; + /* End mostly DIO-specific members */ +}; + +struct nfs_local_fsync_ctx { + struct nfsd_file *localio; + struct nfs_commit_data *data; + struct work_struct work; + struct completion *done; +}; + +static bool localio_enabled __read_mostly = true; +module_param(localio_enabled, bool, 0644); + +static void nfs_local_do_read(struct nfs_local_kiocb *iocb, + const struct rpc_call_ops *call_ops); +static void nfs_local_do_write(struct nfs_local_kiocb *iocb, + const struct rpc_call_ops *call_ops); + +static inline bool nfs_client_is_local(const struct nfs_client *clp) +{ + return !!rcu_access_pointer(clp->cl_uuid.net); +} + +bool nfs_server_is_local(const struct nfs_client *clp) +{ + return nfs_client_is_local(clp) && localio_enabled; +} +EXPORT_SYMBOL_GPL(nfs_server_is_local); + +/* + * UUID_IS_LOCAL XDR functions + */ + +static void localio_xdr_enc_uuidargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const u8 *uuid = data; + + encode_opaque_fixed(xdr, uuid, UUID_SIZE); +} + +static int localio_xdr_dec_uuidres(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + /* void return */ + return 0; +} + +static const struct rpc_procinfo nfs_localio_procedures[] = { + [LOCALIOPROC_UUID_IS_LOCAL] = { + .p_proc = LOCALIOPROC_UUID_IS_LOCAL, + .p_encode = localio_xdr_enc_uuidargs, + .p_decode = localio_xdr_dec_uuidres, + .p_arglen = XDR_QUADLEN(UUID_SIZE), + .p_replen = 0, + .p_statidx = LOCALIOPROC_UUID_IS_LOCAL, + .p_name = "UUID_IS_LOCAL", + }, +}; + +static unsigned int nfs_localio_counts[ARRAY_SIZE(nfs_localio_procedures)]; +static const struct rpc_version nfslocalio_version1 = { + .number = 1, + .nrprocs = ARRAY_SIZE(nfs_localio_procedures), + .procs = nfs_localio_procedures, + .counts = nfs_localio_counts, +}; + +static const struct rpc_version *nfslocalio_version[] = { + [1] = &nfslocalio_version1, +}; + +extern const struct rpc_program nfslocalio_program; +static struct rpc_stat nfslocalio_rpcstat = { &nfslocalio_program }; + +const struct rpc_program nfslocalio_program = { + .name = "nfslocalio", + .number = NFS_LOCALIO_PROGRAM, + .nrvers = ARRAY_SIZE(nfslocalio_version), + .version = nfslocalio_version, + .stats = &nfslocalio_rpcstat, +}; + +/* + * nfs_init_localioclient - Initialise an NFS localio client connection + */ +static struct rpc_clnt *nfs_init_localioclient(struct nfs_client *clp) +{ + struct rpc_clnt *rpcclient_localio; + + rpcclient_localio = rpc_bind_new_program(clp->cl_rpcclient, + &nfslocalio_program, 1); + + dprintk_rcu("%s: server (%s) %s NFS LOCALIO.\n", + __func__, rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR), + (IS_ERR(rpcclient_localio) ? "does not support" : "supports")); + + return rpcclient_localio; +} + +static bool nfs_server_uuid_is_local(struct nfs_client *clp) +{ + u8 uuid[UUID_SIZE]; + struct rpc_message msg = { + .rpc_argp = &uuid, + }; + struct rpc_clnt *rpcclient_localio; + int status; + + rpcclient_localio = nfs_init_localioclient(clp); + if (IS_ERR(rpcclient_localio)) + return false; + + export_uuid(uuid, &clp->cl_uuid.uuid); + + msg.rpc_proc = &nfs_localio_procedures[LOCALIOPROC_UUID_IS_LOCAL]; + status = rpc_call_sync(rpcclient_localio, &msg, 0); + dprintk("%s: NFS reply UUID_IS_LOCAL: status=%d\n", + __func__, status); + rpc_shutdown_client(rpcclient_localio); + + /* Server is only local if it initialized required struct members */ + if (status || !rcu_access_pointer(clp->cl_uuid.net) || !clp->cl_uuid.dom) + return false; + + return true; +} + +/* + * nfs_local_probe - probe local i/o support for an nfs_server and nfs_client + * - called after alloc_client and init_client (so cl_rpcclient exists) + * - this function is idempotent, it can be called for old or new clients + */ +static void nfs_local_probe(struct nfs_client *clp) +{ + /* Disallow localio if disabled via sysfs or AUTH_SYS isn't used */ + if (!localio_enabled || + clp->cl_rpcclient->cl_auth->au_flavor != RPC_AUTH_UNIX) { + nfs_localio_disable_client(clp); + return; + } + + if (nfs_client_is_local(clp)) + return; + + if (!nfs_uuid_begin(&clp->cl_uuid)) + return; + if (nfs_server_uuid_is_local(clp)) + nfs_localio_enable_client(clp); + nfs_uuid_end(&clp->cl_uuid); +} + +void nfs_local_probe_async_work(struct work_struct *work) +{ + struct nfs_client *clp = + container_of(work, struct nfs_client, cl_local_probe_work); + + if (!refcount_inc_not_zero(&clp->cl_count)) + return; + nfs_local_probe(clp); + nfs_put_client(clp); +} + +void nfs_local_probe_async(struct nfs_client *clp) +{ + queue_work(nfsiod_workqueue, &clp->cl_local_probe_work); +} +EXPORT_SYMBOL_GPL(nfs_local_probe_async); + +static inline void nfs_local_file_put(struct nfsd_file *localio) +{ + /* nfs_to_nfsd_file_put_local() expects an __rcu pointer + * but we have a __kernel pointer. It is always safe + * to cast a __kernel pointer to an __rcu pointer + * because the cast only weakens what is known about the pointer. + */ + struct nfsd_file __rcu *nf = (struct nfsd_file __rcu*) localio; + + nfs_to_nfsd_file_put_local(&nf); +} + +/* + * __nfs_local_open_fh - open a local filehandle in terms of nfsd_file. + * + * Returns a pointer to a struct nfsd_file or ERR_PTR. + * Caller must release returned nfsd_file with nfs_to_nfsd_file_put_local(). + */ +static struct nfsd_file * +__nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, + struct nfs_fh *fh, struct nfs_file_localio *nfl, + struct nfsd_file __rcu **pnf, + const fmode_t mode) +{ + int status = 0; + struct nfsd_file *localio; + + localio = nfs_open_local_fh(&clp->cl_uuid, clp->cl_rpcclient, + cred, fh, nfl, pnf, mode); + if (IS_ERR(localio)) { + status = PTR_ERR(localio); + switch (status) { + case -ENOMEM: + case -ENXIO: + case -ENOENT: + /* Revalidate localio */ + nfs_localio_disable_client(clp); + nfs_local_probe(clp); + } + } + trace_nfs_local_open_fh(fh, mode, status); + return localio; +} + +/* + * nfs_local_open_fh - open a local filehandle in terms of nfsd_file. + * First checking if the open nfsd_file is already cached, otherwise + * must __nfs_local_open_fh and insert the nfsd_file in nfs_file_localio. + * + * Returns a pointer to a struct nfsd_file or NULL. + */ +struct nfsd_file * +nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, + struct nfs_fh *fh, struct nfs_file_localio *nfl, + const fmode_t mode) +{ + struct nfsd_file *nf, __rcu **pnf; + + if (!nfs_server_is_local(clp)) + return NULL; + if (mode & ~(FMODE_READ | FMODE_WRITE)) + return NULL; + + if (mode & FMODE_WRITE) + pnf = &nfl->rw_file; + else + pnf = &nfl->ro_file; + + nf = __nfs_local_open_fh(clp, cred, fh, nfl, pnf, mode); + if (IS_ERR(nf)) + return NULL; + return nf; +} +EXPORT_SYMBOL_GPL(nfs_local_open_fh); + +/* + * Ensure all page cache allocations are done from GFP_NOFS context to + * prevent direct reclaim recursion back into NFS via nfs_writepages. + */ +static void +nfs_local_mapping_set_gfp_nofs_context(struct address_space *m) +{ + gfp_t gfp_mask = mapping_gfp_mask(m); + + mapping_set_gfp_mask(m, (gfp_mask & ~(__GFP_FS))); +} + +static void +nfs_local_iocb_free(struct nfs_local_kiocb *iocb) +{ + kfree(iocb->bvec); + kfree(iocb); +} + +static struct nfs_local_kiocb * +nfs_local_iocb_alloc(struct nfs_pgio_header *hdr, + struct file *file, gfp_t flags) +{ + struct nfs_local_kiocb *iocb; + + iocb = kzalloc_obj(*iocb, flags); + if (iocb == NULL) + return NULL; + + iocb->bvec = kmalloc_objs(struct bio_vec, hdr->page_array.npages, flags); + if (iocb->bvec == NULL) { + kfree(iocb); + return NULL; + } + + nfs_local_mapping_set_gfp_nofs_context(file->f_mapping); + init_sync_kiocb(&iocb->kiocb, file); + + iocb->hdr = hdr; + iocb->kiocb.ki_pos = hdr->args.offset; + iocb->kiocb.ki_flags &= ~IOCB_APPEND; + iocb->kiocb.ki_complete = NULL; + iocb->aio_complete_work = NULL; + + iocb->end_iter_index = -1; + + return iocb; +} + +static bool +nfs_is_local_dio_possible(struct nfs_local_kiocb *iocb, int rw, + size_t len, struct nfs_local_dio *local_dio) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + loff_t offset = hdr->args.offset; + u32 nf_dio_mem_align, nf_dio_offset_align, nf_dio_read_offset_align; + loff_t start_end, orig_end, middle_end; + + nfs_to->nfsd_file_dio_alignment(iocb->localio, &nf_dio_mem_align, + &nf_dio_offset_align, &nf_dio_read_offset_align); + if (rw == ITER_DEST) + nf_dio_offset_align = nf_dio_read_offset_align; + + if (unlikely(!nf_dio_mem_align || !nf_dio_offset_align)) + return false; + if (unlikely(len < nf_dio_offset_align)) + return false; + + local_dio->mem_align = nf_dio_mem_align; + local_dio->offset_align = nf_dio_offset_align; + + start_end = round_up(offset, nf_dio_offset_align); + orig_end = offset + len; + middle_end = round_down(orig_end, nf_dio_offset_align); + + local_dio->middle_offset = start_end; + local_dio->end_offset = middle_end; + + local_dio->start_len = start_end - offset; + local_dio->middle_len = middle_end - start_end; + local_dio->end_len = orig_end - middle_end; + + if (rw == ITER_DEST) + trace_nfs_local_dio_read(hdr->inode, offset, len, local_dio); + else + trace_nfs_local_dio_write(hdr->inode, offset, len, local_dio); + return true; +} + +static bool nfs_iov_iter_aligned_bvec(const struct iov_iter *i, + unsigned int addr_mask, unsigned int len_mask) +{ + const struct bio_vec *bvec = i->bvec; + size_t skip = i->iov_offset; + size_t size = i->count; + + if (size & len_mask) + return false; + do { + size_t len = bvec->bv_len; + + if (len > size) + len = size; + if ((unsigned long)(bvec->bv_offset + skip) & addr_mask) + return false; + bvec++; + size -= len; + skip = 0; + } while (size); + + return true; +} + +static void +nfs_local_iter_setup(struct iov_iter *iter, int rw, struct bio_vec *bvec, + unsigned int nvecs, unsigned long total, + size_t start, size_t len) +{ + iov_iter_bvec(iter, rw, bvec, nvecs, total); + if (start) + iov_iter_advance(iter, start); + iov_iter_truncate(iter, len); +} + +/* + * Setup as many as 3 iov_iter based on extents described by @local_dio. + * Returns the number of iov_iter that were setup. + */ +static int +nfs_local_iters_setup_dio(struct nfs_local_kiocb *iocb, int rw, + unsigned int nvecs, unsigned long total, + struct nfs_local_dio *local_dio) +{ + int n_iters = 0; + struct iov_iter *iters = iocb->iters; + + /* Setup misaligned start? */ + if (local_dio->start_len) { + nfs_local_iter_setup(&iters[n_iters], rw, iocb->bvec, + nvecs, total, 0, local_dio->start_len); + ++n_iters; + } + + /* + * Setup DIO-aligned middle, if there is no misaligned end (below) + * then AIO completion is used, see nfs_local_call_{read,write} + */ + nfs_local_iter_setup(&iters[n_iters], rw, iocb->bvec, nvecs, + total, local_dio->start_len, local_dio->middle_len); + + iocb->iter_is_dio_aligned[n_iters] = + nfs_iov_iter_aligned_bvec(&iters[n_iters], + local_dio->mem_align-1, local_dio->offset_align-1); + + if (unlikely(!iocb->iter_is_dio_aligned[n_iters])) { + trace_nfs_local_dio_misaligned(iocb->hdr->inode, + local_dio->start_len, local_dio->middle_len, local_dio); + return 0; /* no DIO-aligned IO possible */ + } + iocb->end_iter_index = n_iters; + ++n_iters; + + /* Setup misaligned end? */ + if (local_dio->end_len) { + nfs_local_iter_setup(&iters[n_iters], rw, iocb->bvec, + nvecs, total, local_dio->start_len + + local_dio->middle_len, local_dio->end_len); + iocb->end_iter_index = n_iters; + ++n_iters; + } + + atomic_set(&iocb->n_iters, n_iters); + return n_iters; +} + +static noinline_for_stack void +nfs_local_iters_init(struct nfs_local_kiocb *iocb, int rw) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + struct page **pagevec = hdr->page_array.pagevec; + unsigned long v, total; + unsigned int base; + size_t len; + + v = 0; + total = hdr->args.count; + base = hdr->args.pgbase; + pagevec += base >> PAGE_SHIFT; + base &= ~PAGE_MASK; + while (total && v < hdr->page_array.npages) { + len = min_t(size_t, total, PAGE_SIZE - base); + bvec_set_page(&iocb->bvec[v], *pagevec, len, base); + total -= len; + ++pagevec; + ++v; + base = 0; + } + len = hdr->args.count - total; + + /* + * For each iocb, iocb->n_iters is always at least 1 and we always + * end io after first nfs_local_pgio_done call unless misaligned DIO. + */ + atomic_set(&iocb->n_iters, 1); + + if (test_bit(NFS_IOHDR_ODIRECT, &hdr->flags)) { + struct nfs_local_dio local_dio; + + if (nfs_is_local_dio_possible(iocb, rw, len, &local_dio) && + nfs_local_iters_setup_dio(iocb, rw, v, len, &local_dio) != 0) { + /* Ensure DIO WRITE's IO on stable storage upon completion */ + if (rw == ITER_SOURCE) + iocb->kiocb.ki_flags |= IOCB_DSYNC|IOCB_SYNC; + return; /* is DIO-aligned */ + } + } + + /* Use buffered IO */ + iov_iter_bvec(&iocb->iters[0], rw, iocb->bvec, v, len); +} + +static void +nfs_local_hdr_release(struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops) +{ + call_ops->rpc_call_done(&hdr->task, hdr); + call_ops->rpc_release(hdr); +} + +static void +nfs_local_pgio_init(struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops) +{ + hdr->task.tk_ops = call_ops; + if (!hdr->task.tk_start) + hdr->task.tk_start = ktime_get(); +} + +static bool nfs_local_pgio_done(struct nfs_local_kiocb *iocb, long status) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + + /* Must handle partial completions */ + if (status >= 0) { + hdr->res.count += status; + /* @hdr was initialized to 0 (zeroed during allocation) */ + if (hdr->task.tk_status == 0) + hdr->res.op_status = NFS4_OK; + } else { + hdr->res.op_status = nfs_localio_errno_to_nfs4_stat(status); + hdr->task.tk_status = status; + } + + BUG_ON(atomic_read(&iocb->n_iters) <= 0); + return atomic_dec_and_test(&iocb->n_iters); +} + +static void +nfs_local_iocb_release(struct nfs_local_kiocb *iocb) +{ + nfs_local_file_put(iocb->localio); + nfs_local_iocb_free(iocb); +} + +static void nfs_local_pgio_restart(struct nfs_local_kiocb *iocb, + struct nfs_pgio_header *hdr) +{ + int status = 0; + + iocb->kiocb.ki_pos = hdr->args.offset; + iocb->kiocb.ki_flags &= ~(IOCB_DSYNC | IOCB_SYNC | IOCB_DIRECT); + iocb->kiocb.ki_complete = NULL; + iocb->aio_complete_work = NULL; + iocb->end_iter_index = -1; + + switch (hdr->rw_mode) { + case FMODE_READ: + nfs_local_iters_init(iocb, ITER_DEST); + nfs_local_do_read(iocb, hdr->task.tk_ops); + break; + case FMODE_WRITE: + nfs_local_iters_init(iocb, ITER_SOURCE); + nfs_local_do_write(iocb, hdr->task.tk_ops); + break; + default: + status = -EOPNOTSUPP; + } + + if (unlikely(status != 0)) { + nfs_local_iocb_release(iocb); + hdr->task.tk_status = status; + nfs_local_hdr_release(hdr, hdr->task.tk_ops); + } +} + +static void nfs_local_pgio_release(struct nfs_local_kiocb *iocb) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + struct rpc_task *task = &hdr->task; + + task->tk_action = NULL; + task->tk_ops->rpc_call_done(task, hdr); + + if (task->tk_action == NULL) { + nfs_local_iocb_release(iocb); + task->tk_ops->rpc_release(hdr); + } else + nfs_local_pgio_restart(iocb, hdr); +} + +/* + * Complete the I/O from iocb->kiocb.ki_complete() + * + * Note that this function can be called from a bottom half context, + * hence we need to queue the rpc_call_done() etc to a workqueue + */ +static inline void nfs_local_pgio_aio_complete(struct nfs_local_kiocb *iocb) +{ + INIT_WORK(&iocb->work, iocb->aio_complete_work); + queue_work(nfsiod_workqueue, &iocb->work); +} + +static void nfs_local_read_done(struct nfs_local_kiocb *iocb) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + struct file *filp = iocb->kiocb.ki_filp; + long status = hdr->task.tk_status; + + if ((iocb->kiocb.ki_flags & IOCB_DIRECT) && status == -EINVAL) { + /* Underlying FS will return -EINVAL if misaligned DIO is attempted. */ + pr_info_ratelimited("nfs: Unexpected direct I/O read alignment failure\n"); + } + + /* + * Must clear replen otherwise NFSv3 data corruption will occur + * if/when switching from LOCALIO back to using normal RPC. + */ + hdr->res.replen = 0; + + /* nfs_readpage_result() handles short read */ + + if (hdr->args.offset + hdr->res.count >= i_size_read(file_inode(filp))) + hdr->res.eof = true; + + dprintk("%s: read %ld bytes eof %d.\n", __func__, + status > 0 ? status : 0, hdr->res.eof); +} + +static inline void nfs_local_read_iocb_done(struct nfs_local_kiocb *iocb) +{ + nfs_local_read_done(iocb); + nfs_local_pgio_release(iocb); +} + +static void nfs_local_read_aio_complete_work(struct work_struct *work) +{ + struct nfs_local_kiocb *iocb = + container_of(work, struct nfs_local_kiocb, work); + + nfs_local_read_iocb_done(iocb); +} + +static void nfs_local_read_aio_complete(struct kiocb *kiocb, long ret) +{ + struct nfs_local_kiocb *iocb = + container_of(kiocb, struct nfs_local_kiocb, kiocb); + + /* AIO completion of DIO read should always be last to complete */ + if (unlikely(!nfs_local_pgio_done(iocb, ret))) + return; + + nfs_local_pgio_aio_complete(iocb); /* Calls nfs_local_read_aio_complete_work */ +} + +static void nfs_local_call_read(struct work_struct *work) +{ + struct nfs_local_kiocb *iocb = + container_of(work, struct nfs_local_kiocb, work); + struct file *filp = iocb->kiocb.ki_filp; + ssize_t status; + int n_iters; + + n_iters = atomic_read(&iocb->n_iters); + for (int i = 0; i < n_iters ; i++) { + if (iocb->iter_is_dio_aligned[i]) { + iocb->kiocb.ki_flags |= IOCB_DIRECT; + /* Only use AIO completion if DIO-aligned segment is last */ + if (i == iocb->end_iter_index) { + iocb->kiocb.ki_complete = nfs_local_read_aio_complete; + iocb->aio_complete_work = nfs_local_read_aio_complete_work; + } + } else + iocb->kiocb.ki_flags &= ~IOCB_DIRECT; + + scoped_with_creds(filp->f_cred) + status = filp->f_op->read_iter(&iocb->kiocb, &iocb->iters[i]); + + if (status == -EIOCBQUEUED) + continue; + /* Break on completion, errors, or short reads */ + if (nfs_local_pgio_done(iocb, status) || status < 0 || + (size_t)status < iov_iter_count(&iocb->iters[i])) { + nfs_local_read_iocb_done(iocb); + break; + } + } +} + +static void nfs_local_do_read(struct nfs_local_kiocb *iocb, + const struct rpc_call_ops *call_ops) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + + dprintk("%s: vfs_read count=%u pos=%llu\n", + __func__, hdr->args.count, hdr->args.offset); + + nfs_local_pgio_init(hdr, call_ops); + hdr->res.eof = false; + + INIT_WORK(&iocb->work, nfs_local_call_read); + queue_work(nfslocaliod_workqueue, &iocb->work); +} + +static void +nfs_copy_boot_verifier(struct nfs_write_verifier *verifier, struct inode *inode) +{ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + u32 *verf = (u32 *)verifier->data; + unsigned int seq; + + do { + seq = read_seqbegin(&clp->cl_boot_lock); + verf[0] = (u32)clp->cl_nfssvc_boot.tv_sec; + verf[1] = (u32)clp->cl_nfssvc_boot.tv_nsec; + } while (read_seqretry(&clp->cl_boot_lock, seq)); +} + +static void +nfs_reset_boot_verifier(struct inode *inode) +{ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + + write_seqlock(&clp->cl_boot_lock); + ktime_get_real_ts64(&clp->cl_nfssvc_boot); + write_sequnlock(&clp->cl_boot_lock); +} + +static void +nfs_set_local_verifier(struct inode *inode, + struct nfs_writeverf *verf, + enum nfs3_stable_how how) +{ + nfs_copy_boot_verifier(&verf->verifier, inode); + verf->committed = how; +} + +/* Factored out from fs/nfsd/vfs.h:fh_getattr() */ +static int __vfs_getattr(const struct path *p, struct kstat *stat, int version) +{ + u32 request_mask = STATX_BASIC_STATS; + + if (version == 4) + request_mask |= (STATX_BTIME | STATX_CHANGE_COOKIE); + return vfs_getattr(p, stat, request_mask, AT_STATX_SYNC_AS_STAT); +} + +/* Copied from fs/nfsd/nfsfh.c:nfsd4_change_attribute() */ +static u64 __nfsd4_change_attribute(const struct kstat *stat, + const struct inode *inode) +{ + u64 chattr; + + if (stat->result_mask & STATX_CHANGE_COOKIE) { + chattr = stat->change_cookie; + if (S_ISREG(inode->i_mode) && + !(stat->attributes & STATX_ATTR_CHANGE_MONOTONIC)) { + chattr += (u64)stat->ctime.tv_sec << 30; + chattr += stat->ctime.tv_nsec; + } + } else { + chattr = time_to_chattr(&stat->ctime); + } + return chattr; +} + +static void nfs_local_vfs_getattr(struct nfs_local_kiocb *iocb) +{ + struct kstat stat; + struct file *filp = iocb->kiocb.ki_filp; + struct nfs_pgio_header *hdr = iocb->hdr; + struct nfs_fattr *fattr = hdr->res.fattr; + int version = NFS_PROTO(hdr->inode)->version; + + if (unlikely(!fattr) || __vfs_getattr(&filp->f_path, &stat, version)) + return; + + fattr->valid = (NFS_ATTR_FATTR_FILEID | + NFS_ATTR_FATTR_CHANGE | + NFS_ATTR_FATTR_SIZE | + NFS_ATTR_FATTR_ATIME | + NFS_ATTR_FATTR_MTIME | + NFS_ATTR_FATTR_CTIME | + NFS_ATTR_FATTR_SPACE_USED); + + fattr->fileid = stat.ino; + fattr->size = stat.size; + fattr->atime = stat.atime; + fattr->mtime = stat.mtime; + fattr->ctime = stat.ctime; + if (version == 4) { + fattr->change_attr = + __nfsd4_change_attribute(&stat, file_inode(filp)); + } else + fattr->change_attr = nfs_timespec_to_change_attr(&fattr->ctime); + fattr->du.nfs3.used = stat.blocks << 9; +} + +static void nfs_local_write_done(struct nfs_local_kiocb *iocb) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + long status = hdr->task.tk_status; + + dprintk("%s: wrote %ld bytes.\n", __func__, status > 0 ? status : 0); + + if ((iocb->kiocb.ki_flags & IOCB_DIRECT) && status == -EINVAL) { + /* Underlying FS will return -EINVAL if misaligned DIO is attempted. */ + pr_info_ratelimited("nfs: Unexpected direct I/O write alignment failure\n"); + } + + if (status < 0) + nfs_reset_boot_verifier(hdr->inode); +} + +static inline void nfs_local_write_iocb_done(struct nfs_local_kiocb *iocb) +{ + nfs_local_write_done(iocb); + nfs_local_vfs_getattr(iocb); + nfs_local_pgio_release(iocb); +} + +static void nfs_local_write_aio_complete_work(struct work_struct *work) +{ + struct nfs_local_kiocb *iocb = + container_of(work, struct nfs_local_kiocb, work); + + nfs_local_write_iocb_done(iocb); +} + +static void nfs_local_write_aio_complete(struct kiocb *kiocb, long ret) +{ + struct nfs_local_kiocb *iocb = + container_of(kiocb, struct nfs_local_kiocb, kiocb); + + /* AIO completion of DIO write should always be last to complete */ + if (unlikely(!nfs_local_pgio_done(iocb, ret))) + return; + + nfs_local_pgio_aio_complete(iocb); /* Calls nfs_local_write_aio_complete_work */ +} + +static void nfs_local_call_write(struct work_struct *work) +{ + struct nfs_local_kiocb *iocb = + container_of(work, struct nfs_local_kiocb, work); + struct file *filp = iocb->kiocb.ki_filp; + unsigned long old_flags = current->flags; + ssize_t status; + int n_iters; + + current->flags |= PF_LOCAL_THROTTLE | PF_MEMALLOC_NOIO; + + file_start_write(filp); + n_iters = atomic_read(&iocb->n_iters); + for (int i = 0; i < n_iters ; i++) { + size_t icount; + + if (iocb->iter_is_dio_aligned[i]) { + iocb->kiocb.ki_flags |= IOCB_DIRECT; + /* Only use AIO completion if DIO-aligned segment is last */ + if (i == iocb->end_iter_index) { + iocb->kiocb.ki_complete = nfs_local_write_aio_complete; + iocb->aio_complete_work = nfs_local_write_aio_complete_work; + } + } else + iocb->kiocb.ki_flags &= ~IOCB_DIRECT; + + scoped_with_creds(filp->f_cred) + status = filp->f_op->write_iter(&iocb->kiocb, &iocb->iters[i]); + + if (status == -EIOCBQUEUED) + continue; + /* Break on completion, errors, or short writes */ + icount = iov_iter_count(&iocb->iters[i]); + if (nfs_local_pgio_done(iocb, status) || status < 0 || + (size_t)status < icount) { + if ((size_t)status < icount) { + struct nfs_lock_context *ctx = + iocb->hdr->req->wb_lock_context; + + set_bit(NFS_CONTEXT_WRITE_SYNC, + &ctx->open_context->flags); + } + nfs_local_write_iocb_done(iocb); + break; + } + } + file_end_write(filp); + + current->flags = old_flags; +} + +static void nfs_local_do_write(struct nfs_local_kiocb *iocb, + const struct rpc_call_ops *call_ops) +{ + struct nfs_pgio_header *hdr = iocb->hdr; + + dprintk("%s: vfs_write count=%u pos=%llu %s\n", + __func__, hdr->args.count, hdr->args.offset, + (hdr->args.stable == NFS_UNSTABLE) ? "unstable" : "stable"); + + if (test_bit(NFS_CONTEXT_WRITE_SYNC, + &hdr->req->wb_lock_context->open_context->flags)) + hdr->args.stable = NFS_FILE_SYNC; + switch (hdr->args.stable) { + default: + break; + case NFS_DATA_SYNC: + iocb->kiocb.ki_flags |= IOCB_DSYNC; + break; + case NFS_FILE_SYNC: + iocb->kiocb.ki_flags |= IOCB_DSYNC|IOCB_SYNC; + } + + nfs_local_pgio_init(hdr, call_ops); + + nfs_set_local_verifier(hdr->inode, hdr->res.verf, hdr->args.stable); + + INIT_WORK(&iocb->work, nfs_local_call_write); + queue_work(nfslocaliod_workqueue, &iocb->work); +} + +static struct nfs_local_kiocb * +nfs_local_iocb_init(struct nfs_pgio_header *hdr, struct nfsd_file *localio) +{ + struct file *file = nfs_to->nfsd_file_file(localio); + struct nfs_local_kiocb *iocb; + gfp_t gfp_mask; + int rw; + + if (hdr->rw_mode & FMODE_READ) { + if (!file->f_op->read_iter) + return ERR_PTR(-EOPNOTSUPP); + gfp_mask = GFP_KERNEL; + rw = ITER_DEST; + } else { + if (!file->f_op->write_iter) + return ERR_PTR(-EOPNOTSUPP); + gfp_mask = GFP_NOIO; + rw = ITER_SOURCE; + } + + iocb = nfs_local_iocb_alloc(hdr, file, gfp_mask); + if (iocb == NULL) + return ERR_PTR(-ENOMEM); + iocb->hdr = hdr; + iocb->localio = localio; + + nfs_local_iters_init(iocb, rw); + + return iocb; +} + +int nfs_local_doio(struct nfs_client *clp, struct nfsd_file *localio, + struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops) +{ + struct nfs_local_kiocb *iocb; + int status = 0; + + if (!hdr->args.count) + return 0; + + iocb = nfs_local_iocb_init(hdr, localio); + if (IS_ERR(iocb)) + return PTR_ERR(iocb); + + switch (hdr->rw_mode) { + case FMODE_READ: + nfs_local_do_read(iocb, call_ops); + break; + case FMODE_WRITE: + nfs_local_do_write(iocb, call_ops); + break; + default: + dprintk("%s: invalid mode: %d\n", __func__, + hdr->rw_mode); + status = -EOPNOTSUPP; + } + + if (unlikely(status != 0)) { + nfs_local_iocb_release(iocb); + hdr->task.tk_status = status; + nfs_local_hdr_release(hdr, call_ops); + } + return status; +} + +static void +nfs_local_init_commit(struct nfs_commit_data *data, + const struct rpc_call_ops *call_ops) +{ + data->task.tk_ops = call_ops; +} + +static int +nfs_local_run_commit(struct file *filp, struct nfs_commit_data *data) +{ + loff_t start = data->args.offset; + loff_t end = LLONG_MAX; + + if (data->args.count > 0) { + end = start + data->args.count - 1; + if (end < start) + end = LLONG_MAX; + } + + nfs_local_mapping_set_gfp_nofs_context(filp->f_mapping); + + dprintk("%s: commit %llu - %llu\n", __func__, start, end); + return vfs_fsync_range(filp, start, end, 0); +} + +static void +nfs_local_commit_done(struct nfs_commit_data *data, int status) +{ + if (status >= 0) { + nfs_set_local_verifier(data->inode, + data->res.verf, + NFS_FILE_SYNC); + data->res.op_status = NFS4_OK; + data->task.tk_status = 0; + } else { + nfs_reset_boot_verifier(data->inode); + data->res.op_status = nfs_localio_errno_to_nfs4_stat(status); + data->task.tk_status = status; + } +} + +static void +nfs_local_release_commit_data(struct nfsd_file *localio, + struct nfs_commit_data *data, + const struct rpc_call_ops *call_ops) +{ + nfs_local_file_put(localio); + call_ops->rpc_call_done(&data->task, data); + call_ops->rpc_release(data); +} + +static void +nfs_local_fsync_ctx_free(struct nfs_local_fsync_ctx *ctx) +{ + nfs_local_release_commit_data(ctx->localio, ctx->data, + ctx->data->task.tk_ops); + kfree(ctx); +} + +static void +nfs_local_fsync_work(struct work_struct *work) +{ + unsigned long old_flags = current->flags; + struct nfs_local_fsync_ctx *ctx; + int status; + + ctx = container_of(work, struct nfs_local_fsync_ctx, work); + + current->flags |= PF_LOCAL_THROTTLE | PF_MEMALLOC_NOIO; + + status = nfs_local_run_commit(nfs_to->nfsd_file_file(ctx->localio), + ctx->data); + nfs_local_commit_done(ctx->data, status); + if (ctx->done != NULL) + complete(ctx->done); + nfs_local_fsync_ctx_free(ctx); + + current->flags = old_flags; +} + +static struct nfs_local_fsync_ctx * +nfs_local_fsync_ctx_alloc(struct nfs_commit_data *data, + struct nfsd_file *localio, gfp_t flags) +{ + struct nfs_local_fsync_ctx *ctx = kmalloc_obj(*ctx, flags); + + if (ctx != NULL) { + ctx->localio = localio; + ctx->data = data; + INIT_WORK(&ctx->work, nfs_local_fsync_work); + ctx->done = NULL; + } + return ctx; +} + +int nfs_local_commit(struct nfsd_file *localio, + struct nfs_commit_data *data, + const struct rpc_call_ops *call_ops, int how) +{ + struct nfs_local_fsync_ctx *ctx; + + ctx = nfs_local_fsync_ctx_alloc(data, localio, GFP_NOIO); + if (!ctx) { + nfs_local_commit_done(data, -ENOMEM); + nfs_local_release_commit_data(localio, data, call_ops); + return -ENOMEM; + } + + nfs_local_init_commit(data, call_ops); + + if (how & FLUSH_SYNC) { + DECLARE_COMPLETION_ONSTACK(done); + ctx->done = &done; + queue_work(nfslocaliod_workqueue, &ctx->work); + wait_for_completion(&done); + } else + queue_work(nfslocaliod_workqueue, &ctx->work); + + return 0; +} diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c new file mode 100644 index 000000000..db8dfb920 --- /dev/null +++ b/fs/nfs/mount_clnt.c @@ -0,0 +1,471 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * In-kernel MOUNT protocol client + * + * Copyright (C) 1997, Olaf Kirch + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_MOUNT + +/* + * Defined by RFC 1094, section A.3; and RFC 1813, section 5.1.4 + */ +#define MNTPATHLEN (1024) + +/* + * XDR data type sizes + */ +#define encode_dirpath_sz (1 + XDR_QUADLEN(MNTPATHLEN)) +#define MNT_status_sz (1) +#define MNT_fhandle_sz XDR_QUADLEN(NFS2_FHSIZE) +#define MNT_fhandlev3_sz XDR_QUADLEN(NFS3_FHSIZE) +#define MNT_authflav3_sz (1 + NFS_MAX_SECFLAVORS) + +/* + * XDR argument and result sizes + */ +#define MNT_enc_dirpath_sz encode_dirpath_sz +#define MNT_dec_mountres_sz (MNT_status_sz + MNT_fhandle_sz) +#define MNT_dec_mountres3_sz (MNT_status_sz + MNT_fhandlev3_sz + \ + MNT_authflav3_sz) + +/* + * Defined by RFC 1094, section A.5 + */ +enum { + MOUNTPROC_NULL = 0, + MOUNTPROC_MNT = 1, + MOUNTPROC_DUMP = 2, + MOUNTPROC_UMNT = 3, + MOUNTPROC_UMNTALL = 4, + MOUNTPROC_EXPORT = 5, +}; + +/* + * Defined by RFC 1813, section 5.2 + */ +enum { + MOUNTPROC3_NULL = 0, + MOUNTPROC3_MNT = 1, + MOUNTPROC3_DUMP = 2, + MOUNTPROC3_UMNT = 3, + MOUNTPROC3_UMNTALL = 4, + MOUNTPROC3_EXPORT = 5, +}; + +static const struct rpc_program mnt_program; + +/* + * Defined by OpenGroup XNFS Version 3W, chapter 8 + */ +enum mountstat { + MNT_OK = 0, + MNT_EPERM = 1, + MNT_ENOENT = 2, + MNT_EACCES = 13, + MNT_EINVAL = 22, +}; + +static struct { + u32 status; + int errno; +} mnt_errtbl[] = { + { .status = MNT_OK, .errno = 0, }, + { .status = MNT_EPERM, .errno = -EPERM, }, + { .status = MNT_ENOENT, .errno = -ENOENT, }, + { .status = MNT_EACCES, .errno = -EACCES, }, + { .status = MNT_EINVAL, .errno = -EINVAL, }, +}; + +/* + * Defined by RFC 1813, section 5.1.5 + */ +enum mountstat3 { + MNT3_OK = 0, /* no error */ + MNT3ERR_PERM = 1, /* Not owner */ + MNT3ERR_NOENT = 2, /* No such file or directory */ + MNT3ERR_IO = 5, /* I/O error */ + MNT3ERR_ACCES = 13, /* Permission denied */ + MNT3ERR_NOTDIR = 20, /* Not a directory */ + MNT3ERR_INVAL = 22, /* Invalid argument */ + MNT3ERR_NAMETOOLONG = 63, /* Filename too long */ + MNT3ERR_NOTSUPP = 10004, /* Operation not supported */ + MNT3ERR_SERVERFAULT = 10006, /* A failure on the server */ +}; + +static struct { + u32 status; + int errno; +} mnt3_errtbl[] = { + { .status = MNT3_OK, .errno = 0, }, + { .status = MNT3ERR_PERM, .errno = -EPERM, }, + { .status = MNT3ERR_NOENT, .errno = -ENOENT, }, + { .status = MNT3ERR_IO, .errno = -EIO, }, + { .status = MNT3ERR_ACCES, .errno = -EACCES, }, + { .status = MNT3ERR_NOTDIR, .errno = -ENOTDIR, }, + { .status = MNT3ERR_INVAL, .errno = -EINVAL, }, + { .status = MNT3ERR_NAMETOOLONG, .errno = -ENAMETOOLONG, }, + { .status = MNT3ERR_NOTSUPP, .errno = -ENOTSUPP, }, + { .status = MNT3ERR_SERVERFAULT, .errno = -EREMOTEIO, }, +}; + +struct mountres { + int errno; + struct nfs_fh *fh; + unsigned int *auth_count; + rpc_authflavor_t *auth_flavors; +}; + +/** + * nfs_mount - Obtain an NFS file handle for the given host and path + * @info: pointer to mount request arguments + * @timeo: deciseconds the mount waits for a response before it retries + * @retrans: number of times the mount retries a request + * + * Uses timeout parameters specified by caller. On successful return, the + * auth_flavs list and auth_flav_len will be populated with the list from the + * server or a faked-up list if the server didn't provide one. + */ +int nfs_mount(struct nfs_mount_request *info, int timeo, int retrans) +{ + struct rpc_timeout mnt_timeout; + struct mountres result = { + .fh = info->fh, + .auth_count = info->auth_flav_len, + .auth_flavors = info->auth_flavs, + }; + struct rpc_message msg = { + .rpc_argp = info->dirpath, + .rpc_resp = &result, + }; + struct rpc_create_args args = { + .net = info->net, + .protocol = info->protocol, + .address = (struct sockaddr *)info->sap, + .addrsize = info->salen, + .timeout = &mnt_timeout, + .servername = info->hostname, + .program = &mnt_program, + .version = info->version, + .authflavor = RPC_AUTH_UNIX, + .cred = current_cred(), + }; + struct rpc_clnt *mnt_clnt; + int status; + + dprintk("NFS: sending MNT request for %s:%s\n", + (info->hostname ? info->hostname : "server"), + info->dirpath); + + if (strlen(info->dirpath) > MNTPATHLEN) + return -ENAMETOOLONG; + + if (info->noresvport) + args.flags |= RPC_CLNT_CREATE_NONPRIVPORT; + + nfs_init_timeout_values(&mnt_timeout, info->protocol, timeo, retrans); + mnt_clnt = rpc_create(&args); + if (IS_ERR(mnt_clnt)) + goto out_clnt_err; + + if (info->version == NFS_MNT3_VERSION) + msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC3_MNT]; + else + msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC_MNT]; + + status = rpc_call_sync(mnt_clnt, &msg, RPC_TASK_SOFT|RPC_TASK_TIMEOUT); + rpc_shutdown_client(mnt_clnt); + + if (status < 0) + goto out_call_err; + if (result.errno != 0) + goto out_mnt_err; + + dprintk("NFS: MNT request succeeded\n"); + status = 0; + + /* + * If the server didn't provide a flavor list, allow the + * client to try any flavor. + */ + if (info->version != NFS_MNT3_VERSION || *info->auth_flav_len == 0) { + dprintk("NFS: Faking up auth_flavs list\n"); + info->auth_flavs[0] = RPC_AUTH_NULL; + *info->auth_flav_len = 1; + } +out: + return status; + +out_clnt_err: + status = PTR_ERR(mnt_clnt); + dprintk("NFS: failed to create MNT RPC client, status=%d\n", status); + goto out; + +out_call_err: + dprintk("NFS: MNT request failed, status=%d\n", status); + goto out; + +out_mnt_err: + dprintk("NFS: MNT server returned result %d\n", result.errno); + status = result.errno; + goto out; +} + +/* + * XDR encode/decode functions for MOUNT + */ + +static void encode_mntdirpath(struct xdr_stream *xdr, const char *pathname) +{ + const u32 pathname_len = strlen(pathname); + __be32 *p; + + p = xdr_reserve_space(xdr, 4 + pathname_len); + xdr_encode_opaque(p, pathname, pathname_len); +} + +static void mnt_xdr_enc_dirpath(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *dirpath) +{ + encode_mntdirpath(xdr, dirpath); +} + +/* + * RFC 1094: "A non-zero status indicates some sort of error. In this + * case, the status is a UNIX error number." This can be problematic + * if the server and client use different errno values for the same + * error. + * + * However, the OpenGroup XNFS spec provides a simple mapping that is + * independent of local errno values on the server and the client. + */ +static int decode_status(struct xdr_stream *xdr, struct mountres *res) +{ + unsigned int i; + u32 status; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return -EIO; + status = be32_to_cpup(p); + + for (i = 0; i < ARRAY_SIZE(mnt_errtbl); i++) { + if (mnt_errtbl[i].status == status) { + res->errno = mnt_errtbl[i].errno; + return 0; + } + } + + dprintk("NFS: unrecognized MNT status code: %u\n", status); + res->errno = -EACCES; + return 0; +} + +static int decode_fhandle(struct xdr_stream *xdr, struct mountres *res) +{ + struct nfs_fh *fh = res->fh; + __be32 *p; + + p = xdr_inline_decode(xdr, NFS2_FHSIZE); + if (unlikely(p == NULL)) + return -EIO; + + fh->size = NFS2_FHSIZE; + memcpy(fh->data, p, NFS2_FHSIZE); + return 0; +} + +static int mnt_xdr_dec_mountres(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct mountres *res = data; + int status; + + status = decode_status(xdr, res); + if (unlikely(status != 0 || res->errno != 0)) + return status; + return decode_fhandle(xdr, res); +} + +static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res) +{ + unsigned int i; + u32 status; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return -EIO; + status = be32_to_cpup(p); + + for (i = 0; i < ARRAY_SIZE(mnt3_errtbl); i++) { + if (mnt3_errtbl[i].status == status) { + res->errno = mnt3_errtbl[i].errno; + return 0; + } + } + + dprintk("NFS: unrecognized MNT3 status code: %u\n", status); + res->errno = -EACCES; + return 0; +} + +static int decode_fhandle3(struct xdr_stream *xdr, struct mountres *res) +{ + struct nfs_fh *fh = res->fh; + u32 size; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return -EIO; + + size = be32_to_cpup(p); + if (size > NFS3_FHSIZE || size == 0) + return -EIO; + + p = xdr_inline_decode(xdr, size); + if (unlikely(p == NULL)) + return -EIO; + + fh->size = size; + memcpy(fh->data, p, size); + return 0; +} + +static int decode_auth_flavors(struct xdr_stream *xdr, struct mountres *res) +{ + rpc_authflavor_t *flavors = res->auth_flavors; + unsigned int *count = res->auth_count; + u32 entries, i; + __be32 *p; + + if (*count == 0) + return 0; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return -EIO; + entries = be32_to_cpup(p); + dprintk("NFS: received %u auth flavors\n", entries); + if (entries > NFS_MAX_SECFLAVORS) + entries = NFS_MAX_SECFLAVORS; + + p = xdr_inline_decode(xdr, 4 * entries); + if (unlikely(p == NULL)) + return -EIO; + + if (entries > *count) + entries = *count; + + for (i = 0; i < entries; i++) { + flavors[i] = be32_to_cpup(p++); + dprintk("NFS: auth flavor[%u]: %d\n", i, flavors[i]); + } + *count = i; + + return 0; +} + +static int mnt_xdr_dec_mountres3(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct mountres *res = data; + int status; + + status = decode_fhs_status(xdr, res); + if (unlikely(status != 0 || res->errno != 0)) + return status; + status = decode_fhandle3(xdr, res); + if (unlikely(status != 0)) { + res->errno = -EBADHANDLE; + return 0; + } + return decode_auth_flavors(xdr, res); +} + +static const struct rpc_procinfo mnt_procedures[] = { + [MOUNTPROC_MNT] = { + .p_proc = MOUNTPROC_MNT, + .p_encode = mnt_xdr_enc_dirpath, + .p_decode = mnt_xdr_dec_mountres, + .p_arglen = MNT_enc_dirpath_sz, + .p_replen = MNT_dec_mountres_sz, + .p_statidx = MOUNTPROC_MNT, + .p_name = "MOUNT", + }, + [MOUNTPROC_UMNT] = { + .p_proc = MOUNTPROC_UMNT, + .p_encode = mnt_xdr_enc_dirpath, + .p_arglen = MNT_enc_dirpath_sz, + .p_statidx = MOUNTPROC_UMNT, + .p_name = "UMOUNT", + }, +}; + +static const struct rpc_procinfo mnt3_procedures[] = { + [MOUNTPROC3_MNT] = { + .p_proc = MOUNTPROC3_MNT, + .p_encode = mnt_xdr_enc_dirpath, + .p_decode = mnt_xdr_dec_mountres3, + .p_arglen = MNT_enc_dirpath_sz, + .p_replen = MNT_dec_mountres3_sz, + .p_statidx = MOUNTPROC3_MNT, + .p_name = "MOUNT", + }, + [MOUNTPROC3_UMNT] = { + .p_proc = MOUNTPROC3_UMNT, + .p_encode = mnt_xdr_enc_dirpath, + .p_arglen = MNT_enc_dirpath_sz, + .p_statidx = MOUNTPROC3_UMNT, + .p_name = "UMOUNT", + }, +}; + +static unsigned int mnt_counts[ARRAY_SIZE(mnt_procedures)]; +static const struct rpc_version mnt_version1 = { + .number = 1, + .nrprocs = ARRAY_SIZE(mnt_procedures), + .procs = mnt_procedures, + .counts = mnt_counts, +}; + +static unsigned int mnt3_counts[ARRAY_SIZE(mnt3_procedures)]; +static const struct rpc_version mnt_version3 = { + .number = 3, + .nrprocs = ARRAY_SIZE(mnt3_procedures), + .procs = mnt3_procedures, + .counts = mnt3_counts, +}; + +static const struct rpc_version *mnt_version[] = { + NULL, + &mnt_version1, + NULL, + &mnt_version3, +}; + +static struct rpc_stat mnt_stats; + +static const struct rpc_program mnt_program = { + .name = "mount", + .number = NFS_MNT_PROGRAM, + .nrvers = ARRAY_SIZE(mnt_version), + .version = mnt_version, + .stats = &mnt_stats, +}; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c new file mode 100644 index 000000000..6d0073c24 --- /dev/null +++ b/fs/nfs/namespace.c @@ -0,0 +1,386 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/namespace.c + * + * Copyright (C) 2005 Trond Myklebust + * - Modified by David Howells + * + * NFS namespace + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" +#include "nfs.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static void nfs_expire_automounts(struct work_struct *work); + +static LIST_HEAD(nfs_automount_list); +static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts); +int nfs_mountpoint_expiry_timeout = 500 * HZ; + +/* + * nfs_path - reconstruct the path given an arbitrary dentry + * @base - used to return pointer to the end of devname part of path + * @dentry_in - pointer to dentry + * @buffer - result buffer + * @buflen_in - length of buffer + * @flags - options (see below) + * + * Helper function for constructing the server pathname + * by arbitrary hashed dentry. + * + * This is mainly for use in figuring out the path on the + * server side when automounting on top of an existing partition + * and in generating /proc/mounts and friends. + * + * Supported flags: + * NFS_PATH_CANONICAL: ensure there is exactly one slash after + * the original device (export) name + * (if unset, the original name is returned verbatim) + */ +char *nfs_path(char **p, struct dentry *dentry_in, char *buffer, + ssize_t buflen_in, unsigned flags) +{ + char *end; + int namelen; + unsigned seq; + const char *base; + struct dentry *dentry; + ssize_t buflen; + +rename_retry: + buflen = buflen_in; + dentry = dentry_in; + end = buffer+buflen; + *--end = '\0'; + buflen--; + + seq = read_seqbegin(&rename_lock); + rcu_read_lock(); + while (1) { + spin_lock(&dentry->d_lock); + if (IS_ROOT(dentry)) + break; + namelen = dentry->d_name.len; + buflen -= namelen + 1; + if (buflen < 0) + goto Elong_unlock; + end -= namelen; + memcpy(end, dentry->d_name.name, namelen); + *--end = '/'; + spin_unlock(&dentry->d_lock); + dentry = dentry->d_parent; + } + if (read_seqretry(&rename_lock, seq)) { + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + goto rename_retry; + } + if ((flags & NFS_PATH_CANONICAL) && *end != '/') { + if (--buflen < 0) { + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + goto Elong; + } + *--end = '/'; + } + *p = end; + base = dentry->d_fsdata; + if (!base) { + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + WARN_ON(1); + return end; + } + namelen = strlen(base); + if (*end == '/') { + /* Strip off excess slashes in base string */ + while (namelen > 0 && base[namelen - 1] == '/') + namelen--; + } + buflen -= namelen; + if (buflen < 0) { + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + goto Elong; + } + end -= namelen; + memcpy(end, base, namelen); + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + return end; +Elong_unlock: + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + if (read_seqretry(&rename_lock, seq)) + goto rename_retry; +Elong: + return ERR_PTR(-ENAMETOOLONG); +} +EXPORT_SYMBOL_GPL(nfs_path); + +/* + * nfs_d_automount - Handle crossing a mountpoint on the server + * @path - The mountpoint + * + * When we encounter a mountpoint on the server, we want to set up + * a mountpoint on the client too, to prevent inode numbers from + * colliding, and to allow "df" to work properly. + * On NFSv4, we also want to allow for the fact that different + * filesystems may be migrated to different servers in a failover + * situation, and that different filesystems may want to use + * different security flavours. + */ +struct vfsmount *nfs_d_automount(struct path *path) +{ + struct nfs_fs_context *ctx; + struct fs_context *fc; + struct vfsmount *mnt = ERR_PTR(-ENOMEM); + struct nfs_server *server = NFS_SB(path->dentry->d_sb); + struct nfs_client *client = server->nfs_client; + unsigned long s_flags = path->dentry->d_sb->s_flags; + int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout); + int ret; + + if (IS_ROOT(path->dentry)) + return ERR_PTR(-ESTALE); + + /* Open a new filesystem context, transferring parameters from the + * parent superblock, including the network namespace. + */ + fc = fs_context_for_submount(path->mnt->mnt_sb->s_type, path->dentry); + if (IS_ERR(fc)) + return ERR_CAST(fc); + + ctx = nfs_fc2context(fc); + ctx->clone_data.dentry = path->dentry; + ctx->clone_data.sb = path->dentry->d_sb; + ctx->clone_data.fattr = nfs_alloc_fattr(); + if (!ctx->clone_data.fattr) + goto out_fc; + + if (fc->cred != server->cred) { + put_cred(fc->cred); + fc->cred = get_cred(server->cred); + } + + if (fc->net_ns != client->cl_net) { + put_net(fc->net_ns); + fc->net_ns = get_net(client->cl_net); + } + + /* Inherit the flags covered by NFS_SB_MASK */ + fc->sb_flags_mask |= NFS_SB_MASK; + fc->sb_flags &= ~NFS_SB_MASK; + fc->sb_flags |= s_flags & NFS_SB_MASK; + + /* for submounts we want the same server; referrals will reassign */ + memcpy(&ctx->nfs_server._address, &client->cl_addr, client->cl_addrlen); + ctx->nfs_server.addrlen = client->cl_addrlen; + ctx->nfs_server.port = server->port; + + ctx->version = client->rpc_ops->version; + ctx->minorversion = client->cl_minorversion; + ctx->nfs_mod = client->cl_nfs_mod; + get_nfs_version(ctx->nfs_mod); + + /* Inherit block sizes if they were specified as mount parameters */ + if (server->automount_inherit & NFS_AUTOMOUNT_INHERIT_BSIZE) + ctx->bsize = server->bsize; + + ret = client->rpc_ops->submount(fc, server); + if (ret < 0) { + mnt = ERR_PTR(ret); + goto out_fc; + } + + up_write(&fc->root->d_sb->s_umount); + mnt = vfs_create_mount(fc); + if (IS_ERR(mnt)) + goto out_fc; + + if (timeout <= 0) + goto out_fc; + + mnt_set_expiry(mnt, &nfs_automount_list); + schedule_delayed_work(&nfs_automount_task, timeout); + +out_fc: + put_fs_context(fc); + return mnt; +} + +static int +nfs_namespace_getattr(struct mnt_idmap *idmap, + const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int query_flags) +{ + if (NFS_FH(d_inode(path->dentry))->size != 0) + return nfs_getattr(idmap, path, stat, request_mask, + query_flags); + generic_fillattr(&nop_mnt_idmap, request_mask, d_inode(path->dentry), + stat); + return 0; +} + +static int +nfs_namespace_setattr(struct mnt_idmap *idmap, struct dentry *dentry, + struct iattr *attr) +{ + if (NFS_FH(d_inode(dentry))->size != 0) + return nfs_setattr(idmap, dentry, attr); + return -EACCES; +} + +const struct inode_operations nfs_mountpoint_inode_operations = { + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +}; + +const struct inode_operations nfs_referral_inode_operations = { + .getattr = nfs_namespace_getattr, + .setattr = nfs_namespace_setattr, + .fileattr_get = nfs_fileattr_get, +}; + +static void nfs_expire_automounts(struct work_struct *work) +{ + struct list_head *list = &nfs_automount_list; + int timeout = READ_ONCE(nfs_mountpoint_expiry_timeout); + + mark_mounts_for_expiry(list); + if (!list_empty(list) && timeout > 0) + schedule_delayed_work(&nfs_automount_task, timeout); +} + +void nfs_release_automount_timer(void) +{ + if (list_empty(&nfs_automount_list)) + cancel_delayed_work(&nfs_automount_task); +} + +/** + * nfs_do_submount - set up mountpoint when crossing a filesystem boundary + * @fc: pointer to struct nfs_fs_context + * + */ +int nfs_do_submount(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct dentry *dentry = ctx->clone_data.dentry; + struct nfs_server *server; + char *buffer, *p; + int ret; + + /* create a new volume representation */ + server = ctx->nfs_mod->rpc_ops->clone_server(NFS_SB(ctx->clone_data.sb), + ctx->mntfh, + ctx->clone_data.fattr, + ctx->selected_flavor); + + if (IS_ERR(server)) + return PTR_ERR(server); + + ctx->server = server; + + buffer = kmalloc(4096, GFP_USER); + if (!buffer) + return -ENOMEM; + + ctx->internal = true; + + p = nfs_devname(dentry, buffer, 4096); + if (IS_ERR(p)) { + nfs_errorf(fc, "NFS: Couldn't determine submount pathname"); + ret = PTR_ERR(p); + } else { + ret = vfs_parse_fs_qstr(fc, "source", + &QSTR_LEN(p, buffer + 4096 - p)); + if (!ret) + ret = vfs_get_tree(fc); + } + kfree(buffer); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_do_submount); + +int nfs_submount(struct fs_context *fc, struct nfs_server *server) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct dentry *dentry = ctx->clone_data.dentry; + struct dentry *parent = dget_parent(dentry); + int err; + + /* Look it up again to get its attributes */ + err = server->nfs_client->rpc_ops->lookup(d_inode(parent), dentry, &dentry->d_name, + ctx->mntfh, ctx->clone_data.fattr); + dput(parent); + if (err != 0) + return err; + + ctx->selected_flavor = server->client->cl_auth->au_flavor; + return nfs_do_submount(fc); +} +EXPORT_SYMBOL_GPL(nfs_submount); + +static int param_set_nfs_timeout(const char *val, const struct kernel_param *kp) +{ + long num; + int ret; + + if (!val) + return -EINVAL; + ret = kstrtol(val, 0, &num); + if (ret) + return -EINVAL; + if (num > 0) { + if (num >= INT_MAX / HZ) + num = INT_MAX; + else + num *= HZ; + *((int *)kp->arg) = num; + if (!list_empty(&nfs_automount_list)) + mod_delayed_work(system_percpu_wq, &nfs_automount_task, num); + } else { + *((int *)kp->arg) = -1*HZ; + cancel_delayed_work(&nfs_automount_task); + } + return 0; +} + +static int param_get_nfs_timeout(char *buffer, const struct kernel_param *kp) +{ + long num = *((int *)kp->arg); + + if (num > 0) { + if (num >= INT_MAX - (HZ - 1)) + num = INT_MAX / HZ; + else + num = (num + (HZ - 1)) / HZ; + } else + num = -1; + return sysfs_emit(buffer, "%li\n", num); +} + +static const struct kernel_param_ops param_ops_nfs_timeout = { + .set = param_set_nfs_timeout, + .get = param_get_nfs_timeout, +}; +#define param_check_nfs_timeout(name, p) __param_check(name, p, int) + +module_param(nfs_mountpoint_expiry_timeout, nfs_timeout, 0644); +MODULE_PARM_DESC(nfs_mountpoint_expiry_timeout, + "Set the NFS automounted mountpoint timeout value (seconds)." + "Values <= 0 turn expiration off."); diff --git a/fs/nfs/netns.h b/fs/nfs/netns.h new file mode 100644 index 000000000..366585791 --- /dev/null +++ b/fs/nfs/netns.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * NFS-private data for each "struct net". Accessed with net_generic(). + */ + +#ifndef __NFS_NETNS_H__ +#define __NFS_NETNS_H__ + +#include +#include +#include +#include + +struct bl_dev_msg { + int32_t status; + uint32_t major, minor; +}; + +struct nfs_netns_client; + +struct nfs_net { + struct cache_detail *nfs_dns_resolve; + struct rpc_pipe *bl_device_pipe; + struct bl_dev_msg bl_mount_reply; + wait_queue_head_t bl_wq; + struct mutex bl_mutex; + struct list_head nfs_client_list; + struct list_head nfs_volume_list; +#if IS_ENABLED(CONFIG_NFS_V4) + struct idr cb_ident_idr; /* Protected by nfs_client_lock */ + unsigned short nfs_callback_tcpport; + unsigned short nfs_callback_tcpport6; + int cb_users[NFS4_MAX_MINOR_VERSION + 1]; + struct list_head nfs4_data_server_cache; + spinlock_t nfs4_data_server_lock; +#endif /* CONFIG_NFS_V4 */ + struct nfs_netns_client *nfs_client; + spinlock_t nfs_client_lock; + ktime_t boot_time; + struct rpc_stat rpcstats; +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *proc_nfsfs; +#endif +}; + +extern unsigned int nfs_net_id; + +#endif diff --git a/fs/nfs/nfs.h b/fs/nfs/nfs.h new file mode 100644 index 000000000..8a5f51be0 --- /dev/null +++ b/fs/nfs/nfs.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2012 Netapp, Inc. All rights reserved. + * + * Function and structures exported by the NFS module + * for use by NFS version-specific modules. + */ +#ifndef __LINUX_INTERNAL_NFS_H +#define __LINUX_INTERNAL_NFS_H + +#include +#include +#include + +struct nfs_subversion { + struct module *owner; /* THIS_MODULE pointer */ + struct file_system_type *nfs_fs; /* NFS filesystem type */ + const struct rpc_version *rpc_vers; /* NFS version information */ + const struct nfs_rpc_ops *rpc_ops; /* NFS operations */ + const struct super_operations *sops; /* NFS Super operations */ + const struct xattr_handler * const *xattr; /* NFS xattr handlers */ +}; + +struct nfs_subversion *find_nfs_version(unsigned int); +int get_nfs_version(struct nfs_subversion *); +void put_nfs_version(struct nfs_subversion *); +void register_nfs_version(struct nfs_subversion *); +void unregister_nfs_version(struct nfs_subversion *); + +#endif /* __LINUX_INTERNAL_NFS_H */ diff --git a/fs/nfs/nfs2super.c b/fs/nfs/nfs2super.c new file mode 100644 index 000000000..b1badc70b --- /dev/null +++ b/fs/nfs/nfs2super.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012 Netapp, Inc. All rights reserved. + */ +#include +#include +#include "internal.h" +#include "nfs.h" + +static struct nfs_subversion nfs_v2 = { + .owner = THIS_MODULE, + .nfs_fs = &nfs_fs_type, + .rpc_vers = &nfs_version2, + .rpc_ops = &nfs_v2_clientops, + .sops = &nfs_sops, +}; + +static int __init init_nfs_v2(void) +{ + register_nfs_version(&nfs_v2); + return 0; +} + +static void __exit exit_nfs_v2(void) +{ + unregister_nfs_version(&nfs_v2); +} + +MODULE_DESCRIPTION("NFSv2 client support"); +MODULE_LICENSE("GPL"); + +module_init(init_nfs_v2); +module_exit(exit_nfs_v2); diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c new file mode 100644 index 000000000..9eff09158 --- /dev/null +++ b/fs/nfs/nfs2xdr.c @@ -0,0 +1,1088 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/nfs2xdr.c + * + * XDR functions to encode/decode NFS RPC arguments and results. + * + * Copyright (C) 1992, 1993, 1994 Rick Sladkey + * Copyright (C) 1996 Olaf Kirch + * 04 Aug 1998 Ion Badulescu + * FIFO's need special handling in NFSv2 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_XDR + +/* + * Declare the space requirements for NFS arguments and replies as + * number of 32bit-words + */ +#define NFS_pagepad_sz (1) /* Page padding */ +#define NFS_fhandle_sz (8) +#define NFS_sattr_sz (8) +#define NFS_filename_sz (1+(NFS2_MAXNAMLEN>>2)) +#define NFS_path_sz (1+(NFS2_MAXPATHLEN>>2)) +#define NFS_fattr_sz (17) +#define NFS_info_sz (5) +#define NFS_entry_sz (NFS_filename_sz+3) + +#define NFS_diropargs_sz (NFS_fhandle_sz+NFS_filename_sz) +#define NFS_removeargs_sz (NFS_fhandle_sz+NFS_filename_sz) +#define NFS_sattrargs_sz (NFS_fhandle_sz+NFS_sattr_sz) +#define NFS_readlinkargs_sz (NFS_fhandle_sz) +#define NFS_readargs_sz (NFS_fhandle_sz+3) +#define NFS_writeargs_sz (NFS_fhandle_sz+4) +#define NFS_createargs_sz (NFS_diropargs_sz+NFS_sattr_sz) +#define NFS_renameargs_sz (NFS_diropargs_sz+NFS_diropargs_sz) +#define NFS_linkargs_sz (NFS_fhandle_sz+NFS_diropargs_sz) +#define NFS_symlinkargs_sz (NFS_diropargs_sz+1+NFS_sattr_sz) +#define NFS_readdirargs_sz (NFS_fhandle_sz+2) + +#define NFS_attrstat_sz (1+NFS_fattr_sz) +#define NFS_diropres_sz (1+NFS_fhandle_sz+NFS_fattr_sz) +#define NFS_readlinkres_sz (2+NFS_pagepad_sz) +#define NFS_readres_sz (1+NFS_fattr_sz+1+NFS_pagepad_sz) +#define NFS_writeres_sz (NFS_attrstat_sz) +#define NFS_stat_sz (1) +#define NFS_readdirres_sz (1+NFS_pagepad_sz) +#define NFS_statfsres_sz (1+NFS_info_sz) + +/* + * Encode/decode NFSv2 basic data types + * + * Basic NFSv2 data types are defined in section 2.3 of RFC 1094: + * "NFS: Network File System Protocol Specification". + * + * Not all basic data types have their own encoding and decoding + * functions. For run-time efficiency, some data types are encoded + * or decoded inline. + */ + +static struct user_namespace *rpc_userns(const struct rpc_clnt *clnt) +{ + if (clnt && clnt->cl_cred) + return clnt->cl_cred->user_ns; + return &init_user_ns; +} + +static struct user_namespace *rpc_rqst_userns(const struct rpc_rqst *rqstp) +{ + if (rqstp->rq_task) + return rpc_userns(rqstp->rq_task->tk_client); + return &init_user_ns; +} + +/* + * typedef opaque nfsdata<>; + */ +static int decode_nfsdata(struct xdr_stream *xdr, struct nfs_pgio_res *result) +{ + u32 recvd, count; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p); + recvd = xdr_read_pages(xdr, count); + if (unlikely(count > recvd)) + goto out_cheating; +out: + result->eof = 0; /* NFSv2 does not pass EOF flag on the wire. */ + result->count = count; + return count; +out_cheating: + dprintk("NFS: server cheating in read result: " + "count %u > recvd %u\n", count, recvd); + count = recvd; + goto out; +} + +/* + * enum stat { + * NFS_OK = 0, + * NFSERR_PERM = 1, + * NFSERR_NOENT = 2, + * NFSERR_IO = 5, + * NFSERR_NXIO = 6, + * NFSERR_ACCES = 13, + * NFSERR_EXIST = 17, + * NFSERR_NODEV = 19, + * NFSERR_NOTDIR = 20, + * NFSERR_ISDIR = 21, + * NFSERR_FBIG = 27, + * NFSERR_NOSPC = 28, + * NFSERR_ROFS = 30, + * NFSERR_NAMETOOLONG = 63, + * NFSERR_NOTEMPTY = 66, + * NFSERR_DQUOT = 69, + * NFSERR_STALE = 70, + * NFSERR_WFLUSH = 99 + * }; + */ +static int decode_stat(struct xdr_stream *xdr, enum nfs_stat *status) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + if (unlikely(*p != cpu_to_be32(NFS_OK))) + goto out_status; + *status = 0; + return 0; +out_status: + *status = be32_to_cpup(p); + trace_nfs_xdr_status(xdr, (int)*status); + return 0; +} + +/* + * 2.3.2. ftype + * + * enum ftype { + * NFNON = 0, + * NFREG = 1, + * NFDIR = 2, + * NFBLK = 3, + * NFCHR = 4, + * NFLNK = 5 + * }; + * + */ +static __be32 *xdr_decode_ftype(__be32 *p, u32 *type) +{ + *type = be32_to_cpup(p++); + if (unlikely(*type > NF2FIFO)) + *type = NFBAD; + return p; +} + +/* + * 2.3.3. fhandle + * + * typedef opaque fhandle[FHSIZE]; + */ +static void encode_fhandle(struct xdr_stream *xdr, const struct nfs_fh *fh) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, NFS2_FHSIZE); + memcpy(p, fh->data, NFS2_FHSIZE); +} + +static int decode_fhandle(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS2_FHSIZE); + if (unlikely(!p)) + return -EIO; + fh->size = NFS2_FHSIZE; + memcpy(fh->data, p, NFS2_FHSIZE); + return 0; +} + +/* + * 2.3.4. timeval + * + * struct timeval { + * unsigned int seconds; + * unsigned int useconds; + * }; + */ +static __be32 *xdr_encode_time(__be32 *p, const struct timespec64 *timep) +{ + *p++ = cpu_to_be32((u32)timep->tv_sec); + if (timep->tv_nsec != 0) + *p++ = cpu_to_be32(timep->tv_nsec / NSEC_PER_USEC); + else + *p++ = cpu_to_be32(0); + return p; +} + +/* + * Passing the invalid value useconds=1000000 is a Sun convention for + * "set to current server time". It's needed to make permissions checks + * for the "touch" program across v2 mounts to Solaris and Irix servers + * work correctly. See description of sattr in section 6.1 of "NFS + * Illustrated" by Brent Callaghan, Addison-Wesley, ISBN 0-201-32750-5. + */ +static __be32 *xdr_encode_current_server_time(__be32 *p, + const struct timespec64 *timep) +{ + *p++ = cpu_to_be32(timep->tv_sec); + *p++ = cpu_to_be32(1000000); + return p; +} + +static __be32 *xdr_decode_time(__be32 *p, struct timespec64 *timep) +{ + timep->tv_sec = be32_to_cpup(p++); + timep->tv_nsec = be32_to_cpup(p++) * NSEC_PER_USEC; + return p; +} + +/* + * 2.3.5. fattr + * + * struct fattr { + * ftype type; + * unsigned int mode; + * unsigned int nlink; + * unsigned int uid; + * unsigned int gid; + * unsigned int size; + * unsigned int blocksize; + * unsigned int rdev; + * unsigned int blocks; + * unsigned int fsid; + * unsigned int fileid; + * timeval atime; + * timeval mtime; + * timeval ctime; + * }; + * + */ +static int decode_fattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) +{ + u32 rdev, type; + __be32 *p; + + p = xdr_inline_decode(xdr, NFS_fattr_sz << 2); + if (unlikely(!p)) + return -EIO; + + fattr->valid |= NFS_ATTR_FATTR_V2; + + p = xdr_decode_ftype(p, &type); + + fattr->mode = be32_to_cpup(p++); + fattr->nlink = be32_to_cpup(p++); + fattr->uid = make_kuid(userns, be32_to_cpup(p++)); + if (!uid_valid(fattr->uid)) + goto out_uid; + fattr->gid = make_kgid(userns, be32_to_cpup(p++)); + if (!gid_valid(fattr->gid)) + goto out_gid; + + fattr->size = be32_to_cpup(p++); + fattr->du.nfs2.blocksize = be32_to_cpup(p++); + + rdev = be32_to_cpup(p++); + fattr->rdev = new_decode_dev(rdev); + if (type == (u32)NFCHR && rdev == (u32)NFS2_FIFO_DEV) { + fattr->mode = (fattr->mode & ~S_IFMT) | S_IFIFO; + fattr->rdev = 0; + } + + fattr->du.nfs2.blocks = be32_to_cpup(p++); + fattr->fsid.major = be32_to_cpup(p++); + fattr->fsid.minor = 0; + fattr->fileid = be32_to_cpup(p++); + + p = xdr_decode_time(p, &fattr->atime); + p = xdr_decode_time(p, &fattr->mtime); + xdr_decode_time(p, &fattr->ctime); + fattr->change_attr = nfs_timespec_to_change_attr(&fattr->ctime); + + return 0; +out_uid: + dprintk("NFS: returned invalid uid\n"); + return -EINVAL; +out_gid: + dprintk("NFS: returned invalid gid\n"); + return -EINVAL; +} + +/* + * 2.3.6. sattr + * + * struct sattr { + * unsigned int mode; + * unsigned int uid; + * unsigned int gid; + * unsigned int size; + * timeval atime; + * timeval mtime; + * }; + */ + +#define NFS2_SATTR_NOT_SET (0xffffffff) + +static __be32 *xdr_time_not_set(__be32 *p) +{ + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + return p; +} + +static void encode_sattr(struct xdr_stream *xdr, const struct iattr *attr, + struct user_namespace *userns) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, NFS_sattr_sz << 2); + + if (attr->ia_valid & ATTR_MODE) + *p++ = cpu_to_be32(attr->ia_mode); + else + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + if (attr->ia_valid & ATTR_UID) + *p++ = cpu_to_be32(from_kuid_munged(userns, attr->ia_uid)); + else + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + if (attr->ia_valid & ATTR_GID) + *p++ = cpu_to_be32(from_kgid_munged(userns, attr->ia_gid)); + else + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + if (attr->ia_valid & ATTR_SIZE) + *p++ = cpu_to_be32((u32)attr->ia_size); + else + *p++ = cpu_to_be32(NFS2_SATTR_NOT_SET); + + if (attr->ia_valid & ATTR_ATIME_SET) + p = xdr_encode_time(p, &attr->ia_atime); + else if (attr->ia_valid & ATTR_ATIME) + p = xdr_encode_current_server_time(p, &attr->ia_atime); + else + p = xdr_time_not_set(p); + if (attr->ia_valid & ATTR_MTIME_SET) + xdr_encode_time(p, &attr->ia_mtime); + else if (attr->ia_valid & ATTR_MTIME) + xdr_encode_current_server_time(p, &attr->ia_mtime); + else + xdr_time_not_set(p); +} + +/* + * 2.3.7. filename + * + * typedef string filename; + */ +static void encode_filename(struct xdr_stream *xdr, + const char *name, u32 length) +{ + __be32 *p; + + WARN_ON_ONCE(length > NFS2_MAXNAMLEN); + p = xdr_reserve_space(xdr, 4 + length); + xdr_encode_opaque(p, name, length); +} + +static int decode_filename_inline(struct xdr_stream *xdr, + const char **name, u32 *length) +{ + __be32 *p; + u32 count; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p); + if (count > NFS3_MAXNAMLEN) + goto out_nametoolong; + p = xdr_inline_decode(xdr, count); + if (unlikely(!p)) + return -EIO; + *name = (const char *)p; + *length = count; + return 0; +out_nametoolong: + dprintk("NFS: returned filename too long: %u\n", count); + return -ENAMETOOLONG; +} + +/* + * 2.3.8. path + * + * typedef string path; + */ +static void encode_path(struct xdr_stream *xdr, struct page **pages, u32 length) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 4); + *p = cpu_to_be32(length); + xdr_write_pages(xdr, pages, 0, length); +} + +static int decode_path(struct xdr_stream *xdr) +{ + u32 length, recvd; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + length = be32_to_cpup(p); + if (unlikely(length >= xdr->buf->page_len || length > NFS_MAXPATHLEN)) + goto out_size; + recvd = xdr_read_pages(xdr, length); + if (unlikely(length > recvd)) + goto out_cheating; + xdr_terminate_string(xdr->buf, length); + return 0; +out_size: + dprintk("NFS: returned pathname too long: %u\n", length); + return -ENAMETOOLONG; +out_cheating: + dprintk("NFS: server cheating in pathname result: " + "length %u > received %u\n", length, recvd); + return -EIO; +} + +/* + * 2.3.9. attrstat + * + * union attrstat switch (stat status) { + * case NFS_OK: + * fattr attributes; + * default: + * void; + * }; + */ +static int decode_attrstat(struct xdr_stream *xdr, struct nfs_fattr *result, + __u32 *op_status, + struct user_namespace *userns) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (op_status) + *op_status = status; + if (status != NFS_OK) + goto out_default; + error = decode_fattr(xdr, result, userns); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 2.3.10. diropargs + * + * struct diropargs { + * fhandle dir; + * filename name; + * }; + */ +static void encode_diropargs(struct xdr_stream *xdr, const struct nfs_fh *fh, + const char *name, u32 length) +{ + encode_fhandle(xdr, fh); + encode_filename(xdr, name, length); +} + +/* + * 2.3.11. diropres + * + * union diropres switch (stat status) { + * case NFS_OK: + * struct { + * fhandle file; + * fattr attributes; + * } diropok; + * default: + * void; + * }; + */ +static int decode_diropok(struct xdr_stream *xdr, struct nfs_diropok *result, + struct user_namespace *userns) +{ + int error; + + error = decode_fhandle(xdr, result->fh); + if (unlikely(error)) + goto out; + error = decode_fattr(xdr, result->fattr, userns); +out: + return error; +} + +static int decode_diropres(struct xdr_stream *xdr, struct nfs_diropok *result, + struct user_namespace *userns) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS_OK) + goto out_default; + error = decode_diropok(xdr, result, userns); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + + +/* + * NFSv2 XDR encode functions + * + * NFSv2 argument types are defined in section 2.2 of RFC 1094: + * "NFS: Network File System Protocol Specification". + */ + +static void nfs2_xdr_enc_fhandle(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_fh *fh = data; + + encode_fhandle(xdr, fh); +} + +/* + * 2.2.3. sattrargs + * + * struct sattrargs { + * fhandle file; + * sattr attributes; + * }; + */ +static void nfs2_xdr_enc_sattrargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_sattrargs *args = data; + + encode_fhandle(xdr, args->fh); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); +} + +static void nfs2_xdr_enc_diropargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_diropargs *args = data; + + encode_diropargs(xdr, args->fh, args->name, args->len); +} + +static void nfs2_xdr_enc_readlinkargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_readlinkargs *args = data; + + encode_fhandle(xdr, args->fh); + rpc_prepare_reply_pages(req, args->pages, args->pgbase, args->pglen, + NFS_readlinkres_sz - NFS_pagepad_sz); +} + +/* + * 2.2.7. readargs + * + * struct readargs { + * fhandle file; + * unsigned offset; + * unsigned count; + * unsigned totalcount; + * }; + */ +static void encode_readargs(struct xdr_stream *xdr, + const struct nfs_pgio_args *args) +{ + u32 offset = args->offset; + u32 count = args->count; + __be32 *p; + + encode_fhandle(xdr, args->fh); + + p = xdr_reserve_space(xdr, 4 + 4 + 4); + *p++ = cpu_to_be32(offset); + *p++ = cpu_to_be32(count); + *p = cpu_to_be32(count); +} + +static void nfs2_xdr_enc_readargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + + encode_readargs(xdr, args); + rpc_prepare_reply_pages(req, args->pages, args->pgbase, args->count, + NFS_readres_sz - NFS_pagepad_sz); + req->rq_rcv_buf.flags |= XDRBUF_READ; +} + +/* + * 2.2.9. writeargs + * + * struct writeargs { + * fhandle file; + * unsigned beginoffset; + * unsigned offset; + * unsigned totalcount; + * nfsdata data; + * }; + */ +static void encode_writeargs(struct xdr_stream *xdr, + const struct nfs_pgio_args *args) +{ + u32 offset = args->offset; + u32 count = args->count; + __be32 *p; + + encode_fhandle(xdr, args->fh); + + p = xdr_reserve_space(xdr, 4 + 4 + 4 + 4); + *p++ = cpu_to_be32(offset); + *p++ = cpu_to_be32(offset); + *p++ = cpu_to_be32(count); + + /* nfsdata */ + *p = cpu_to_be32(count); + xdr_write_pages(xdr, args->pages, args->pgbase, count); +} + +static void nfs2_xdr_enc_writeargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + + encode_writeargs(xdr, args); + xdr->buf->flags |= XDRBUF_WRITE; +} + +/* + * 2.2.10. createargs + * + * struct createargs { + * diropargs where; + * sattr attributes; + * }; + */ +static void nfs2_xdr_enc_createargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_createargs *args = data; + + encode_diropargs(xdr, args->fh, args->name, args->len); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); +} + +static void nfs2_xdr_enc_removeargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_removeargs *args = data; + + encode_diropargs(xdr, args->fh, args->name.name, args->name.len); +} + +/* + * 2.2.12. renameargs + * + * struct renameargs { + * diropargs from; + * diropargs to; + * }; + */ +static void nfs2_xdr_enc_renameargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_renameargs *args = data; + const struct qstr *old = args->old_name; + const struct qstr *new = args->new_name; + + encode_diropargs(xdr, args->old_dir, old->name, old->len); + encode_diropargs(xdr, args->new_dir, new->name, new->len); +} + +/* + * 2.2.13. linkargs + * + * struct linkargs { + * fhandle from; + * diropargs to; + * }; + */ +static void nfs2_xdr_enc_linkargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_linkargs *args = data; + + encode_fhandle(xdr, args->fromfh); + encode_diropargs(xdr, args->tofh, args->toname, args->tolen); +} + +/* + * 2.2.14. symlinkargs + * + * struct symlinkargs { + * diropargs from; + * path to; + * sattr attributes; + * }; + */ +static void nfs2_xdr_enc_symlinkargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_symlinkargs *args = data; + + encode_diropargs(xdr, args->fromfh, args->fromname, args->fromlen); + encode_path(xdr, args->pages, args->pathlen); + encode_sattr(xdr, args->sattr, rpc_rqst_userns(req)); +} + +/* + * 2.2.17. readdirargs + * + * struct readdirargs { + * fhandle dir; + * nfscookie cookie; + * unsigned count; + * }; + */ +static void encode_readdirargs(struct xdr_stream *xdr, + const struct nfs_readdirargs *args) +{ + __be32 *p; + + encode_fhandle(xdr, args->fh); + + p = xdr_reserve_space(xdr, 4 + 4); + *p++ = cpu_to_be32(args->cookie); + *p = cpu_to_be32(args->count); +} + +static void nfs2_xdr_enc_readdirargs(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_readdirargs *args = data; + + encode_readdirargs(xdr, args); + rpc_prepare_reply_pages(req, args->pages, 0, args->count, + NFS_readdirres_sz - NFS_pagepad_sz); +} + +/* + * NFSv2 XDR decode functions + * + * NFSv2 result types are defined in section 2.2 of RFC 1094: + * "NFS: Network File System Protocol Specification". + */ + +static int nfs2_xdr_dec_stat(struct rpc_rqst *req, struct xdr_stream *xdr, + void *__unused) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS_OK) + goto out_default; +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +static int nfs2_xdr_dec_attrstat(struct rpc_rqst *req, struct xdr_stream *xdr, + void *result) +{ + return decode_attrstat(xdr, result, NULL, rpc_rqst_userns(req)); +} + +static int nfs2_xdr_dec_diropres(struct rpc_rqst *req, struct xdr_stream *xdr, + void *result) +{ + return decode_diropres(xdr, result, rpc_rqst_userns(req)); +} + +/* + * 2.2.6. readlinkres + * + * union readlinkres switch (stat status) { + * case NFS_OK: + * path data; + * default: + * void; + * }; + */ +static int nfs2_xdr_dec_readlinkres(struct rpc_rqst *req, + struct xdr_stream *xdr, void *__unused) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS_OK) + goto out_default; + error = decode_path(xdr); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 2.2.7. readres + * + * union readres switch (stat status) { + * case NFS_OK: + * fattr attributes; + * nfsdata data; + * default: + * void; + * }; + */ +static int nfs2_xdr_dec_readres(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *result = data; + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + result->op_status = status; + if (status != NFS_OK) + goto out_default; + error = decode_fattr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + error = decode_nfsdata(xdr, result); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +static int nfs2_xdr_dec_writeres(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *result = data; + + /* All NFSv2 writes are "file sync" writes */ + result->verf->committed = NFS_FILE_SYNC; + return decode_attrstat(xdr, result->fattr, &result->op_status, + rpc_rqst_userns(req)); +} + +/** + * nfs2_decode_dirent - Decode a single NFSv2 directory entry stored in + * the local page cache. + * @xdr: XDR stream where entry resides + * @entry: buffer to fill in with entry data + * @plus: boolean indicating whether this should be a readdirplus entry + * + * Returns zero if successful, otherwise a negative errno value is + * returned. + * + * This function is not invoked during READDIR reply decoding, but + * rather whenever an application invokes the getdents(2) system call + * on a directory already in our cache. + * + * 2.2.17. entry + * + * struct entry { + * unsigned fileid; + * filename name; + * nfscookie cookie; + * entry *nextentry; + * }; + */ +int nfs2_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, + bool plus) +{ + __be32 *p; + int error; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p++ == xdr_zero) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p++ == xdr_zero) + return -EAGAIN; + entry->eof = 1; + return -EBADCOOKIE; + } + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + entry->ino = be32_to_cpup(p); + + error = decode_filename_inline(xdr, &entry->name, &entry->len); + if (unlikely(error)) + return error == -ENAMETOOLONG ? -ENAMETOOLONG : -EAGAIN; + + /* + * The type (size and byte order) of nfscookie isn't defined in + * RFC 1094. This implementation assumes that it's an XDR uint32. + */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + entry->cookie = be32_to_cpup(p); + + entry->d_type = DT_UNKNOWN; + + return 0; +} + +/* + * 2.2.17. readdirres + * + * union readdirres switch (stat status) { + * case NFS_OK: + * struct { + * entry *entries; + * bool eof; + * } readdirok; + * default: + * void; + * }; + * + * Read the directory contents into the page cache, but don't + * touch them. The actual decoding is done by nfs2_decode_dirent() + * during subsequent nfs_readdir() calls. + */ +static int decode_readdirok(struct xdr_stream *xdr) +{ + return xdr_read_pages(xdr, xdr->buf->page_len); +} + +static int nfs2_xdr_dec_readdirres(struct rpc_rqst *req, + struct xdr_stream *xdr, void *__unused) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS_OK) + goto out_default; + error = decode_readdirok(xdr); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 2.2.18. statfsres + * + * union statfsres (stat status) { + * case NFS_OK: + * struct { + * unsigned tsize; + * unsigned bsize; + * unsigned blocks; + * unsigned bfree; + * unsigned bavail; + * } info; + * default: + * void; + * }; + */ +static int decode_info(struct xdr_stream *xdr, struct nfs2_fsstat *result) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS_info_sz << 2); + if (unlikely(!p)) + return -EIO; + result->tsize = be32_to_cpup(p++); + result->bsize = be32_to_cpup(p++); + result->blocks = be32_to_cpup(p++); + result->bfree = be32_to_cpup(p++); + result->bavail = be32_to_cpup(p); + return 0; +} + +static int nfs2_xdr_dec_statfsres(struct rpc_rqst *req, struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_stat(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS_OK) + goto out_default; + error = decode_info(xdr, result); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +#define PROC(proc, argtype, restype, timer) \ +[NFSPROC_##proc] = { \ + .p_proc = NFSPROC_##proc, \ + .p_encode = nfs2_xdr_enc_##argtype, \ + .p_decode = nfs2_xdr_dec_##restype, \ + .p_arglen = NFS_##argtype##_sz, \ + .p_replen = NFS_##restype##_sz, \ + .p_timer = timer, \ + .p_statidx = NFSPROC_##proc, \ + .p_name = #proc, \ + } +const struct rpc_procinfo nfs_procedures[] = { + PROC(GETATTR, fhandle, attrstat, 1), + PROC(SETATTR, sattrargs, attrstat, 0), + PROC(LOOKUP, diropargs, diropres, 2), + PROC(READLINK, readlinkargs, readlinkres, 3), + PROC(READ, readargs, readres, 3), + PROC(WRITE, writeargs, writeres, 4), + PROC(CREATE, createargs, diropres, 0), + PROC(REMOVE, removeargs, stat, 0), + PROC(RENAME, renameargs, stat, 0), + PROC(LINK, linkargs, stat, 0), + PROC(SYMLINK, symlinkargs, stat, 0), + PROC(MKDIR, createargs, diropres, 0), + PROC(RMDIR, diropargs, stat, 0), + PROC(READDIR, readdirargs, readdirres, 3), + PROC(STATFS, fhandle, statfsres, 0), +}; + +static unsigned int nfs_version2_counts[ARRAY_SIZE(nfs_procedures)]; +const struct rpc_version nfs_version2 = { + .number = 2, + .nrprocs = ARRAY_SIZE(nfs_procedures), + .procs = nfs_procedures, + .counts = nfs_version2_counts, +}; diff --git a/fs/nfs/nfs3_fs.h b/fs/nfs/nfs3_fs.h new file mode 100644 index 000000000..b333ea119 --- /dev/null +++ b/fs/nfs/nfs3_fs.h @@ -0,0 +1,37 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2014 Anna Schumaker. + * + * NFSv3-specific filesystem definitions and declarations + */ +#ifndef __LINUX_FS_NFS_NFS3_FS_H +#define __LINUX_FS_NFS_NFS3_FS_H + +/* + * nfs3acl.c + */ +#ifdef CONFIG_NFS_V3_ACL +extern struct posix_acl *nfs3_get_acl(struct inode *inode, int type, bool rcu); +extern int nfs3_set_acl(struct mnt_idmap *idmap, struct dentry *dentry, + struct posix_acl *acl, int type); +extern int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl); +extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t); +#else +static inline int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + return 0; +} +#define nfs3_listxattr NULL +#endif /* CONFIG_NFS_V3_ACL */ + +/* nfs3client.c */ +struct nfs_server *nfs3_create_server(struct fs_context *); +struct nfs_server *nfs3_clone_server(struct nfs_server *, struct nfs_fh *, + struct nfs_fattr *, rpc_authflavor_t); + +/* nfs3super.c */ +extern struct nfs_subversion nfs_v3; + +#endif /* __LINUX_FS_NFS_NFS3_FS_H */ diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c new file mode 100644 index 000000000..a126eb31f --- /dev/null +++ b/fs/nfs/nfs3acl.c @@ -0,0 +1,343 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "nfs3_fs.h" + +#define NFSDBG_FACILITY NFSDBG_PROC + +/* + * nfs3_prepare_get_acl, nfs3_complete_get_acl, nfs3_abort_get_acl: Helpers for + * caching get_acl results in a race-free way. See fs/posix_acl.c:get_acl() + * for explanations. + */ +static void nfs3_prepare_get_acl(struct posix_acl **p) +{ + struct posix_acl *sentinel = uncached_acl_sentinel(current); + + /* If the ACL isn't being read yet, set our sentinel. */ + cmpxchg(p, ACL_NOT_CACHED, sentinel); +} + +static void nfs3_complete_get_acl(struct posix_acl **p, struct posix_acl *acl) +{ + struct posix_acl *sentinel = uncached_acl_sentinel(current); + + /* Only cache the ACL if our sentinel is still in place. */ + posix_acl_dup(acl); + if (cmpxchg(p, sentinel, acl) != sentinel) + posix_acl_release(acl); +} + +static void nfs3_abort_get_acl(struct posix_acl **p) +{ + struct posix_acl *sentinel = uncached_acl_sentinel(current); + + /* Remove our sentinel upon failure. */ + cmpxchg(p, sentinel, ACL_NOT_CACHED); +} + +struct posix_acl *nfs3_get_acl(struct inode *inode, int type, bool rcu) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct page *pages[NFSACL_MAXPAGES] = { }; + struct nfs3_getaclargs args = { + .fh = NFS_FH(inode), + /* The xdr layer may allocate pages here. */ + .pages = pages, + }; + struct nfs3_getaclres res = { + NULL, + }; + struct rpc_message msg = { + .rpc_argp = &args, + .rpc_resp = &res, + }; + int status, count; + + if (rcu) + return ERR_PTR(-ECHILD); + + if (!nfs_server_capable(inode, NFS_CAP_ACLS)) + return ERR_PTR(-EOPNOTSUPP); + + status = nfs_revalidate_inode(inode, NFS_INO_INVALID_CHANGE); + if (status < 0) + return ERR_PTR(status); + + /* + * Only get the access acl when explicitly requested: We don't + * need it for access decisions, and only some applications use + * it. Applications which request the access acl first are not + * penalized from this optimization. + */ + if (type == ACL_TYPE_ACCESS) + args.mask |= NFS_ACLCNT|NFS_ACL; + if (S_ISDIR(inode->i_mode)) + args.mask |= NFS_DFACLCNT|NFS_DFACL; + if (args.mask == 0) + return NULL; + + dprintk("NFS call getacl\n"); + msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_GETACL]; + res.fattr = nfs_alloc_fattr(); + if (res.fattr == NULL) + return ERR_PTR(-ENOMEM); + + if (args.mask & NFS_ACL) + nfs3_prepare_get_acl(&inode->i_acl); + if (args.mask & NFS_DFACL) + nfs3_prepare_get_acl(&inode->i_default_acl); + + status = rpc_call_sync(server->client_acl, &msg, 0); + dprintk("NFS reply getacl: %d\n", status); + + /* pages may have been allocated at the xdr layer. */ + for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++) + __free_page(args.pages[count]); + + switch (status) { + case 0: + nfs_refresh_inode(inode, res.fattr); + break; + case -EPFNOSUPPORT: + case -EPROTONOSUPPORT: + dprintk("NFS_V3_ACL extension not supported; disabling\n"); + server->caps &= ~NFS_CAP_ACLS; + fallthrough; + case -ENOTSUPP: + status = -EOPNOTSUPP; + goto getout; + default: + goto getout; + } + if ((args.mask & res.mask) != args.mask) { + status = -EIO; + goto getout; + } + + if (res.acl_access != NULL) { + if ((posix_acl_equiv_mode(res.acl_access, NULL) == 0) || + res.acl_access->a_count == 0) { + posix_acl_release(res.acl_access); + res.acl_access = NULL; + } + } + + if (res.mask & NFS_ACL) + nfs3_complete_get_acl(&inode->i_acl, res.acl_access); + else + forget_cached_acl(inode, ACL_TYPE_ACCESS); + + if (res.mask & NFS_DFACL) + nfs3_complete_get_acl(&inode->i_default_acl, res.acl_default); + else + forget_cached_acl(inode, ACL_TYPE_DEFAULT); + + nfs_free_fattr(res.fattr); + if (type == ACL_TYPE_ACCESS) { + posix_acl_release(res.acl_default); + return res.acl_access; + } else { + posix_acl_release(res.acl_access); + return res.acl_default; + } + +getout: + nfs3_abort_get_acl(&inode->i_acl); + nfs3_abort_get_acl(&inode->i_default_acl); + posix_acl_release(res.acl_access); + posix_acl_release(res.acl_default); + nfs_free_fattr(res.fattr); + return ERR_PTR(status); +} + +static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_fattr *fattr; + struct page *pages[NFSACL_MAXPAGES]; + struct nfs3_setaclargs args = { + .inode = inode, + .mask = NFS_ACL, + .acl_access = acl, + .pages = pages, + }; + struct rpc_message msg = { + .rpc_argp = &args, + .rpc_resp = &fattr, + }; + int status = 0; + + if (acl == NULL && (!S_ISDIR(inode->i_mode) || dfacl == NULL)) + goto out; + + status = -EOPNOTSUPP; + if (!nfs_server_capable(inode, NFS_CAP_ACLS)) + goto out; + + /* We are doing this here because XDR marshalling does not + * return any results, it BUGs. */ + status = -ENOSPC; + if (acl != NULL && acl->a_count > NFS_ACL_MAX_ENTRIES) + goto out; + if (dfacl != NULL && dfacl->a_count > NFS_ACL_MAX_ENTRIES) + goto out; + if (S_ISDIR(inode->i_mode)) { + args.mask |= NFS_DFACL; + args.acl_default = dfacl; + args.len = nfsacl_size(acl, dfacl); + } else + args.len = nfsacl_size(acl, NULL); + + if (args.len > NFS_ACL_INLINE_BUFSIZE) { + unsigned int npages = 1 + ((args.len - 1) >> PAGE_SHIFT); + + status = -ENOMEM; + do { + args.pages[args.npages] = alloc_page(GFP_KERNEL); + if (args.pages[args.npages] == NULL) + goto out_freepages; + args.npages++; + } while (args.npages < npages); + } + + dprintk("NFS call setacl\n"); + status = -ENOMEM; + fattr = nfs_alloc_fattr(); + if (fattr == NULL) + goto out_freepages; + + msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_SETACL]; + msg.rpc_resp = fattr; + status = rpc_call_sync(server->client_acl, &msg, 0); + nfs_access_zap_cache(inode); + nfs_zap_acl_cache(inode); + dprintk("NFS reply setacl: %d\n", status); + + switch (status) { + case 0: + status = nfs_refresh_inode(inode, fattr); + break; + case -EPFNOSUPPORT: + case -EPROTONOSUPPORT: + dprintk("NFS_V3_ACL SETACL RPC not supported" + "(will not retry)\n"); + server->caps &= ~NFS_CAP_ACLS; + fallthrough; + case -ENOTSUPP: + status = -EOPNOTSUPP; + } + nfs_free_fattr(fattr); +out_freepages: + while (args.npages != 0) { + args.npages--; + __free_page(args.pages[args.npages]); + } +out: + return status; +} + +int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + int ret; + ret = __nfs3_proc_setacls(inode, acl, dfacl); + return (ret == -EOPNOTSUPP) ? 0 : ret; + +} + +int nfs3_set_acl(struct mnt_idmap *idmap, struct dentry *dentry, + struct posix_acl *acl, int type) +{ + struct posix_acl *orig = acl, *dfacl = NULL, *alloc; + struct inode *inode = d_inode(dentry); + int status; + + if (S_ISDIR(inode->i_mode)) { + switch(type) { + case ACL_TYPE_ACCESS: + alloc = get_inode_acl(inode, ACL_TYPE_DEFAULT); + if (IS_ERR(alloc)) + goto fail; + dfacl = alloc; + break; + + case ACL_TYPE_DEFAULT: + alloc = get_inode_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(alloc)) + goto fail; + dfacl = acl; + acl = alloc; + break; + } + } + + if (acl == NULL) { + alloc = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); + if (IS_ERR(alloc)) + goto fail; + acl = alloc; + } + status = __nfs3_proc_setacls(inode, acl, dfacl); +out: + if (acl != orig) + posix_acl_release(acl); + if (dfacl != orig) + posix_acl_release(dfacl); + return status; + +fail: + status = PTR_ERR(alloc); + goto out; +} + +static int +nfs3_list_one_acl(struct inode *inode, int type, const char *name, void *data, + size_t size, ssize_t *result) +{ + struct posix_acl *acl; + char *p = data + *result; + + acl = get_inode_acl(inode, type); + if (IS_ERR_OR_NULL(acl)) + return 0; + + posix_acl_release(acl); + + *result += strlen(name); + *result += 1; + if (!size) + return 0; + if (*result > size) + return -ERANGE; + + strcpy(p, name); + return 0; +} + +ssize_t +nfs3_listxattr(struct dentry *dentry, char *data, size_t size) +{ + struct inode *inode = d_inode(dentry); + ssize_t result = 0; + int error; + + error = nfs3_list_one_acl(inode, ACL_TYPE_ACCESS, + XATTR_NAME_POSIX_ACL_ACCESS, data, size, &result); + if (error) + return error; + + error = nfs3_list_one_acl(inode, ACL_TYPE_DEFAULT, + XATTR_NAME_POSIX_ACL_DEFAULT, data, size, &result); + if (error) + return error; + return result; +} diff --git a/fs/nfs/nfs3client.c b/fs/nfs/nfs3client.c new file mode 100644 index 000000000..5d97c1d38 --- /dev/null +++ b/fs/nfs/nfs3client.c @@ -0,0 +1,144 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include +#include +#include +#include "internal.h" +#include "nfs3_fs.h" +#include "netns.h" +#include "sysfs.h" + +#ifdef CONFIG_NFS_V3_ACL +static struct rpc_stat nfsacl_rpcstat = { &nfsacl_program }; +static const struct rpc_version *nfsacl_version[] = { + [3] = &nfsacl_version3, +}; + +const struct rpc_program nfsacl_program = { + .name = "nfsacl", + .number = NFS_ACL_PROGRAM, + .nrvers = ARRAY_SIZE(nfsacl_version), + .version = nfsacl_version, + .stats = &nfsacl_rpcstat, +}; + +/* + * Initialise an NFSv3 ACL client connection + */ +static void nfs_init_server_aclclient(struct nfs_server *server) +{ + if (server->flags & NFS_MOUNT_NOACL) + goto out_noacl; + + server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3); + if (IS_ERR(server->client_acl)) + goto out_noacl; + + nfs_sysfs_link_rpc_client(server, server->client_acl, NULL); + + /* No errors! Assume that Sun nfsacls are supported */ + server->caps |= NFS_CAP_ACLS; + return; + +out_noacl: + server->caps &= ~NFS_CAP_ACLS; +} +#else +static inline void nfs_init_server_aclclient(struct nfs_server *server) +{ + server->flags &= ~NFS_MOUNT_NOACL; + server->caps &= ~NFS_CAP_ACLS; +} +#endif + +struct nfs_server *nfs3_create_server(struct fs_context *fc) +{ + struct nfs_server *server = nfs_create_server(fc); + + /* Create a client RPC handle for the NFS v3 ACL management interface */ + if (!IS_ERR(server)) + nfs_init_server_aclclient(server); + return server; +} + +struct nfs_server *nfs3_clone_server(struct nfs_server *source, + struct nfs_fh *fh, + struct nfs_fattr *fattr, + rpc_authflavor_t flavor) +{ + struct nfs_server *server = nfs_clone_server(source, fh, fattr, flavor); + if (!IS_ERR(server) && !IS_ERR(source->client_acl)) + nfs_init_server_aclclient(server); + return server; +} + +/* + * Set up a pNFS Data Server client over NFSv3. + * + * Return any existing nfs_client that matches server address,port,version + * and minorversion. + * + * For a new nfs_client, use a soft mount (default), a low retrans and a + * low timeout interval so that if a connection is lost, we retry through + * the MDS. + */ +struct nfs_client *nfs3_set_ds_client(struct nfs_server *mds_srv, + const struct sockaddr_storage *ds_addr, int ds_addrlen, + int ds_proto, unsigned int ds_timeo, unsigned int ds_retrans) +{ + struct rpc_timeout ds_timeout; + unsigned long connect_timeout = ds_timeo * (ds_retrans + 1) * HZ / 10; + struct nfs_client *mds_clp = mds_srv->nfs_client; + struct nfs_client_initdata cl_init = { + .addr = ds_addr, + .addrlen = ds_addrlen, + .nodename = mds_clp->cl_rpcclient->cl_nodename, + .ip_addr = mds_clp->cl_ipaddr, + .nfs_mod = &nfs_v3, + .proto = ds_proto, + .net = mds_clp->cl_net, + .timeparms = &ds_timeout, + .cred = mds_srv->cred, + .xprtsec = { + .policy = RPC_XPRTSEC_NONE, + .cert_serial = TLS_NO_CERT, + .privkey_serial = TLS_NO_PRIVKEY, + }, + .connect_timeout = connect_timeout, + .reconnect_timeout = connect_timeout, + }; + struct nfs_client *clp; + char buf[INET6_ADDRSTRLEN + 1]; + + /* fake a hostname because lockd wants it */ + if (rpc_ntop((struct sockaddr *)ds_addr, buf, sizeof(buf)) <= 0) + return ERR_PTR(-EINVAL); + cl_init.hostname = buf; + + switch (ds_proto) { + case XPRT_TRANSPORT_TCP_TLS: + if (mds_clp->cl_xprtsec.policy != RPC_XPRTSEC_NONE) + cl_init.xprtsec = mds_clp->cl_xprtsec; + else + ds_proto = XPRT_TRANSPORT_TCP; + fallthrough; + case XPRT_TRANSPORT_RDMA: + case XPRT_TRANSPORT_TCP: + if (mds_clp->cl_nconnect > 1) + cl_init.nconnect = mds_clp->cl_nconnect; + } + + if (mds_srv->flags & NFS_MOUNT_NORESVPORT) + __set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); + if (test_bit(NFS_CS_NETUNREACH_FATAL, &mds_clp->cl_flags)) + __set_bit(NFS_CS_NETUNREACH_FATAL, &cl_init.init_flags); + + __set_bit(NFS_CS_DS, &cl_init.init_flags); + + /* Use the MDS nfs_client cl_ipaddr. */ + nfs_init_timeout_values(&ds_timeout, ds_proto, ds_timeo, ds_retrans); + clp = nfs_get_client(&cl_init); + + return clp; +} +EXPORT_SYMBOL_GPL(nfs3_set_ds_client); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c new file mode 100644 index 000000000..b80d0c5ef --- /dev/null +++ b/fs/nfs/nfs3proc.c @@ -0,0 +1,1128 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/nfs3proc.c + * + * Client-side NFSv3 procedures stubs. + * + * Copyright (C) 1997, Olaf Kirch + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "iostat.h" +#include "internal.h" +#include "nfs3_fs.h" + +#define NFSDBG_FACILITY NFSDBG_PROC + +/* A wrapper to handle the EJUKEBOX error messages */ +static int +nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) +{ + int res; + do { + res = rpc_call_sync(clnt, msg, flags); + if (res != -EJUKEBOX) + break; + __set_current_state(TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(NFS_JUKEBOX_RETRY_TIME); + res = -ERESTARTSYS; + } while (!fatal_signal_pending(current) && !nfs_current_task_exiting()); + return res; +} + +#define rpc_call_sync(clnt, msg, flags) nfs3_rpc_wrapper(clnt, msg, flags) + +static int +nfs3_async_handle_jukebox(struct rpc_task *task, struct inode *inode) +{ + if (task->tk_status != -EJUKEBOX) + return 0; + nfs_inc_stats(inode, NFSIOS_DELAY); + task->tk_status = 0; + rpc_restart_call(task); + rpc_delay(task, NFS_JUKEBOX_RETRY_TIME); + return 1; +} + +static int +do_proc_get_root(struct rpc_clnt *client, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_FSINFO], + .rpc_argp = fhandle, + .rpc_resp = info, + }; + int status; + + dprintk("%s: call fsinfo\n", __func__); + nfs_fattr_init(info->fattr); + status = rpc_call_sync(client, &msg, 0); + dprintk("%s: reply fsinfo: %d\n", __func__, status); + if (status == 0 && !(info->fattr->valid & NFS_ATTR_FATTR)) { + msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR]; + msg.rpc_resp = info->fattr; + status = rpc_call_sync(client, &msg, 0); + dprintk("%s: reply getattr: %d\n", __func__, status); + } + return status; +} + +/* + * Bare-bones access to getattr: this is for nfs_get_root/nfs_get_sb + */ +static int +nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + int status; + + status = do_proc_get_root(server->client, fhandle, info); + if (status && server->nfs_client->cl_rpcclient != server->client) + status = do_proc_get_root(server->nfs_client->cl_rpcclient, fhandle, info); + return status; +} + +/* + * One function for each procedure in the NFS protocol. + */ +static int +nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode) +{ + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR], + .rpc_argp = fhandle, + .rpc_resp = fattr, + }; + int status; + unsigned short task_flags = 0; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (inode && (server->flags & NFS_MOUNT_SOFTREVAL)) + task_flags |= RPC_TASK_TIMEOUT; + + dprintk("NFS call getattr\n"); + nfs_fattr_init(fattr); + status = rpc_call_sync(server->client, &msg, task_flags); + dprintk("NFS reply getattr: %d\n", status); + return status; +} + +static int +nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, + struct iattr *sattr) +{ + struct inode *inode = d_inode(dentry); + struct nfs3_sattrargs arg = { + .fh = NFS_FH(inode), + .sattr = sattr, + }; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_SETATTR], + .rpc_argp = &arg, + .rpc_resp = fattr, + }; + int status; + + dprintk("NFS call setattr\n"); + if (sattr->ia_valid & ATTR_FILE) + msg.rpc_cred = nfs_file_cred(sattr->ia_file); + nfs_fattr_init(fattr); + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + if (status == 0) { + nfs_setattr_update_inode(inode, sattr, fattr); + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); + } + dprintk("NFS reply setattr: %d\n", status); + return status; +} + +static int +__nfs3_proc_lookup(struct inode *dir, const char *name, size_t len, + struct nfs_fh *fhandle, struct nfs_fattr *fattr, + unsigned short task_flags) +{ + struct nfs3_diropargs arg = { + .fh = NFS_FH(dir), + .name = name, + .len = len + }; + struct nfs3_diropres res = { + .fh = fhandle, + .fattr = fattr + }; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_LOOKUP], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status; + + res.dir_attr = nfs_alloc_fattr(); + if (res.dir_attr == NULL) + return -ENOMEM; + + nfs_fattr_init(fattr); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags); + nfs_refresh_inode(dir, res.dir_attr); + if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) { + msg.rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR]; + msg.rpc_argp = fhandle; + msg.rpc_resp = fattr; + status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags); + } + nfs_free_fattr(res.dir_attr); + dprintk("NFS reply lookup: %d\n", status); + return status; +} + +static int +nfs3_proc_lookup(struct inode *dir, struct dentry *dentry, const struct qstr *name, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + unsigned short task_flags = 0; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (nfs_lookup_is_soft_revalidate(dentry)) + task_flags |= RPC_TASK_TIMEOUT; + + dprintk("NFS call lookup %pd2\n", dentry); + return __nfs3_proc_lookup(dir, name->name, name->len, fhandle, fattr, + task_flags); +} + +static int nfs3_proc_lookupp(struct inode *inode, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + const char dotdot[] = ".."; + const size_t len = strlen(dotdot); + unsigned short task_flags = 0; + + if (NFS_SERVER(inode)->flags & NFS_MOUNT_SOFTREVAL) + task_flags |= RPC_TASK_TIMEOUT; + + return __nfs3_proc_lookup(inode, dotdot, len, fhandle, fattr, + task_flags); +} + +static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry, + const struct cred *cred) +{ + struct nfs3_accessargs arg = { + .fh = NFS_FH(inode), + .access = entry->mask, + }; + struct nfs3_accessres res; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_ACCESS], + .rpc_argp = &arg, + .rpc_resp = &res, + .rpc_cred = cred, + }; + int status = -ENOMEM; + + dprintk("NFS call access\n"); + res.fattr = nfs_alloc_fattr(); + if (res.fattr == NULL) + goto out; + + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + nfs_refresh_inode(inode, res.fattr); + if (status == 0) + nfs_access_set_mask(entry, res.access); + nfs_free_fattr(res.fattr); +out: + dprintk("NFS reply access: %d\n", status); + return status; +} + +static int nfs3_proc_readlink(struct inode *inode, struct page *page, + unsigned int pgbase, unsigned int pglen) +{ + struct nfs_fattr *fattr; + struct nfs3_readlinkargs args = { + .fh = NFS_FH(inode), + .pgbase = pgbase, + .pglen = pglen, + .pages = &page + }; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_READLINK], + .rpc_argp = &args, + }; + int status = -ENOMEM; + + dprintk("NFS call readlink\n"); + fattr = nfs_alloc_fattr(); + if (fattr == NULL) + goto out; + msg.rpc_resp = fattr; + + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + nfs_refresh_inode(inode, fattr); + nfs_free_fattr(fattr); +out: + dprintk("NFS reply readlink: %d\n", status); + return status; +} + +struct nfs3_createdata { + struct rpc_message msg; + union { + struct nfs3_createargs create; + struct nfs3_mkdirargs mkdir; + struct nfs3_symlinkargs symlink; + struct nfs3_mknodargs mknod; + } arg; + struct nfs3_diropres res; + struct nfs_fh fh; + struct nfs_fattr fattr; + struct nfs_fattr dir_attr; +}; + +static struct nfs3_createdata *nfs3_alloc_createdata(void) +{ + struct nfs3_createdata *data; + + data = kzalloc_obj(*data); + if (data != NULL) { + data->msg.rpc_argp = &data->arg; + data->msg.rpc_resp = &data->res; + data->res.fh = &data->fh; + data->res.fattr = &data->fattr; + data->res.dir_attr = &data->dir_attr; + nfs_fattr_init(data->res.fattr); + nfs_fattr_init(data->res.dir_attr); + } + return data; +} + +static struct dentry * +nfs3_do_create(struct inode *dir, struct dentry *dentry, struct nfs3_createdata *data) +{ + int status; + + status = rpc_call_sync(NFS_CLIENT(dir), &data->msg, 0); + nfs_post_op_update_inode(dir, data->res.dir_attr); + if (status != 0) + return ERR_PTR(status); + + return nfs_add_or_obtain(dentry, data->res.fh, data->res.fattr); +} + +static void nfs3_free_createdata(struct nfs3_createdata *data) +{ + kfree(data); +} + +/* + * Create a regular file. + */ +static int +nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, + int flags) +{ + struct posix_acl *default_acl, *acl; + struct nfs3_createdata *data; + struct dentry *d_alias; + int status = -ENOMEM; + + dprintk("NFS call create %pd\n", dentry); + + data = nfs3_alloc_createdata(); + if (data == NULL) + goto out; + + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_CREATE]; + data->arg.create.fh = NFS_FH(dir); + data->arg.create.name = dentry->d_name.name; + data->arg.create.len = dentry->d_name.len; + data->arg.create.sattr = sattr; + + data->arg.create.createmode = NFS3_CREATE_UNCHECKED; + if (flags & O_EXCL) { + data->arg.create.createmode = NFS3_CREATE_EXCLUSIVE; + data->arg.create.verifier[0] = cpu_to_be32(jiffies); + data->arg.create.verifier[1] = cpu_to_be32(current->pid); + } + + status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl); + if (status) + goto out; + + for (;;) { + d_alias = nfs3_do_create(dir, dentry, data); + status = PTR_ERR_OR_ZERO(d_alias); + + if (status != -ENOTSUPP) + break; + /* If the server doesn't support the exclusive creation + * semantics, try again with simple 'guarded' mode. */ + switch (data->arg.create.createmode) { + case NFS3_CREATE_EXCLUSIVE: + data->arg.create.createmode = NFS3_CREATE_GUARDED; + break; + + case NFS3_CREATE_GUARDED: + data->arg.create.createmode = NFS3_CREATE_UNCHECKED; + break; + + case NFS3_CREATE_UNCHECKED: + goto out_release_acls; + } + nfs_fattr_init(data->res.dir_attr); + nfs_fattr_init(data->res.fattr); + } + + if (status != 0) + goto out_release_acls; + + if (d_alias) { + if (d_is_dir(d_alias)) { + status = -EISDIR; + goto out_dput; + } + dentry = d_alias; + } + + /* When we created the file with exclusive semantics, make + * sure we set the attributes afterwards. */ + if (data->arg.create.createmode == NFS3_CREATE_EXCLUSIVE) { + dprintk("NFS call setattr (post-create)\n"); + + if (!(sattr->ia_valid & ATTR_ATIME_SET)) + sattr->ia_valid |= ATTR_ATIME; + if (!(sattr->ia_valid & ATTR_MTIME_SET)) + sattr->ia_valid |= ATTR_MTIME; + + /* Note: we could use a guarded setattr here, but I'm + * not sure this buys us anything (and I'd have + * to revamp the NFSv3 XDR code) */ + status = nfs3_proc_setattr(dentry, data->res.fattr, sattr); + nfs_post_op_update_inode(d_inode(dentry), data->res.fattr); + dprintk("NFS reply setattr (post-create): %d\n", status); + if (status != 0) + goto out_dput; + } + + status = nfs3_proc_setacls(d_inode(dentry), acl, default_acl); + +out_dput: + dput(d_alias); +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); +out: + nfs3_free_createdata(data); + dprintk("NFS reply create: %d\n", status); + return status; +} + +static int +nfs3_proc_remove(struct inode *dir, struct dentry *dentry) +{ + struct nfs_removeargs arg = { + .fh = NFS_FH(dir), + .name = dentry->d_name, + }; + struct nfs_removeres res; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status = -ENOMEM; + + dprintk("NFS call remove %pd2\n", dentry); + res.dir_attr = nfs_alloc_fattr(); + if (res.dir_attr == NULL) + goto out; + + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_post_op_update_inode(dir, res.dir_attr); + nfs_free_fattr(res.dir_attr); +out: + dprintk("NFS reply remove: %d\n", status); + return status; +} + +static void +nfs3_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) +{ + msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE]; +} + +static void nfs3_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data) +{ + rpc_call_start(task); +} + +static int +nfs3_proc_unlink_done(struct rpc_task *task, struct inode *dir) +{ + struct nfs_removeres *res; + if (nfs3_async_handle_jukebox(task, dir)) + return 0; + res = task->tk_msg.rpc_resp; + nfs_post_op_update_inode(dir, res->dir_attr); + return 1; +} + +static void +nfs3_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry, + struct inode *same_parent) +{ + msg->rpc_proc = &nfs3_procedures[NFS3PROC_RENAME]; +} + +static void nfs3_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data) +{ + rpc_call_start(task); +} + +static int +nfs3_proc_rename_done(struct rpc_task *task, struct inode *old_dir, + struct inode *new_dir) +{ + struct nfs_renameres *res; + + if (nfs3_async_handle_jukebox(task, old_dir)) + return 0; + res = task->tk_msg.rpc_resp; + + nfs_post_op_update_inode(old_dir, res->old_fattr); + nfs_post_op_update_inode(new_dir, res->new_fattr); + return 1; +} + +static int +nfs3_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) +{ + struct nfs3_linkargs arg = { + .fromfh = NFS_FH(inode), + .tofh = NFS_FH(dir), + .toname = name->name, + .tolen = name->len + }; + struct nfs3_linkres res; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_LINK], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status = -ENOMEM; + + dprintk("NFS call link %s\n", name->name); + res.fattr = nfs_alloc_fattr(); + res.dir_attr = nfs_alloc_fattr(); + if (res.fattr == NULL || res.dir_attr == NULL) + goto out; + + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + nfs_post_op_update_inode(dir, res.dir_attr); + nfs_post_op_update_inode(inode, res.fattr); +out: + nfs_free_fattr(res.dir_attr); + nfs_free_fattr(res.fattr); + dprintk("NFS reply link: %d\n", status); + return status; +} + +static int +nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct folio *folio, + unsigned int len, struct iattr *sattr) +{ + struct page *page = &folio->page; + struct nfs3_createdata *data; + struct dentry *d_alias; + int status = -ENOMEM; + + if (len > NFS3_MAXPATHLEN) + return -ENAMETOOLONG; + + dprintk("NFS call symlink %pd\n", dentry); + + data = nfs3_alloc_createdata(); + if (data == NULL) + goto out; + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_SYMLINK]; + data->arg.symlink.fromfh = NFS_FH(dir); + data->arg.symlink.fromname = dentry->d_name.name; + data->arg.symlink.fromlen = dentry->d_name.len; + data->arg.symlink.pages = &page; + data->arg.symlink.pathlen = len; + data->arg.symlink.sattr = sattr; + + d_alias = nfs3_do_create(dir, dentry, data); + status = PTR_ERR_OR_ZERO(d_alias); + + if (status == 0) + dput(d_alias); + + nfs3_free_createdata(data); +out: + dprintk("NFS reply symlink: %d\n", status); + return status; +} + +static struct dentry * +nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) +{ + struct posix_acl *default_acl, *acl; + struct nfs3_createdata *data; + struct dentry *ret = ERR_PTR(-ENOMEM); + int status; + + dprintk("NFS call mkdir %pd\n", dentry); + + data = nfs3_alloc_createdata(); + if (data == NULL) + goto out; + + ret = ERR_PTR(posix_acl_create(dir, &sattr->ia_mode, + &default_acl, &acl)); + if (IS_ERR(ret)) + goto out; + + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKDIR]; + data->arg.mkdir.fh = NFS_FH(dir); + data->arg.mkdir.name = dentry->d_name.name; + data->arg.mkdir.len = dentry->d_name.len; + data->arg.mkdir.sattr = sattr; + + ret = nfs3_do_create(dir, dentry, data); + + if (IS_ERR(ret)) + goto out_release_acls; + + if (ret) + dentry = ret; + + status = nfs3_proc_setacls(d_inode(dentry), acl, default_acl); + if (status) { + dput(ret); + ret = ERR_PTR(status); + } + +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); +out: + nfs3_free_createdata(data); + dprintk("NFS reply mkdir: %d\n", PTR_ERR_OR_ZERO(ret)); + return ret; +} + +static int +nfs3_proc_rmdir(struct inode *dir, const struct qstr *name) +{ + struct nfs_fattr *dir_attr; + struct nfs3_diropargs arg = { + .fh = NFS_FH(dir), + .name = name->name, + .len = name->len + }; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_RMDIR], + .rpc_argp = &arg, + }; + int status = -ENOMEM; + + dprintk("NFS call rmdir %s\n", name->name); + dir_attr = nfs_alloc_fattr(); + if (dir_attr == NULL) + goto out; + + msg.rpc_resp = dir_attr; + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_post_op_update_inode(dir, dir_attr); + nfs_free_fattr(dir_attr); +out: + dprintk("NFS reply rmdir: %d\n", status); + return status; +} + +/* + * The READDIR implementation is somewhat hackish - we pass the user buffer + * to the encode function, which installs it in the receive iovec. + * The decode function itself doesn't perform any decoding, it just makes + * sure the reply is syntactically correct. + * + * Also note that this implementation handles both plain readdir and + * readdirplus. + */ +static int nfs3_proc_readdir(struct nfs_readdir_arg *nr_arg, + struct nfs_readdir_res *nr_res) +{ + struct inode *dir = d_inode(nr_arg->dentry); + struct nfs3_readdirargs arg = { + .fh = NFS_FH(dir), + .cookie = nr_arg->cookie, + .plus = nr_arg->plus, + .count = nr_arg->page_len, + .pages = nr_arg->pages + }; + struct nfs3_readdirres res = { + .verf = nr_res->verf, + .plus = nr_arg->plus, + }; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_READDIR], + .rpc_argp = &arg, + .rpc_resp = &res, + .rpc_cred = nr_arg->cred, + }; + int status = -ENOMEM; + + if (nr_arg->plus) + msg.rpc_proc = &nfs3_procedures[NFS3PROC_READDIRPLUS]; + if (arg.cookie) + memcpy(arg.verf, nr_arg->verf, sizeof(arg.verf)); + + dprintk("NFS call readdir%s %llu\n", nr_arg->plus ? "plus" : "", + (unsigned long long)nr_arg->cookie); + + res.dir_attr = nfs_alloc_fattr(); + if (res.dir_attr == NULL) + goto out; + + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + + nfs_invalidate_atime(dir); + nfs_refresh_inode(dir, res.dir_attr); + + nfs_free_fattr(res.dir_attr); +out: + dprintk("NFS reply readdir%s: %d\n", nr_arg->plus ? "plus" : "", + status); + return status; +} + +static int +nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, + dev_t rdev) +{ + struct posix_acl *default_acl, *acl; + struct nfs3_createdata *data; + struct dentry *d_alias; + int status = -ENOMEM; + + dprintk("NFS call mknod %pd %u:%u\n", dentry, + MAJOR(rdev), MINOR(rdev)); + + data = nfs3_alloc_createdata(); + if (data == NULL) + goto out; + + status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl); + if (status) + goto out; + + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKNOD]; + data->arg.mknod.fh = NFS_FH(dir); + data->arg.mknod.name = dentry->d_name.name; + data->arg.mknod.len = dentry->d_name.len; + data->arg.mknod.sattr = sattr; + data->arg.mknod.rdev = rdev; + + switch (sattr->ia_mode & S_IFMT) { + case S_IFBLK: + data->arg.mknod.type = NF3BLK; + break; + case S_IFCHR: + data->arg.mknod.type = NF3CHR; + break; + case S_IFIFO: + data->arg.mknod.type = NF3FIFO; + break; + case S_IFSOCK: + data->arg.mknod.type = NF3SOCK; + break; + default: + status = -EINVAL; + goto out_release_acls; + } + + d_alias = nfs3_do_create(dir, dentry, data); + status = PTR_ERR_OR_ZERO(d_alias); + if (status != 0) + goto out_release_acls; + + if (d_alias) + dentry = d_alias; + + status = nfs3_proc_setacls(d_inode(dentry), acl, default_acl); + + dput(d_alias); +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); +out: + nfs3_free_createdata(data); + dprintk("NFS reply mknod: %d\n", status); + return status; +} + +static int +nfs3_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsstat *stat) +{ + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_FSSTAT], + .rpc_argp = fhandle, + .rpc_resp = stat, + }; + int status; + + dprintk("NFS call fsstat\n"); + nfs_fattr_init(stat->fattr); + status = rpc_call_sync(server->client, &msg, 0); + dprintk("NFS reply fsstat: %d\n", status); + return status; +} + +static int +do_proc_fsinfo(struct rpc_clnt *client, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_FSINFO], + .rpc_argp = fhandle, + .rpc_resp = info, + }; + int status; + + dprintk("NFS call fsinfo\n"); + nfs_fattr_init(info->fattr); + status = rpc_call_sync(client, &msg, 0); + dprintk("NFS reply fsinfo: %d\n", status); + return status; +} + +/* + * Bare-bones access to fsinfo: this is for nfs_get_root/nfs_get_sb via + * nfs_create_server + */ +static int +nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + int status; + + status = do_proc_fsinfo(server->client, fhandle, info); + if (status && server->nfs_client->cl_rpcclient != server->client) + status = do_proc_fsinfo(server->nfs_client->cl_rpcclient, fhandle, info); + return status; +} + +static int +nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_pathconf *info) +{ + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_PATHCONF], + .rpc_argp = fhandle, + .rpc_resp = info, + }; + int status; + + dprintk("NFS call pathconf\n"); + nfs_fattr_init(info->fattr); + status = rpc_call_sync(server->client, &msg, 0); + dprintk("NFS reply pathconf: %d\n", status); + return status; +} + +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + +static unsigned nfs3_localio_probe_throttle __read_mostly = 0; +module_param(nfs3_localio_probe_throttle, uint, 0644); +MODULE_PARM_DESC(nfs3_localio_probe_throttle, + "Probe for NFSv3 LOCALIO every N IO requests. Must be power-of-2, defaults to 0 (probing disabled)."); + +static void nfs3_localio_probe(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + + /* Throttled to reduce nfs_local_probe_async() frequency */ + if (!nfs3_localio_probe_throttle || nfs_server_is_local(clp)) + return; + + /* + * Try (re)enabling LOCALIO if isn't enabled -- admin deems + * it worthwhile to periodically check if LOCALIO possible by + * setting the 'nfs3_localio_probe_throttle' module parameter. + * + * This is useful if LOCALIO was previously enabled, but was + * disabled due to server restart, and IO has successfully + * completed in terms of normal RPC. + */ + if ((clp->cl_uuid.nfs3_localio_probe_count++ & + (nfs3_localio_probe_throttle - 1)) == 0) { + if (!nfs_server_is_local(clp)) + nfs_local_probe_async(clp); + } +} + +#else +static void nfs3_localio_probe(struct nfs_server *server) {} +#endif + +static int nfs3_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + struct inode *inode = hdr->inode; + struct nfs_server *server = NFS_SERVER(inode); + + if (hdr->pgio_done_cb != NULL) + return hdr->pgio_done_cb(task, hdr); + + if (nfs3_async_handle_jukebox(task, inode)) + return -EAGAIN; + + if (task->tk_status >= 0) { + if (!server->read_hdrsize) + cmpxchg(&server->read_hdrsize, 0, hdr->res.replen); + nfs3_localio_probe(server); + } + + nfs_invalidate_atime(inode); + nfs_refresh_inode(inode, &hdr->fattr); + return 0; +} + +static void nfs3_proc_read_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg) +{ + msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ]; + hdr->args.replen = NFS_SERVER(hdr->inode)->read_hdrsize; +} + +static int nfs3_proc_pgio_rpc_prepare(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + rpc_call_start(task); + return 0; +} + +static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + struct inode *inode = hdr->inode; + + if (hdr->pgio_done_cb != NULL) + return hdr->pgio_done_cb(task, hdr); + + if (nfs3_async_handle_jukebox(task, inode)) + return -EAGAIN; + if (task->tk_status >= 0) { + nfs_writeback_update_inode(hdr); + nfs3_localio_probe(NFS_SERVER(inode)); + } + return 0; +} + +static void nfs3_proc_write_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE]; +} + +static void nfs3_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) +{ + rpc_call_start(task); +} + +static int nfs3_commit_done(struct rpc_task *task, struct nfs_commit_data *data) +{ + if (data->commit_done_cb != NULL) + return data->commit_done_cb(task, data); + + if (nfs3_async_handle_jukebox(task, data->inode)) + return -EAGAIN; + nfs_refresh_inode(data->inode, data->res.fattr); + return 0; +} + +static void nfs3_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + msg->rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT]; +} + +static void nfs3_nlm_alloc_call(void *data) +{ + struct nfs_lock_context *l_ctx = data; + if (l_ctx && test_bit(NFS_CONTEXT_UNLOCK, &l_ctx->open_context->flags)) { + get_nfs_open_context(l_ctx->open_context); + nfs_get_lock_context(l_ctx->open_context); + } +} + +static bool nfs3_nlm_unlock_prepare(struct rpc_task *task, void *data) +{ + struct nfs_lock_context *l_ctx = data; + if (l_ctx && test_bit(NFS_CONTEXT_UNLOCK, &l_ctx->open_context->flags)) + return nfs_async_iocounter_wait(task, l_ctx); + return false; + +} + +static void nfs3_nlm_release_call(void *data) +{ + struct nfs_lock_context *l_ctx = data; + struct nfs_open_context *ctx; + if (l_ctx && test_bit(NFS_CONTEXT_UNLOCK, &l_ctx->open_context->flags)) { + ctx = l_ctx->open_context; + nfs_put_lock_context(l_ctx); + put_nfs_open_context(ctx); + } +} + +static const struct nlmclnt_operations nlmclnt_fl_close_lock_ops = { + .nlmclnt_alloc_call = nfs3_nlm_alloc_call, + .nlmclnt_unlock_prepare = nfs3_nlm_unlock_prepare, + .nlmclnt_release_call = nfs3_nlm_release_call, +}; + +static int +nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl) +{ + struct inode *inode = file_inode(filp); + struct nfs_lock_context *l_ctx = NULL; + struct nfs_open_context *ctx = nfs_file_open_context(filp); + int status; + + if (fl->c.flc_flags & FL_CLOSE) { + l_ctx = nfs_get_lock_context(ctx); + if (IS_ERR(l_ctx)) + l_ctx = NULL; + else + set_bit(NFS_CONTEXT_UNLOCK, &ctx->flags); + } + + status = nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl, l_ctx); + + if (l_ctx) + nfs_put_lock_context(l_ctx); + + return status; +} + +static int nfs3_have_delegation(struct inode *inode, fmode_t type, int flags) +{ + return 0; +} + +static void nfs3_return_delegation(struct inode *inode) +{ + if (S_ISREG(inode->i_mode)) + nfs_wb_all(inode); +} + +static const struct inode_operations nfs3_dir_inode_operations = { + .create = nfs_create, + .atomic_open = nfs_atomic_open_v23, + .lookup = nfs_lookup, + .link = nfs_link, + .unlink = nfs_unlink, + .symlink = nfs_symlink, + .mkdir = nfs_mkdir, + .rmdir = nfs_rmdir, + .mknod = nfs_mknod, + .rename = nfs_rename, + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +#ifdef CONFIG_NFS_V3_ACL + .listxattr = nfs3_listxattr, + .get_inode_acl = nfs3_get_acl, + .set_acl = nfs3_set_acl, +#endif +}; + +static const struct inode_operations nfs3_file_inode_operations = { + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +#ifdef CONFIG_NFS_V3_ACL + .listxattr = nfs3_listxattr, + .get_inode_acl = nfs3_get_acl, + .set_acl = nfs3_set_acl, +#endif +}; + +const struct nfs_rpc_ops nfs_v3_clientops = { + .version = 3, /* protocol version */ + .dentry_ops = &nfs_dentry_operations, + .dir_inode_ops = &nfs3_dir_inode_operations, + .file_inode_ops = &nfs3_file_inode_operations, + .file_ops = &nfs_file_operations, + .nlmclnt_ops = &nlmclnt_fl_close_lock_ops, + .getroot = nfs3_proc_get_root, + .submount = nfs_submount, + .try_get_tree = nfs_try_get_tree, + .getattr = nfs3_proc_getattr, + .setattr = nfs3_proc_setattr, + .lookup = nfs3_proc_lookup, + .lookupp = nfs3_proc_lookupp, + .access = nfs3_proc_access, + .readlink = nfs3_proc_readlink, + .create = nfs3_proc_create, + .remove = nfs3_proc_remove, + .unlink_setup = nfs3_proc_unlink_setup, + .unlink_rpc_prepare = nfs3_proc_unlink_rpc_prepare, + .unlink_done = nfs3_proc_unlink_done, + .rename_setup = nfs3_proc_rename_setup, + .rename_rpc_prepare = nfs3_proc_rename_rpc_prepare, + .rename_done = nfs3_proc_rename_done, + .link = nfs3_proc_link, + .symlink = nfs3_proc_symlink, + .mkdir = nfs3_proc_mkdir, + .rmdir = nfs3_proc_rmdir, + .readdir = nfs3_proc_readdir, + .mknod = nfs3_proc_mknod, + .statfs = nfs3_proc_statfs, + .fsinfo = nfs3_proc_fsinfo, + .pathconf = nfs3_proc_pathconf, + .decode_dirent = nfs3_decode_dirent, + .pgio_rpc_prepare = nfs3_proc_pgio_rpc_prepare, + .read_setup = nfs3_proc_read_setup, + .read_done = nfs3_read_done, + .write_setup = nfs3_proc_write_setup, + .write_done = nfs3_write_done, + .commit_setup = nfs3_proc_commit_setup, + .commit_rpc_prepare = nfs3_proc_commit_rpc_prepare, + .commit_done = nfs3_commit_done, + .lock = nfs3_proc_lock, + .clear_acl_cache = forget_all_cached_acls, + .close_context = nfs_close_context, + .have_delegation = nfs3_have_delegation, + .return_delegation = nfs3_return_delegation, + .alloc_client = nfs_alloc_client, + .init_client = nfs_init_client, + .free_client = nfs_free_client, + .create_server = nfs3_create_server, + .clone_server = nfs3_clone_server, +}; diff --git a/fs/nfs/nfs3super.c b/fs/nfs/nfs3super.c new file mode 100644 index 000000000..20a804784 --- /dev/null +++ b/fs/nfs/nfs3super.c @@ -0,0 +1,34 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012 Netapp, Inc. All rights reserved. + */ +#include +#include +#include "internal.h" +#include "nfs3_fs.h" +#include "nfs.h" + +struct nfs_subversion nfs_v3 = { + .owner = THIS_MODULE, + .nfs_fs = &nfs_fs_type, + .rpc_vers = &nfs_version3, + .rpc_ops = &nfs_v3_clientops, + .sops = &nfs_sops, +}; + +static int __init init_nfs_v3(void) +{ + register_nfs_version(&nfs_v3); + return 0; +} + +static void __exit exit_nfs_v3(void) +{ + unregister_nfs_version(&nfs_v3); +} + +MODULE_DESCRIPTION("NFSv3 client support"); +MODULE_LICENSE("GPL"); + +module_init(init_nfs_v3); +module_exit(exit_nfs_v3); diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c new file mode 100644 index 000000000..e745e78fa --- /dev/null +++ b/fs/nfs/nfs3xdr.c @@ -0,0 +1,2516 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/nfs3xdr.c + * + * XDR functions to encode/decode NFSv3 RPC arguments and results. + * + * Copyright (C) 1996, 1997 Olaf Kirch + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_XDR + +/* + * Declare the space requirements for NFS arguments and replies as + * number of 32bit-words + */ +#define NFS3_pagepad_sz (1) /* Page padding */ +#define NFS3_fhandle_sz (1+16) +#define NFS3_fh_sz (NFS3_fhandle_sz) /* shorthand */ +#define NFS3_post_op_fh_sz (1+NFS3_fh_sz) +#define NFS3_sattr_sz (15) +#define NFS3_filename_sz (1+(NFS3_MAXNAMLEN>>2)) +#define NFS3_path_sz (1+(NFS3_MAXPATHLEN>>2)) +#define NFS3_fattr_sz (21) +#define NFS3_cookieverf_sz (NFS3_COOKIEVERFSIZE>>2) +#define NFS3_wcc_attr_sz (6) +#define NFS3_pre_op_attr_sz (1+NFS3_wcc_attr_sz) +#define NFS3_post_op_attr_sz (1+NFS3_fattr_sz) +#define NFS3_wcc_data_sz (NFS3_pre_op_attr_sz+NFS3_post_op_attr_sz) +#define NFS3_diropargs_sz (NFS3_fh_sz+NFS3_filename_sz) + +#define NFS3_getattrargs_sz (NFS3_fh_sz) +#define NFS3_setattrargs_sz (NFS3_fh_sz+NFS3_sattr_sz+3) +#define NFS3_lookupargs_sz (NFS3_fh_sz+NFS3_filename_sz) +#define NFS3_accessargs_sz (NFS3_fh_sz+1) +#define NFS3_readlinkargs_sz (NFS3_fh_sz) +#define NFS3_readargs_sz (NFS3_fh_sz+3) +#define NFS3_writeargs_sz (NFS3_fh_sz+5) +#define NFS3_createargs_sz (NFS3_diropargs_sz+NFS3_sattr_sz) +#define NFS3_mkdirargs_sz (NFS3_diropargs_sz+NFS3_sattr_sz) +#define NFS3_symlinkargs_sz (NFS3_diropargs_sz+1+NFS3_sattr_sz) +#define NFS3_mknodargs_sz (NFS3_diropargs_sz+2+NFS3_sattr_sz) +#define NFS3_removeargs_sz (NFS3_fh_sz+NFS3_filename_sz) +#define NFS3_renameargs_sz (NFS3_diropargs_sz+NFS3_diropargs_sz) +#define NFS3_linkargs_sz (NFS3_fh_sz+NFS3_diropargs_sz) +#define NFS3_readdirargs_sz (NFS3_fh_sz+NFS3_cookieverf_sz+3) +#define NFS3_readdirplusargs_sz (NFS3_fh_sz+NFS3_cookieverf_sz+4) +#define NFS3_commitargs_sz (NFS3_fh_sz+3) + +#define NFS3_getattrres_sz (1+NFS3_fattr_sz) +#define NFS3_setattrres_sz (1+NFS3_wcc_data_sz) +#define NFS3_removeres_sz (NFS3_setattrres_sz) +#define NFS3_lookupres_sz (1+NFS3_fh_sz+(2 * NFS3_post_op_attr_sz)) +#define NFS3_accessres_sz (1+NFS3_post_op_attr_sz+1) +#define NFS3_readlinkres_sz (1+NFS3_post_op_attr_sz+1+NFS3_pagepad_sz) +#define NFS3_readres_sz (1+NFS3_post_op_attr_sz+3+NFS3_pagepad_sz) +#define NFS3_writeres_sz (1+NFS3_wcc_data_sz+4) +#define NFS3_createres_sz (1+NFS3_post_op_fh_sz+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) +#define NFS3_renameres_sz (1+(2 * NFS3_wcc_data_sz)) +#define NFS3_linkres_sz (1+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) +#define NFS3_readdirres_sz (1+NFS3_post_op_attr_sz+2+NFS3_pagepad_sz) +#define NFS3_fsstatres_sz (1+NFS3_post_op_attr_sz+13) +#define NFS3_fsinfores_sz (1+NFS3_post_op_attr_sz+12) +#define NFS3_pathconfres_sz (1+NFS3_post_op_attr_sz+6) +#define NFS3_commitres_sz (1+NFS3_wcc_data_sz+2) + +#define ACL3_getaclargs_sz (NFS3_fh_sz+1) +#define ACL3_setaclargs_sz (NFS3_fh_sz+1+ \ + XDR_QUADLEN(NFS_ACL_INLINE_BUFSIZE)) +#define ACL3_getaclres_sz (1+NFS3_post_op_attr_sz+1+ \ + XDR_QUADLEN(NFS_ACL_INLINE_BUFSIZE)+\ + NFS3_pagepad_sz) +#define ACL3_setaclres_sz (1+NFS3_post_op_attr_sz) + +/* + * Map file type to S_IFMT bits + */ +static const umode_t nfs_type2fmt[] = { + [NF3BAD] = 0, + [NF3REG] = S_IFREG, + [NF3DIR] = S_IFDIR, + [NF3BLK] = S_IFBLK, + [NF3CHR] = S_IFCHR, + [NF3LNK] = S_IFLNK, + [NF3SOCK] = S_IFSOCK, + [NF3FIFO] = S_IFIFO, +}; + +static struct user_namespace *rpc_userns(const struct rpc_clnt *clnt) +{ + if (clnt && clnt->cl_cred) + return clnt->cl_cred->user_ns; + return &init_user_ns; +} + +static struct user_namespace *rpc_rqst_userns(const struct rpc_rqst *rqstp) +{ + if (rqstp->rq_task) + return rpc_userns(rqstp->rq_task->tk_client); + return &init_user_ns; +} + +/* + * Encode/decode NFSv3 basic data types + * + * Basic NFSv3 data types are defined in section 2.5 of RFC 1813: + * "NFS Version 3 Protocol Specification". + * + * Not all basic data types have their own encoding and decoding + * functions. For run-time efficiency, some data types are encoded + * or decoded inline. + */ + +static void encode_uint32(struct xdr_stream *xdr, u32 value) +{ + __be32 *p = xdr_reserve_space(xdr, 4); + *p = cpu_to_be32(value); +} + +static int decode_uint32(struct xdr_stream *xdr, u32 *value) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *value = be32_to_cpup(p); + return 0; +} + +static int decode_uint64(struct xdr_stream *xdr, u64 *value) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, value); + return 0; +} + +/* + * fileid3 + * + * typedef uint64 fileid3; + */ +static __be32 *xdr_decode_fileid3(__be32 *p, u64 *fileid) +{ + return xdr_decode_hyper(p, fileid); +} + +static int decode_fileid3(struct xdr_stream *xdr, u64 *fileid) +{ + return decode_uint64(xdr, fileid); +} + +/* + * filename3 + * + * typedef string filename3<>; + */ +static void encode_filename3(struct xdr_stream *xdr, + const char *name, u32 length) +{ + __be32 *p; + + WARN_ON_ONCE(length > NFS3_MAXNAMLEN); + p = xdr_reserve_space(xdr, 4 + length); + xdr_encode_opaque(p, name, length); +} + +static int decode_inline_filename3(struct xdr_stream *xdr, + const char **name, u32 *length) +{ + __be32 *p; + u32 count; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p); + if (count > NFS3_MAXNAMLEN) + goto out_nametoolong; + p = xdr_inline_decode(xdr, count); + if (unlikely(!p)) + return -EIO; + *name = (const char *)p; + *length = count; + return 0; + +out_nametoolong: + dprintk("NFS: returned filename too long: %u\n", count); + return -ENAMETOOLONG; +} + +/* + * nfspath3 + * + * typedef string nfspath3<>; + */ +static void encode_nfspath3(struct xdr_stream *xdr, struct page **pages, + const u32 length) +{ + encode_uint32(xdr, length); + xdr_write_pages(xdr, pages, 0, length); +} + +static int decode_nfspath3(struct xdr_stream *xdr) +{ + u32 recvd, count; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p); + if (unlikely(count >= xdr->buf->page_len || count > NFS3_MAXPATHLEN)) + goto out_nametoolong; + recvd = xdr_read_pages(xdr, count); + if (unlikely(count > recvd)) + goto out_cheating; + xdr_terminate_string(xdr->buf, count); + return 0; + +out_nametoolong: + dprintk("NFS: returned pathname too long: %u\n", count); + return -ENAMETOOLONG; +out_cheating: + dprintk("NFS: server cheating in pathname result: " + "count %u > recvd %u\n", count, recvd); + return -EIO; +} + +/* + * cookie3 + * + * typedef uint64 cookie3 + */ +static __be32 *xdr_encode_cookie3(__be32 *p, u64 cookie) +{ + return xdr_encode_hyper(p, cookie); +} + +static int decode_cookie3(struct xdr_stream *xdr, u64 *cookie) +{ + return decode_uint64(xdr, cookie); +} + +/* + * cookieverf3 + * + * typedef opaque cookieverf3[NFS3_COOKIEVERFSIZE]; + */ +static __be32 *xdr_encode_cookieverf3(__be32 *p, const __be32 *verifier) +{ + memcpy(p, verifier, NFS3_COOKIEVERFSIZE); + return p + XDR_QUADLEN(NFS3_COOKIEVERFSIZE); +} + +static int decode_cookieverf3(struct xdr_stream *xdr, __be32 *verifier) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS3_COOKIEVERFSIZE); + if (unlikely(!p)) + return -EIO; + memcpy(verifier, p, NFS3_COOKIEVERFSIZE); + return 0; +} + +/* + * createverf3 + * + * typedef opaque createverf3[NFS3_CREATEVERFSIZE]; + */ +static void encode_createverf3(struct xdr_stream *xdr, const __be32 *verifier) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, NFS3_CREATEVERFSIZE); + memcpy(p, verifier, NFS3_CREATEVERFSIZE); +} + +static int decode_writeverf3(struct xdr_stream *xdr, struct nfs_write_verifier *verifier) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS3_WRITEVERFSIZE); + if (unlikely(!p)) + return -EIO; + memcpy(verifier->data, p, NFS3_WRITEVERFSIZE); + return 0; +} + +/* + * size3 + * + * typedef uint64 size3; + */ +static __be32 *xdr_decode_size3(__be32 *p, u64 *size) +{ + return xdr_decode_hyper(p, size); +} + +/* + * nfsstat3 + * + * enum nfsstat3 { + * NFS3_OK = 0, + * ... + * } + */ +#define NFS3_OK NFS_OK + +static int decode_nfsstat3(struct xdr_stream *xdr, enum nfs_stat *status) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + if (unlikely(*p != cpu_to_be32(NFS3_OK))) + goto out_status; + *status = 0; + return 0; +out_status: + *status = be32_to_cpup(p); + trace_nfs_xdr_status(xdr, (int)*status); + return 0; +} + +/* + * ftype3 + * + * enum ftype3 { + * NF3REG = 1, + * NF3DIR = 2, + * NF3BLK = 3, + * NF3CHR = 4, + * NF3LNK = 5, + * NF3SOCK = 6, + * NF3FIFO = 7 + * }; + */ +static void encode_ftype3(struct xdr_stream *xdr, const u32 type) +{ + encode_uint32(xdr, type); +} + +static __be32 *xdr_decode_ftype3(__be32 *p, umode_t *mode) +{ + u32 type; + + type = be32_to_cpup(p++); + if (type > NF3FIFO) + type = NF3NON; + *mode = nfs_type2fmt[type]; + return p; +} + +/* + * specdata3 + * + * struct specdata3 { + * uint32 specdata1; + * uint32 specdata2; + * }; + */ +static void encode_specdata3(struct xdr_stream *xdr, const dev_t rdev) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 8); + *p++ = cpu_to_be32(MAJOR(rdev)); + *p = cpu_to_be32(MINOR(rdev)); +} + +static __be32 *xdr_decode_specdata3(__be32 *p, dev_t *rdev) +{ + unsigned int major, minor; + + major = be32_to_cpup(p++); + minor = be32_to_cpup(p++); + *rdev = MKDEV(major, minor); + if (MAJOR(*rdev) != major || MINOR(*rdev) != minor) + *rdev = 0; + return p; +} + +/* + * nfs_fh3 + * + * struct nfs_fh3 { + * opaque data; + * }; + */ +static void encode_nfs_fh3(struct xdr_stream *xdr, const struct nfs_fh *fh) +{ + __be32 *p; + + WARN_ON_ONCE(fh->size > NFS3_FHSIZE); + p = xdr_reserve_space(xdr, 4 + fh->size); + xdr_encode_opaque(p, fh->data, fh->size); +} + +static int decode_nfs_fh3(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + u32 length; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + length = be32_to_cpup(p++); + if (unlikely(length > NFS3_FHSIZE || length == 0)) + goto out_toobig; + p = xdr_inline_decode(xdr, length); + if (unlikely(!p)) + return -EIO; + fh->size = length; + memcpy(fh->data, p, length); + return 0; +out_toobig: + trace_nfs_xdr_bad_filehandle(xdr, NFSERR_BADHANDLE); + return -E2BIG; +} + +static void zero_nfs_fh3(struct nfs_fh *fh) +{ + memset(fh, 0, sizeof(*fh)); +} + +/* + * nfstime3 + * + * struct nfstime3 { + * uint32 seconds; + * uint32 nseconds; + * }; + */ +static __be32 *xdr_encode_nfstime3(__be32 *p, const struct timespec64 *timep) +{ + *p++ = cpu_to_be32((u32)timep->tv_sec); + *p++ = cpu_to_be32(timep->tv_nsec); + return p; +} + +static __be32 *xdr_decode_nfstime3(__be32 *p, struct timespec64 *timep) +{ + timep->tv_sec = be32_to_cpup(p++); + timep->tv_nsec = be32_to_cpup(p++); + return p; +} + +/* + * sattr3 + * + * enum time_how { + * DONT_CHANGE = 0, + * SET_TO_SERVER_TIME = 1, + * SET_TO_CLIENT_TIME = 2 + * }; + * + * union set_mode3 switch (bool set_it) { + * case TRUE: + * mode3 mode; + * default: + * void; + * }; + * + * union set_uid3 switch (bool set_it) { + * case TRUE: + * uid3 uid; + * default: + * void; + * }; + * + * union set_gid3 switch (bool set_it) { + * case TRUE: + * gid3 gid; + * default: + * void; + * }; + * + * union set_size3 switch (bool set_it) { + * case TRUE: + * size3 size; + * default: + * void; + * }; + * + * union set_atime switch (time_how set_it) { + * case SET_TO_CLIENT_TIME: + * nfstime3 atime; + * default: + * void; + * }; + * + * union set_mtime switch (time_how set_it) { + * case SET_TO_CLIENT_TIME: + * nfstime3 mtime; + * default: + * void; + * }; + * + * struct sattr3 { + * set_mode3 mode; + * set_uid3 uid; + * set_gid3 gid; + * set_size3 size; + * set_atime atime; + * set_mtime mtime; + * }; + */ +static void encode_sattr3(struct xdr_stream *xdr, const struct iattr *attr, + struct user_namespace *userns) +{ + u32 nbytes; + __be32 *p; + + /* + * In order to make only a single xdr_reserve_space() call, + * pre-compute the total number of bytes to be reserved. + * Six boolean values, one for each set_foo field, are always + * present in the encoded result, so start there. + */ + nbytes = 6 * 4; + if (attr->ia_valid & ATTR_MODE) + nbytes += 4; + if (attr->ia_valid & ATTR_UID) + nbytes += 4; + if (attr->ia_valid & ATTR_GID) + nbytes += 4; + if (attr->ia_valid & ATTR_SIZE) + nbytes += 8; + if (attr->ia_valid & ATTR_ATIME_SET) + nbytes += 8; + if (attr->ia_valid & ATTR_MTIME_SET) + nbytes += 8; + p = xdr_reserve_space(xdr, nbytes); + + if (attr->ia_valid & ATTR_MODE) { + *p++ = xdr_one; + *p++ = cpu_to_be32(attr->ia_mode & S_IALLUGO); + } else + *p++ = xdr_zero; + + if (attr->ia_valid & ATTR_UID) { + *p++ = xdr_one; + *p++ = cpu_to_be32(from_kuid_munged(userns, attr->ia_uid)); + } else + *p++ = xdr_zero; + + if (attr->ia_valid & ATTR_GID) { + *p++ = xdr_one; + *p++ = cpu_to_be32(from_kgid_munged(userns, attr->ia_gid)); + } else + *p++ = xdr_zero; + + if (attr->ia_valid & ATTR_SIZE) { + *p++ = xdr_one; + p = xdr_encode_hyper(p, (u64)attr->ia_size); + } else + *p++ = xdr_zero; + + if (attr->ia_valid & ATTR_ATIME_SET) { + *p++ = xdr_two; + p = xdr_encode_nfstime3(p, &attr->ia_atime); + } else if (attr->ia_valid & ATTR_ATIME) { + *p++ = xdr_one; + } else + *p++ = xdr_zero; + + if (attr->ia_valid & ATTR_MTIME_SET) { + *p++ = xdr_two; + xdr_encode_nfstime3(p, &attr->ia_mtime); + } else if (attr->ia_valid & ATTR_MTIME) { + *p = xdr_one; + } else + *p = xdr_zero; +} + +/* + * fattr3 + * + * struct fattr3 { + * ftype3 type; + * mode3 mode; + * uint32 nlink; + * uid3 uid; + * gid3 gid; + * size3 size; + * size3 used; + * specdata3 rdev; + * uint64 fsid; + * fileid3 fileid; + * nfstime3 atime; + * nfstime3 mtime; + * nfstime3 ctime; + * }; + */ +static int decode_fattr3(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) +{ + umode_t fmode; + __be32 *p; + + p = xdr_inline_decode(xdr, NFS3_fattr_sz << 2); + if (unlikely(!p)) + return -EIO; + + p = xdr_decode_ftype3(p, &fmode); + + fattr->mode = (be32_to_cpup(p++) & ~S_IFMT) | fmode; + fattr->nlink = be32_to_cpup(p++); + fattr->uid = make_kuid(userns, be32_to_cpup(p++)); + if (!uid_valid(fattr->uid)) + goto out_uid; + fattr->gid = make_kgid(userns, be32_to_cpup(p++)); + if (!gid_valid(fattr->gid)) + goto out_gid; + + p = xdr_decode_size3(p, &fattr->size); + p = xdr_decode_size3(p, &fattr->du.nfs3.used); + p = xdr_decode_specdata3(p, &fattr->rdev); + + p = xdr_decode_hyper(p, &fattr->fsid.major); + fattr->fsid.minor = 0; + + p = xdr_decode_fileid3(p, &fattr->fileid); + p = xdr_decode_nfstime3(p, &fattr->atime); + p = xdr_decode_nfstime3(p, &fattr->mtime); + xdr_decode_nfstime3(p, &fattr->ctime); + fattr->change_attr = nfs_timespec_to_change_attr(&fattr->ctime); + + fattr->valid |= NFS_ATTR_FATTR_V3; + return 0; +out_uid: + dprintk("NFS: returned invalid uid\n"); + return -EINVAL; +out_gid: + dprintk("NFS: returned invalid gid\n"); + return -EINVAL; +} + +/* + * post_op_attr + * + * union post_op_attr switch (bool attributes_follow) { + * case TRUE: + * fattr3 attributes; + * case FALSE: + * void; + * }; + */ +static int decode_post_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + if (*p != xdr_zero) + return decode_fattr3(xdr, fattr, userns); + return 0; +} + +/* + * wcc_attr + * struct wcc_attr { + * size3 size; + * nfstime3 mtime; + * nfstime3 ctime; + * }; + */ +static int decode_wcc_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, NFS3_wcc_attr_sz << 2); + if (unlikely(!p)) + return -EIO; + + fattr->valid |= NFS_ATTR_FATTR_PRESIZE + | NFS_ATTR_FATTR_PRECHANGE + | NFS_ATTR_FATTR_PREMTIME + | NFS_ATTR_FATTR_PRECTIME; + + p = xdr_decode_size3(p, &fattr->pre_size); + p = xdr_decode_nfstime3(p, &fattr->pre_mtime); + xdr_decode_nfstime3(p, &fattr->pre_ctime); + fattr->pre_change_attr = nfs_timespec_to_change_attr(&fattr->pre_ctime); + + return 0; +} + +/* + * pre_op_attr + * union pre_op_attr switch (bool attributes_follow) { + * case TRUE: + * wcc_attr attributes; + * case FALSE: + * void; + * }; + * + * wcc_data + * + * struct wcc_data { + * pre_op_attr before; + * post_op_attr after; + * }; + */ +static int decode_pre_op_attr(struct xdr_stream *xdr, struct nfs_fattr *fattr) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + if (*p != xdr_zero) + return decode_wcc_attr(xdr, fattr); + return 0; +} + +static int decode_wcc_data(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct user_namespace *userns) +{ + int error; + + error = decode_pre_op_attr(xdr, fattr); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, fattr, userns); +out: + return error; +} + +/* + * post_op_fh3 + * + * union post_op_fh3 switch (bool handle_follows) { + * case TRUE: + * nfs_fh3 handle; + * case FALSE: + * void; + * }; + */ +static int decode_post_op_fh3(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + __be32 *p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + if (*p != xdr_zero) + return decode_nfs_fh3(xdr, fh); + zero_nfs_fh3(fh); + return 0; +} + +/* + * diropargs3 + * + * struct diropargs3 { + * nfs_fh3 dir; + * filename3 name; + * }; + */ +static void encode_diropargs3(struct xdr_stream *xdr, const struct nfs_fh *fh, + const char *name, u32 length) +{ + encode_nfs_fh3(xdr, fh); + encode_filename3(xdr, name, length); +} + + +/* + * NFSv3 XDR encode functions + * + * NFSv3 argument types are defined in section 3.3 of RFC 1813: + * "NFS Version 3 Protocol Specification". + */ + +/* + * 3.3.1 GETATTR3args + * + * struct GETATTR3args { + * nfs_fh3 object; + * }; + */ +static void nfs3_xdr_enc_getattr3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_fh *fh = data; + + encode_nfs_fh3(xdr, fh); +} + +/* + * 3.3.2 SETATTR3args + * + * union sattrguard3 switch (bool check) { + * case TRUE: + * nfstime3 obj_ctime; + * case FALSE: + * void; + * }; + * + * struct SETATTR3args { + * nfs_fh3 object; + * sattr3 new_attributes; + * sattrguard3 guard; + * }; + */ +static void encode_sattrguard3(struct xdr_stream *xdr, + const struct nfs3_sattrargs *args) +{ + __be32 *p; + + if (args->guard) { + p = xdr_reserve_space(xdr, 4 + 8); + *p++ = xdr_one; + xdr_encode_nfstime3(p, &args->guardtime); + } else { + p = xdr_reserve_space(xdr, 4); + *p = xdr_zero; + } +} + +static void nfs3_xdr_enc_setattr3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_sattrargs *args = data; + encode_nfs_fh3(xdr, args->fh); + encode_sattr3(xdr, args->sattr, rpc_rqst_userns(req)); + encode_sattrguard3(xdr, args); +} + +/* + * 3.3.3 LOOKUP3args + * + * struct LOOKUP3args { + * diropargs3 what; + * }; + */ +static void nfs3_xdr_enc_lookup3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_diropargs *args = data; + + encode_diropargs3(xdr, args->fh, args->name, args->len); +} + +/* + * 3.3.4 ACCESS3args + * + * struct ACCESS3args { + * nfs_fh3 object; + * uint32 access; + * }; + */ +static void encode_access3args(struct xdr_stream *xdr, + const struct nfs3_accessargs *args) +{ + encode_nfs_fh3(xdr, args->fh); + encode_uint32(xdr, args->access); +} + +static void nfs3_xdr_enc_access3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_accessargs *args = data; + + encode_access3args(xdr, args); +} + +/* + * 3.3.5 READLINK3args + * + * struct READLINK3args { + * nfs_fh3 symlink; + * }; + */ +static void nfs3_xdr_enc_readlink3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_readlinkargs *args = data; + + encode_nfs_fh3(xdr, args->fh); + rpc_prepare_reply_pages(req, args->pages, args->pgbase, args->pglen, + NFS3_readlinkres_sz - NFS3_pagepad_sz); +} + +/* + * 3.3.6 READ3args + * + * struct READ3args { + * nfs_fh3 file; + * offset3 offset; + * count3 count; + * }; + */ +static void encode_read3args(struct xdr_stream *xdr, + const struct nfs_pgio_args *args) +{ + __be32 *p; + + encode_nfs_fh3(xdr, args->fh); + + p = xdr_reserve_space(xdr, 8 + 4); + p = xdr_encode_hyper(p, args->offset); + *p = cpu_to_be32(args->count); +} + +static void nfs3_xdr_enc_read3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + unsigned int replen = args->replen ? args->replen : + NFS3_readres_sz - NFS3_pagepad_sz; + + encode_read3args(xdr, args); + rpc_prepare_reply_pages(req, args->pages, args->pgbase, + args->count, replen); + req->rq_rcv_buf.flags |= XDRBUF_READ; +} + +/* + * 3.3.7 WRITE3args + * + * enum stable_how { + * UNSTABLE = 0, + * DATA_SYNC = 1, + * FILE_SYNC = 2 + * }; + * + * struct WRITE3args { + * nfs_fh3 file; + * offset3 offset; + * count3 count; + * stable_how stable; + * opaque data<>; + * }; + */ +static void encode_write3args(struct xdr_stream *xdr, + const struct nfs_pgio_args *args) +{ + __be32 *p; + + encode_nfs_fh3(xdr, args->fh); + + p = xdr_reserve_space(xdr, 8 + 4 + 4 + 4); + p = xdr_encode_hyper(p, args->offset); + *p++ = cpu_to_be32(args->count); + *p++ = cpu_to_be32(args->stable); + *p = cpu_to_be32(args->count); + xdr_write_pages(xdr, args->pages, args->pgbase, args->count); +} + +static void nfs3_xdr_enc_write3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + + encode_write3args(xdr, args); + xdr->buf->flags |= XDRBUF_WRITE; +} + +/* + * 3.3.8 CREATE3args + * + * enum createmode3 { + * UNCHECKED = 0, + * GUARDED = 1, + * EXCLUSIVE = 2 + * }; + * + * union createhow3 switch (createmode3 mode) { + * case UNCHECKED: + * case GUARDED: + * sattr3 obj_attributes; + * case EXCLUSIVE: + * createverf3 verf; + * }; + * + * struct CREATE3args { + * diropargs3 where; + * createhow3 how; + * }; + */ +static void encode_createhow3(struct xdr_stream *xdr, + const struct nfs3_createargs *args, + struct user_namespace *userns) +{ + encode_uint32(xdr, args->createmode); + switch (args->createmode) { + case NFS3_CREATE_UNCHECKED: + case NFS3_CREATE_GUARDED: + encode_sattr3(xdr, args->sattr, userns); + break; + case NFS3_CREATE_EXCLUSIVE: + encode_createverf3(xdr, args->verifier); + break; + default: + BUG(); + } +} + +static void nfs3_xdr_enc_create3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_createargs *args = data; + + encode_diropargs3(xdr, args->fh, args->name, args->len); + encode_createhow3(xdr, args, rpc_rqst_userns(req)); +} + +/* + * 3.3.9 MKDIR3args + * + * struct MKDIR3args { + * diropargs3 where; + * sattr3 attributes; + * }; + */ +static void nfs3_xdr_enc_mkdir3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_mkdirargs *args = data; + + encode_diropargs3(xdr, args->fh, args->name, args->len); + encode_sattr3(xdr, args->sattr, rpc_rqst_userns(req)); +} + +/* + * 3.3.10 SYMLINK3args + * + * struct symlinkdata3 { + * sattr3 symlink_attributes; + * nfspath3 symlink_data; + * }; + * + * struct SYMLINK3args { + * diropargs3 where; + * symlinkdata3 symlink; + * }; + */ +static void encode_symlinkdata3(struct xdr_stream *xdr, + const void *data, + struct user_namespace *userns) +{ + const struct nfs3_symlinkargs *args = data; + + encode_sattr3(xdr, args->sattr, userns); + encode_nfspath3(xdr, args->pages, args->pathlen); +} + +static void nfs3_xdr_enc_symlink3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_symlinkargs *args = data; + + encode_diropargs3(xdr, args->fromfh, args->fromname, args->fromlen); + encode_symlinkdata3(xdr, args, rpc_rqst_userns(req)); + xdr->buf->flags |= XDRBUF_WRITE; +} + +/* + * 3.3.11 MKNOD3args + * + * struct devicedata3 { + * sattr3 dev_attributes; + * specdata3 spec; + * }; + * + * union mknoddata3 switch (ftype3 type) { + * case NF3CHR: + * case NF3BLK: + * devicedata3 device; + * case NF3SOCK: + * case NF3FIFO: + * sattr3 pipe_attributes; + * default: + * void; + * }; + * + * struct MKNOD3args { + * diropargs3 where; + * mknoddata3 what; + * }; + */ +static void encode_devicedata3(struct xdr_stream *xdr, + const struct nfs3_mknodargs *args, + struct user_namespace *userns) +{ + encode_sattr3(xdr, args->sattr, userns); + encode_specdata3(xdr, args->rdev); +} + +static void encode_mknoddata3(struct xdr_stream *xdr, + const struct nfs3_mknodargs *args, + struct user_namespace *userns) +{ + encode_ftype3(xdr, args->type); + switch (args->type) { + case NF3CHR: + case NF3BLK: + encode_devicedata3(xdr, args, userns); + break; + case NF3SOCK: + case NF3FIFO: + encode_sattr3(xdr, args->sattr, userns); + break; + case NF3REG: + case NF3DIR: + break; + default: + BUG(); + } +} + +static void nfs3_xdr_enc_mknod3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_mknodargs *args = data; + + encode_diropargs3(xdr, args->fh, args->name, args->len); + encode_mknoddata3(xdr, args, rpc_rqst_userns(req)); +} + +/* + * 3.3.12 REMOVE3args + * + * struct REMOVE3args { + * diropargs3 object; + * }; + */ +static void nfs3_xdr_enc_remove3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_removeargs *args = data; + + encode_diropargs3(xdr, args->fh, args->name.name, args->name.len); +} + +/* + * 3.3.14 RENAME3args + * + * struct RENAME3args { + * diropargs3 from; + * diropargs3 to; + * }; + */ +static void nfs3_xdr_enc_rename3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_renameargs *args = data; + const struct qstr *old = args->old_name; + const struct qstr *new = args->new_name; + + encode_diropargs3(xdr, args->old_dir, old->name, old->len); + encode_diropargs3(xdr, args->new_dir, new->name, new->len); +} + +/* + * 3.3.15 LINK3args + * + * struct LINK3args { + * nfs_fh3 file; + * diropargs3 link; + * }; + */ +static void nfs3_xdr_enc_link3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_linkargs *args = data; + + encode_nfs_fh3(xdr, args->fromfh); + encode_diropargs3(xdr, args->tofh, args->toname, args->tolen); +} + +/* + * 3.3.16 READDIR3args + * + * struct READDIR3args { + * nfs_fh3 dir; + * cookie3 cookie; + * cookieverf3 cookieverf; + * count3 count; + * }; + */ +static void encode_readdir3args(struct xdr_stream *xdr, + const struct nfs3_readdirargs *args) +{ + __be32 *p; + + encode_nfs_fh3(xdr, args->fh); + + p = xdr_reserve_space(xdr, 8 + NFS3_COOKIEVERFSIZE + 4); + p = xdr_encode_cookie3(p, args->cookie); + p = xdr_encode_cookieverf3(p, args->verf); + *p = cpu_to_be32(args->count); +} + +static void nfs3_xdr_enc_readdir3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_readdirargs *args = data; + + encode_readdir3args(xdr, args); + rpc_prepare_reply_pages(req, args->pages, 0, args->count, + NFS3_readdirres_sz - NFS3_pagepad_sz); +} + +/* + * 3.3.17 READDIRPLUS3args + * + * struct READDIRPLUS3args { + * nfs_fh3 dir; + * cookie3 cookie; + * cookieverf3 cookieverf; + * count3 dircount; + * count3 maxcount; + * }; + */ +static void encode_readdirplus3args(struct xdr_stream *xdr, + const struct nfs3_readdirargs *args) +{ + uint32_t dircount = args->count; + uint32_t maxcount = args->count; + __be32 *p; + + encode_nfs_fh3(xdr, args->fh); + + p = xdr_reserve_space(xdr, 8 + NFS3_COOKIEVERFSIZE + 4 + 4); + p = xdr_encode_cookie3(p, args->cookie); + p = xdr_encode_cookieverf3(p, args->verf); + + /* + * readdirplus: need dircount + buffer size. + * We just make sure we make dircount big enough + */ + *p++ = cpu_to_be32(dircount); + *p = cpu_to_be32(maxcount); +} + +static void nfs3_xdr_enc_readdirplus3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_readdirargs *args = data; + + encode_readdirplus3args(xdr, args); + rpc_prepare_reply_pages(req, args->pages, 0, args->count, + NFS3_readdirres_sz - NFS3_pagepad_sz); +} + +/* + * 3.3.21 COMMIT3args + * + * struct COMMIT3args { + * nfs_fh3 file; + * offset3 offset; + * count3 count; + * }; + */ +static void encode_commit3args(struct xdr_stream *xdr, + const struct nfs_commitargs *args) +{ + __be32 *p; + + encode_nfs_fh3(xdr, args->fh); + + p = xdr_reserve_space(xdr, 8 + 4); + p = xdr_encode_hyper(p, args->offset); + *p = cpu_to_be32(args->count); +} + +static void nfs3_xdr_enc_commit3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_commitargs *args = data; + + encode_commit3args(xdr, args); +} + +#ifdef CONFIG_NFS_V3_ACL + +static void nfs3_xdr_enc_getacl3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_getaclargs *args = data; + + encode_nfs_fh3(xdr, args->fh); + encode_uint32(xdr, args->mask); + if (args->mask & (NFS_ACL | NFS_DFACL)) { + rpc_prepare_reply_pages(req, args->pages, 0, + NFSACL_MAXPAGES << PAGE_SHIFT, + ACL3_getaclres_sz - NFS3_pagepad_sz); + req->rq_rcv_buf.flags |= XDRBUF_SPARSE_PAGES; + } +} + +static void nfs3_xdr_enc_setacl3args(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs3_setaclargs *args = data; + unsigned int base; + int error; + + encode_nfs_fh3(xdr, NFS_FH(args->inode)); + encode_uint32(xdr, args->mask); + + base = req->rq_slen; + if (args->npages != 0) + xdr_write_pages(xdr, args->pages, 0, args->len); + else + xdr_reserve_space(xdr, args->len); + + error = nfsacl_encode(xdr->buf, base, args->inode, + (args->mask & NFS_ACL) ? + args->acl_access : NULL, 1, 0); + /* FIXME: this is just broken */ + BUG_ON(error < 0); + error = nfsacl_encode(xdr->buf, base + error, args->inode, + (args->mask & NFS_DFACL) ? + args->acl_default : NULL, 1, + NFS_ACL_DEFAULT); + BUG_ON(error < 0); +} + +#endif /* CONFIG_NFS_V3_ACL */ + +/* + * NFSv3 XDR decode functions + * + * NFSv3 result types are defined in section 3.3 of RFC 1813: + * "NFS Version 3 Protocol Specification". + */ + +/* + * 3.3.1 GETATTR3res + * + * struct GETATTR3resok { + * fattr3 obj_attributes; + * }; + * + * union GETATTR3res switch (nfsstat3 status) { + * case NFS3_OK: + * GETATTR3resok resok; + * default: + * void; + * }; + */ +static int nfs3_xdr_dec_getattr3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_fattr3(xdr, result, rpc_rqst_userns(req)); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.2 SETATTR3res + * + * struct SETATTR3resok { + * wcc_data obj_wcc; + * }; + * + * struct SETATTR3resfail { + * wcc_data obj_wcc; + * }; + * + * union SETATTR3res switch (nfsstat3 status) { + * case NFS3_OK: + * SETATTR3resok resok; + * default: + * SETATTR3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_setattr3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.3 LOOKUP3res + * + * struct LOOKUP3resok { + * nfs_fh3 object; + * post_op_attr obj_attributes; + * post_op_attr dir_attributes; + * }; + * + * struct LOOKUP3resfail { + * post_op_attr dir_attributes; + * }; + * + * union LOOKUP3res switch (nfsstat3 status) { + * case NFS3_OK: + * LOOKUP3resok resok; + * default: + * LOOKUP3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_lookup3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct user_namespace *userns = rpc_rqst_userns(req); + struct nfs3_diropres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_nfs_fh3(xdr, result->fh); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, userns); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->dir_attr, userns); +out: + return error; +out_default: + error = decode_post_op_attr(xdr, result->dir_attr, userns); + if (unlikely(error)) + goto out; + return nfs_stat_to_errno(status); +} + +/* + * 3.3.4 ACCESS3res + * + * struct ACCESS3resok { + * post_op_attr obj_attributes; + * uint32 access; + * }; + * + * struct ACCESS3resfail { + * post_op_attr obj_attributes; + * }; + * + * union ACCESS3res switch (nfsstat3 status) { + * case NFS3_OK: + * ACCESS3resok resok; + * default: + * ACCESS3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_access3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs3_accessres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_uint32(xdr, &result->access); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.5 READLINK3res + * + * struct READLINK3resok { + * post_op_attr symlink_attributes; + * nfspath3 data; + * }; + * + * struct READLINK3resfail { + * post_op_attr symlink_attributes; + * }; + * + * union READLINK3res switch (nfsstat3 status) { + * case NFS3_OK: + * READLINK3resok resok; + * default: + * READLINK3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_readlink3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_nfspath3(xdr); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.6 READ3res + * + * struct READ3resok { + * post_op_attr file_attributes; + * count3 count; + * bool eof; + * opaque data<>; + * }; + * + * struct READ3resfail { + * post_op_attr file_attributes; + * }; + * + * union READ3res switch (nfsstat3 status) { + * case NFS3_OK: + * READ3resok resok; + * default: + * READ3resfail resfail; + * }; + */ +static int decode_read3resok(struct xdr_stream *xdr, + struct nfs_pgio_res *result) +{ + u32 eof, count, ocount, recvd; + __be32 *p; + + p = xdr_inline_decode(xdr, 4 + 4 + 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p++); + eof = be32_to_cpup(p++); + ocount = be32_to_cpup(p++); + if (unlikely(ocount != count)) + goto out_mismatch; + recvd = xdr_read_pages(xdr, count); + if (unlikely(count > recvd)) + goto out_cheating; +out: + result->eof = eof; + result->count = count; + return count; +out_mismatch: + dprintk("NFS: READ count doesn't match length of opaque: " + "count %u != ocount %u\n", count, ocount); + return -EIO; +out_cheating: + dprintk("NFS: server cheating in read result: " + "count %u > recvd %u\n", count, recvd); + count = recvd; + eof = 0; + goto out; +} + +static int nfs3_xdr_dec_read3res(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *result = data; + unsigned int pos; + enum nfs_stat status; + int error; + + pos = xdr_stream_pos(xdr); + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + result->op_status = status; + if (status != NFS3_OK) + goto out_status; + result->replen = 3 + ((xdr_stream_pos(xdr) - pos) >> 2); + error = decode_read3resok(xdr, result); +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.7 WRITE3res + * + * enum stable_how { + * UNSTABLE = 0, + * DATA_SYNC = 1, + * FILE_SYNC = 2 + * }; + * + * struct WRITE3resok { + * wcc_data file_wcc; + * count3 count; + * stable_how committed; + * writeverf3 verf; + * }; + * + * struct WRITE3resfail { + * wcc_data file_wcc; + * }; + * + * union WRITE3res switch (nfsstat3 status) { + * case NFS3_OK: + * WRITE3resok resok; + * default: + * WRITE3resfail resfail; + * }; + */ +static int decode_write3resok(struct xdr_stream *xdr, + struct nfs_pgio_res *result) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4 + 4); + if (unlikely(!p)) + return -EIO; + result->count = be32_to_cpup(p++); + result->verf->committed = be32_to_cpup(p++); + if (unlikely(result->verf->committed > NFS_FILE_SYNC)) + goto out_badvalue; + if (decode_writeverf3(xdr, &result->verf->verifier)) + return -EIO; + return result->count; +out_badvalue: + dprintk("NFS: bad stable_how value: %u\n", result->verf->committed); + return -EIO; +} + +static int nfs3_xdr_dec_write3res(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + result->op_status = status; + if (status != NFS3_OK) + goto out_status; + error = decode_write3resok(xdr, result); +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.8 CREATE3res + * + * struct CREATE3resok { + * post_op_fh3 obj; + * post_op_attr obj_attributes; + * wcc_data dir_wcc; + * }; + * + * struct CREATE3resfail { + * wcc_data dir_wcc; + * }; + * + * union CREATE3res switch (nfsstat3 status) { + * case NFS3_OK: + * CREATE3resok resok; + * default: + * CREATE3resfail resfail; + * }; + */ +static int decode_create3resok(struct xdr_stream *xdr, + struct nfs3_diropres *result, + struct user_namespace *userns) +{ + int error; + + error = decode_post_op_fh3(xdr, result->fh); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, userns); + if (unlikely(error)) + goto out; + /* The server isn't required to return a file handle. + * If it didn't, force the client to perform a LOOKUP + * to determine the correct file handle and attribute + * values for the new object. */ + if (result->fh->size == 0) + result->fattr->valid = 0; + error = decode_wcc_data(xdr, result->dir_attr, userns); +out: + return error; +} + +static int nfs3_xdr_dec_create3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct user_namespace *userns = rpc_rqst_userns(req); + struct nfs3_diropres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_create3resok(xdr, result, userns); +out: + return error; +out_default: + error = decode_wcc_data(xdr, result->dir_attr, userns); + if (unlikely(error)) + goto out; + return nfs_stat_to_errno(status); +} + +/* + * 3.3.12 REMOVE3res + * + * struct REMOVE3resok { + * wcc_data dir_wcc; + * }; + * + * struct REMOVE3resfail { + * wcc_data dir_wcc; + * }; + * + * union REMOVE3res switch (nfsstat3 status) { + * case NFS3_OK: + * REMOVE3resok resok; + * default: + * REMOVE3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_remove3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_removeres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->dir_attr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.14 RENAME3res + * + * struct RENAME3resok { + * wcc_data fromdir_wcc; + * wcc_data todir_wcc; + * }; + * + * struct RENAME3resfail { + * wcc_data fromdir_wcc; + * wcc_data todir_wcc; + * }; + * + * union RENAME3res switch (nfsstat3 status) { + * case NFS3_OK: + * RENAME3resok resok; + * default: + * RENAME3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_rename3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct user_namespace *userns = rpc_rqst_userns(req); + struct nfs_renameres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->old_fattr, userns); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->new_fattr, userns); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.15 LINK3res + * + * struct LINK3resok { + * post_op_attr file_attributes; + * wcc_data linkdir_wcc; + * }; + * + * struct LINK3resfail { + * post_op_attr file_attributes; + * wcc_data linkdir_wcc; + * }; + * + * union LINK3res switch (nfsstat3 status) { + * case NFS3_OK: + * LINK3resok resok; + * default: + * LINK3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_link3res(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct user_namespace *userns = rpc_rqst_userns(req); + struct nfs3_linkres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, userns); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->dir_attr, userns); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/** + * nfs3_decode_dirent - Decode a single NFSv3 directory entry stored in + * the local page cache + * @xdr: XDR stream where entry resides + * @entry: buffer to fill in with entry data + * @plus: boolean indicating whether this should be a readdirplus entry + * + * Returns zero if successful, otherwise a negative errno value is + * returned. + * + * This function is not invoked during READDIR reply decoding, but + * rather whenever an application invokes the getdents(2) system call + * on a directory already in our cache. + * + * 3.3.16 entry3 + * + * struct entry3 { + * fileid3 fileid; + * filename3 name; + * cookie3 cookie; + * fhandle3 filehandle; + * post_op_attr3 attributes; + * entry3 *nextentry; + * }; + * + * 3.3.17 entryplus3 + * struct entryplus3 { + * fileid3 fileid; + * filename3 name; + * cookie3 cookie; + * post_op_attr name_attributes; + * post_op_fh3 name_handle; + * entryplus3 *nextentry; + * }; + */ +int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, + bool plus) +{ + struct user_namespace *userns = rpc_userns(entry->server->client); + __be32 *p; + int error; + u64 new_cookie; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p == xdr_zero) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p == xdr_zero) + return -EAGAIN; + entry->eof = 1; + return -EBADCOOKIE; + } + + error = decode_fileid3(xdr, &entry->ino); + if (unlikely(error)) + return -EAGAIN; + + error = decode_inline_filename3(xdr, &entry->name, &entry->len); + if (unlikely(error)) + return error == -ENAMETOOLONG ? -ENAMETOOLONG : -EAGAIN; + + error = decode_cookie3(xdr, &new_cookie); + if (unlikely(error)) + return -EAGAIN; + + entry->d_type = DT_UNKNOWN; + + if (plus) { + entry->fattr->valid = 0; + error = decode_post_op_attr(xdr, entry->fattr, userns); + if (unlikely(error)) + return -EAGAIN; + if (entry->fattr->valid & NFS_ATTR_FATTR_V3) + entry->d_type = nfs_umode_to_dtype(entry->fattr->mode); + + if (entry->fattr->fileid != entry->ino) { + entry->fattr->mounted_on_fileid = entry->ino; + entry->fattr->valid |= NFS_ATTR_FATTR_MOUNTED_ON_FILEID; + } + + /* In fact, a post_op_fh3: */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p != xdr_zero) { + error = decode_nfs_fh3(xdr, entry->fh); + if (unlikely(error)) + return -EAGAIN; + } else + zero_nfs_fh3(entry->fh); + } + + entry->cookie = new_cookie; + + return 0; +} + +/* + * 3.3.16 READDIR3res + * + * struct dirlist3 { + * entry3 *entries; + * bool eof; + * }; + * + * struct READDIR3resok { + * post_op_attr dir_attributes; + * cookieverf3 cookieverf; + * dirlist3 reply; + * }; + * + * struct READDIR3resfail { + * post_op_attr dir_attributes; + * }; + * + * union READDIR3res switch (nfsstat3 status) { + * case NFS3_OK: + * READDIR3resok resok; + * default: + * READDIR3resfail resfail; + * }; + * + * Read the directory contents into the page cache, but otherwise + * don't touch them. The actual decoding is done by nfs3_decode_entry() + * during subsequent nfs_readdir() calls. + */ +static int decode_dirlist3(struct xdr_stream *xdr) +{ + return xdr_read_pages(xdr, xdr->buf->page_len); +} + +static int decode_readdir3resok(struct xdr_stream *xdr, + struct nfs3_readdirres *result, + struct user_namespace *userns) +{ + int error; + + error = decode_post_op_attr(xdr, result->dir_attr, userns); + if (unlikely(error)) + goto out; + /* XXX: do we need to check if result->verf != NULL ? */ + error = decode_cookieverf3(xdr, result->verf); + if (unlikely(error)) + goto out; + error = decode_dirlist3(xdr); +out: + return error; +} + +static int nfs3_xdr_dec_readdir3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs3_readdirres *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_readdir3resok(xdr, result, rpc_rqst_userns(req)); +out: + return error; +out_default: + error = decode_post_op_attr(xdr, result->dir_attr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + return nfs_stat_to_errno(status); +} + +/* + * 3.3.18 FSSTAT3res + * + * struct FSSTAT3resok { + * post_op_attr obj_attributes; + * size3 tbytes; + * size3 fbytes; + * size3 abytes; + * size3 tfiles; + * size3 ffiles; + * size3 afiles; + * uint32 invarsec; + * }; + * + * struct FSSTAT3resfail { + * post_op_attr obj_attributes; + * }; + * + * union FSSTAT3res switch (nfsstat3 status) { + * case NFS3_OK: + * FSSTAT3resok resok; + * default: + * FSSTAT3resfail resfail; + * }; + */ +static int decode_fsstat3resok(struct xdr_stream *xdr, + struct nfs_fsstat *result) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 8 * 6 + 4); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_size3(p, &result->tbytes); + p = xdr_decode_size3(p, &result->fbytes); + p = xdr_decode_size3(p, &result->abytes); + p = xdr_decode_size3(p, &result->tfiles); + p = xdr_decode_size3(p, &result->ffiles); + xdr_decode_size3(p, &result->afiles); + /* ignore invarsec */ + return 0; +} + +static int nfs3_xdr_dec_fsstat3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_fsstat *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; + error = decode_fsstat3resok(xdr, result); +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.19 FSINFO3res + * + * struct FSINFO3resok { + * post_op_attr obj_attributes; + * uint32 rtmax; + * uint32 rtpref; + * uint32 rtmult; + * uint32 wtmax; + * uint32 wtpref; + * uint32 wtmult; + * uint32 dtpref; + * size3 maxfilesize; + * nfstime3 time_delta; + * uint32 properties; + * }; + * + * struct FSINFO3resfail { + * post_op_attr obj_attributes; + * }; + * + * union FSINFO3res switch (nfsstat3 status) { + * case NFS3_OK: + * FSINFO3resok resok; + * default: + * FSINFO3resfail resfail; + * }; + */ +static int decode_fsinfo3resok(struct xdr_stream *xdr, + struct nfs_fsinfo *result) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4 * 7 + 8 + 8 + 4); + if (unlikely(!p)) + return -EIO; + result->rtmax = be32_to_cpup(p++); + result->rtpref = be32_to_cpup(p++); + result->rtmult = be32_to_cpup(p++); + result->wtmax = be32_to_cpup(p++); + result->wtpref = be32_to_cpup(p++); + result->wtmult = be32_to_cpup(p++); + result->dtpref = be32_to_cpup(p++); + p = xdr_decode_size3(p, &result->maxfilesize); + xdr_decode_nfstime3(p, &result->time_delta); + + /* ignore properties */ + result->lease_time = 0; + result->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED; + result->xattr_support = 0; + return 0; +} + +static int nfs3_xdr_dec_fsinfo3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_fsinfo *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; + error = decode_fsinfo3resok(xdr, result); +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.20 PATHCONF3res + * + * struct PATHCONF3resok { + * post_op_attr obj_attributes; + * uint32 linkmax; + * uint32 name_max; + * bool no_trunc; + * bool chown_restricted; + * bool case_insensitive; + * bool case_preserving; + * }; + * + * struct PATHCONF3resfail { + * post_op_attr obj_attributes; + * }; + * + * union PATHCONF3res switch (nfsstat3 status) { + * case NFS3_OK: + * PATHCONF3resok resok; + * default: + * PATHCONF3resfail resfail; + * }; + */ +static int decode_pathconf3resok(struct xdr_stream *xdr, + struct nfs_pathconf *result) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4 * 6); + if (unlikely(!p)) + return -EIO; + result->max_link = be32_to_cpup(p++); + result->max_namelen = be32_to_cpup(p++); + p++; /* ignore no_trunc */ + p++; /* ignore chown_restricted */ + result->case_insensitive = be32_to_cpup(p++) != 0; + result->case_preserving = be32_to_cpup(p) != 0; + return 0; +} + +static int nfs3_xdr_dec_pathconf3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_pathconf *result = data; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_post_op_attr(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_status; + error = decode_pathconf3resok(xdr, result); +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +/* + * 3.3.21 COMMIT3res + * + * struct COMMIT3resok { + * wcc_data file_wcc; + * writeverf3 verf; + * }; + * + * struct COMMIT3resfail { + * wcc_data file_wcc; + * }; + * + * union COMMIT3res switch (nfsstat3 status) { + * case NFS3_OK: + * COMMIT3resok resok; + * default: + * COMMIT3resfail resfail; + * }; + */ +static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_commitres *result = data; + struct nfs_writeverf *verf = result->verf; + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + error = decode_wcc_data(xdr, result->fattr, rpc_rqst_userns(req)); + if (unlikely(error)) + goto out; + result->op_status = status; + if (status != NFS3_OK) + goto out_status; + error = decode_writeverf3(xdr, &verf->verifier); + if (!error) + verf->committed = NFS_FILE_SYNC; +out: + return error; +out_status: + return nfs_stat_to_errno(status); +} + +#ifdef CONFIG_NFS_V3_ACL + +static inline int decode_getacl3resok(struct xdr_stream *xdr, + struct nfs3_getaclres *result, + struct user_namespace *userns) +{ + struct posix_acl **acl; + unsigned int *aclcnt; + size_t hdrlen; + int error; + + error = decode_post_op_attr(xdr, result->fattr, userns); + if (unlikely(error)) + goto out; + error = decode_uint32(xdr, &result->mask); + if (unlikely(error)) + goto out; + error = -EINVAL; + if (result->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT)) + goto out; + + hdrlen = xdr_stream_pos(xdr); + + acl = NULL; + if (result->mask & NFS_ACL) + acl = &result->acl_access; + aclcnt = NULL; + if (result->mask & NFS_ACLCNT) + aclcnt = &result->acl_access_count; + error = nfsacl_decode(xdr->buf, hdrlen, aclcnt, acl); + if (unlikely(error <= 0)) + goto out; + + acl = NULL; + if (result->mask & NFS_DFACL) + acl = &result->acl_default; + aclcnt = NULL; + if (result->mask & NFS_DFACLCNT) + aclcnt = &result->acl_default_count; + error = nfsacl_decode(xdr->buf, hdrlen + error, aclcnt, acl); + if (unlikely(error <= 0)) + return error; + error = 0; +out: + return error; +} + +static int nfs3_xdr_dec_getacl3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_getacl3resok(xdr, result, rpc_rqst_userns(req)); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +static int nfs3_xdr_dec_setacl3res(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *result) +{ + enum nfs_stat status; + int error; + + error = decode_nfsstat3(xdr, &status); + if (unlikely(error)) + goto out; + if (status != NFS3_OK) + goto out_default; + error = decode_post_op_attr(xdr, result, rpc_rqst_userns(req)); +out: + return error; +out_default: + return nfs_stat_to_errno(status); +} + +#endif /* CONFIG_NFS_V3_ACL */ + +#define PROC(proc, argtype, restype, timer) \ +[NFS3PROC_##proc] = { \ + .p_proc = NFS3PROC_##proc, \ + .p_encode = nfs3_xdr_enc_##argtype##3args, \ + .p_decode = nfs3_xdr_dec_##restype##3res, \ + .p_arglen = NFS3_##argtype##args_sz, \ + .p_replen = NFS3_##restype##res_sz, \ + .p_timer = timer, \ + .p_statidx = NFS3PROC_##proc, \ + .p_name = #proc, \ + } + +const struct rpc_procinfo nfs3_procedures[] = { + PROC(GETATTR, getattr, getattr, 1), + PROC(SETATTR, setattr, setattr, 0), + PROC(LOOKUP, lookup, lookup, 2), + PROC(ACCESS, access, access, 1), + PROC(READLINK, readlink, readlink, 3), + PROC(READ, read, read, 3), + PROC(WRITE, write, write, 4), + PROC(CREATE, create, create, 0), + PROC(MKDIR, mkdir, create, 0), + PROC(SYMLINK, symlink, create, 0), + PROC(MKNOD, mknod, create, 0), + PROC(REMOVE, remove, remove, 0), + PROC(RMDIR, lookup, setattr, 0), + PROC(RENAME, rename, rename, 0), + PROC(LINK, link, link, 0), + PROC(READDIR, readdir, readdir, 3), + PROC(READDIRPLUS, readdirplus, readdir, 3), + PROC(FSSTAT, getattr, fsstat, 0), + PROC(FSINFO, getattr, fsinfo, 0), + PROC(PATHCONF, getattr, pathconf, 0), + PROC(COMMIT, commit, commit, 5), +}; + +static unsigned int nfs_version3_counts[ARRAY_SIZE(nfs3_procedures)]; +const struct rpc_version nfs_version3 = { + .number = 3, + .nrprocs = ARRAY_SIZE(nfs3_procedures), + .procs = nfs3_procedures, + .counts = nfs_version3_counts, +}; + +#ifdef CONFIG_NFS_V3_ACL +static const struct rpc_procinfo nfs3_acl_procedures[] = { + [ACLPROC3_GETACL] = { + .p_proc = ACLPROC3_GETACL, + .p_encode = nfs3_xdr_enc_getacl3args, + .p_decode = nfs3_xdr_dec_getacl3res, + .p_arglen = ACL3_getaclargs_sz, + .p_replen = ACL3_getaclres_sz, + .p_timer = 1, + .p_name = "GETACL", + }, + [ACLPROC3_SETACL] = { + .p_proc = ACLPROC3_SETACL, + .p_encode = nfs3_xdr_enc_setacl3args, + .p_decode = nfs3_xdr_dec_setacl3res, + .p_arglen = ACL3_setaclargs_sz, + .p_replen = ACL3_setaclres_sz, + .p_timer = 0, + .p_name = "SETACL", + }, +}; + +static unsigned int nfs3_acl_counts[ARRAY_SIZE(nfs3_acl_procedures)]; +const struct rpc_version nfsacl_version3 = { + .number = 3, + .nrprocs = ARRAY_SIZE(nfs3_acl_procedures), + .procs = nfs3_acl_procedures, + .counts = nfs3_acl_counts, +}; +#endif /* CONFIG_NFS_V3_ACL */ diff --git a/fs/nfs/nfs40.h b/fs/nfs/nfs40.h new file mode 100644 index 000000000..ee09aac73 --- /dev/null +++ b/fs/nfs/nfs40.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __LINUX_FS_NFS_NFS4_0_H +#define __LINUX_FS_NFS_NFS4_0_H + + +/* nfs40client.c */ +void nfs40_shutdown_client(struct nfs_client *); +int nfs40_init_client(struct nfs_client *); +void nfs40_handle_cb_pathdown(struct nfs_client *clp); + +/* nfs40proc.c */ +extern const struct nfs4_minor_version_ops nfs_v4_0_minor_ops; + +/* nfs40state.c */ +int nfs40_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result, + const struct cred *cred); + +#endif /* __LINUX_FS_NFS_NFS4_0_H */ diff --git a/fs/nfs/nfs40client.c b/fs/nfs/nfs40client.c new file mode 100644 index 000000000..f44d68fae --- /dev/null +++ b/fs/nfs/nfs40client.c @@ -0,0 +1,245 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include +#include "nfs4_fs.h" +#include "nfs4session.h" +#include "callback.h" +#include "delegation.h" +#include "internal.h" +#include "netns.h" +#include "nfs40.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +/* + * SETCLIENTID just did a callback update with the callback ident in + * "drop," but server trunking discovery claims "drop" and "keep" are + * actually the same server. Swap the callback IDs so that "keep" + * will continue to use the callback ident the server now knows about, + * and so that "keep"'s original callback ident is destroyed when + * "drop" is freed. + */ +static void nfs4_swap_callback_idents(struct nfs_client *keep, + struct nfs_client *drop) +{ + struct nfs_net *nn = net_generic(keep->cl_net, nfs_net_id); + unsigned int save = keep->cl_cb_ident; + + if (keep->cl_cb_ident == drop->cl_cb_ident) + return; + + dprintk("%s: keeping callback ident %u and dropping ident %u\n", + __func__, keep->cl_cb_ident, drop->cl_cb_ident); + + spin_lock(&nn->nfs_client_lock); + + idr_replace(&nn->cb_ident_idr, keep, drop->cl_cb_ident); + keep->cl_cb_ident = drop->cl_cb_ident; + + idr_replace(&nn->cb_ident_idr, drop, save); + drop->cl_cb_ident = save; + + spin_unlock(&nn->nfs_client_lock); +} + +static bool nfs4_same_verifier(nfs4_verifier *v1, nfs4_verifier *v2) +{ + return memcmp(v1->data, v2->data, sizeof(v1->data)) == 0; +} + +void nfs40_shutdown_client(struct nfs_client *clp) +{ + if (clp->cl_slot_tbl) { + nfs4_shutdown_slot_table(clp->cl_slot_tbl); + kfree(clp->cl_slot_tbl); + } +} + +/** + * nfs40_init_client - nfs_client initialization tasks for NFSv4.0 + * @clp: nfs_client to initialize + * + * Returns zero on success, or a negative errno if some error occurred. + */ +int nfs40_init_client(struct nfs_client *clp) +{ + struct nfs4_slot_table *tbl; + int ret; + + tbl = kzalloc_obj(*tbl, GFP_NOFS); + if (tbl == NULL) + return -ENOMEM; + + ret = nfs4_setup_slot_table(tbl, NFS4_MAX_SLOT_TABLE, + "NFSv4.0 transport Slot table"); + if (ret) { + nfs4_shutdown_slot_table(tbl); + kfree(tbl); + return ret; + } + + clp->cl_slot_tbl = tbl; + return 0; +} + +/* + * nfs40_handle_cb_pathdown - return all delegations after NFS4ERR_CB_PATH_DOWN + * @clp: client to process + * + * Set the NFS4CLNT_LEASE_EXPIRED state in order to force a + * resend of the SETCLIENTID and hence re-establish the + * callback channel. Then return all existing delegations. + */ +void nfs40_handle_cb_pathdown(struct nfs_client *clp) +{ + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + nfs_expire_all_delegations(clp); + dprintk("%s: handling CB_PATHDOWN recovery for server %s\n", __func__, + clp->cl_hostname); +} + +void nfs4_schedule_path_down_recovery(struct nfs_client *clp) +{ + nfs40_handle_cb_pathdown(clp); + nfs4_schedule_state_manager(clp); +} + +/** + * nfs40_walk_client_list - Find server that recognizes a client ID + * + * @new: nfs_client with client ID to test + * @result: OUT: found nfs_client, or new + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in "result." + * + * NB: nfs40_walk_client_list() relies on the new nfs_client being + * the last nfs_client on the list. + */ +static int nfs40_walk_client_list(struct nfs_client *new, + struct nfs_client **result, + const struct cred *cred) +{ + struct nfs_net *nn = net_generic(new->cl_net, nfs_net_id); + struct nfs_client *pos, *prev = NULL; + struct nfs4_setclientid_res clid = { + .clientid = new->cl_clientid, + .confirm = new->cl_confirm, + }; + int status = -NFS4ERR_STALE_CLIENTID; + + spin_lock(&nn->nfs_client_lock); + list_for_each_entry(pos, &nn->nfs_client_list, cl_share_link) { + + if (pos == new) + goto found; + + status = nfs4_match_client(pos, new, &prev, nn); + if (status < 0) + goto out_unlock; + if (status != 0) + continue; + /* + * We just sent a new SETCLIENTID, which should have + * caused the server to return a new cl_confirm. So if + * cl_confirm is the same, then this is a different + * server that just returned the same cl_confirm by + * coincidence: + */ + if ((new != pos) && nfs4_same_verifier(&pos->cl_confirm, + &new->cl_confirm)) + continue; + /* + * But if the cl_confirm's are different, then the only + * way that a SETCLIENTID_CONFIRM to pos can succeed is + * if new and pos point to the same server: + */ +found: + refcount_inc(&pos->cl_count); + spin_unlock(&nn->nfs_client_lock); + + nfs_put_client(prev); + prev = pos; + + status = nfs4_proc_setclientid_confirm(pos, &clid, cred); + switch (status) { + case -NFS4ERR_STALE_CLIENTID: + break; + case 0: + nfs4_swap_callback_idents(pos, new); + pos->cl_confirm = new->cl_confirm; + nfs_mark_client_ready(pos, NFS_CS_READY); + + prev = NULL; + *result = pos; + goto out; + case -ERESTARTSYS: + case -ETIMEDOUT: + /* The callback path may have been inadvertently + * changed. Schedule recovery! + */ + nfs4_schedule_path_down_recovery(pos); + goto out; + default: + goto out; + } + + spin_lock(&nn->nfs_client_lock); + } +out_unlock: + spin_unlock(&nn->nfs_client_lock); + + /* No match found. The server lost our clientid */ +out: + nfs_put_client(prev); + return status; +} + +/** + * nfs40_discover_server_trunking - Detect server IP address trunking (mv0) + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in + * "result". + * + * Note: The returned client may not yet be marked ready. + */ +int nfs40_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result, + const struct cred *cred) +{ + struct nfs4_setclientid_res clid = { + .clientid = clp->cl_clientid, + .confirm = clp->cl_confirm, + }; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + unsigned short port; + int status; + + port = nn->nfs_callback_tcpport; + if (clp->cl_addr.ss_family == AF_INET6) + port = nn->nfs_callback_tcpport6; + + status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid); + if (status != 0) + goto out; + clp->cl_clientid = clid.clientid; + clp->cl_confirm = clid.confirm; + + status = nfs40_walk_client_list(clp, result, cred); + if (status == 0) { + /* Sustain the lease, even if it's empty. If the clientid4 + * goes stale it's of no use for trunking discovery. */ + nfs4_schedule_state_renewal(*result); + + /* If the client state need to recover, do it. */ + if (clp->cl_state) + nfs4_schedule_state_manager(clp); + } +out: + return status; +} diff --git a/fs/nfs/nfs40proc.c b/fs/nfs/nfs40proc.c new file mode 100644 index 000000000..b07f2d6cc --- /dev/null +++ b/fs/nfs/nfs40proc.c @@ -0,0 +1,395 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include +#include +#include +#include +#include "internal.h" +#include "nfs4_fs.h" +#include "nfs40.h" +#include "nfs4session.h" +#include "nfs4trace.h" + +static void nfs40_call_sync_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + nfs4_setup_sequence(data->seq_server->nfs_client, + data->seq_args, data->seq_res, task); +} + +static void nfs40_call_sync_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + nfs4_sequence_done(task, data->seq_res); +} + +static void nfs40_sequence_free_slot(struct nfs4_sequence_res *res) +{ + struct nfs4_slot *slot = res->sr_slot; + struct nfs4_slot_table *tbl; + + tbl = slot->table; + spin_lock(&tbl->slot_tbl_lock); + if (!nfs41_wake_and_assign_slot(tbl, slot)) + nfs4_free_slot(tbl, slot); + spin_unlock(&tbl->slot_tbl_lock); + + res->sr_slot = NULL; +} + +static int nfs40_sequence_done(struct rpc_task *task, + struct nfs4_sequence_res *res) +{ + if (res->sr_slot != NULL) + nfs40_sequence_free_slot(res); + return 1; +} + +static void nfs40_clear_delegation_stateid(struct nfs4_state *state) +{ + if (rcu_access_pointer(NFS_I(state->inode)->delegation) != NULL) + nfs_finish_clear_delegation_stateid(state, NULL); +} + +static int nfs40_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state) +{ + /* NFSv4.0 doesn't allow for delegation recovery on open expire */ + nfs40_clear_delegation_stateid(state); + nfs_state_clear_open_state_flags(state); + return nfs4_open_expired(sp, state); +} + +struct nfs4_renewdata { + struct nfs_client *client; + unsigned long timestamp; +}; + +/* + * nfs4_proc_async_renew(): This is not one of the nfs_rpc_ops; it is a special + * standalone procedure for queueing an asynchronous RENEW. + */ +static void nfs4_renew_release(void *calldata) +{ + struct nfs4_renewdata *data = calldata; + struct nfs_client *clp = data->client; + + if (refcount_read(&clp->cl_count) > 1) + nfs4_schedule_state_renewal(clp); + nfs_put_client(clp); + kfree(data); +} + +static void nfs4_renew_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_renewdata *data = calldata; + struct nfs_client *clp = data->client; + unsigned long timestamp = data->timestamp; + + trace_nfs4_renew_async(clp, task->tk_status); + switch (task->tk_status) { + case 0: + break; + case -NFS4ERR_LEASE_MOVED: + nfs4_schedule_lease_moved_recovery(clp); + break; + default: + /* Unless we're shutting down, schedule state recovery! */ + if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) == 0) + return; + if (task->tk_status != NFS4ERR_CB_PATH_DOWN) { + nfs4_schedule_lease_recovery(clp); + return; + } + nfs4_schedule_path_down_recovery(clp); + } + do_renew_lease(clp, timestamp); +} + +static const struct rpc_call_ops nfs4_renew_ops = { + .rpc_call_done = nfs4_renew_done, + .rpc_release = nfs4_renew_release, +}; + +static int nfs4_proc_async_renew(struct nfs_client *clp, const struct cred *cred, unsigned renew_flags) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], + .rpc_argp = clp, + .rpc_cred = cred, + }; + struct nfs4_renewdata *data; + + if (renew_flags == 0) + return 0; + if (!refcount_inc_not_zero(&clp->cl_count)) + return -EIO; + data = kmalloc_obj(*data, GFP_NOFS); + if (data == NULL) { + nfs_put_client(clp); + return -ENOMEM; + } + data->client = clp; + data->timestamp = jiffies; + return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT, + &nfs4_renew_ops, data); +} + +static int nfs4_proc_renew(struct nfs_client *clp, const struct cred *cred) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], + .rpc_argp = clp, + .rpc_cred = cred, + }; + unsigned long now = jiffies; + int status; + + status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + if (status < 0) + return status; + do_renew_lease(clp, now); + return 0; +} + +static int nfs40_test_and_free_expired_stateid(struct nfs_server *server, + nfs4_stateid *stateid, + const struct cred *cred) +{ + return -NFS4ERR_BAD_STATEID; +} + +/* + * This operation also signals the server that this client is + * performing migration recovery. The server can stop returning + * NFS4ERR_LEASE_MOVED to this client. A RENEW operation is + * appended to this compound to identify the client ID which is + * performing recovery. + */ +static int _nfs40_proc_get_locations(struct nfs_server *server, + struct nfs_fh *fhandle, + struct nfs4_fs_locations *locations, + struct page *page, const struct cred *cred) +{ + struct rpc_clnt *clnt = server->client; + struct nfs_client *clp = server->nfs_client; + u32 bitmask[2] = { + [0] = FATTR4_WORD0_FSID | FATTR4_WORD0_FS_LOCATIONS, + }; + struct nfs4_fs_locations_arg args = { + .clientid = clp->cl_clientid, + .fh = fhandle, + .page = page, + .bitmask = bitmask, + .migration = 1, /* skip LOOKUP */ + .renew = 1, /* append RENEW */ + }; + struct nfs4_fs_locations_res res = { + .fs_locations = locations, + .migration = 1, + .renew = 1, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + unsigned long now = jiffies; + int status; + + nfs_fattr_init(locations->fattr); + locations->server = server; + locations->nlocations = 0; + + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 1); + status = nfs4_call_sync_sequence(clnt, server, &msg, + &args.seq_args, &res.seq_res); + if (status) + return status; + + renew_lease(server, now); + return 0; +} + +/* + * This operation also signals the server that this client is + * performing "lease moved" recovery. The server can stop + * returning NFS4ERR_LEASE_MOVED to this client. A RENEW operation + * is appended to this compound to identify the client ID which is + * performing recovery. + */ +static int _nfs40_proc_fsid_present(struct inode *inode, const struct cred *cred) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + struct rpc_clnt *clnt = server->client; + struct nfs4_fsid_present_arg args = { + .fh = NFS_FH(inode), + .clientid = clp->cl_clientid, + .renew = 1, /* append RENEW */ + }; + struct nfs4_fsid_present_res res = { + .renew = 1, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSID_PRESENT], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + unsigned long now = jiffies; + int status; + + res.fh = nfs_alloc_fhandle(); + if (res.fh == NULL) + return -ENOMEM; + + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 1); + status = nfs4_call_sync_sequence(clnt, server, &msg, + &args.seq_args, &res.seq_res); + nfs_free_fhandle(res.fh); + if (status) + return status; + + do_renew_lease(clp, now); + return 0; +} + +struct nfs_release_lockowner_data { + struct nfs4_lock_state *lsp; + struct nfs_server *server; + struct nfs_release_lockowner_args args; + struct nfs_release_lockowner_res res; + unsigned long timestamp; +}; + +static void nfs4_release_lockowner_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_release_lockowner_data *data = calldata; + struct nfs_server *server = data->server; + nfs4_setup_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, task); + data->args.lock_owner.clientid = server->nfs_client->cl_clientid; + data->timestamp = jiffies; +} + +static void nfs4_release_lockowner_done(struct rpc_task *task, void *calldata) +{ + struct nfs_release_lockowner_data *data = calldata; + struct nfs_server *server = data->server; + + nfs40_sequence_done(task, &data->res.seq_res); + + switch (task->tk_status) { + case 0: + renew_lease(server, data->timestamp); + break; + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_EXPIRED: + nfs4_schedule_lease_recovery(server->nfs_client); + break; + case -NFS4ERR_LEASE_MOVED: + case -NFS4ERR_DELAY: + if (nfs4_async_handle_error(task, server, + NULL, NULL) == -EAGAIN) + rpc_restart_call_prepare(task); + } +} + +static void nfs4_release_lockowner_release(void *calldata) +{ + struct nfs_release_lockowner_data *data = calldata; + nfs4_free_lock_state(data->server, data->lsp); + kfree(calldata); +} + +static const struct rpc_call_ops nfs4_release_lockowner_ops = { + .rpc_call_prepare = nfs4_release_lockowner_prepare, + .rpc_call_done = nfs4_release_lockowner_done, + .rpc_release = nfs4_release_lockowner_release, +}; + +static void +nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp) +{ + struct nfs_release_lockowner_data *data; + struct nfs_client *clp = server->nfs_client; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RELEASE_LOCKOWNER], + }; + + if (clp->cl_mvops->minor_version != 0) + return; + + data = kmalloc_obj(*data); + if (!data) + return; + data->lsp = lsp; + data->server = server; + data->args.lock_owner.clientid = clp->cl_clientid; + data->args.lock_owner.id = lsp->ls_seqid.owner_id; + data->args.lock_owner.s_dev = server->s_dev; + + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + nfs4_init_sequence(clp, &data->args.seq_args, &data->res.seq_res, 0, 0); + rpc_call_async(server->client, &msg, 0, &nfs4_release_lockowner_ops, data); +} + +static const struct rpc_call_ops nfs40_call_sync_ops = { + .rpc_call_prepare = nfs40_call_sync_prepare, + .rpc_call_done = nfs40_call_sync_done, +}; + +static const struct nfs4_sequence_slot_ops nfs40_sequence_slot_ops = { + .process = nfs40_sequence_done, + .done = nfs40_sequence_done, + .free_slot = nfs40_sequence_free_slot, +}; + +static const struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT, + .state_flag_bit = NFS_STATE_RECLAIM_REBOOT, + .recover_open = nfs4_open_reclaim, + .recover_lock = nfs4_lock_reclaim, + .establish_clid = nfs4_init_clientid, + .detect_trunking = nfs40_discover_server_trunking, +}; + +static const struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE, + .state_flag_bit = NFS_STATE_RECLAIM_NOGRACE, + .recover_open = nfs40_open_expired, + .recover_lock = nfs4_lock_expired, + .establish_clid = nfs4_init_clientid, +}; + +static const struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = { + .sched_state_renewal = nfs4_proc_async_renew, + .get_state_renewal_cred = nfs4_get_renew_cred, + .renew_lease = nfs4_proc_renew, +}; + +static const struct nfs4_mig_recovery_ops nfs40_mig_recovery_ops = { + .get_locations = _nfs40_proc_get_locations, + .fsid_present = _nfs40_proc_fsid_present, +}; + +const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = { + .minor_version = 0, + .init_caps = NFS_CAP_READDIRPLUS + | NFS_CAP_ATOMIC_OPEN + | NFS_CAP_POSIX_LOCK, + .init_client = nfs40_init_client, + .shutdown_client = nfs40_shutdown_client, + .match_stateid = nfs4_match_stateid, + .find_root_sec = nfs4_find_root_sec, + .free_lock_state = nfs4_release_lockowner, + .test_and_free_expired = nfs40_test_and_free_expired_stateid, + .alloc_seqid = nfs_alloc_seqid, + .call_sync_ops = &nfs40_call_sync_ops, + .sequence_slot_ops = &nfs40_sequence_slot_ops, + .reboot_recovery_ops = &nfs40_reboot_recovery_ops, + .nograce_recovery_ops = &nfs40_nograce_recovery_ops, + .state_renewal_ops = &nfs40_state_renewal_ops, + .mig_recovery_ops = &nfs40_mig_recovery_ops, +}; diff --git a/fs/nfs/nfs42.h b/fs/nfs/nfs42.h new file mode 100644 index 000000000..aafd15a4a --- /dev/null +++ b/fs/nfs/nfs42.h @@ -0,0 +1,69 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2014 Anna Schumaker + */ + +#ifndef __LINUX_FS_NFS_NFS4_2_H +#define __LINUX_FS_NFS_NFS4_2_H + +#include + +/* + * FIXME: four LAYOUTSTATS calls per compound at most! Do we need to support + * more? Need to consider not to pre-alloc too much for a compound. + */ +#define PNFS_LAYOUTSTATS_MAXDEV (4) +#define READ_PLUS_SCRATCH_SIZE (16) + +/* nfs4.2proc.c */ +#ifdef CONFIG_NFS_V4_2 +int nfs42_proc_allocate(struct file *, loff_t, loff_t); +ssize_t nfs42_proc_copy(struct file *, loff_t, struct file *, loff_t, size_t, + struct nl4_server *, nfs4_stateid *, bool); +int nfs42_proc_deallocate(struct file *, loff_t, loff_t); +int nfs42_proc_zero_range(struct file *, loff_t, loff_t); +loff_t nfs42_proc_llseek(struct file *, loff_t, int); +int nfs42_proc_layoutstats_generic(struct nfs_server *, + struct nfs42_layoutstat_data *); +int nfs42_proc_clone(struct file *, struct file *, loff_t, loff_t, loff_t); +int nfs42_proc_layouterror(struct pnfs_layout_segment *lseg, + const struct nfs42_layout_error *errors, + size_t n); +int nfs42_proc_copy_notify(struct file *, struct file *, + struct nfs42_copy_notify_res *); +static inline bool nfs42_files_from_same_server(struct file *in, + struct file *out) +{ + struct nfs_client *c_in = (NFS_SERVER(file_inode(in)))->nfs_client; + struct nfs_client *c_out = (NFS_SERVER(file_inode(out)))->nfs_client; + + return nfs4_check_serverowner_major_id(c_in->cl_serverowner, + c_out->cl_serverowner); +} + +ssize_t nfs42_proc_getxattr(struct inode *inode, const char *name, + void *buf, size_t buflen); +int nfs42_proc_setxattr(struct inode *inode, const char *name, + const void *buf, size_t buflen, int flags); +ssize_t nfs42_proc_listxattrs(struct inode *inode, void *buf, + size_t buflen, u64 *cookiep, bool *eofp); +int nfs42_proc_removexattr(struct inode *inode, const char *name); + +/* + * Maximum XDR buffer size needed for a listxattr buffer of buflen size. + * + * The upper boundary is a buffer with all 1-byte sized attribute names. + * They would be 7 bytes long in the eventual buffer ("user.x\0"), and + * 8 bytes long XDR-encoded. + * + * Include the trailing eof word as well and make the result a multiple + * of 4 bytes. + */ +static inline u32 nfs42_listxattr_xdrsize(u32 buflen) +{ + u32 size = 8 * buflen / (XATTR_USER_PREFIX_LEN + 2) + 4; + + return (size + 3) & ~3; +} +#endif /* CONFIG_NFS_V4_2 */ +#endif /* __LINUX_FS_NFS_NFS4_2_H */ diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c new file mode 100644 index 000000000..ab86246fc --- /dev/null +++ b/fs/nfs/nfs42proc.c @@ -0,0 +1,1687 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2014 Anna Schumaker + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include "nfs4_fs.h" +#include "nfs42.h" +#include "iostat.h" +#include "pnfs.h" +#include "nfs4session.h" +#include "internal.h" +#include "delegation.h" +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PROC +static int nfs42_do_offload_cancel_async(struct file *dst, nfs4_stateid *std); +static int nfs42_proc_offload_status(struct file *file, nfs4_stateid *stateid, + u64 *copied); + +static void nfs42_set_netaddr(struct file *filep, struct nfs42_netaddr *naddr) +{ + struct nfs_client *clp = (NFS_SERVER(file_inode(filep)))->nfs_client; + unsigned short port = 2049; + + rcu_read_lock(); + naddr->netid_len = scnprintf(naddr->netid, + sizeof(naddr->netid), "%s", + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_NETID)); + naddr->addr_len = scnprintf(naddr->addr, + sizeof(naddr->addr), + "%s.%u.%u", + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR), + port >> 8, port & 255); + rcu_read_unlock(); +} + +static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep, + struct nfs_lock_context *lock, loff_t offset, loff_t len) +{ + struct inode *inode = file_inode(filep); + struct nfs_server *server = NFS_SERVER(inode); + u32 bitmask[NFS_BITMASK_SZ]; + struct nfs42_falloc_args args = { + .falloc_fh = NFS_FH(inode), + .falloc_offset = offset, + .falloc_length = len, + .falloc_bitmask = bitmask, + }; + struct nfs42_falloc_res res = { + .falloc_server = server, + }; + int status; + + msg->rpc_argp = &args; + msg->rpc_resp = &res; + + status = nfs4_set_rw_stateid(&args.falloc_stateid, lock->open_context, + lock, FMODE_WRITE); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + + nfs4_bitmask_set(bitmask, server->cache_consistency_bitmask, inode, + NFS_INO_INVALID_BLOCKS); + + res.falloc_fattr = nfs_alloc_fattr(); + if (!res.falloc_fattr) + return -ENOMEM; + + status = nfs4_call_sync(server->client, server, msg, + &args.seq_args, &res.seq_res, 0); + if (status == 0) { + loff_t newsize = offset + len; + + spin_lock(&inode->i_lock); + if (newsize > i_size_read(inode)) + i_size_write(inode, newsize); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + if (nfs_should_remove_suid(inode)) + nfs_set_cache_invalid(inode, + NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_MODE); + spin_unlock(&inode->i_lock); + status = nfs_post_op_update_inode_force_wcc(inode, + res.falloc_fattr); + } + if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE]) + trace_nfs4_fallocate(inode, &args, status); + else + trace_nfs4_deallocate(inode, &args, status); + kfree(res.falloc_fattr); + return status; +} + +static int nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep, + loff_t offset, loff_t len) +{ + struct inode *inode = file_inode(filep); + struct nfs_server *server = NFS_SERVER(inode); + struct nfs4_exception exception = { }; + struct nfs_lock_context *lock; + int err; + + lock = nfs_get_lock_context(nfs_file_open_context(filep)); + if (IS_ERR(lock)) + return PTR_ERR(lock); + + exception.inode = inode; + exception.state = lock->open_context->state; + + err = nfs_sync_inode(inode); + if (err) + goto out; + + do { + err = _nfs42_proc_fallocate(msg, filep, lock, offset, len); + if (err == -ENOTSUPP) { + err = -EOPNOTSUPP; + break; + } + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); +out: + nfs_put_lock_context(lock); + return err; +} + +int nfs42_proc_allocate(struct file *filep, loff_t offset, loff_t len) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE], + }; + struct inode *inode = file_inode(filep); + loff_t oldsize; + int err; + + if (!nfs_server_capable(inode, NFS_CAP_ALLOCATE)) + return -EOPNOTSUPP; + + err = nfs_start_io_write(inode); + if (err) + return err; + + oldsize = i_size_read(inode); + + err = nfs42_proc_fallocate(&msg, filep, offset, len); + + if (err == 0) + nfs_truncate_last_folio(inode->i_mapping, oldsize, + offset + len); + else if (err == -EOPNOTSUPP) + NFS_SERVER(inode)->caps &= ~(NFS_CAP_ALLOCATE | + NFS_CAP_ZERO_RANGE); + + nfs_end_io_write(inode); + return err; +} + +int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DEALLOCATE], + }; + struct inode *inode = file_inode(filep); + int err; + + if (!nfs_server_capable(inode, NFS_CAP_DEALLOCATE)) + return -EOPNOTSUPP; + + err = nfs_start_io_write(inode); + if (err) + return err; + + err = nfs42_proc_fallocate(&msg, filep, offset, len); + if (err == 0) + truncate_pagecache_range(inode, offset, (offset + len) -1); + if (err == -EOPNOTSUPP) + NFS_SERVER(inode)->caps &= ~(NFS_CAP_DEALLOCATE | + NFS_CAP_ZERO_RANGE); + + nfs_end_io_write(inode); + return err; +} + +int nfs42_proc_zero_range(struct file *filep, loff_t offset, loff_t len) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ZERO_RANGE], + }; + struct inode *inode = file_inode(filep); + loff_t oldsize; + int err; + + if (!nfs_server_capable(inode, NFS_CAP_ZERO_RANGE)) + return -EOPNOTSUPP; + + err = nfs_start_io_write(inode); + if (err) + return err; + + oldsize = i_size_read(inode); + err = nfs42_proc_fallocate(&msg, filep, offset, len); + if (err == 0) { + nfs_truncate_last_folio(inode->i_mapping, oldsize, + offset + len); + truncate_pagecache_range(inode, offset, (offset + len) -1); + } else if (err == -EOPNOTSUPP) + NFS_SERVER(inode)->caps &= ~NFS_CAP_ZERO_RANGE; + + nfs_end_io_write(inode); + return err; +} + +static void nfs4_copy_dequeue_callback(struct nfs_server *dst_server, + struct nfs_server *src_server, + struct nfs4_copy_state *copy) +{ + spin_lock(&dst_server->nfs_client->cl_lock); + list_del_init(©->copies); + spin_unlock(&dst_server->nfs_client->cl_lock); + if (dst_server != src_server) { + spin_lock(&src_server->nfs_client->cl_lock); + list_del_init(©->src_copies); + spin_unlock(&src_server->nfs_client->cl_lock); + } +} + +static int handle_async_copy(struct nfs42_copy_res *res, + struct nfs_server *dst_server, + struct nfs_server *src_server, + struct file *src, + struct file *dst, + nfs4_stateid *src_stateid, + bool *restart) +{ + struct nfs4_copy_state *copy, *tmp_copy = NULL, *iter; + struct nfs_open_context *dst_ctx = nfs_file_open_context(dst); + struct nfs_open_context *src_ctx = nfs_file_open_context(src); + struct nfs_client *clp = dst_server->nfs_client; + unsigned long timeout = 3 * HZ; + int status = NFS4_OK; + u64 copied; + + copy = kzalloc_obj(struct nfs4_copy_state); + if (!copy) + return -ENOMEM; + + spin_lock(&dst_server->nfs_client->cl_lock); + list_for_each_entry(iter, + &dst_server->nfs_client->pending_cb_stateids, + copies) { + if (memcmp(&res->write_res.stateid, &iter->stateid, + NFS4_STATEID_SIZE)) + continue; + tmp_copy = iter; + list_del(&iter->copies); + break; + } + if (tmp_copy) { + spin_unlock(&dst_server->nfs_client->cl_lock); + kfree(copy); + copy = tmp_copy; + goto out; + } + + memcpy(©->stateid, &res->write_res.stateid, NFS4_STATEID_SIZE); + init_completion(©->completion); + copy->parent_dst_state = dst_ctx->state; + copy->parent_src_state = src_ctx->state; + + list_add_tail(©->copies, &dst_server->ss_copies); + spin_unlock(&dst_server->nfs_client->cl_lock); + + if (dst_server != src_server) { + spin_lock(&src_server->nfs_client->cl_lock); + list_add_tail(©->src_copies, &src_server->ss_src_copies); + spin_unlock(&src_server->nfs_client->cl_lock); + } + +wait: + status = wait_for_completion_interruptible_timeout(©->completion, + timeout); + if (!status) + goto timeout; + nfs4_copy_dequeue_callback(dst_server, src_server, copy); + if (status == -ERESTARTSYS) { + goto out_cancel; + } else if (copy->flags || copy->error == NFS4ERR_PARTNER_NO_AUTH) { + status = -EAGAIN; + *restart = true; + goto out_cancel; + } +out: + res->write_res.count = copy->count; + /* Copy out the updated write verifier provided by CB_OFFLOAD. */ + memcpy(&res->write_res.verifier, ©->verf, sizeof(copy->verf)); + status = -copy->error; + +out_free: + kfree(copy); + return status; +out_cancel: + nfs42_do_offload_cancel_async(dst, ©->stateid); + if (!nfs42_files_from_same_server(src, dst)) + nfs42_do_offload_cancel_async(src, src_stateid); + goto out_free; +timeout: + timeout <<= 1; + if (timeout > (clp->cl_lease_time >> 1)) + timeout = clp->cl_lease_time >> 1; + status = nfs42_proc_offload_status(dst, ©->stateid, &copied); + if (status == -EINPROGRESS) + goto wait; + nfs4_copy_dequeue_callback(dst_server, src_server, copy); + switch (status) { + case 0: + /* The server recognized the copy stateid, so it hasn't + * rebooted. Don't overwrite the verifier returned in the + * COPY result. */ + res->write_res.count = copied; + goto out_free; + case -EREMOTEIO: + /* COPY operation failed on the server. */ + status = -EOPNOTSUPP; + res->write_res.count = copied; + goto out_free; + case -EBADF: + /* Server did not recognize the copy stateid. It has + * probably restarted and lost the plot. */ + res->write_res.count = 0; + status = -EOPNOTSUPP; + break; + case -EOPNOTSUPP: + /* RFC 7862 REQUIREs server to support OFFLOAD_STATUS when + * it has signed up for an async COPY, so server is not + * spec-compliant. */ + res->write_res.count = 0; + } + goto out_free; +} + +static int process_copy_commit(struct file *dst, loff_t pos_dst, + struct nfs42_copy_res *res) +{ + struct nfs_commitres cres; + int status = -ENOMEM; + + cres.verf = kzalloc_obj(struct nfs_writeverf); + if (!cres.verf) + goto out; + + status = nfs4_proc_commit(dst, pos_dst, res->write_res.count, &cres); + if (status) + goto out_free; + if (nfs_write_verifier_cmp(&res->write_res.verifier.verifier, + &cres.verf->verifier)) { + dprintk("commit verf differs from copy verf\n"); + status = -EAGAIN; + } +out_free: + kfree(cres.verf); +out: + return status; +} + +/** + * nfs42_copy_dest_done - perform inode cache updates after clone/copy offload + * @file: pointer to destination file + * @pos: destination offset + * @len: copy length + * @oldsize: length of the file prior to clone/copy + * + * Punch a hole in the inode page cache, so that the NFS client will + * know to retrieve new data. + * Update the file size if necessary, and then mark the inode as having + * invalid cached values for change attribute, ctime, mtime and space used. + */ +static void nfs42_copy_dest_done(struct file *file, loff_t pos, loff_t len, + loff_t oldsize) +{ + struct inode *inode = file_inode(file); + struct address_space *mapping = file->f_mapping; + loff_t newsize = pos + len; + loff_t end = newsize - 1; + + nfs_truncate_last_folio(mapping, oldsize, pos); + WARN_ON_ONCE(invalidate_inode_pages2_range(mapping, pos >> PAGE_SHIFT, + end >> PAGE_SHIFT)); + + spin_lock(&inode->i_lock); + if (newsize > i_size_read(inode)) + i_size_write(inode, newsize); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_MTIME | + NFS_INO_INVALID_BLOCKS); + spin_unlock(&inode->i_lock); + nfs_update_delegated_mtime(inode); +} + +static ssize_t _nfs42_proc_copy(struct file *src, + struct nfs_lock_context *src_lock, + struct file *dst, + struct nfs_lock_context *dst_lock, + struct nfs42_copy_args *args, + struct nfs42_copy_res *res, + struct nl4_server *nss, + nfs4_stateid *cnr_stateid, + bool *restart) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COPY], + .rpc_argp = args, + .rpc_resp = res, + }; + struct inode *dst_inode = file_inode(dst); + struct inode *src_inode = file_inode(src); + struct nfs_server *dst_server = NFS_SERVER(dst_inode); + struct nfs_server *src_server = NFS_SERVER(src_inode); + loff_t pos_src = args->src_pos; + loff_t pos_dst = args->dst_pos; + loff_t oldsize_dst; + size_t count = args->count; + ssize_t status; + + if (nss) { + args->cp_src = nss; + nfs4_stateid_copy(&args->src_stateid, cnr_stateid); + } else { + status = nfs4_set_rw_stateid(&args->src_stateid, + src_lock->open_context, src_lock, FMODE_READ); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + } + status = nfs_filemap_write_and_wait_range(src->f_mapping, + pos_src, pos_src + (loff_t)count - 1); + if (status) + return status; + + status = nfs4_set_rw_stateid(&args->dst_stateid, dst_lock->open_context, + dst_lock, FMODE_WRITE); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + + nfs_file_block_o_direct(NFS_I(dst_inode)); + status = nfs_sync_inode(dst_inode); + if (status) + return status; + + res->commit_res.verf = NULL; + if (args->sync) { + res->commit_res.verf = + kzalloc_obj(struct nfs_writeverf); + if (!res->commit_res.verf) + return -ENOMEM; + } + set_bit(NFS_CLNT_SRC_SSC_COPY_STATE, + &src_lock->open_context->state->flags); + set_bit(NFS_CLNT_DST_SSC_COPY_STATE, + &dst_lock->open_context->state->flags); + oldsize_dst = i_size_read(dst_inode); + + status = nfs4_call_sync(dst_server->client, dst_server, &msg, + &args->seq_args, &res->seq_res, 0); + trace_nfs4_copy(src_inode, dst_inode, args, res, nss, status); + if (status == -ENOTSUPP) + dst_server->caps &= ~NFS_CAP_COPY; + if (status) + goto out; + + if (args->sync && + nfs_write_verifier_cmp(&res->write_res.verifier.verifier, + &res->commit_res.verf->verifier)) { + status = -EAGAIN; + goto out; + } + + if (!res->synchronous) { + status = handle_async_copy(res, dst_server, src_server, src, + dst, &args->src_stateid, restart); + if (status) + goto out; + } + + if ((!res->synchronous || !args->sync) && + res->write_res.verifier.committed != NFS_FILE_SYNC) { + status = process_copy_commit(dst, pos_dst, res); + if (status) + goto out; + } + + nfs42_copy_dest_done(dst, pos_dst, res->write_res.count, oldsize_dst); + nfs_invalidate_atime(src_inode); + status = res->write_res.count; +out: + if (args->sync) + kfree(res->commit_res.verf); + return status; +} + +ssize_t nfs42_proc_copy(struct file *src, loff_t pos_src, + struct file *dst, loff_t pos_dst, size_t count, + struct nl4_server *nss, + nfs4_stateid *cnr_stateid, bool sync) +{ + struct nfs_server *server = NFS_SERVER(file_inode(dst)); + struct nfs_lock_context *src_lock; + struct nfs_lock_context *dst_lock; + struct nfs42_copy_args args = { + .src_fh = NFS_FH(file_inode(src)), + .src_pos = pos_src, + .dst_fh = NFS_FH(file_inode(dst)), + .dst_pos = pos_dst, + .count = count, + .sync = sync, + }; + struct nfs42_copy_res res; + struct nfs4_exception src_exception = { + .inode = file_inode(src), + .stateid = &args.src_stateid, + }; + struct nfs4_exception dst_exception = { + .inode = file_inode(dst), + .stateid = &args.dst_stateid, + }; + ssize_t err, err2; + bool restart = false; + + src_lock = nfs_get_lock_context(nfs_file_open_context(src)); + if (IS_ERR(src_lock)) + return PTR_ERR(src_lock); + + src_exception.state = src_lock->open_context->state; + + dst_lock = nfs_get_lock_context(nfs_file_open_context(dst)); + if (IS_ERR(dst_lock)) { + err = PTR_ERR(dst_lock); + goto out_put_src_lock; + } + + dst_exception.state = dst_lock->open_context->state; + + do { + inode_lock(file_inode(dst)); + err = _nfs42_proc_copy(src, src_lock, + dst, dst_lock, + &args, &res, + nss, cnr_stateid, &restart); + inode_unlock(file_inode(dst)); + + if (err >= 0) + break; + if ((err == -ENOTSUPP || + err == -NFS4ERR_OFFLOAD_DENIED) && + nfs42_files_from_same_server(src, dst)) { + err = -EOPNOTSUPP; + break; + } else if (err == -EAGAIN) { + if (!restart) { + dst_exception.retry = 1; + continue; + } + break; + } else if (err == -NFS4ERR_OFFLOAD_NO_REQS && + args.sync != res.synchronous) { + args.sync = res.synchronous; + dst_exception.retry = 1; + continue; + } else if ((err == -ESTALE || + err == -NFS4ERR_OFFLOAD_DENIED || + err == -ENOTSUPP) && + !nfs42_files_from_same_server(src, dst)) { + nfs42_do_offload_cancel_async(src, &args.src_stateid); + err = -EOPNOTSUPP; + break; + } + + err2 = nfs4_handle_exception(server, err, &src_exception); + err = nfs4_handle_exception(server, err, &dst_exception); + if (!err) + err = err2; + } while (src_exception.retry || dst_exception.retry); + + nfs_put_lock_context(dst_lock); +out_put_src_lock: + nfs_put_lock_context(src_lock); + return err; +} + +struct nfs42_offload_data { + struct nfs_server *seq_server; + struct nfs42_offload_status_args args; + struct nfs42_offload_status_res res; +}; + +static void nfs42_offload_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs42_offload_data *data = calldata; + + nfs4_setup_sequence(data->seq_server->nfs_client, + &data->args.osa_seq_args, + &data->res.osr_seq_res, task); +} + +static void nfs42_offload_cancel_done(struct rpc_task *task, void *calldata) +{ + struct nfs42_offload_data *data = calldata; + + trace_nfs4_offload_cancel(&data->args, task->tk_status); + nfs41_sequence_done(task, &data->res.osr_seq_res); + if (task->tk_status && + nfs4_async_handle_error(task, data->seq_server, NULL, + NULL) == -EAGAIN) + rpc_restart_call_prepare(task); +} + +static void nfs42_offload_release(void *data) +{ + kfree(data); +} + +static const struct rpc_call_ops nfs42_offload_cancel_ops = { + .rpc_call_prepare = nfs42_offload_prepare, + .rpc_call_done = nfs42_offload_cancel_done, + .rpc_release = nfs42_offload_release, +}; + +static int nfs42_do_offload_cancel_async(struct file *dst, + nfs4_stateid *stateid) +{ + struct nfs_server *dst_server = NFS_SERVER(file_inode(dst)); + struct nfs42_offload_data *data = NULL; + struct nfs_open_context *ctx = nfs_file_open_context(dst); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OFFLOAD_CANCEL], + .rpc_cred = ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = dst_server->client, + .rpc_message = &msg, + .callback_ops = &nfs42_offload_cancel_ops, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_MOVEABLE, + }; + int status; + + if (!(dst_server->caps & NFS_CAP_OFFLOAD_CANCEL)) + return -EOPNOTSUPP; + + data = kzalloc_obj(struct nfs42_offload_data); + if (data == NULL) + return -ENOMEM; + + data->seq_server = dst_server; + data->args.osa_src_fh = NFS_FH(file_inode(dst)); + memcpy(&data->args.osa_stateid, stateid, + sizeof(data->args.osa_stateid)); + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + task_setup_data.callback_data = data; + nfs4_init_sequence(dst_server->nfs_client, &data->args.osa_seq_args, + &data->res.osr_seq_res, 1, 0); + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + status = rpc_wait_for_completion_task(task); + if (status == -ENOTSUPP) + dst_server->caps &= ~NFS_CAP_OFFLOAD_CANCEL; + rpc_put_task(task); + return status; +} + +static int +_nfs42_proc_offload_status(struct nfs_server *server, struct file *file, + struct nfs42_offload_data *data) +{ + struct nfs_open_context *ctx = nfs_file_open_context(file); + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OFFLOAD_STATUS], + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = ctx->cred, + }; + int status; + + status = nfs4_call_sync(server->client, server, &msg, + &data->args.osa_seq_args, + &data->res.osr_seq_res, 1); + trace_nfs4_offload_status(&data->args, status); + switch (status) { + case 0: + break; + + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_OLD_STATEID: + /* + * Server does not recognize the COPY stateid. CB_OFFLOAD + * could have purged it, or server might have rebooted. + * Since COPY stateids don't have an associated inode, + * avoid triggering state recovery. + */ + status = -EBADF; + break; + case -NFS4ERR_NOTSUPP: + case -ENOTSUPP: + case -EOPNOTSUPP: + server->caps &= ~NFS_CAP_OFFLOAD_STATUS; + status = -EOPNOTSUPP; + break; + } + + return status; +} + +/** + * nfs42_proc_offload_status - Poll completion status of an async copy operation + * @dst: handle of file being copied into + * @stateid: copy stateid (from async COPY result) + * @copied: OUT: number of bytes copied so far + * + * Return values: + * %0: Server returned an NFS4_OK completion status + * %-EINPROGRESS: Server returned no completion status + * %-EREMOTEIO: Server returned an error completion status + * %-EBADF: Server did not recognize the copy stateid + * %-EOPNOTSUPP: Server does not support OFFLOAD_STATUS + * %-ERESTARTSYS: Wait interrupted by signal + * + * Other negative errnos indicate the client could not complete the + * request. + */ +static int +nfs42_proc_offload_status(struct file *dst, nfs4_stateid *stateid, u64 *copied) +{ + struct inode *inode = file_inode(dst); + struct nfs_server *server = NFS_SERVER(inode); + struct nfs4_exception exception = { + .inode = inode, + }; + struct nfs42_offload_data *data; + int status; + + if (!(server->caps & NFS_CAP_OFFLOAD_STATUS)) + return -EOPNOTSUPP; + + data = kzalloc_obj(*data); + if (!data) + return -ENOMEM; + data->seq_server = server; + data->args.osa_src_fh = NFS_FH(inode); + memcpy(&data->args.osa_stateid, stateid, + sizeof(data->args.osa_stateid)); + exception.stateid = &data->args.osa_stateid; + do { + status = _nfs42_proc_offload_status(server, dst, data); + if (status == -EOPNOTSUPP) + goto out; + status = nfs4_handle_exception(server, status, &exception); + } while (exception.retry); + if (status) + goto out; + + *copied = data->res.osr_count; + if (!data->res.complete_count) + status = -EINPROGRESS; + else if (data->res.osr_complete != NFS_OK) + status = -EREMOTEIO; + +out: + kfree(data); + return status; +} + +static int _nfs42_proc_copy_notify(struct file *src, struct file *dst, + struct nfs42_copy_notify_args *args, + struct nfs42_copy_notify_res *res) +{ + struct nfs_server *src_server = NFS_SERVER(file_inode(src)); + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COPY_NOTIFY], + .rpc_argp = args, + .rpc_resp = res, + }; + int status; + struct nfs_open_context *ctx; + struct nfs_lock_context *l_ctx; + + ctx = get_nfs_open_context(nfs_file_open_context(src)); + l_ctx = nfs_get_lock_context(ctx); + if (IS_ERR(l_ctx)) { + status = PTR_ERR(l_ctx); + goto out; + } + + status = nfs4_set_rw_stateid(&args->cna_src_stateid, ctx, l_ctx, + FMODE_READ); + nfs_put_lock_context(l_ctx); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + goto out; + } + + status = nfs4_call_sync(src_server->client, src_server, &msg, + &args->cna_seq_args, &res->cnr_seq_res, 0); + trace_nfs4_copy_notify(file_inode(src), args, res, status); + if (status == -ENOTSUPP) + src_server->caps &= ~NFS_CAP_COPY_NOTIFY; + +out: + put_nfs_open_context(nfs_file_open_context(src)); + return status; +} + +int nfs42_proc_copy_notify(struct file *src, struct file *dst, + struct nfs42_copy_notify_res *res) +{ + struct nfs_server *src_server = NFS_SERVER(file_inode(src)); + struct nfs42_copy_notify_args *args; + struct nfs4_exception exception = { + .inode = file_inode(src), + }; + int status; + + if (!(src_server->caps & NFS_CAP_COPY_NOTIFY)) + return -EOPNOTSUPP; + + args = kzalloc_obj(struct nfs42_copy_notify_args); + if (args == NULL) + return -ENOMEM; + + args->cna_src_fh = NFS_FH(file_inode(src)), + args->cna_dst.nl4_type = NL4_NETADDR; + nfs42_set_netaddr(dst, &args->cna_dst.u.nl4_addr); + exception.stateid = &args->cna_src_stateid; + + do { + status = _nfs42_proc_copy_notify(src, dst, args, res); + if (status == -ENOTSUPP) { + status = -EOPNOTSUPP; + goto out; + } + status = nfs4_handle_exception(src_server, status, &exception); + } while (exception.retry); + +out: + kfree(args); + return status; +} + +static loff_t _nfs42_proc_llseek(struct file *filep, + struct nfs_lock_context *lock, loff_t offset, int whence) +{ + struct inode *inode = file_inode(filep); + struct nfs42_seek_args args = { + .sa_fh = NFS_FH(inode), + .sa_offset = offset, + .sa_what = (whence == SEEK_HOLE) ? + NFS4_CONTENT_HOLE : NFS4_CONTENT_DATA, + }; + struct nfs42_seek_res res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEEK], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct nfs_server *server = NFS_SERVER(inode); + int status; + + if (!nfs_server_capable(inode, NFS_CAP_SEEK)) + return -ENOTSUPP; + + status = nfs4_set_rw_stateid(&args.sa_stateid, lock->open_context, + lock, FMODE_READ); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + + status = nfs_filemap_write_and_wait_range(inode->i_mapping, + offset, LLONG_MAX); + if (status) + return status; + + status = nfs4_call_sync(server->client, server, &msg, + &args.seq_args, &res.seq_res, 0); + trace_nfs4_llseek(inode, &args, &res, status); + if (status == -ENOTSUPP) + server->caps &= ~NFS_CAP_SEEK; + if (status) + return status; + + if (whence == SEEK_DATA && res.sr_eof) + return -NFS4ERR_NXIO; + else + return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes); +} + +loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence) +{ + struct nfs_server *server = NFS_SERVER(file_inode(filep)); + struct nfs4_exception exception = { }; + struct nfs_lock_context *lock; + loff_t err; + + lock = nfs_get_lock_context(nfs_file_open_context(filep)); + if (IS_ERR(lock)) + return PTR_ERR(lock); + + exception.inode = file_inode(filep); + exception.state = lock->open_context->state; + + do { + err = _nfs42_proc_llseek(filep, lock, offset, whence); + if (err >= 0) + break; + if (err == -ENOTSUPP) { + err = -EOPNOTSUPP; + break; + } + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + + nfs_put_lock_context(lock); + return err; +} + + +static void +nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs42_layoutstat_data *data = calldata; + struct inode *inode = data->inode; + struct nfs_server *server = NFS_SERVER(inode); + struct pnfs_layout_hdr *lo; + + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (!pnfs_layout_is_valid(lo)) { + spin_unlock(&inode->i_lock); + rpc_exit(task, 0); + return; + } + nfs4_stateid_copy(&data->args.stateid, &lo->plh_stateid); + spin_unlock(&inode->i_lock); + nfs4_setup_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, task); +} + +static void +nfs42_layoutstat_done(struct rpc_task *task, void *calldata) +{ + struct nfs42_layoutstat_data *data = calldata; + struct inode *inode = data->inode; + struct pnfs_layout_hdr *lo; + + if (!nfs4_sequence_done(task, &data->res.seq_res)) + return; + + switch (task->tk_status) { + case 0: + return; + case -NFS4ERR_BADHANDLE: + case -ESTALE: + pnfs_destroy_layout(NFS_I(inode)); + break; + case -NFS4ERR_EXPIRED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_BAD_STATEID: + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (pnfs_layout_is_valid(lo) && + nfs4_stateid_match(&data->args.stateid, + &lo->plh_stateid)) { + LIST_HEAD(head); + + /* + * Mark the bad layout state as invalid, then retry + * with the current stateid. + */ + pnfs_mark_layout_stateid_invalid(lo, &head); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&head); + nfs_commit_inode(inode, 0); + } else + spin_unlock(&inode->i_lock); + break; + case -NFS4ERR_OLD_STATEID: + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (pnfs_layout_is_valid(lo) && + nfs4_stateid_match_other(&data->args.stateid, + &lo->plh_stateid)) { + /* Do we need to delay before resending? */ + if (!nfs4_stateid_is_newer(&lo->plh_stateid, + &data->args.stateid)) + rpc_delay(task, HZ); + rpc_restart_call_prepare(task); + } + spin_unlock(&inode->i_lock); + break; + case -ENOTSUPP: + case -EOPNOTSUPP: + NFS_SERVER(inode)->caps &= ~NFS_CAP_LAYOUTSTATS; + } + + trace_nfs4_layoutstats(inode, &data->args.stateid, task->tk_status); +} + +static void +nfs42_layoutstat_release(void *calldata) +{ + struct nfs42_layoutstat_data *data = calldata; + struct nfs42_layoutstat_devinfo *devinfo = data->args.devinfo; + int i; + + for (i = 0; i < data->args.num_dev; i++) { + if (devinfo[i].ld_private.ops && devinfo[i].ld_private.ops->free) + devinfo[i].ld_private.ops->free(&devinfo[i].ld_private); + } + + pnfs_put_layout_hdr(NFS_I(data->args.inode)->layout); + smp_mb__before_atomic(); + clear_bit(NFS_INO_LAYOUTSTATS, &NFS_I(data->args.inode)->flags); + smp_mb__after_atomic(); + nfs_iput_and_deactive(data->inode); + kfree(data->args.devinfo); + kfree(data); +} + +static const struct rpc_call_ops nfs42_layoutstat_ops = { + .rpc_call_prepare = nfs42_layoutstat_prepare, + .rpc_call_done = nfs42_layoutstat_done, + .rpc_release = nfs42_layoutstat_release, +}; + +int nfs42_proc_layoutstats_generic(struct nfs_server *server, + struct nfs42_layoutstat_data *data) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTSTATS], + .rpc_argp = &data->args, + .rpc_resp = &data->res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs42_layoutstat_ops, + .callback_data = data, + .flags = RPC_TASK_ASYNC | RPC_TASK_MOVEABLE, + }; + struct rpc_task *task; + + data->inode = nfs_igrab_and_active(data->args.inode); + if (!data->inode) { + nfs42_layoutstat_release(data); + return -EAGAIN; + } + nfs4_init_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, 0, 0); + task = rpc_run_task(&task_setup); + if (IS_ERR(task)) + return PTR_ERR(task); + rpc_put_task(task); + return 0; +} + +static struct nfs42_layouterror_data * +nfs42_alloc_layouterror_data(struct pnfs_layout_segment *lseg, gfp_t gfp_flags) +{ + struct nfs42_layouterror_data *data; + struct inode *inode = lseg->pls_layout->plh_inode; + + data = kzalloc_obj(*data, gfp_flags); + if (data) { + data->args.inode = data->inode = nfs_igrab_and_active(inode); + if (data->inode) { + data->lseg = pnfs_get_lseg(lseg); + if (data->lseg) + return data; + nfs_iput_and_deactive(data->inode); + } + kfree(data); + } + return NULL; +} + +static void +nfs42_free_layouterror_data(struct nfs42_layouterror_data *data) +{ + pnfs_put_lseg(data->lseg); + nfs_iput_and_deactive(data->inode); + kfree(data); +} + +static void +nfs42_layouterror_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs42_layouterror_data *data = calldata; + struct inode *inode = data->inode; + struct nfs_server *server = NFS_SERVER(inode); + struct pnfs_layout_hdr *lo = data->lseg->pls_layout; + unsigned i; + + spin_lock(&inode->i_lock); + if (!pnfs_layout_is_valid(lo)) { + spin_unlock(&inode->i_lock); + rpc_exit(task, 0); + return; + } + for (i = 0; i < data->args.num_errors; i++) + nfs4_stateid_copy(&data->args.errors[i].stateid, + &lo->plh_stateid); + spin_unlock(&inode->i_lock); + nfs4_setup_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, task); +} + +static void +nfs42_layouterror_done(struct rpc_task *task, void *calldata) +{ + struct nfs42_layouterror_data *data = calldata; + struct inode *inode = data->inode; + struct pnfs_layout_hdr *lo = data->lseg->pls_layout; + + if (!nfs4_sequence_done(task, &data->res.seq_res)) + return; + + switch (task->tk_status) { + case 0: + return; + case -NFS4ERR_BADHANDLE: + case -ESTALE: + pnfs_destroy_layout(NFS_I(inode)); + break; + case -NFS4ERR_EXPIRED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_BAD_STATEID: + spin_lock(&inode->i_lock); + if (pnfs_layout_is_valid(lo) && + nfs4_stateid_match(&data->args.errors[0].stateid, + &lo->plh_stateid)) { + LIST_HEAD(head); + + /* + * Mark the bad layout state as invalid, then retry + * with the current stateid. + */ + pnfs_mark_layout_stateid_invalid(lo, &head); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&head); + nfs_commit_inode(inode, 0); + } else + spin_unlock(&inode->i_lock); + break; + case -NFS4ERR_OLD_STATEID: + spin_lock(&inode->i_lock); + if (pnfs_layout_is_valid(lo) && + nfs4_stateid_match_other(&data->args.errors[0].stateid, + &lo->plh_stateid)) { + /* Do we need to delay before resending? */ + if (!nfs4_stateid_is_newer(&lo->plh_stateid, + &data->args.errors[0].stateid)) + rpc_delay(task, HZ); + rpc_restart_call_prepare(task); + } + spin_unlock(&inode->i_lock); + break; + case -ENOTSUPP: + case -EOPNOTSUPP: + NFS_SERVER(inode)->caps &= ~NFS_CAP_LAYOUTERROR; + } + + trace_nfs4_layouterror(inode, &data->args.errors[0].stateid, + task->tk_status); +} + +static void +nfs42_layouterror_release(void *calldata) +{ + struct nfs42_layouterror_data *data = calldata; + + nfs42_free_layouterror_data(data); +} + +static const struct rpc_call_ops nfs42_layouterror_ops = { + .rpc_call_prepare = nfs42_layouterror_prepare, + .rpc_call_done = nfs42_layouterror_done, + .rpc_release = nfs42_layouterror_release, +}; + +int nfs42_proc_layouterror(struct pnfs_layout_segment *lseg, + const struct nfs42_layout_error *errors, size_t n) +{ + struct inode *inode = lseg->pls_layout->plh_inode; + struct nfs_server *server = NFS_SERVER(inode); + struct nfs42_layouterror_data *data; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTERROR], + }; + struct rpc_task_setup task_setup = { + .rpc_message = &msg, + .callback_ops = &nfs42_layouterror_ops, + .flags = RPC_TASK_ASYNC | RPC_TASK_MOVEABLE, + }; + unsigned int i; + + if (!nfs_server_capable(inode, NFS_CAP_LAYOUTERROR)) + return -EOPNOTSUPP; + if (n > NFS42_LAYOUTERROR_MAX) + return -EINVAL; + data = nfs42_alloc_layouterror_data(lseg, nfs_io_gfp_mask()); + if (!data) + return -ENOMEM; + for (i = 0; i < n; i++) { + data->args.errors[i] = errors[i]; + data->args.num_errors++; + data->res.num_errors++; + } + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + task_setup.callback_data = data; + task_setup.rpc_client = server->client; + nfs4_init_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, 0, 0); + task = rpc_run_task(&task_setup); + if (IS_ERR(task)) + return PTR_ERR(task); + rpc_put_task(task); + return 0; +} +EXPORT_SYMBOL_GPL(nfs42_proc_layouterror); + +static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f, + struct file *dst_f, struct nfs_lock_context *src_lock, + struct nfs_lock_context *dst_lock, loff_t src_offset, + loff_t dst_offset, loff_t count) +{ + struct inode *src_inode = file_inode(src_f); + struct inode *dst_inode = file_inode(dst_f); + struct nfs_server *server = NFS_SERVER(dst_inode); + __u32 dst_bitmask[NFS_BITMASK_SZ]; + struct nfs42_clone_args args = { + .src_fh = NFS_FH(src_inode), + .dst_fh = NFS_FH(dst_inode), + .src_offset = src_offset, + .dst_offset = dst_offset, + .count = count, + .dst_bitmask = dst_bitmask, + }; + struct nfs42_clone_res res = { + .server = server, + }; + loff_t oldsize_dst = i_size_read(dst_inode); + int status; + + msg->rpc_argp = &args; + msg->rpc_resp = &res; + + status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context, + src_lock, FMODE_READ); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context, + dst_lock, FMODE_WRITE); + if (status) { + if (status == -EAGAIN) + status = -NFS4ERR_BAD_STATEID; + return status; + } + + res.dst_fattr = nfs_alloc_fattr(); + if (!res.dst_fattr) + return -ENOMEM; + + nfs4_bitmask_set(dst_bitmask, server->cache_consistency_bitmask, + dst_inode, NFS_INO_INVALID_BLOCKS); + + status = nfs4_call_sync(server->client, server, msg, + &args.seq_args, &res.seq_res, 0); + trace_nfs4_clone(src_inode, dst_inode, &args, status); + if (status == 0) { + /* a zero-length count means clone to EOF in src */ + if (count == 0 && res.dst_fattr->valid & NFS_ATTR_FATTR_SIZE) + count = nfs_size_to_loff_t(res.dst_fattr->size) - dst_offset; + nfs42_copy_dest_done(dst_f, dst_offset, count, oldsize_dst); + status = nfs_post_op_update_inode(dst_inode, res.dst_fattr); + } + + kfree(res.dst_fattr); + return status; +} + +int nfs42_proc_clone(struct file *src_f, struct file *dst_f, + loff_t src_offset, loff_t dst_offset, loff_t count) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLONE], + }; + struct inode *inode = file_inode(src_f); + struct nfs_server *server = NFS_SERVER(file_inode(src_f)); + struct nfs_lock_context *src_lock; + struct nfs_lock_context *dst_lock; + struct nfs4_exception src_exception = { }; + struct nfs4_exception dst_exception = { }; + int err, err2; + + if (!nfs_server_capable(inode, NFS_CAP_CLONE)) + return -EOPNOTSUPP; + + src_lock = nfs_get_lock_context(nfs_file_open_context(src_f)); + if (IS_ERR(src_lock)) + return PTR_ERR(src_lock); + + src_exception.inode = file_inode(src_f); + src_exception.state = src_lock->open_context->state; + + dst_lock = nfs_get_lock_context(nfs_file_open_context(dst_f)); + if (IS_ERR(dst_lock)) { + err = PTR_ERR(dst_lock); + goto out_put_src_lock; + } + + dst_exception.inode = file_inode(dst_f); + dst_exception.state = dst_lock->open_context->state; + + do { + err = _nfs42_proc_clone(&msg, src_f, dst_f, src_lock, dst_lock, + src_offset, dst_offset, count); + if (err == -ENOTSUPP || err == -EOPNOTSUPP) { + NFS_SERVER(inode)->caps &= ~NFS_CAP_CLONE; + err = -EOPNOTSUPP; + break; + } + + err2 = nfs4_handle_exception(server, err, &src_exception); + err = nfs4_handle_exception(server, err, &dst_exception); + if (!err) + err = err2; + } while (src_exception.retry || dst_exception.retry); + + nfs_put_lock_context(dst_lock); +out_put_src_lock: + nfs_put_lock_context(src_lock); + return err; +} + +#define NFS4XATTR_MAXPAGES DIV_ROUND_UP(XATTR_SIZE_MAX, PAGE_SIZE) + +static int _nfs42_proc_removexattr(struct inode *inode, const char *name) +{ + struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS_BITMASK_SZ]; + struct nfs42_removexattrargs args = { + .fh = NFS_FH(inode), + .bitmask = bitmask, + .xattr_name = name, + }; + struct nfs42_removexattrres res = { + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVEXATTR], + .rpc_argp = &args, + .rpc_resp = &res, + }; + int ret; + unsigned long timestamp = jiffies; + + res.fattr = nfs_alloc_fattr(); + if (!res.fattr) + return -ENOMEM; + + nfs4_bitmask_set(bitmask, server->cache_consistency_bitmask, + inode, NFS_INO_INVALID_CHANGE); + + ret = nfs4_call_sync(server->client, server, &msg, &args.seq_args, + &res.seq_res, 1); + trace_nfs4_removexattr(inode, name, ret); + if (!ret) { + nfs4_update_changeattr(inode, &res.cinfo, timestamp, 0); + ret = nfs_post_op_update_inode(inode, res.fattr); + } + + kfree(res.fattr); + return ret; +} + +static int _nfs42_proc_setxattr(struct inode *inode, const char *name, + const void *buf, size_t buflen, int flags) +{ + struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS_BITMASK_SZ]; + struct page *pages[NFS4XATTR_MAXPAGES]; + struct nfs42_setxattrargs arg = { + .fh = NFS_FH(inode), + .bitmask = bitmask, + .xattr_pages = pages, + .xattr_len = buflen, + .xattr_name = name, + .xattr_flags = flags, + }; + struct nfs42_setxattrres res = { + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETXATTR], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int ret, np; + unsigned long timestamp = jiffies; + + if (buflen > server->sxasize) + return -ERANGE; + + res.fattr = nfs_alloc_fattr(); + if (!res.fattr) + return -ENOMEM; + + if (buflen > 0) { + np = nfs4_buf_to_pages_noslab(buf, buflen, arg.xattr_pages); + if (np < 0) { + ret = np; + goto out; + } + } else + np = 0; + + nfs4_bitmask_set(bitmask, server->cache_consistency_bitmask, + inode, NFS_INO_INVALID_CHANGE); + + ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, + &res.seq_res, 1); + trace_nfs4_setxattr(inode, name, ret); + + for (; np > 0; np--) + put_page(pages[np - 1]); + + if (!ret) { + nfs4_update_changeattr(inode, &res.cinfo, timestamp, 0); + ret = nfs_post_op_update_inode(inode, res.fattr); + } + +out: + kfree(res.fattr); + return ret; +} + +static ssize_t _nfs42_proc_getxattr(struct inode *inode, const char *name, + void *buf, size_t buflen, struct page **pages, + size_t plen) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs42_getxattrargs arg = { + .fh = NFS_FH(inode), + .xattr_name = name, + }; + struct nfs42_getxattrres res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETXATTR], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + ssize_t ret; + + arg.xattr_len = plen; + arg.xattr_pages = pages; + + ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, + &res.seq_res, 0); + trace_nfs4_getxattr(inode, name, ret); + if (ret < 0) + return ret; + + /* + * Normally, the caching is done one layer up, but for successful + * RPCS, always cache the result here, even if the caller was + * just querying the length, or if the reply was too big for + * the caller. This avoids a second RPC in the case of the + * common query-alloc-retrieve cycle for xattrs. + * + * Note that xattr_len is always capped to XATTR_SIZE_MAX. + */ + + nfs4_xattr_cache_add(inode, name, NULL, pages, res.xattr_len); + + if (buflen) { + if (res.xattr_len > buflen) + return -ERANGE; + _copy_from_pages(buf, pages, 0, res.xattr_len); + } + + return res.xattr_len; +} + +static ssize_t _nfs42_proc_listxattrs(struct inode *inode, void *buf, + size_t buflen, u64 *cookiep, bool *eofp) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct page **pages; + struct nfs42_listxattrsargs arg = { + .fh = NFS_FH(inode), + .cookie = *cookiep, + }; + struct nfs42_listxattrsres res = { + .eof = false, + .xattr_buf = buf, + .xattr_len = buflen, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LISTXATTRS], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + u32 xdrlen; + int ret, np, i; + + + ret = -ENOMEM; + res.scratch = folio_alloc(GFP_KERNEL, 0); + if (!res.scratch) + goto out; + + xdrlen = nfs42_listxattr_xdrsize(buflen); + if (xdrlen > server->lxasize) + xdrlen = server->lxasize; + np = xdrlen / PAGE_SIZE + 1; + + pages = kzalloc_objs(struct page *, np); + if (!pages) + goto out_free_scratch; + for (i = 0; i < np; i++) { + pages[i] = alloc_page(GFP_KERNEL); + if (!pages[i]) + goto out_free_pages; + } + + arg.xattr_pages = pages; + arg.count = xdrlen; + + ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, + &res.seq_res, 0); + trace_nfs4_listxattr(inode, ret); + + if (ret >= 0) { + ret = res.copied; + *cookiep = res.cookie; + *eofp = res.eof; + } + +out_free_pages: + while (--np >= 0) { + if (pages[np]) + __free_page(pages[np]); + } + kfree(pages); +out_free_scratch: + folio_put(res.scratch); +out: + return ret; + +} + +ssize_t nfs42_proc_getxattr(struct inode *inode, const char *name, + void *buf, size_t buflen) +{ + struct nfs4_exception exception = { }; + ssize_t err, np, i; + struct page **pages; + + np = nfs_page_array_len(0, buflen ?: XATTR_SIZE_MAX); + pages = kmalloc_objs(*pages, np); + if (!pages) + return -ENOMEM; + + for (i = 0; i < np; i++) { + pages[i] = alloc_page(GFP_KERNEL); + if (!pages[i]) { + err = -ENOMEM; + goto out; + } + } + + /* + * The GETXATTR op has no length field in the call, and the + * xattr data is at the end of the reply. + * + * There is no downside in using the page-aligned length. It will + * allow receiving and caching xattrs that are too large for the + * caller but still fit in the page-rounded value. + */ + do { + err = _nfs42_proc_getxattr(inode, name, buf, buflen, + pages, np * PAGE_SIZE); + if (err >= 0) + break; + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + +out: + while (--i >= 0) + __free_page(pages[i]); + kfree(pages); + + return err; +} + +int nfs42_proc_setxattr(struct inode *inode, const char *name, + const void *buf, size_t buflen, int flags) +{ + struct nfs4_exception exception = { }; + int err; + + do { + err = _nfs42_proc_setxattr(inode, name, buf, buflen, flags); + if (!err) + break; + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + + return err; +} + +ssize_t nfs42_proc_listxattrs(struct inode *inode, void *buf, + size_t buflen, u64 *cookiep, bool *eofp) +{ + struct nfs4_exception exception = { }; + ssize_t err; + + do { + err = _nfs42_proc_listxattrs(inode, buf, buflen, + cookiep, eofp); + if (err >= 0) + break; + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + + return err; +} + +int nfs42_proc_removexattr(struct inode *inode, const char *name) +{ + struct nfs4_exception exception = { }; + int err; + + do { + err = _nfs42_proc_removexattr(inode, name); + if (!err) + break; + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + + return err; +} diff --git a/fs/nfs/nfs42xattr.c b/fs/nfs/nfs42xattr.c new file mode 100644 index 000000000..37d79400e --- /dev/null +++ b/fs/nfs/nfs42xattr.c @@ -0,0 +1,1067 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Copyright 2019, 2020 Amazon.com, Inc. or its affiliates. All rights reserved. + * + * User extended attribute client side cache functions. + * + * Author: Frank van der Linden + */ +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "internal.h" + +/* + * User extended attributes client side caching is implemented by having + * a cache structure attached to NFS inodes. This structure is allocated + * when needed, and freed when the cache is zapped. + * + * The cache structure contains as hash table of entries, and a pointer + * to a special-cased entry for the listxattr cache. + * + * Accessing and allocating / freeing the caches is done via reference + * counting. The cache entries use a similar refcounting scheme. + * + * This makes freeing a cache, both from the shrinker and from the + * zap cache path, easy. It also means that, in current use cases, + * the large majority of inodes will not waste any memory, as they + * will never have any user extended attributes assigned to them. + * + * Attribute entries are hashed in to a simple hash table. They are + * also part of an LRU. + * + * There are three shrinkers. + * + * Two shrinkers deal with the cache entries themselves: one for + * large entries (> PAGE_SIZE), and one for smaller entries. The + * shrinker for the larger entries works more aggressively than + * those for the smaller entries. + * + * The other shrinker frees the cache structures themselves. + */ + +/* + * 64 buckets is a good default. There is likely no reasonable + * workload that uses more than even 64 user extended attributes. + * You can certainly add a lot more - but you get what you ask for + * in those circumstances. + */ +#define NFS4_XATTR_HASH_SIZE 64 + +#define NFSDBG_FACILITY NFSDBG_XATTRCACHE + +struct nfs4_xattr_cache; +struct nfs4_xattr_entry; + +struct nfs4_xattr_bucket { + spinlock_t lock; + struct hlist_head hlist; + struct nfs4_xattr_cache *cache; + bool draining; +}; + +struct nfs4_xattr_cache { + struct kref ref; + struct nfs4_xattr_bucket buckets[NFS4_XATTR_HASH_SIZE]; + struct list_head lru; + struct list_head dispose; + atomic_long_t nent; + spinlock_t listxattr_lock; + struct inode *inode; + struct nfs4_xattr_entry *listxattr; +}; + +struct nfs4_xattr_entry { + struct kref ref; + struct hlist_node hnode; + struct list_head lru; + struct list_head dispose; + char *xattr_name; + void *xattr_value; + size_t xattr_size; + struct nfs4_xattr_bucket *bucket; + uint32_t flags; +}; + +#define NFS4_XATTR_ENTRY_EXTVAL 0x0001 + +/* + * LRU list of NFS inodes that have xattr caches. + */ +static struct list_lru nfs4_xattr_cache_lru; +static struct list_lru nfs4_xattr_entry_lru; +static struct list_lru nfs4_xattr_large_entry_lru; + +static struct kmem_cache *nfs4_xattr_cache_cachep; + +/* + * Hashing helper functions. + */ +static void +nfs4_xattr_hash_init(struct nfs4_xattr_cache *cache) +{ + unsigned int i; + + for (i = 0; i < NFS4_XATTR_HASH_SIZE; i++) { + INIT_HLIST_HEAD(&cache->buckets[i].hlist); + spin_lock_init(&cache->buckets[i].lock); + cache->buckets[i].cache = cache; + cache->buckets[i].draining = false; + } +} + +/* + * Locking order: + * 1. inode i_lock or bucket lock + * 2. list_lru lock (taken by list_lru_* functions) + */ + +/* + * Wrapper functions to add a cache entry to the right LRU. + */ +static bool +nfs4_xattr_entry_lru_add(struct nfs4_xattr_entry *entry) +{ + struct list_lru *lru; + + lru = (entry->flags & NFS4_XATTR_ENTRY_EXTVAL) ? + &nfs4_xattr_large_entry_lru : &nfs4_xattr_entry_lru; + + return list_lru_add_obj(lru, &entry->lru); +} + +static bool +nfs4_xattr_entry_lru_del(struct nfs4_xattr_entry *entry) +{ + struct list_lru *lru; + + lru = (entry->flags & NFS4_XATTR_ENTRY_EXTVAL) ? + &nfs4_xattr_large_entry_lru : &nfs4_xattr_entry_lru; + + return list_lru_del_obj(lru, &entry->lru); +} + +/* + * This function allocates cache entries. They are the normal + * extended attribute name/value pairs, but may also be a listxattr + * cache. Those allocations use the same entry so that they can be + * treated as one by the memory shrinker. + * + * xattr cache entries are allocated together with names. If the + * value fits in to one page with the entry structure and the name, + * it will also be part of the same allocation (kmalloc). This is + * expected to be the vast majority of cases. Larger allocations + * have a value pointer that is allocated separately by kvmalloc. + * + * Parameters: + * + * @name: Name of the extended attribute. NULL for listxattr cache + * entry. + * @value: Value of attribute, or listxattr cache. NULL if the + * value is to be copied from pages instead. + * @pages: Pages to copy the value from, if not NULL. Passed in to + * make it easier to copy the value after an RPC, even if + * the value will not be passed up to application (e.g. + * for a 'query' getxattr with NULL buffer). + * @len: Length of the value. Can be 0 for zero-length attributes. + * @value and @pages will be NULL if @len is 0. + */ +static struct nfs4_xattr_entry * +nfs4_xattr_alloc_entry(const char *name, const void *value, + struct page **pages, size_t len) +{ + struct nfs4_xattr_entry *entry; + void *valp; + char *namep; + size_t alloclen, slen; + char *buf; + uint32_t flags; + + BUILD_BUG_ON(sizeof(struct nfs4_xattr_entry) + + XATTR_NAME_MAX + 1 > PAGE_SIZE); + + alloclen = sizeof(struct nfs4_xattr_entry); + if (name != NULL) { + slen = strlen(name) + 1; + alloclen += slen; + } else + slen = 0; + + if (alloclen + len <= PAGE_SIZE) { + alloclen += len; + flags = 0; + } else { + flags = NFS4_XATTR_ENTRY_EXTVAL; + } + + buf = kmalloc(alloclen, GFP_KERNEL); + if (buf == NULL) + return NULL; + entry = (struct nfs4_xattr_entry *)buf; + + if (name != NULL) { + namep = buf + sizeof(struct nfs4_xattr_entry); + memcpy(namep, name, slen); + } else { + namep = NULL; + } + + + if (flags & NFS4_XATTR_ENTRY_EXTVAL) { + valp = kvmalloc(len, GFP_KERNEL); + if (valp == NULL) { + kfree(buf); + return NULL; + } + } else if (len != 0) { + valp = buf + sizeof(struct nfs4_xattr_entry) + slen; + } else + valp = NULL; + + if (valp != NULL) { + if (value != NULL) + memcpy(valp, value, len); + else + _copy_from_pages(valp, pages, 0, len); + } + + entry->flags = flags; + entry->xattr_value = valp; + kref_init(&entry->ref); + entry->xattr_name = namep; + entry->xattr_size = len; + entry->bucket = NULL; + INIT_LIST_HEAD(&entry->lru); + INIT_LIST_HEAD(&entry->dispose); + INIT_HLIST_NODE(&entry->hnode); + + return entry; +} + +static void +nfs4_xattr_free_entry(struct nfs4_xattr_entry *entry) +{ + if (entry->flags & NFS4_XATTR_ENTRY_EXTVAL) + kvfree(entry->xattr_value); + kfree(entry); +} + +static void +nfs4_xattr_free_entry_cb(struct kref *kref) +{ + struct nfs4_xattr_entry *entry; + + entry = container_of(kref, struct nfs4_xattr_entry, ref); + + if (WARN_ON(!list_empty(&entry->lru))) + return; + + nfs4_xattr_free_entry(entry); +} + +static void +nfs4_xattr_free_cache_cb(struct kref *kref) +{ + struct nfs4_xattr_cache *cache; + int i; + + cache = container_of(kref, struct nfs4_xattr_cache, ref); + + for (i = 0; i < NFS4_XATTR_HASH_SIZE; i++) { + if (WARN_ON(!hlist_empty(&cache->buckets[i].hlist))) + return; + cache->buckets[i].draining = false; + } + + cache->listxattr = NULL; + + kmem_cache_free(nfs4_xattr_cache_cachep, cache); + +} + +static struct nfs4_xattr_cache * +nfs4_xattr_alloc_cache(void) +{ + struct nfs4_xattr_cache *cache; + + cache = kmem_cache_alloc(nfs4_xattr_cache_cachep, GFP_KERNEL); + if (cache == NULL) + return NULL; + + kref_init(&cache->ref); + atomic_long_set(&cache->nent, 0); + + return cache; +} + +/* + * Set the listxattr cache, which is a special-cased cache entry. + * The special value ERR_PTR(-ESTALE) is used to indicate that + * the cache is being drained - this prevents a new listxattr + * cache from being added to what is now a stale cache. + */ +static int +nfs4_xattr_set_listcache(struct nfs4_xattr_cache *cache, + struct nfs4_xattr_entry *new) +{ + struct nfs4_xattr_entry *old; + int ret = 1; + + spin_lock(&cache->listxattr_lock); + + old = cache->listxattr; + + if (old == ERR_PTR(-ESTALE)) { + ret = 0; + goto out; + } + + cache->listxattr = new; + if (new != NULL && new != ERR_PTR(-ESTALE)) + nfs4_xattr_entry_lru_add(new); + + if (old != NULL) { + nfs4_xattr_entry_lru_del(old); + kref_put(&old->ref, nfs4_xattr_free_entry_cb); + } +out: + spin_unlock(&cache->listxattr_lock); + + return ret; +} + +/* + * Unlink a cache from its parent inode, clearing out an invalid + * cache. Must be called with i_lock held. + */ +static struct nfs4_xattr_cache * +nfs4_xattr_cache_unlink(struct inode *inode) +{ + struct nfs_inode *nfsi; + struct nfs4_xattr_cache *oldcache; + + nfsi = NFS_I(inode); + + oldcache = nfsi->xattr_cache; + if (oldcache != NULL) { + list_lru_del_obj(&nfs4_xattr_cache_lru, &oldcache->lru); + oldcache->inode = NULL; + } + nfsi->xattr_cache = NULL; + nfsi->cache_validity &= ~NFS_INO_INVALID_XATTR; + + return oldcache; + +} + +/* + * Discard a cache. Called by get_cache() if there was an old, + * invalid cache. Can also be called from a shrinker callback. + * + * The cache is dead, it has already been unlinked from its inode, + * and no longer appears on the cache LRU list. + * + * Mark all buckets as draining, so that no new entries are added. This + * could still happen in the unlikely, but possible case that another + * thread had grabbed a reference before it was unlinked from the inode, + * and is still holding it for an add operation. + * + * Remove all entries from the LRU lists, so that there is no longer + * any way to 'find' this cache. Then, remove the entries from the hash + * table. + * + * At that point, the cache will remain empty and can be freed when the final + * reference drops, which is very likely the kref_put at the end of + * this function, or the one called immediately afterwards in the + * shrinker callback. + */ +static void +nfs4_xattr_discard_cache(struct nfs4_xattr_cache *cache) +{ + unsigned int i; + struct nfs4_xattr_entry *entry; + struct nfs4_xattr_bucket *bucket; + struct hlist_node *n; + + nfs4_xattr_set_listcache(cache, ERR_PTR(-ESTALE)); + + for (i = 0; i < NFS4_XATTR_HASH_SIZE; i++) { + bucket = &cache->buckets[i]; + + spin_lock(&bucket->lock); + bucket->draining = true; + hlist_for_each_entry_safe(entry, n, &bucket->hlist, hnode) { + nfs4_xattr_entry_lru_del(entry); + hlist_del_init(&entry->hnode); + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + } + spin_unlock(&bucket->lock); + } + + atomic_long_set(&cache->nent, 0); + + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); +} + +/* + * Get a referenced copy of the cache structure. Avoid doing allocs + * while holding i_lock. Which means that we do some optimistic allocation, + * and might have to free the result in rare cases. + * + * This function only checks the NFS_INO_INVALID_XATTR cache validity bit + * and acts accordingly, replacing the cache when needed. For the read case + * (!add), this means that the caller must make sure that the cache + * is valid before caling this function. getxattr and listxattr call + * revalidate_inode to do this. The attribute cache timeout (for the + * non-delegated case) is expected to be dealt with in the revalidate + * call. + */ + +static struct nfs4_xattr_cache * +nfs4_xattr_get_cache(struct inode *inode, int add) +{ + struct nfs_inode *nfsi; + struct nfs4_xattr_cache *cache, *oldcache, *newcache; + + nfsi = NFS_I(inode); + + cache = oldcache = NULL; + + spin_lock(&inode->i_lock); + + if (nfsi->cache_validity & NFS_INO_INVALID_XATTR) + oldcache = nfs4_xattr_cache_unlink(inode); + else + cache = nfsi->xattr_cache; + + if (cache != NULL) + kref_get(&cache->ref); + + spin_unlock(&inode->i_lock); + + if (add && cache == NULL) { + newcache = NULL; + + cache = nfs4_xattr_alloc_cache(); + if (cache == NULL) + goto out; + + spin_lock(&inode->i_lock); + if (nfsi->cache_validity & NFS_INO_INVALID_XATTR) { + /* + * The cache was invalidated again. Give up, + * since what we want to enter is now likely + * outdated anyway. + */ + spin_unlock(&inode->i_lock); + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); + cache = NULL; + goto out; + } + + /* + * Check if someone beat us to it. + */ + if (nfsi->xattr_cache != NULL) { + newcache = nfsi->xattr_cache; + kref_get(&newcache->ref); + } else { + kref_get(&cache->ref); + nfsi->xattr_cache = cache; + cache->inode = inode; + list_lru_add_obj(&nfs4_xattr_cache_lru, &cache->lru); + } + + spin_unlock(&inode->i_lock); + + /* + * If there was a race, throw away the cache we just + * allocated, and use the new one allocated by someone + * else. + */ + if (newcache != NULL) { + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); + cache = newcache; + } + } + +out: + /* + * Discard the now orphaned old cache. + */ + if (oldcache != NULL) + nfs4_xattr_discard_cache(oldcache); + + return cache; +} + +static inline struct nfs4_xattr_bucket * +nfs4_xattr_hash_bucket(struct nfs4_xattr_cache *cache, const char *name) +{ + return &cache->buckets[jhash(name, strlen(name), 0) & + (ARRAY_SIZE(cache->buckets) - 1)]; +} + +static struct nfs4_xattr_entry * +nfs4_xattr_get_entry(struct nfs4_xattr_bucket *bucket, const char *name) +{ + struct nfs4_xattr_entry *entry; + + entry = NULL; + + hlist_for_each_entry(entry, &bucket->hlist, hnode) { + if (!strcmp(entry->xattr_name, name)) + break; + } + + return entry; +} + +static int +nfs4_xattr_hash_add(struct nfs4_xattr_cache *cache, + struct nfs4_xattr_entry *entry) +{ + struct nfs4_xattr_bucket *bucket; + struct nfs4_xattr_entry *oldentry = NULL; + int ret = 1; + + bucket = nfs4_xattr_hash_bucket(cache, entry->xattr_name); + entry->bucket = bucket; + + spin_lock(&bucket->lock); + + if (bucket->draining) { + ret = 0; + goto out; + } + + oldentry = nfs4_xattr_get_entry(bucket, entry->xattr_name); + if (oldentry != NULL) { + hlist_del_init(&oldentry->hnode); + nfs4_xattr_entry_lru_del(oldentry); + } else { + atomic_long_inc(&cache->nent); + } + + hlist_add_head(&entry->hnode, &bucket->hlist); + nfs4_xattr_entry_lru_add(entry); + +out: + spin_unlock(&bucket->lock); + + if (oldentry != NULL) + kref_put(&oldentry->ref, nfs4_xattr_free_entry_cb); + + return ret; +} + +static void +nfs4_xattr_hash_remove(struct nfs4_xattr_cache *cache, const char *name) +{ + struct nfs4_xattr_bucket *bucket; + struct nfs4_xattr_entry *entry; + + bucket = nfs4_xattr_hash_bucket(cache, name); + + spin_lock(&bucket->lock); + + entry = nfs4_xattr_get_entry(bucket, name); + if (entry != NULL) { + hlist_del_init(&entry->hnode); + nfs4_xattr_entry_lru_del(entry); + atomic_long_dec(&cache->nent); + } + + spin_unlock(&bucket->lock); + + if (entry != NULL) + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); +} + +static struct nfs4_xattr_entry * +nfs4_xattr_hash_find(struct nfs4_xattr_cache *cache, const char *name) +{ + struct nfs4_xattr_bucket *bucket; + struct nfs4_xattr_entry *entry; + + bucket = nfs4_xattr_hash_bucket(cache, name); + + spin_lock(&bucket->lock); + + entry = nfs4_xattr_get_entry(bucket, name); + if (entry != NULL) + kref_get(&entry->ref); + + spin_unlock(&bucket->lock); + + return entry; +} + +/* + * Entry point to retrieve an entry from the cache. + */ +ssize_t nfs4_xattr_cache_get(struct inode *inode, const char *name, char *buf, + ssize_t buflen) +{ + struct nfs4_xattr_cache *cache; + struct nfs4_xattr_entry *entry; + ssize_t ret; + + cache = nfs4_xattr_get_cache(inode, 0); + if (cache == NULL) + return -ENOENT; + + ret = 0; + entry = nfs4_xattr_hash_find(cache, name); + + if (entry != NULL) { + dprintk("%s: cache hit '%s', len %lu\n", __func__, + entry->xattr_name, (unsigned long)entry->xattr_size); + if (buflen == 0) { + /* Length probe only */ + ret = entry->xattr_size; + } else if (buflen < entry->xattr_size) + ret = -ERANGE; + else { + memcpy(buf, entry->xattr_value, entry->xattr_size); + ret = entry->xattr_size; + } + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + } else { + dprintk("%s: cache miss '%s'\n", __func__, name); + ret = -ENOENT; + } + + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); + + return ret; +} + +/* + * Retrieve a cached list of xattrs from the cache. + */ +ssize_t nfs4_xattr_cache_list(struct inode *inode, char *buf, ssize_t buflen) +{ + struct nfs4_xattr_cache *cache; + struct nfs4_xattr_entry *entry; + ssize_t ret; + + cache = nfs4_xattr_get_cache(inode, 0); + if (cache == NULL) + return -ENOENT; + + spin_lock(&cache->listxattr_lock); + + entry = cache->listxattr; + + if (entry != NULL && entry != ERR_PTR(-ESTALE)) { + if (buflen == 0) { + /* Length probe only */ + ret = entry->xattr_size; + } else if (entry->xattr_size > buflen) + ret = -ERANGE; + else { + memcpy(buf, entry->xattr_value, entry->xattr_size); + ret = entry->xattr_size; + } + } else { + ret = -ENOENT; + } + + spin_unlock(&cache->listxattr_lock); + + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); + + return ret; +} + +/* + * Add an xattr to the cache. + * + * This also invalidates the xattr list cache. + */ +void nfs4_xattr_cache_add(struct inode *inode, const char *name, + const char *buf, struct page **pages, ssize_t buflen) +{ + struct nfs4_xattr_cache *cache; + struct nfs4_xattr_entry *entry; + + dprintk("%s: add '%s' len %lu\n", __func__, + name, (unsigned long)buflen); + + cache = nfs4_xattr_get_cache(inode, 1); + if (cache == NULL) + return; + + entry = nfs4_xattr_alloc_entry(name, buf, pages, buflen); + if (entry == NULL) + goto out; + + (void)nfs4_xattr_set_listcache(cache, NULL); + + if (!nfs4_xattr_hash_add(cache, entry)) + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + +out: + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); +} + + +/* + * Remove an xattr from the cache. + * + * This also invalidates the xattr list cache. + */ +void nfs4_xattr_cache_remove(struct inode *inode, const char *name) +{ + struct nfs4_xattr_cache *cache; + + dprintk("%s: remove '%s'\n", __func__, name); + + cache = nfs4_xattr_get_cache(inode, 0); + if (cache == NULL) + return; + + (void)nfs4_xattr_set_listcache(cache, NULL); + nfs4_xattr_hash_remove(cache, name); + + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); +} + +/* + * Cache listxattr output, replacing any possible old one. + */ +void nfs4_xattr_cache_set_list(struct inode *inode, const char *buf, + ssize_t buflen) +{ + struct nfs4_xattr_cache *cache; + struct nfs4_xattr_entry *entry; + + cache = nfs4_xattr_get_cache(inode, 1); + if (cache == NULL) + return; + + entry = nfs4_xattr_alloc_entry(NULL, buf, NULL, buflen); + if (entry == NULL) + goto out; + + /* + * This is just there to be able to get to bucket->cache, + * which is obviously the same for all buckets, so just + * use bucket 0. + */ + entry->bucket = &cache->buckets[0]; + + if (!nfs4_xattr_set_listcache(cache, entry)) + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + +out: + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); +} + +/* + * Zap the entire cache. Called when an inode is evicted. + */ +void nfs4_xattr_cache_zap(struct inode *inode) +{ + struct nfs4_xattr_cache *oldcache; + + spin_lock(&inode->i_lock); + oldcache = nfs4_xattr_cache_unlink(inode); + spin_unlock(&inode->i_lock); + + if (oldcache) + nfs4_xattr_discard_cache(oldcache); +} + +/* + * The entry LRU is shrunk more aggressively than the cache LRU, + * by settings @seeks to 1. + * + * Cache structures are freed only when they've become empty, after + * pruning all but one entry. + */ + +static unsigned long nfs4_xattr_cache_count(struct shrinker *shrink, + struct shrink_control *sc); +static unsigned long nfs4_xattr_entry_count(struct shrinker *shrink, + struct shrink_control *sc); +static unsigned long nfs4_xattr_cache_scan(struct shrinker *shrink, + struct shrink_control *sc); +static unsigned long nfs4_xattr_entry_scan(struct shrinker *shrink, + struct shrink_control *sc); + +static struct shrinker *nfs4_xattr_cache_shrinker; +static struct shrinker *nfs4_xattr_entry_shrinker; +static struct shrinker *nfs4_xattr_large_entry_shrinker; + +static enum lru_status +cache_lru_isolate(struct list_head *item, + struct list_lru_one *lru, void *arg) +{ + struct list_head *dispose = arg; + struct inode *inode; + struct nfs4_xattr_cache *cache = container_of(item, + struct nfs4_xattr_cache, lru); + + if (atomic_long_read(&cache->nent) > 1) + return LRU_SKIP; + + /* + * If a cache structure is on the LRU list, we know that + * its inode is valid. Try to lock it to break the link. + * Since we're inverting the lock order here, only try. + */ + inode = cache->inode; + + if (!spin_trylock(&inode->i_lock)) + return LRU_SKIP; + + kref_get(&cache->ref); + + cache->inode = NULL; + NFS_I(inode)->xattr_cache = NULL; + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_XATTR; + list_lru_isolate(lru, &cache->lru); + + spin_unlock(&inode->i_lock); + + list_add_tail(&cache->dispose, dispose); + return LRU_REMOVED; +} + +static unsigned long +nfs4_xattr_cache_scan(struct shrinker *shrink, struct shrink_control *sc) +{ + LIST_HEAD(dispose); + unsigned long freed; + struct nfs4_xattr_cache *cache; + + freed = list_lru_shrink_walk(&nfs4_xattr_cache_lru, sc, + cache_lru_isolate, &dispose); + while (!list_empty(&dispose)) { + cache = list_first_entry(&dispose, struct nfs4_xattr_cache, + dispose); + list_del_init(&cache->dispose); + nfs4_xattr_discard_cache(cache); + kref_put(&cache->ref, nfs4_xattr_free_cache_cb); + } + + return freed; +} + + +static unsigned long +nfs4_xattr_cache_count(struct shrinker *shrink, struct shrink_control *sc) +{ + unsigned long count; + + count = list_lru_shrink_count(&nfs4_xattr_cache_lru, sc); + return vfs_pressure_ratio(count); +} + +static enum lru_status +entry_lru_isolate(struct list_head *item, + struct list_lru_one *lru, void *arg) +{ + struct list_head *dispose = arg; + struct nfs4_xattr_bucket *bucket; + struct nfs4_xattr_cache *cache; + struct nfs4_xattr_entry *entry = container_of(item, + struct nfs4_xattr_entry, lru); + + bucket = entry->bucket; + cache = bucket->cache; + + /* + * Unhook the entry from its parent (either a cache bucket + * or a cache structure if it's a listxattr buf), so that + * it's no longer found. Then add it to the isolate list, + * to be freed later. + * + * In both cases, we're reverting lock order, so use + * trylock and skip the entry if we can't get the lock. + */ + if (entry->xattr_name != NULL) { + /* Regular cache entry */ + if (!spin_trylock(&bucket->lock)) + return LRU_SKIP; + + kref_get(&entry->ref); + + hlist_del_init(&entry->hnode); + atomic_long_dec(&cache->nent); + list_lru_isolate(lru, &entry->lru); + + spin_unlock(&bucket->lock); + } else { + /* Listxattr cache entry */ + if (!spin_trylock(&cache->listxattr_lock)) + return LRU_SKIP; + + kref_get(&entry->ref); + + cache->listxattr = NULL; + list_lru_isolate(lru, &entry->lru); + + spin_unlock(&cache->listxattr_lock); + } + + list_add_tail(&entry->dispose, dispose); + return LRU_REMOVED; +} + +static unsigned long +nfs4_xattr_entry_scan(struct shrinker *shrink, struct shrink_control *sc) +{ + LIST_HEAD(dispose); + unsigned long freed; + struct nfs4_xattr_entry *entry; + struct list_lru *lru; + + lru = (shrink == nfs4_xattr_large_entry_shrinker) ? + &nfs4_xattr_large_entry_lru : &nfs4_xattr_entry_lru; + + freed = list_lru_shrink_walk(lru, sc, entry_lru_isolate, &dispose); + + while (!list_empty(&dispose)) { + entry = list_first_entry(&dispose, struct nfs4_xattr_entry, + dispose); + list_del_init(&entry->dispose); + + /* + * Drop two references: the one that we just grabbed + * in entry_lru_isolate, and the one that was set + * when the entry was first allocated. + */ + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + kref_put(&entry->ref, nfs4_xattr_free_entry_cb); + } + + return freed; +} + +static unsigned long +nfs4_xattr_entry_count(struct shrinker *shrink, struct shrink_control *sc) +{ + unsigned long count; + struct list_lru *lru; + + lru = (shrink == nfs4_xattr_large_entry_shrinker) ? + &nfs4_xattr_large_entry_lru : &nfs4_xattr_entry_lru; + + count = list_lru_shrink_count(lru, sc); + return vfs_pressure_ratio(count); +} + + +static void nfs4_xattr_cache_init_once(void *p) +{ + struct nfs4_xattr_cache *cache = p; + + spin_lock_init(&cache->listxattr_lock); + atomic_long_set(&cache->nent, 0); + nfs4_xattr_hash_init(cache); + cache->listxattr = NULL; + INIT_LIST_HEAD(&cache->lru); + INIT_LIST_HEAD(&cache->dispose); +} + +typedef unsigned long (*count_objects_cb)(struct shrinker *s, + struct shrink_control *sc); +typedef unsigned long (*scan_objects_cb)(struct shrinker *s, + struct shrink_control *sc); + +static int __init nfs4_xattr_shrinker_init(struct shrinker **shrinker, + struct list_lru *lru, const char *name, + count_objects_cb count, + scan_objects_cb scan, long batch, int seeks) +{ + int ret; + + *shrinker = shrinker_alloc(SHRINKER_MEMCG_AWARE, name); + if (!*shrinker) + return -ENOMEM; + + ret = list_lru_init_memcg(lru, *shrinker); + if (ret) { + shrinker_free(*shrinker); + return ret; + } + + (*shrinker)->count_objects = count; + (*shrinker)->scan_objects = scan; + (*shrinker)->batch = batch; + (*shrinker)->seeks = seeks; + + shrinker_register(*shrinker); + + return ret; +} + +static void nfs4_xattr_shrinker_destroy(struct shrinker *shrinker, + struct list_lru *lru) +{ + shrinker_free(shrinker); + list_lru_destroy(lru); +} + +int __init nfs4_xattr_cache_init(void) +{ + int ret = 0; + + nfs4_xattr_cache_cachep = kmem_cache_create("nfs4_xattr_cache_cache", + sizeof(struct nfs4_xattr_cache), 0, + (SLAB_RECLAIM_ACCOUNT), + nfs4_xattr_cache_init_once); + if (nfs4_xattr_cache_cachep == NULL) + return -ENOMEM; + + ret = nfs4_xattr_shrinker_init(&nfs4_xattr_cache_shrinker, + &nfs4_xattr_cache_lru, "nfs-xattr_cache", + nfs4_xattr_cache_count, + nfs4_xattr_cache_scan, 0, DEFAULT_SEEKS); + if (ret) + goto out1; + + ret = nfs4_xattr_shrinker_init(&nfs4_xattr_entry_shrinker, + &nfs4_xattr_entry_lru, "nfs-xattr_entry", + nfs4_xattr_entry_count, + nfs4_xattr_entry_scan, 512, DEFAULT_SEEKS); + if (ret) + goto out2; + + ret = nfs4_xattr_shrinker_init(&nfs4_xattr_large_entry_shrinker, + &nfs4_xattr_large_entry_lru, + "nfs-xattr_large_entry", + nfs4_xattr_entry_count, + nfs4_xattr_entry_scan, 512, 1); + if (!ret) + return 0; + + nfs4_xattr_shrinker_destroy(nfs4_xattr_entry_shrinker, + &nfs4_xattr_entry_lru); +out2: + nfs4_xattr_shrinker_destroy(nfs4_xattr_cache_shrinker, + &nfs4_xattr_cache_lru); +out1: + kmem_cache_destroy(nfs4_xattr_cache_cachep); + + return ret; +} + +void nfs4_xattr_cache_exit(void) +{ + nfs4_xattr_shrinker_destroy(nfs4_xattr_large_entry_shrinker, + &nfs4_xattr_large_entry_lru); + nfs4_xattr_shrinker_destroy(nfs4_xattr_entry_shrinker, + &nfs4_xattr_entry_lru); + nfs4_xattr_shrinker_destroy(nfs4_xattr_cache_shrinker, + &nfs4_xattr_cache_lru); + kmem_cache_destroy(nfs4_xattr_cache_cachep); +} diff --git a/fs/nfs/nfs42xdr.c b/fs/nfs/nfs42xdr.c new file mode 100644 index 000000000..ec105c62f --- /dev/null +++ b/fs/nfs/nfs42xdr.c @@ -0,0 +1,1830 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2014 Anna Schumaker + */ +#ifndef __LINUX_FS_NFS_NFS4_2XDR_H +#define __LINUX_FS_NFS_NFS4_2XDR_H + +#include "nfs42.h" + +/* Not limited by NFS itself, limited by the generic xattr code */ +#define nfs4_xattr_name_maxsz XDR_QUADLEN(XATTR_NAME_MAX) + +#define encode_fallocate_maxsz (encode_stateid_maxsz + \ + 2 /* offset */ + \ + 2 /* length */) +#define NFS42_WRITE_RES_SIZE (1 /* wr_callback_id size */ +\ + XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + 2 /* wr_count */ + \ + 1 /* wr_committed */ + \ + XDR_QUADLEN(NFS4_VERIFIER_SIZE)) +#define encode_allocate_maxsz (op_encode_hdr_maxsz + \ + encode_fallocate_maxsz) +#define decode_allocate_maxsz (op_decode_hdr_maxsz) +#define encode_copy_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + 2 + 2 + 2 + 1 + 1 + 1 +\ + 1 + /* One cnr_source_server */\ + 1 + /* nl4_type */ \ + 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT)) +#define decode_copy_maxsz (op_decode_hdr_maxsz + \ + NFS42_WRITE_RES_SIZE + \ + 1 /* cr_consecutive */ + \ + 1 /* cr_synchronous */) +#define encode_offload_cancel_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_offload_cancel_maxsz (op_decode_hdr_maxsz) +#define encode_offload_status_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_offload_status_maxsz (op_decode_hdr_maxsz + \ + 2 /* osr_count */ + \ + 2 /* osr_complete */) +#define encode_copy_notify_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + 1 + /* nl4_type */ \ + 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT)) +#define decode_copy_notify_maxsz (op_decode_hdr_maxsz + \ + 3 + /* cnr_lease_time */\ + XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + 1 + /* Support 1 cnr_source_server */\ + 1 + /* nl4_type */ \ + 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT)) +#define encode_deallocate_maxsz (op_encode_hdr_maxsz + \ + encode_fallocate_maxsz) +#define decode_deallocate_maxsz (op_decode_hdr_maxsz) +#define encode_read_plus_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define NFS42_READ_PLUS_DATA_SEGMENT_SIZE \ + (1 /* data_content4 */ + \ + 2 /* data_info4.di_offset */ + \ + 1 /* data_info4.di_length */) +#define NFS42_READ_PLUS_HOLE_SEGMENT_SIZE \ + (1 /* data_content4 */ + \ + 2 /* data_info4.di_offset */ + \ + 2 /* data_info4.di_length */) +#define READ_PLUS_SEGMENT_SIZE_DIFF (NFS42_READ_PLUS_HOLE_SEGMENT_SIZE - \ + NFS42_READ_PLUS_DATA_SEGMENT_SIZE) +#define decode_read_plus_maxsz (op_decode_hdr_maxsz + \ + 1 /* rpr_eof */ + \ + 1 /* rpr_contents count */ + \ + NFS42_READ_PLUS_HOLE_SEGMENT_SIZE) +#define encode_seek_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + 2 /* offset */ + \ + 1 /* whence */) +#define decode_seek_maxsz (op_decode_hdr_maxsz + \ + 1 /* eof */ + \ + 1 /* whence */ + \ + 2 /* offset */ + \ + 2 /* length */) +#define encode_io_info_maxsz 4 +#define encode_layoutstats_maxsz (op_decode_hdr_maxsz + \ + 2 /* offset */ + \ + 2 /* length */ + \ + encode_stateid_maxsz + \ + encode_io_info_maxsz + \ + encode_io_info_maxsz + \ + 1 /* opaque devaddr4 length */ + \ + XDR_QUADLEN(PNFS_LAYOUTSTATS_MAXSIZE)) +#define decode_layoutstats_maxsz (op_decode_hdr_maxsz) +#define encode_device_error_maxsz (XDR_QUADLEN(NFS4_DEVICEID4_SIZE) + \ + 1 /* status */ + 1 /* opnum */) +#define encode_layouterror_maxsz (op_decode_hdr_maxsz + \ + 2 /* offset */ + \ + 2 /* length */ + \ + encode_stateid_maxsz + \ + 1 /* Array size */ + \ + encode_device_error_maxsz) +#define decode_layouterror_maxsz (op_decode_hdr_maxsz) +#define encode_clone_maxsz (encode_stateid_maxsz + \ + encode_stateid_maxsz + \ + 2 /* src offset */ + \ + 2 /* dst offset */ + \ + 2 /* count */) +#define decode_clone_maxsz (op_decode_hdr_maxsz) +#define encode_getxattr_maxsz (op_encode_hdr_maxsz + 1 + \ + nfs4_xattr_name_maxsz) +#define decode_getxattr_maxsz (op_decode_hdr_maxsz + 1 + pagepad_maxsz) +#define encode_setxattr_maxsz (op_encode_hdr_maxsz + \ + 1 + nfs4_xattr_name_maxsz + 1) +#define decode_setxattr_maxsz (op_decode_hdr_maxsz + decode_change_info_maxsz) +#define encode_listxattrs_maxsz (op_encode_hdr_maxsz + 2 + 1) +#define decode_listxattrs_maxsz (op_decode_hdr_maxsz + 2 + 1 + 1 + 1) +#define encode_removexattr_maxsz (op_encode_hdr_maxsz + 1 + \ + nfs4_xattr_name_maxsz) +#define decode_removexattr_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz) + +#define NFS4_enc_allocate_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_allocate_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_allocate_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_allocate_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_copy_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_savefh_maxsz + \ + encode_putfh_maxsz + \ + encode_copy_maxsz + \ + encode_commit_maxsz) +#define NFS4_dec_copy_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_savefh_maxsz + \ + decode_putfh_maxsz + \ + decode_copy_maxsz + \ + decode_commit_maxsz) +#define NFS4_enc_offload_cancel_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_offload_cancel_maxsz) +#define NFS4_dec_offload_cancel_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_offload_cancel_maxsz) +#define NFS4_enc_offload_status_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_offload_status_maxsz) +#define NFS4_dec_offload_status_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_offload_status_maxsz) +#define NFS4_enc_copy_notify_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_copy_notify_maxsz) +#define NFS4_dec_copy_notify_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_copy_notify_maxsz) +#define NFS4_enc_deallocate_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_deallocate_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_deallocate_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_deallocate_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_zero_range_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_deallocate_maxsz + \ + encode_allocate_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_zero_range_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_deallocate_maxsz + \ + decode_allocate_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_read_plus_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_read_plus_maxsz) +#define NFS4_dec_read_plus_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_read_plus_maxsz) +#define NFS4_enc_seek_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_seek_maxsz) +#define NFS4_dec_seek_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_seek_maxsz) +#define NFS4_enc_layoutstats_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + PNFS_LAYOUTSTATS_MAXDEV * encode_layoutstats_maxsz) +#define NFS4_dec_layoutstats_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + PNFS_LAYOUTSTATS_MAXDEV * decode_layoutstats_maxsz) +#define NFS4_enc_layouterror_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + NFS42_LAYOUTERROR_MAX * \ + encode_layouterror_maxsz) +#define NFS4_dec_layouterror_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + NFS42_LAYOUTERROR_MAX * \ + decode_layouterror_maxsz) +#define NFS4_enc_clone_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_savefh_maxsz + \ + encode_putfh_maxsz + \ + encode_clone_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_clone_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_savefh_maxsz + \ + decode_putfh_maxsz + \ + decode_clone_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_getxattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_getxattr_maxsz) +#define NFS4_dec_getxattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_getxattr_maxsz) +#define NFS4_enc_setxattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_setxattr_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_setxattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_setxattr_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_listxattrs_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_listxattrs_maxsz) +#define NFS4_dec_listxattrs_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_listxattrs_maxsz) +#define NFS4_enc_removexattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_removexattr_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_removexattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_removexattr_maxsz + \ + decode_getattr_maxsz) + +/* + * These values specify the maximum amount of data that is not + * associated with the extended attribute name or extended + * attribute list in the SETXATTR, GETXATTR and LISTXATTR + * respectively. + */ +const u32 nfs42_maxsetxattr_overhead = ((RPC_MAX_HEADER_WITH_AUTH + + compound_encode_hdr_maxsz + + encode_sequence_maxsz + + encode_putfh_maxsz + 1 + + nfs4_xattr_name_maxsz) + * XDR_UNIT); + +const u32 nfs42_maxgetxattr_overhead = ((RPC_MAX_HEADER_WITH_AUTH + + compound_decode_hdr_maxsz + + decode_sequence_maxsz + + decode_putfh_maxsz + 1) * XDR_UNIT); + +const u32 nfs42_maxlistxattrs_overhead = ((RPC_MAX_HEADER_WITH_AUTH + + compound_decode_hdr_maxsz + + decode_sequence_maxsz + + decode_putfh_maxsz + 3) * XDR_UNIT); + +static void encode_fallocate(struct xdr_stream *xdr, + const struct nfs42_falloc_args *args) +{ + encode_nfs4_stateid(xdr, &args->falloc_stateid); + encode_uint64(xdr, args->falloc_offset); + encode_uint64(xdr, args->falloc_length); +} + +static void encode_allocate(struct xdr_stream *xdr, + const struct nfs42_falloc_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_ALLOCATE, decode_allocate_maxsz, hdr); + encode_fallocate(xdr, args); +} + +static void encode_nl4_server(struct xdr_stream *xdr, + const struct nl4_server *ns) +{ + encode_uint32(xdr, ns->nl4_type); + switch (ns->nl4_type) { + case NL4_NAME: + case NL4_URL: + encode_string(xdr, ns->u.nl4_str_sz, ns->u.nl4_str); + break; + case NL4_NETADDR: + encode_string(xdr, ns->u.nl4_addr.netid_len, + ns->u.nl4_addr.netid); + encode_string(xdr, ns->u.nl4_addr.addr_len, + ns->u.nl4_addr.addr); + break; + default: + WARN_ON_ONCE(1); + } +} + +static void encode_copy(struct xdr_stream *xdr, + const struct nfs42_copy_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_COPY, decode_copy_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->src_stateid); + encode_nfs4_stateid(xdr, &args->dst_stateid); + + encode_uint64(xdr, args->src_pos); + encode_uint64(xdr, args->dst_pos); + encode_uint64(xdr, args->count); + + encode_uint32(xdr, 1); /* consecutive = true */ + encode_uint32(xdr, args->sync); + if (args->cp_src == NULL) { /* intra-ssc */ + encode_uint32(xdr, 0); /* no src server list */ + return; + } + encode_uint32(xdr, 1); /* supporting 1 server */ + encode_nl4_server(xdr, args->cp_src); +} + +static void encode_copy_commit(struct xdr_stream *xdr, + const struct nfs42_copy_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_COMMIT, decode_commit_maxsz, hdr); + p = reserve_space(xdr, 12); + p = xdr_encode_hyper(p, args->dst_pos); + *p = cpu_to_be32(args->count); +} + +static void encode_offload_cancel(struct xdr_stream *xdr, + const struct nfs42_offload_status_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_OFFLOAD_CANCEL, decode_offload_cancel_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->osa_stateid); +} + +static void encode_offload_status(struct xdr_stream *xdr, + const struct nfs42_offload_status_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_OFFLOAD_STATUS, decode_offload_status_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->osa_stateid); +} + +static void encode_copy_notify(struct xdr_stream *xdr, + const struct nfs42_copy_notify_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_COPY_NOTIFY, decode_copy_notify_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->cna_src_stateid); + encode_nl4_server(xdr, &args->cna_dst); +} + +static void encode_deallocate(struct xdr_stream *xdr, + const struct nfs42_falloc_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_DEALLOCATE, decode_deallocate_maxsz, hdr); + encode_fallocate(xdr, args); +} + +static void encode_read_plus(struct xdr_stream *xdr, + const struct nfs_pgio_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_READ_PLUS, decode_read_plus_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->stateid); + encode_uint64(xdr, args->offset); + encode_uint32(xdr, args->count); +} + +static void encode_seek(struct xdr_stream *xdr, + const struct nfs42_seek_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SEEK, decode_seek_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->sa_stateid); + encode_uint64(xdr, args->sa_offset); + encode_uint32(xdr, args->sa_what); +} + +static void encode_layoutstats(struct xdr_stream *xdr, + const struct nfs42_layoutstat_args *args, + struct nfs42_layoutstat_devinfo *devinfo, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LAYOUTSTATS, decode_layoutstats_maxsz, hdr); + p = reserve_space(xdr, 8 + 8); + p = xdr_encode_hyper(p, devinfo->offset); + p = xdr_encode_hyper(p, devinfo->length); + encode_nfs4_stateid(xdr, &args->stateid); + p = reserve_space(xdr, 4*8 + NFS4_DEVICEID4_SIZE + 4); + p = xdr_encode_hyper(p, devinfo->read_count); + p = xdr_encode_hyper(p, devinfo->read_bytes); + p = xdr_encode_hyper(p, devinfo->write_count); + p = xdr_encode_hyper(p, devinfo->write_bytes); + p = xdr_encode_opaque_fixed(p, devinfo->dev_id.data, + NFS4_DEVICEID4_SIZE); + /* Encode layoutupdate4 */ + *p++ = cpu_to_be32(devinfo->layout_type); + if (devinfo->ld_private.ops) + devinfo->ld_private.ops->encode(xdr, args, + &devinfo->ld_private); + else + encode_uint32(xdr, 0); +} + +static void encode_clone(struct xdr_stream *xdr, + const struct nfs42_clone_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_CLONE, decode_clone_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->src_stateid); + encode_nfs4_stateid(xdr, &args->dst_stateid); + p = reserve_space(xdr, 3*8); + p = xdr_encode_hyper(p, args->src_offset); + p = xdr_encode_hyper(p, args->dst_offset); + xdr_encode_hyper(p, args->count); +} + +static void encode_device_error(struct xdr_stream *xdr, + const struct nfs42_device_error *error) +{ + __be32 *p; + + p = reserve_space(xdr, NFS4_DEVICEID4_SIZE + 2*4); + p = xdr_encode_opaque_fixed(p, error->dev_id.data, + NFS4_DEVICEID4_SIZE); + *p++ = cpu_to_be32(error->status); + *p = cpu_to_be32(error->opnum); +} + +static void encode_layouterror(struct xdr_stream *xdr, + const struct nfs42_layout_error *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LAYOUTERROR, decode_layouterror_maxsz, hdr); + p = reserve_space(xdr, 8 + 8); + p = xdr_encode_hyper(p, args->offset); + p = xdr_encode_hyper(p, args->length); + encode_nfs4_stateid(xdr, &args->stateid); + p = reserve_space(xdr, 4); + *p = cpu_to_be32(1); + encode_device_error(xdr, &args->errors[0]); +} + +static void encode_setxattr(struct xdr_stream *xdr, + const struct nfs42_setxattrargs *arg, + struct compound_hdr *hdr) +{ + __be32 *p; + + BUILD_BUG_ON(XATTR_CREATE != SETXATTR4_CREATE); + BUILD_BUG_ON(XATTR_REPLACE != SETXATTR4_REPLACE); + + encode_op_hdr(xdr, OP_SETXATTR, decode_setxattr_maxsz, hdr); + p = reserve_space(xdr, 4); + *p = cpu_to_be32(arg->xattr_flags); + encode_string(xdr, strlen(arg->xattr_name), arg->xattr_name); + p = reserve_space(xdr, 4); + *p = cpu_to_be32(arg->xattr_len); + if (arg->xattr_len) + xdr_write_pages(xdr, arg->xattr_pages, 0, arg->xattr_len); +} + +static void encode_getxattr(struct xdr_stream *xdr, const char *name, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_GETXATTR, decode_getxattr_maxsz, hdr); + encode_string(xdr, strlen(name), name); +} + +static void encode_removexattr(struct xdr_stream *xdr, const char *name, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_REMOVEXATTR, decode_removexattr_maxsz, hdr); + encode_string(xdr, strlen(name), name); +} + +static void encode_listxattrs(struct xdr_stream *xdr, + const struct nfs42_listxattrsargs *arg, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LISTXATTRS, decode_listxattrs_maxsz, hdr); + + p = reserve_space(xdr, 12); + if (unlikely(!p)) + return; + + p = xdr_encode_hyper(p, arg->cookie); + /* + * RFC 8276 says to specify the full max length of the LISTXATTRS + * XDR reply. Count is set to the XDR length of the names array + * plus the EOF marker. So, add the cookie and the names count. + */ + *p = cpu_to_be32(arg->count + 8 + 4); +} + +/* + * Encode ALLOCATE request + */ +static void nfs4_xdr_enc_allocate(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_falloc_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->falloc_fh, &hdr); + encode_allocate(xdr, args, &hdr); + encode_getfattr(xdr, args->falloc_bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode COPY request + */ +static void nfs4_xdr_enc_copy(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_copy_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->src_fh, &hdr); + encode_savefh(xdr, &hdr); + encode_putfh(xdr, args->dst_fh, &hdr); + encode_copy(xdr, args, &hdr); + if (args->sync) + encode_copy_commit(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode OFFLOAD_CANCEL request + */ +static void nfs4_xdr_enc_offload_cancel(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_offload_status_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->osa_seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->osa_seq_args, &hdr); + encode_putfh(xdr, args->osa_src_fh, &hdr); + encode_offload_cancel(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode OFFLOAD_STATUS request + */ +static void nfs4_xdr_enc_offload_status(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_offload_status_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->osa_seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->osa_seq_args, &hdr); + encode_putfh(xdr, args->osa_src_fh, &hdr); + encode_offload_status(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode COPY_NOTIFY request + */ +static void nfs4_xdr_enc_copy_notify(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_copy_notify_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->cna_seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->cna_seq_args, &hdr); + encode_putfh(xdr, args->cna_src_fh, &hdr); + encode_copy_notify(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode DEALLOCATE request + */ +static void nfs4_xdr_enc_deallocate(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_falloc_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->falloc_fh, &hdr); + encode_deallocate(xdr, args, &hdr); + encode_getfattr(xdr, args->falloc_bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode ZERO_RANGE request + */ +static void nfs4_xdr_enc_zero_range(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_falloc_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->falloc_fh, &hdr); + encode_deallocate(xdr, args, &hdr); + encode_allocate(xdr, args, &hdr); + encode_getfattr(xdr, args->falloc_bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode READ_PLUS request + */ +static void nfs4_xdr_enc_read_plus(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_read_plus(xdr, args, &hdr); + + rpc_prepare_reply_pages(req, args->pages, args->pgbase, args->count, + hdr.replen - READ_PLUS_SEGMENT_SIZE_DIFF); + encode_nops(&hdr); +} + +/* + * Encode SEEK request + */ +static void nfs4_xdr_enc_seek(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_seek_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->sa_fh, &hdr); + encode_seek(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LAYOUTSTATS request + */ +static void nfs4_xdr_enc_layoutstats(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_layoutstat_args *args = data; + int i; + + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + WARN_ON(args->num_dev > PNFS_LAYOUTSTATS_MAXDEV); + for (i = 0; i < args->num_dev; i++) + encode_layoutstats(xdr, args, &args->devinfo[i], &hdr); + encode_nops(&hdr); +} + +/* + * Encode CLONE request + */ +static void nfs4_xdr_enc_clone(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_clone_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->src_fh, &hdr); + encode_savefh(xdr, &hdr); + encode_putfh(xdr, args->dst_fh, &hdr); + encode_clone(xdr, args, &hdr); + encode_getfattr(xdr, args->dst_bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LAYOUTERROR request + */ +static void nfs4_xdr_enc_layouterror(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_layouterror_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + int i; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, NFS_FH(args->inode), &hdr); + for (i = 0; i < args->num_errors; i++) + encode_layouterror(xdr, &args->errors[i], &hdr); + encode_nops(&hdr); +} + +/* + * Encode SETXATTR request + */ +static void nfs4_xdr_enc_setxattr(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_setxattrargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_setxattr(xdr, args, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode GETXATTR request + */ +static void nfs4_xdr_enc_getxattr(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs42_getxattrargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + uint32_t replen; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + replen = hdr.replen + op_decode_hdr_maxsz + 1; + encode_getxattr(xdr, args->xattr_name, &hdr); + + rpc_prepare_reply_pages(req, args->xattr_pages, 0, args->xattr_len, + replen); + + encode_nops(&hdr); +} + +/* + * Encode LISTXATTR request + */ +static void nfs4_xdr_enc_listxattrs(struct rpc_rqst *req, + struct xdr_stream *xdr, const void *data) +{ + const struct nfs42_listxattrsargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + uint32_t replen; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + replen = hdr.replen + op_decode_hdr_maxsz + 2 + 1; + encode_listxattrs(xdr, args, &hdr); + + rpc_prepare_reply_pages(req, args->xattr_pages, 0, args->count, replen); + + encode_nops(&hdr); +} + +/* + * Encode REMOVEXATTR request + */ +static void nfs4_xdr_enc_removexattr(struct rpc_rqst *req, + struct xdr_stream *xdr, const void *data) +{ + const struct nfs42_removexattrargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_removexattr(xdr, args->xattr_name, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +static int decode_allocate(struct xdr_stream *xdr, struct nfs42_falloc_res *res) +{ + return decode_op_hdr(xdr, OP_ALLOCATE); +} + +static int decode_write_response(struct xdr_stream *xdr, + struct nfs42_write_res *res) +{ + __be32 *p; + int status, count; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + count = be32_to_cpup(p); + if (count > 1) + return -EREMOTEIO; + else if (count == 1) { + status = decode_opaque_fixed(xdr, &res->stateid, + NFS4_STATEID_SIZE); + if (unlikely(status)) + return -EIO; + } + p = xdr_inline_decode(xdr, 8 + 4); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &res->count); + res->verifier.committed = be32_to_cpup(p); + return decode_verifier(xdr, &res->verifier.verifier); +} + +static int decode_nl4_server(struct xdr_stream *xdr, struct nl4_server *ns) +{ + struct nfs42_netaddr *naddr; + uint32_t dummy; + char *dummy_str; + __be32 *p; + int status; + + /* nl_type */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + ns->nl4_type = be32_to_cpup(p); + switch (ns->nl4_type) { + case NL4_NAME: + case NL4_URL: + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + if (unlikely(dummy > NFS4_OPAQUE_LIMIT)) + return -EIO; + memcpy(&ns->u.nl4_str, dummy_str, dummy); + ns->u.nl4_str_sz = dummy; + break; + case NL4_NETADDR: + naddr = &ns->u.nl4_addr; + + /* netid string */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + if (unlikely(dummy > RPCBIND_MAXNETIDLEN)) + return -EIO; + naddr->netid_len = dummy; + memcpy(naddr->netid, dummy_str, naddr->netid_len); + + /* uaddr string */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + if (unlikely(dummy > RPCBIND_MAXUADDRLEN)) + return -EIO; + naddr->addr_len = dummy; + memcpy(naddr->addr, dummy_str, naddr->addr_len); + break; + default: + WARN_ON_ONCE(1); + return -EIO; + } + return 0; +} + +static int decode_copy_requirements(struct xdr_stream *xdr, + struct nfs42_copy_res *res) { + __be32 *p; + + p = xdr_inline_decode(xdr, 4 + 4); + if (unlikely(!p)) + return -EIO; + + res->consecutive = be32_to_cpup(p++); + res->synchronous = be32_to_cpup(p++); + return 0; +} + +static int decode_copy(struct xdr_stream *xdr, struct nfs42_copy_res *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_COPY); + if (status == NFS4ERR_OFFLOAD_NO_REQS) { + status = decode_copy_requirements(xdr, res); + if (status) + return status; + return NFS4ERR_OFFLOAD_NO_REQS; + } else if (status) + return status; + + status = decode_write_response(xdr, &res->write_res); + if (status) + return status; + + return decode_copy_requirements(xdr, res); +} + +static int decode_offload_cancel(struct xdr_stream *xdr, + struct nfs42_offload_status_res *res) +{ + return decode_op_hdr(xdr, OP_OFFLOAD_CANCEL); +} + +static int decode_offload_status(struct xdr_stream *xdr, + struct nfs42_offload_status_res *res) +{ + ssize_t result; + int status; + + status = decode_op_hdr(xdr, OP_OFFLOAD_STATUS); + if (status) + return status; + /* osr_count */ + if (xdr_stream_decode_u64(xdr, &res->osr_count) < 0) + return -EIO; + /* osr_complete<1> */ + result = xdr_stream_decode_uint32_array(xdr, &res->osr_complete, 1); + if (result < 0) + return -EIO; + res->complete_count = result; + return 0; +} + +static int decode_copy_notify(struct xdr_stream *xdr, + struct nfs42_copy_notify_res *res) +{ + __be32 *p; + int status, count; + + status = decode_op_hdr(xdr, OP_COPY_NOTIFY); + if (status) + return status; + /* cnr_lease_time */ + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &res->cnr_lease_time.seconds); + res->cnr_lease_time.nseconds = be32_to_cpup(p); + + status = decode_opaque_fixed(xdr, &res->cnr_stateid, NFS4_STATEID_SIZE); + if (unlikely(status)) + return -EIO; + + /* number of source addresses */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + count = be32_to_cpup(p); + if (count > 1) + pr_warn("NFS: %s: nsvr %d > Supported. Use first servers\n", + __func__, count); + + status = decode_nl4_server(xdr, &res->cnr_src); + if (unlikely(status)) + return -EIO; + return 0; +} + +static int decode_deallocate(struct xdr_stream *xdr, struct nfs42_falloc_res *res) +{ + return decode_op_hdr(xdr, OP_DEALLOCATE); +} + +struct read_plus_segment { + enum data_content4 type; + uint64_t offset; + union { + struct { + uint64_t length; + } hole; + + struct { + uint32_t length; + unsigned int from; + } data; + }; +}; + +static inline uint64_t read_plus_segment_length(struct read_plus_segment *seg) +{ + return seg->type == NFS4_CONTENT_DATA ? seg->data.length : seg->hole.length; +} + +static int decode_read_plus_segment(struct xdr_stream *xdr, + struct read_plus_segment *seg) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (!p) + return -EIO; + seg->type = be32_to_cpup(p++); + + p = xdr_inline_decode(xdr, seg->type == NFS4_CONTENT_DATA ? 12 : 16); + if (!p) + return -EIO; + p = xdr_decode_hyper(p, &seg->offset); + + if (seg->type == NFS4_CONTENT_DATA) { + struct xdr_buf buf; + uint32_t len = be32_to_cpup(p); + + seg->data.length = len; + seg->data.from = xdr_stream_pos(xdr); + + if (!xdr_stream_subsegment(xdr, &buf, xdr_align_size(len))) + return -EIO; + } else if (seg->type == NFS4_CONTENT_HOLE) { + xdr_decode_hyper(p, &seg->hole.length); + } else + return -EINVAL; + return 0; +} + +static int process_read_plus_segment(struct xdr_stream *xdr, + struct nfs_pgio_args *args, + struct nfs_pgio_res *res, + struct read_plus_segment *seg) +{ + unsigned long offset = seg->offset; + unsigned long length = read_plus_segment_length(seg); + unsigned int bufpos; + + if (offset + length < args->offset) + return 0; + else if (offset > args->offset + args->count) { + res->eof = 0; + return 0; + } else if (offset < args->offset) { + length -= (args->offset - offset); + offset = args->offset; + } else if (offset + length > args->offset + args->count) { + length = (args->offset + args->count) - offset; + res->eof = 0; + } + + bufpos = xdr->buf->head[0].iov_len + (offset - args->offset); + if (seg->type == NFS4_CONTENT_HOLE) + return xdr_stream_zero(xdr, bufpos, length); + else + return xdr_stream_move_subsegment(xdr, seg->data.from, bufpos, length); +} + +static int decode_read_plus(struct xdr_stream *xdr, struct nfs_pgio_res *res) +{ + struct nfs_pgio_header *hdr = + container_of(res, struct nfs_pgio_header, res); + struct nfs_pgio_args *args = &hdr->args; + uint32_t segments; + struct read_plus_segment *segs; + int status, i; + __be32 *p; + + status = decode_op_hdr(xdr, OP_READ_PLUS); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4 + 4); + if (unlikely(!p)) + return -EIO; + + res->count = 0; + res->eof = be32_to_cpup(p++); + segments = be32_to_cpup(p++); + if (segments == 0) + return 0; + + segs = kmalloc_objs(*segs, segments); + if (!segs) + return -ENOMEM; + + for (i = 0; i < segments; i++) { + status = decode_read_plus_segment(xdr, &segs[i]); + if (status < 0) + goto out; + } + + xdr_set_pagelen(xdr, xdr_align_size(args->count)); + for (i = segments; i > 0; i--) + res->count += process_read_plus_segment(xdr, args, res, &segs[i-1]); + status = 0; + +out: + kfree(segs); + return status; +} + +static int decode_seek(struct xdr_stream *xdr, struct nfs42_seek_res *res) +{ + int status; + __be32 *p; + + status = decode_op_hdr(xdr, OP_SEEK); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4 + 8); + if (unlikely(!p)) + return -EIO; + + res->sr_eof = be32_to_cpup(p++); + p = xdr_decode_hyper(p, &res->sr_offset); + return 0; +} + +static int decode_layoutstats(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_LAYOUTSTATS); +} + +static int decode_clone(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_CLONE); +} + +static int decode_layouterror(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_LAYOUTERROR); +} + +static int decode_setxattr(struct xdr_stream *xdr, + struct nfs4_change_info *cinfo) +{ + int status; + + status = decode_op_hdr(xdr, OP_SETXATTR); + if (status) + goto out; + status = decode_change_info(xdr, cinfo); +out: + return status; +} + +static int decode_getxattr(struct xdr_stream *xdr, + struct nfs42_getxattrres *res, + struct rpc_rqst *req) +{ + int status; + __be32 *p; + u32 len, rdlen; + + status = decode_op_hdr(xdr, OP_GETXATTR); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + len = be32_to_cpup(p); + + /* + * Only check against the page length here. The actual + * requested length may be smaller, but that is only + * checked against after possibly caching a valid reply. + */ + if (len > req->rq_rcv_buf.page_len) + return -ERANGE; + + res->xattr_len = len; + + if (len > 0) { + rdlen = xdr_read_pages(xdr, len); + if (rdlen < len) + return -EIO; + } + + return 0; +} + +static int decode_removexattr(struct xdr_stream *xdr, + struct nfs4_change_info *cinfo) +{ + int status; + + status = decode_op_hdr(xdr, OP_REMOVEXATTR); + if (status) + goto out; + + status = decode_change_info(xdr, cinfo); +out: + return status; +} + +static int decode_listxattrs(struct xdr_stream *xdr, + struct nfs42_listxattrsres *res) +{ + int status; + __be32 *p; + u32 count, len, ulen; + size_t left, copied; + char *buf; + + status = decode_op_hdr(xdr, OP_LISTXATTRS); + if (status) { + /* + * Special case: for LISTXATTRS, NFS4ERR_TOOSMALL + * should be translated to ERANGE. + */ + if (status == -ETOOSMALL) + status = -ERANGE; + /* + * Special case: for LISTXATTRS, NFS4ERR_NOXATTR + * should be translated to success with zero-length reply. + */ + if (status == -ENODATA) { + res->eof = true; + status = 0; + } + goto out; + } + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + + xdr_decode_hyper(p, &res->cookie); + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + left = res->xattr_len; + buf = res->xattr_buf; + + count = be32_to_cpup(p); + copied = 0; + + /* + * We have asked for enough room to encode the maximum number + * of possible attribute names, so everything should fit. + * + * But, don't rely on that assumption. Just decode entries + * until they don't fit anymore, just in case the server did + * something odd. + */ + while (count--) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + len = be32_to_cpup(p); + if (len > (XATTR_NAME_MAX - XATTR_USER_PREFIX_LEN)) { + status = -ERANGE; + goto out; + } + + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + + ulen = len + XATTR_USER_PREFIX_LEN + 1; + if (buf) { + if (ulen > left) { + status = -ERANGE; + goto out; + } + + memcpy(buf, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN); + memcpy(buf + XATTR_USER_PREFIX_LEN, p, len); + + buf[ulen - 1] = 0; + buf += ulen; + left -= ulen; + } + copied += ulen; + } + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + res->eof = be32_to_cpup(p); + res->copied = copied; + +out: + if (status == -ERANGE && res->xattr_len == XATTR_LIST_MAX) + status = -E2BIG; + + return status; +} + +/* + * Decode ALLOCATE request + */ +static int nfs4_xdr_dec_allocate(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_falloc_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_allocate(xdr, res); + if (status) + goto out; + decode_getfattr(xdr, res->falloc_fattr, res->falloc_server); +out: + return status; +} + +/* + * Decode COPY response + */ +static int nfs4_xdr_dec_copy(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_copy_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_savefh(xdr); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_copy(xdr, res); + if (status) + goto out; + if (res->commit_res.verf) + status = decode_commit(xdr, &res->commit_res); +out: + return status; +} + +/* + * Decode OFFLOAD_CANCEL response + */ +static int nfs4_xdr_dec_offload_cancel(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_offload_status_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->osr_seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_offload_cancel(xdr, res); + +out: + return status; +} + +/* + * Decode OFFLOAD_STATUS response + */ +static int nfs4_xdr_dec_offload_status(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_offload_status_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->osr_seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_offload_status(xdr, res); + +out: + return status; +} + +/* + * Decode COPY_NOTIFY response + */ +static int nfs4_xdr_dec_copy_notify(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_copy_notify_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->cnr_seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_copy_notify(xdr, res); + +out: + return status; +} + +/* + * Decode DEALLOCATE request + */ +static int nfs4_xdr_dec_deallocate(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_falloc_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_deallocate(xdr, res); + if (status) + goto out; + decode_getfattr(xdr, res->falloc_fattr, res->falloc_server); +out: + return status; +} + +/* + * Decode ZERO_RANGE request + */ +static int nfs4_xdr_dec_zero_range(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_falloc_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_deallocate(xdr, res); + if (status) + goto out; + status = decode_allocate(xdr, res); + if (status) + goto out; + decode_getfattr(xdr, res->falloc_fattr, res->falloc_server); +out: + return status; +} + +/* + * Decode READ_PLUS request + */ +static int nfs4_xdr_dec_read_plus(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *res = data; + struct compound_hdr hdr; + int status; + + xdr_set_scratch_buffer(xdr, res->scratch, READ_PLUS_SCRATCH_SIZE); + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_read_plus(xdr, res); + if (!status) + status = res->count; +out: + return status; +} + +/* + * Decode SEEK request + */ +static int nfs4_xdr_dec_seek(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_seek_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_seek(xdr, res); +out: + return status; +} + +/* + * Decode LAYOUTSTATS request + */ +static int nfs4_xdr_dec_layoutstats(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_layoutstat_res *res = data; + struct compound_hdr hdr; + int status, i; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + WARN_ON(res->num_dev > PNFS_LAYOUTSTATS_MAXDEV); + for (i = 0; i < res->num_dev; i++) { + status = decode_layoutstats(xdr); + if (status) + goto out; + } +out: + res->rpc_status = status; + return status; +} + +/* + * Decode CLONE request + */ +static int nfs4_xdr_dec_clone(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_clone_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_savefh(xdr); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_clone(xdr); + if (status) + goto out; + decode_getfattr(xdr, res->dst_fattr, res->server); +out: + res->rpc_status = status; + return status; +} + +/* + * Decode LAYOUTERROR request + */ +static int nfs4_xdr_dec_layouterror(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs42_layouterror_res *res = data; + struct compound_hdr hdr; + int status, i; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + + for (i = 0; i < res->num_errors && status == 0; i++) + status = decode_layouterror(xdr); +out: + res->rpc_status = status; + return status; +} + +/* + * Decode SETXATTR request + */ +static int nfs4_xdr_dec_setxattr(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs42_setxattrres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, req); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_setxattr(xdr, &res->cinfo); + if (status) + goto out; + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode GETXATTR request + */ +static int nfs4_xdr_dec_getxattr(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, void *data) +{ + struct nfs42_getxattrres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_getxattr(xdr, res, rqstp); +out: + return status; +} + +/* + * Decode LISTXATTR request + */ +static int nfs4_xdr_dec_listxattrs(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, void *data) +{ + struct nfs42_listxattrsres *res = data; + struct compound_hdr hdr; + int status; + + xdr_set_scratch_folio(xdr, res->scratch); + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_listxattrs(xdr, res); +out: + return status; +} + +/* + * Decode REMOVEXATTR request + */ +static int nfs4_xdr_dec_removexattr(struct rpc_rqst *req, + struct xdr_stream *xdr, void *data) +{ + struct nfs42_removexattrres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, req); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + + status = decode_removexattr(xdr, &res->cinfo); + if (status) + goto out; + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} +#endif /* __LINUX_FS_NFS_NFS4_2XDR_H */ diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h new file mode 100644 index 000000000..b48e5b87c --- /dev/null +++ b/fs/nfs/nfs4_fs.h @@ -0,0 +1,682 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * linux/fs/nfs/nfs4_fs.h + * + * Copyright (C) 2005 Trond Myklebust + * + * NFSv4-specific filesystem definitions and declarations + */ + +#ifndef __LINUX_FS_NFS_NFS4_FS_H +#define __LINUX_FS_NFS_NFS4_FS_H + +#if defined(CONFIG_NFS_V4_0) +#define NFS4_MIN_MINOR_VERSION 0 +#else +#define NFS4_MIN_MINOR_VERSION 1 +#endif + +#if defined(CONFIG_NFS_V4_2) +#define NFS4_MAX_MINOR_VERSION 2 +#else +#define NFS4_MAX_MINOR_VERSION 1 +#endif + +#if IS_ENABLED(CONFIG_NFS_V4) + +#define NFS4_MAX_LOOP_ON_RECOVER (10) + +#include +#include + +struct idmap; + +enum nfs4_client_state { + NFS4CLNT_MANAGER_RUNNING = 0, + NFS4CLNT_CHECK_LEASE, + NFS4CLNT_LEASE_EXPIRED, + NFS4CLNT_RECLAIM_REBOOT, + NFS4CLNT_RECLAIM_NOGRACE, + NFS4CLNT_DELEGRETURN, + NFS4CLNT_SESSION_RESET, + NFS4CLNT_LEASE_CONFIRM, + NFS4CLNT_SERVER_SCOPE_MISMATCH, + NFS4CLNT_PURGE_STATE, + NFS4CLNT_BIND_CONN_TO_SESSION, + NFS4CLNT_MOVED, + NFS4CLNT_LEASE_MOVED, + NFS4CLNT_DELEGATION_EXPIRED, + NFS4CLNT_RUN_MANAGER, + NFS4CLNT_MANAGER_AVAILABLE, + NFS4CLNT_RECALL_RUNNING, + NFS4CLNT_RECALL_ANY_LAYOUT_READ, + NFS4CLNT_RECALL_ANY_LAYOUT_RW, + NFS4CLNT_DELEGRETURN_DELAYED, +}; + +#define NFS4_RENEW_TIMEOUT 0x01 +#define NFS4_RENEW_DELEGATION_CB 0x02 + +struct nfs_seqid_counter; +struct nfs4_minor_version_ops { + u32 minor_version; + unsigned init_caps; + + int (*init_client)(struct nfs_client *); + void (*shutdown_client)(struct nfs_client *); + bool (*match_stateid)(const nfs4_stateid *, + const nfs4_stateid *); + int (*find_root_sec)(struct nfs_server *, struct nfs_fh *, + struct nfs_fattr *); + void (*free_lock_state)(struct nfs_server *, + struct nfs4_lock_state *); + int (*test_and_free_expired)(struct nfs_server *, + nfs4_stateid *, const struct cred *); + struct nfs_seqid * + (*alloc_seqid)(struct nfs_seqid_counter *, gfp_t); + void (*session_trunk)(struct rpc_clnt *clnt, + struct rpc_xprt *xprt, void *data); + const struct rpc_call_ops *call_sync_ops; + const struct nfs4_sequence_slot_ops *sequence_slot_ops; + const struct nfs4_state_recovery_ops *reboot_recovery_ops; + const struct nfs4_state_recovery_ops *nograce_recovery_ops; + const struct nfs4_state_maintenance_ops *state_renewal_ops; + const struct nfs4_mig_recovery_ops *mig_recovery_ops; +}; + +#define NFS_SEQID_CONFIRMED 1 +struct nfs_seqid_counter { + ktime_t create_time; + u64 owner_id; + int flags; + u32 counter; + spinlock_t lock; /* Protects the list */ + struct list_head list; /* Defines sequence of RPC calls */ + struct rpc_wait_queue wait; /* RPC call delay queue */ +}; + +struct nfs_seqid { + struct nfs_seqid_counter *sequence; + struct list_head list; + struct rpc_task *task; +}; + +static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status) +{ + if (seqid_mutating_err(-status)) + seqid->flags |= NFS_SEQID_CONFIRMED; +} + +/* + * NFS4 state_owners and lock_owners are simply labels for ordered + * sequences of RPC calls. Their sole purpose is to provide once-only + * semantics by allowing the server to identify replayed requests. + */ +struct nfs4_state_owner { + struct nfs_server *so_server; + struct list_head so_lru; + unsigned long so_expires; + struct rb_node so_server_node; + + const struct cred *so_cred; /* Associated cred */ + + spinlock_t so_lock; + atomic_t so_count; + unsigned long so_flags; + struct list_head so_states; + struct nfs_seqid_counter so_seqid; + struct mutex so_delegreturn_mutex; +}; + +enum { + NFS_OWNER_RECLAIM_REBOOT, + NFS_OWNER_RECLAIM_NOGRACE +}; + +#define NFS_LOCK_NEW 0 +#define NFS_LOCK_RECLAIM 1 +#define NFS_LOCK_EXPIRED 2 + +/* + * struct nfs4_state maintains the client-side state for a given + * (state_owner,inode) tuple (OPEN) or state_owner (LOCK). + * + * OPEN: + * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server, + * we need to know how many files are open for reading or writing on a + * given inode. This information too is stored here. + * + * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN) + */ + +struct nfs4_lock_state { + struct list_head ls_locks; /* Other lock stateids */ + struct nfs4_state * ls_state; /* Pointer to open state */ +#define NFS_LOCK_INITIALIZED 0 +#define NFS_LOCK_LOST 1 +#define NFS_LOCK_UNLOCKING 2 + unsigned long ls_flags; + struct nfs_seqid_counter ls_seqid; + nfs4_stateid ls_stateid; + refcount_t ls_count; + fl_owner_t ls_owner; +}; + +/* bits for nfs4_state->flags */ +enum { + LK_STATE_IN_USE, + NFS_DELEGATED_STATE, /* Current stateid is delegation */ + NFS_OPEN_STATE, /* OPEN stateid is set */ + NFS_O_RDONLY_STATE, /* OPEN stateid has read-only state */ + NFS_O_WRONLY_STATE, /* OPEN stateid has write-only state */ + NFS_O_RDWR_STATE, /* OPEN stateid has read/write state */ + NFS_STATE_RECLAIM_REBOOT, /* OPEN stateid server rebooted */ + NFS_STATE_RECLAIM_NOGRACE, /* OPEN stateid needs to recover state */ + NFS_STATE_POSIX_LOCKS, /* Posix locks are supported */ + NFS_STATE_RECOVERY_FAILED, /* OPEN stateid state recovery failed */ + NFS_STATE_MAY_NOTIFY_LOCK, /* server may CB_NOTIFY_LOCK */ + NFS_STATE_CHANGE_WAIT, /* A state changing operation is outstanding */ + NFS_CLNT_DST_SSC_COPY_STATE, /* dst server open state on client*/ + NFS_CLNT_SRC_SSC_COPY_STATE, /* src server open state on client*/ + NFS_SRV_SSC_COPY_STATE, /* ssc state on the dst server */ +}; + +struct nfs4_state { + struct list_head open_states; /* List of states for the same state_owner */ + struct list_head inode_states; /* List of states for the same inode */ + struct list_head lock_states; /* List of subservient lock stateids */ + + struct nfs4_state_owner *owner; /* Pointer to the open owner */ + struct inode *inode; /* Pointer to the inode */ + + unsigned long flags; /* Do we hold any locks? */ + spinlock_t state_lock; /* Protects the lock_states list */ + + seqlock_t seqlock; /* Protects the stateid/open_stateid */ + nfs4_stateid stateid; /* Current stateid: may be delegation */ + nfs4_stateid open_stateid; /* OPEN stateid */ + + /* The following 3 fields are protected by owner->so_lock */ + unsigned int n_rdonly; /* Number of read-only references */ + unsigned int n_wronly; /* Number of write-only references */ + unsigned int n_rdwr; /* Number of read/write references */ + fmode_t state; /* State on the server (R,W, or RW) */ + refcount_t count; + + wait_queue_head_t waitq; + struct rcu_head rcu_head; +}; + + +struct nfs4_exception { + struct nfs4_state *state; + struct inode *inode; + nfs4_stateid *stateid; + long timeout; + unsigned short retrans; + unsigned char task_is_privileged : 1; + unsigned char delay : 1, + recovering : 1, + retry : 1; + bool interruptible; +}; + +struct nfs4_state_recovery_ops { + int owner_flag_bit; + int state_flag_bit; + int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); + int (*recover_lock)(struct nfs4_state *, struct file_lock *); + int (*establish_clid)(struct nfs_client *, const struct cred *); + int (*reclaim_complete)(struct nfs_client *, const struct cred *); + int (*detect_trunking)(struct nfs_client *, struct nfs_client **, + const struct cred *); +}; + +struct nfs4_opendata { + struct kref kref; + struct nfs_openargs o_arg; + struct nfs_openres o_res; + struct nfs_open_confirmargs c_arg; + struct nfs_open_confirmres c_res; + struct nfs4_string owner_name; + struct nfs4_string group_name; + struct nfs4_label *a_label; + struct nfs_fattr f_attr; + struct dentry *dir; + struct dentry *dentry; + struct nfs4_state_owner *owner; + struct nfs4_state *state; + struct iattr attrs; + struct nfs4_layoutget *lgp; + unsigned long timestamp; + bool rpc_done; + bool file_created; + bool is_recover; + bool cancelled; + int rpc_status; +}; + +struct nfs4_add_xprt_data { + struct nfs_client *clp; + const struct cred *cred; +}; + +struct nfs4_sequence_slot_ops { + int (*process)(struct rpc_task *, struct nfs4_sequence_res *); + int (*done)(struct rpc_task *, struct nfs4_sequence_res *); + void (*free_slot)(struct nfs4_sequence_res *); +}; + +struct nfs4_state_maintenance_ops { + int (*sched_state_renewal)(struct nfs_client *, const struct cred *, unsigned); + const struct cred * (*get_state_renewal_cred)(struct nfs_client *); + int (*renew_lease)(struct nfs_client *, const struct cred *); +}; + +struct nfs4_mig_recovery_ops { + int (*get_locations)(struct nfs_server *, struct nfs_fh *, + struct nfs4_fs_locations *, struct page *, const struct cred *); + int (*fsid_present)(struct inode *, const struct cred *); +}; + +extern const struct dentry_operations nfs4_dentry_operations; + +/* dir.c */ +int nfs_atomic_open(struct inode *, struct dentry *, struct file *, + unsigned, umode_t); + +/* fs_context.c */ +extern struct file_system_type nfs4_fs_type; + +/* nfs4client.c */ +struct nfs_net; +int nfs4_match_client(struct nfs_client *pos, struct nfs_client *new, + struct nfs_client **prev, struct nfs_net *nn); + +/* nfs4namespace.c */ +struct rpc_clnt *nfs4_negotiate_security(struct rpc_clnt *, struct inode *, + const struct qstr *); +int nfs4_submount(struct fs_context *, struct nfs_server *); +int nfs4_replace_transport(struct nfs_server *server, + const struct nfs4_fs_locations *locations); +size_t nfs_parse_server_name(char *string, size_t len, struct sockaddr_storage *ss, + size_t salen, struct net *net, int port); +/* nfs4proc.c */ +struct nfs4_call_sync_data { + const struct nfs_server *seq_server; + struct nfs4_sequence_args *seq_args; + struct nfs4_sequence_res *seq_res; +}; + +extern int nfs4_handle_exception(struct nfs_server *, int, struct nfs4_exception *); +extern int nfs4_async_handle_error(struct rpc_task *task, + struct nfs_server *server, + struct nfs4_state *state, long *timeout); +extern int nfs4_call_sync(struct rpc_clnt *, struct nfs_server *, + struct rpc_message *, struct nfs4_sequence_args *, + struct nfs4_sequence_res *, int); +extern int nfs4_call_sync_sequence(struct rpc_clnt *clnt, + struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res); +extern void nfs4_init_sequence(struct nfs_client *clp, struct nfs4_sequence_args *, + struct nfs4_sequence_res *, int, int); +extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, const struct cred *, struct nfs4_setclientid_res *); +extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, const struct cred *); +extern void renew_lease(const struct nfs_server *server, unsigned long timestamp); +extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, + struct nfs_fattr *, bool); +extern int nfs4_proc_bind_conn_to_session(struct nfs_client *, const struct cred *cred); +extern int nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cred); +extern int nfs4_destroy_clientid(struct nfs_client *clp); +extern int nfs4_init_clientid(struct nfs_client *, const struct cred *); +extern int nfs41_init_clientid(struct nfs_client *, const struct cred *); +extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait); +extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle); +extern int nfs4_proc_fs_locations(struct rpc_clnt *, struct inode *, const struct qstr *, + struct nfs4_fs_locations *, struct page *); +extern int nfs4_proc_get_locations(struct nfs_server *, struct nfs_fh *, + struct nfs4_fs_locations *, + struct page *page, const struct cred *); +extern int nfs4_proc_fsid_present(struct inode *, const struct cred *); +extern struct rpc_clnt *nfs4_proc_lookup_mountpoint(struct inode *, + struct dentry *, + struct nfs_fh *, + struct nfs_fattr *); +extern int nfs4_proc_secinfo(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *); +extern const struct xattr_handler * const nfs4_xattr_handlers[]; +extern int nfs4_set_rw_stateid(nfs4_stateid *stateid, + const struct nfs_open_context *ctx, + const struct nfs_lock_context *l_ctx, + fmode_t fmode); +extern void nfs4_bitmask_set(__u32 bitmask[], const __u32 src[], + struct inode *inode, unsigned long cache_validity); +extern int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode); +extern int update_open_stateid(struct nfs4_state *state, + const nfs4_stateid *open_stateid, + const nfs4_stateid *deleg_stateid, + fmode_t fmode); +extern int nfs4_proc_setlease(struct file *file, int arg, + struct file_lease **lease, void **priv); +extern int nfs4_proc_get_lease_time(struct nfs_client *clp, + struct nfs_fsinfo *fsinfo); +extern void nfs4_update_changeattr(struct inode *dir, + struct nfs4_change_info *cinfo, + unsigned long timestamp, + unsigned long cache_validity); +extern int nfs4_buf_to_pages_noslab(const void *buf, size_t buflen, + struct page **pages); +extern int nfs4_open_reclaim(struct nfs4_state_owner *, struct nfs4_state *); +extern int nfs4_open_expired(struct nfs4_state_owner *, struct nfs4_state *); +extern int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request); +extern int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request); +extern void nfs_state_clear_delegation(struct nfs4_state *state); +extern void nfs_finish_clear_delegation_stateid(struct nfs4_state *state, + const nfs4_stateid *stateid); +extern void nfs_state_clear_open_state_flags(struct nfs4_state *state); +extern void do_renew_lease(struct nfs_client *clp, unsigned long timestamp); +extern bool nfs4_match_stateid(const nfs4_stateid *s1, const nfs4_stateid *s2); +extern int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr); + +extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *); +extern int nfs4_proc_create_session(struct nfs_client *, const struct cred *); +extern int nfs4_proc_destroy_session(struct nfs4_session *, const struct cred *); +extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, + bool sync); +extern int nfs4_detect_session_trunking(struct nfs_client *clp, + struct nfs41_exchange_id_res *res, struct rpc_xprt *xprt); + +static inline bool +is_ds_only_client(struct nfs_client *clp) +{ + return (clp->cl_exchange_flags & EXCHGID4_FLAG_MASK_PNFS) == + EXCHGID4_FLAG_USE_PNFS_DS; +} + +static inline bool +is_ds_client(struct nfs_client *clp) +{ + return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS; +} + +static inline bool +_nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode, + struct rpc_clnt **clntp, struct rpc_message *msg) +{ + rpc_authflavor_t flavor; + + if (sp4_mode == NFS_SP4_MACH_CRED_CLEANUP || + sp4_mode == NFS_SP4_MACH_CRED_PNFS_CLEANUP) { + /* Using machine creds for cleanup operations + * is only relevent if the client credentials + * might expire. So don't bother for + * RPC_AUTH_UNIX. If file was only exported to + * sec=sys, the PUTFH would fail anyway. + */ + if ((*clntp)->cl_auth->au_flavor == RPC_AUTH_UNIX) + return false; + } + if (test_bit(sp4_mode, &clp->cl_sp4_flags)) { + msg->rpc_cred = rpc_machine_cred(); + + flavor = clp->cl_rpcclient->cl_auth->au_flavor; + WARN_ON_ONCE(flavor != RPC_AUTH_GSS_KRB5I && + flavor != RPC_AUTH_GSS_KRB5P); + *clntp = clp->cl_rpcclient; + + return true; + } + return false; +} + +/* + * Function responsible for determining if an rpc_message should use the + * machine cred under SP4_MACH_CRED and if so switching the credential and + * authflavor (using the nfs_client's rpc_clnt which will be krb5i/p). + * Should be called before rpc_call_sync/rpc_call_async. + */ +static inline void +nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode, + struct rpc_clnt **clntp, struct rpc_message *msg) +{ + _nfs4_state_protect(clp, sp4_mode, clntp, msg); +} + +/* + * Special wrapper to nfs4_state_protect for write. + * If WRITE can use machine cred but COMMIT cannot, make sure all writes + * that use machine cred use NFS_FILE_SYNC. + */ +static inline void +nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp, + struct rpc_message *msg, struct nfs_pgio_header *hdr) +{ + if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) && + !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags)) + hdr->args.stable = NFS_FILE_SYNC; +} + +extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[]; + +extern const u32 nfs4_fattr_bitmap[3]; +extern const u32 nfs4_statfs_bitmap[3]; +extern const u32 nfs4_pathconf_bitmap[3]; +extern const u32 nfs4_fsinfo_bitmap[3]; +extern const u32 nfs4_fs_locations_bitmap[3]; + +void nfs41_shutdown_client(struct nfs_client *); +int nfs41_init_client(struct nfs_client *); +void nfs4_free_client(struct nfs_client *); + +struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *); + +/* nfs4renewd.c */ +extern void nfs4_schedule_state_renewal(struct nfs_client *); +extern void nfs4_kill_renewd(struct nfs_client *); +extern void nfs4_renew_state(struct work_struct *); +extern void nfs4_set_lease_period(struct nfs_client *clp, u32 period); + +/* nfs4state.c */ +extern const nfs4_stateid current_stateid; + +const struct cred *nfs4_get_clid_cred(struct nfs_client *clp); +const struct cred *nfs4_get_machine_cred(struct nfs_client *clp); +const struct cred *nfs4_get_renew_cred(struct nfs_client *clp); +int nfs4_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **); +int nfs40_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **, const struct cred *); +int nfs41_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **, const struct cred *); +extern void nfs4_schedule_session_recovery(struct nfs4_session *, int); +extern void nfs41_notify_server(struct nfs_client *); +bool nfs4_check_serverowner_major_id(struct nfs41_server_owner *o1, + struct nfs41_server_owner *o2); + +extern struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *, const struct cred *, gfp_t); +extern void nfs4_put_state_owner(struct nfs4_state_owner *); +extern void nfs4_purge_state_owners(struct nfs_server *, struct list_head *); +extern void nfs4_free_state_owners(struct list_head *head); +extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); +extern void nfs4_put_open_state(struct nfs4_state *); +extern void nfs4_close_state(struct nfs4_state *, fmode_t); +extern void nfs4_close_sync(struct nfs4_state *, fmode_t); +extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t); +extern void nfs_inode_find_state_and_recover(struct inode *inode, + const nfs4_stateid *stateid); +extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *, struct nfs4_state *); +extern void nfs4_schedule_lease_recovery(struct nfs_client *); +extern int nfs4_wait_clnt_recover(struct nfs_client *clp); +extern int nfs4_client_recover_expired_lease(struct nfs_client *clp); +extern void nfs4_schedule_state_manager(struct nfs_client *); +extern void nfs4_schedule_path_down_recovery(struct nfs_client *clp); +extern int nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *); +extern int nfs4_schedule_migration_recovery(const struct nfs_server *); +extern void nfs4_schedule_lease_moved_recovery(struct nfs_client *); +extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags, bool); +extern void nfs41_handle_server_scope(struct nfs_client *, + struct nfs41_server_scope **); +extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp); +extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl); +extern int nfs4_select_rw_stateid(struct nfs4_state *, fmode_t, + const struct nfs_lock_context *, nfs4_stateid *, + const struct cred **); +extern bool nfs4_copy_open_stateid(nfs4_stateid *dst, + struct nfs4_state *state); + +extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask); +extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task); +extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid); +extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid); +extern void nfs_release_seqid(struct nfs_seqid *seqid); +extern void nfs_free_seqid(struct nfs_seqid *seqid); +extern int nfs4_setup_sequence(struct nfs_client *client, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct rpc_task *task); +extern int nfs4_sequence_done(struct rpc_task *task, + struct nfs4_sequence_res *res); + +extern void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp); +extern int nfs4_proc_commit(struct file *dst, __u64 offset, __u32 count, struct nfs_commitres *res); +extern const nfs4_stateid zero_stateid; +extern const nfs4_stateid invalid_stateid; + +/* nfs4super.c */ +struct nfs_mount_info; +extern struct nfs_subversion nfs_v4; +extern bool nfs4_disable_idmapping; +extern unsigned short max_session_slots; +extern unsigned short max_session_cb_slots; +extern unsigned short send_implementation_id; +extern bool recover_lost_locks; +extern short nfs_delay_retrans; + +#define NFS4_CLIENT_ID_UNIQ_LEN (64) +extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN]; + +extern int nfs4_try_get_tree(struct fs_context *); +extern int nfs4_get_referral_tree(struct fs_context *); + +/* nfs4sysctl.c */ +#ifdef CONFIG_SYSCTL +int nfs4_register_sysctl(void); +void nfs4_unregister_sysctl(void); +#else +static inline int nfs4_register_sysctl(void) +{ + return 0; +} + +static inline void nfs4_unregister_sysctl(void) +{ +} +#endif + +/* nfs4xdr.c */ +extern const struct rpc_procinfo nfs4_procedures[]; + +#ifdef CONFIG_NFS_V4_2 +extern const u32 nfs42_maxsetxattr_overhead; +extern const u32 nfs42_maxgetxattr_overhead; +extern const u32 nfs42_maxlistxattrs_overhead; +#endif + +struct nfs4_mount_data; + +/* callback_xdr.c */ +extern const struct svc_version nfs4_callback_version1; +extern const struct svc_version nfs4_callback_version4; + +static inline void nfs4_stateid_copy(nfs4_stateid *dst, const nfs4_stateid *src) +{ + memcpy(dst->data, src->data, sizeof(dst->data)); + dst->type = src->type; +} + +static inline bool nfs4_stateid_match(const nfs4_stateid *dst, const nfs4_stateid *src) +{ + if (dst->type != src->type) + return false; + return memcmp(dst->data, src->data, sizeof(dst->data)) == 0; +} + +static inline bool nfs4_stateid_match_other(const nfs4_stateid *dst, const nfs4_stateid *src) +{ + return memcmp(dst->other, src->other, NFS4_STATEID_OTHER_SIZE) == 0; +} + +static inline bool nfs4_stateid_is_newer(const nfs4_stateid *s1, const nfs4_stateid *s2) +{ + return (s32)(be32_to_cpu(s1->seqid) - be32_to_cpu(s2->seqid)) > 0; +} + +static inline bool nfs4_stateid_is_next(const nfs4_stateid *s1, const nfs4_stateid *s2) +{ + u32 seq1 = be32_to_cpu(s1->seqid); + u32 seq2 = be32_to_cpu(s2->seqid); + + return seq2 == seq1 + 1U || (seq2 == 1U && seq1 == 0xffffffffU); +} + +static inline bool nfs4_stateid_match_or_older(const nfs4_stateid *dst, const nfs4_stateid *src) +{ + return nfs4_stateid_match_other(dst, src) && + !(src->seqid && nfs4_stateid_is_newer(dst, src)); +} + +static inline void nfs4_stateid_seqid_inc(nfs4_stateid *s1) +{ + u32 seqid = be32_to_cpu(s1->seqid); + + if (++seqid == 0) + ++seqid; + s1->seqid = cpu_to_be32(seqid); +} + +static inline bool nfs4_valid_open_stateid(const struct nfs4_state *state) +{ + return test_bit(NFS_STATE_RECOVERY_FAILED, &state->flags) == 0; +} + +static inline bool nfs4_state_match_open_stateid_other(const struct nfs4_state *state, + const nfs4_stateid *stateid) +{ + return test_bit(NFS_OPEN_STATE, &state->flags) && + nfs4_stateid_match_other(&state->open_stateid, stateid); +} + +/* nfs42xattr.c */ +#ifdef CONFIG_NFS_V4_2 +extern int __init nfs4_xattr_cache_init(void); +extern void nfs4_xattr_cache_exit(void); +extern void nfs4_xattr_cache_add(struct inode *inode, const char *name, + const char *buf, struct page **pages, + ssize_t buflen); +extern void nfs4_xattr_cache_remove(struct inode *inode, const char *name); +extern ssize_t nfs4_xattr_cache_get(struct inode *inode, const char *name, + char *buf, ssize_t buflen); +extern void nfs4_xattr_cache_set_list(struct inode *inode, const char *buf, + ssize_t buflen); +extern ssize_t nfs4_xattr_cache_list(struct inode *inode, char *buf, + ssize_t buflen); +extern void nfs4_xattr_cache_zap(struct inode *inode); +#else +static inline void nfs4_xattr_cache_zap(struct inode *inode) +{ +} +#endif /* CONFIG_NFS_V4_2 */ + +#else /* CONFIG_NFS_V4 */ + +#define nfs4_close_state(a, b) do { } while (0) +#define nfs4_close_sync(a, b) do { } while (0) +#define nfs4_state_protect(a, b, c, d) do { } while (0) +#define nfs4_state_protect_write(a, b, c, d) do { } while (0) + + +#endif /* CONFIG_NFS_V4 */ +#endif /* __LINUX_FS_NFS_NFS4_FS.H */ diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c new file mode 100644 index 000000000..71c271a17 --- /dev/null +++ b/fs/nfs/nfs4client.c @@ -0,0 +1,1203 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" +#include "callback.h" +#include "delegation.h" +#include "nfs4session.h" +#include "nfs4idmap.h" +#include "pnfs.h" +#include "netns.h" +#include "sysfs.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +/* + * Get a unique NFSv4.0 callback identifier which will be used + * by the V4.0 callback service to lookup the nfs_client struct + */ +static int nfs_get_cb_ident_idr(struct nfs_client *clp, int minorversion) +{ + int ret = 0; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + + if (clp->rpc_ops->version != 4 || minorversion != 0) + return ret; + idr_preload(GFP_KERNEL); + spin_lock(&nn->nfs_client_lock); + ret = idr_alloc(&nn->cb_ident_idr, clp, 1, 0, GFP_NOWAIT); + if (ret >= 0) + clp->cl_cb_ident = ret; + spin_unlock(&nn->nfs_client_lock); + idr_preload_end(); + return ret < 0 ? ret : 0; +} + +/* + * Per auth flavor data server rpc clients + */ +struct nfs4_ds_server { + struct list_head list; /* ds_clp->cl_ds_clients */ + struct rpc_clnt *rpc_clnt; +}; + +/** + * nfs4_find_ds_client - Common lookup case for DS I/O + * @ds_clp: pointer to the DS's nfs_client + * @flavor: rpc auth flavour to match + */ +static struct nfs4_ds_server * +nfs4_find_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor) +{ + struct nfs4_ds_server *dss; + + rcu_read_lock(); + list_for_each_entry_rcu(dss, &ds_clp->cl_ds_clients, list) { + if (dss->rpc_clnt->cl_auth->au_flavor != flavor) + continue; + goto out; + } + dss = NULL; +out: + rcu_read_unlock(); + return dss; +} + +static struct nfs4_ds_server * +nfs4_add_ds_client(struct nfs_client *ds_clp, rpc_authflavor_t flavor, + struct nfs4_ds_server *new) +{ + struct nfs4_ds_server *dss; + + spin_lock(&ds_clp->cl_lock); + list_for_each_entry(dss, &ds_clp->cl_ds_clients, list) { + if (dss->rpc_clnt->cl_auth->au_flavor != flavor) + continue; + goto out; + } + if (new) + list_add_rcu(&new->list, &ds_clp->cl_ds_clients); + dss = new; +out: + spin_unlock(&ds_clp->cl_lock); /* need some lock to protect list */ + return dss; +} + +static struct nfs4_ds_server * +nfs4_alloc_ds_server(struct nfs_client *ds_clp, rpc_authflavor_t flavor) +{ + struct nfs4_ds_server *dss; + + dss = kmalloc_obj(*dss, GFP_NOFS); + if (dss == NULL) + return ERR_PTR(-ENOMEM); + + dss->rpc_clnt = rpc_clone_client_set_auth(ds_clp->cl_rpcclient, flavor); + if (IS_ERR(dss->rpc_clnt)) { + int err = PTR_ERR(dss->rpc_clnt); + kfree (dss); + return ERR_PTR(err); + } + INIT_LIST_HEAD(&dss->list); + + return dss; +} + +static void +nfs4_free_ds_server(struct nfs4_ds_server *dss) +{ + rpc_release_client(dss->rpc_clnt); + kfree(dss); +} + +/** + * nfs4_find_or_create_ds_client - Find or create a DS rpc client + * @ds_clp: pointer to the DS's nfs_client + * @inode: pointer to the inode + * + * Find or create a DS rpc client with th MDS server rpc client auth flavor + * in the nfs_client cl_ds_clients list. + */ +struct rpc_clnt * +nfs4_find_or_create_ds_client(struct nfs_client *ds_clp, struct inode *inode) +{ + struct nfs4_ds_server *dss, *new; + rpc_authflavor_t flavor = NFS_SERVER(inode)->client->cl_auth->au_flavor; + + dss = nfs4_find_ds_client(ds_clp, flavor); + if (dss != NULL) + goto out; + new = nfs4_alloc_ds_server(ds_clp, flavor); + if (IS_ERR(new)) + return ERR_CAST(new); + dss = nfs4_add_ds_client(ds_clp, flavor, new); + if (dss != new) + nfs4_free_ds_server(new); +out: + return dss->rpc_clnt; +} +EXPORT_SYMBOL_GPL(nfs4_find_or_create_ds_client); + +static void +nfs4_shutdown_ds_clients(struct nfs_client *clp) +{ + struct nfs4_ds_server *dss; + + while (!list_empty(&clp->cl_ds_clients)) { + dss = list_entry(clp->cl_ds_clients.next, + struct nfs4_ds_server, list); + list_del(&dss->list); + rpc_shutdown_client(dss->rpc_clnt); + kfree (dss); + } +} + +static void +nfs4_cleanup_callback(struct nfs_client *clp) +{ + struct nfs4_copy_state *cp_state; + + while (!list_empty(&clp->pending_cb_stateids)) { + cp_state = list_entry(clp->pending_cb_stateids.next, + struct nfs4_copy_state, copies); + list_del(&cp_state->copies); + kfree(cp_state); + } +} + +void nfs41_shutdown_client(struct nfs_client *clp) +{ + if (nfs4_has_session(clp)) { + nfs4_cleanup_callback(clp); + nfs4_shutdown_ds_clients(clp); + nfs4_destroy_session(clp->cl_session); + nfs4_destroy_clientid(clp); + } + +} + +struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init) +{ + char buf[INET6_ADDRSTRLEN + 1]; + const char *ip_addr = cl_init->ip_addr; + struct nfs_client *clp = nfs_alloc_client(cl_init); + int err; + + if (IS_ERR(clp)) + return clp; + + err = nfs_get_cb_ident_idr(clp, cl_init->minorversion); + if (err) + goto error; + + if (cl_init->minorversion < NFS4_MIN_MINOR_VERSION || + cl_init->minorversion > NFS4_MAX_MINOR_VERSION) { + err = -EINVAL; + goto error; + } + + spin_lock_init(&clp->cl_lock); + INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state); + INIT_LIST_HEAD(&clp->cl_ds_clients); + rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); + clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; + clp->cl_mvops = nfs_v4_minor_ops[cl_init->minorversion]; + clp->cl_mig_gen = 1; + clp->cl_last_renewal = jiffies; + init_waitqueue_head(&clp->cl_lock_waitq); + INIT_LIST_HEAD(&clp->pending_cb_stateids); + + if (cl_init->minorversion != 0) + __set_bit(NFS_CS_INFINITE_SLOTS, &clp->cl_flags); + __set_bit(NFS_CS_DISCRTRY, &clp->cl_flags); + __set_bit(NFS_CS_NO_RETRANS_TIMEOUT, &clp->cl_flags); + if (test_bit(NFS_CS_PNFS, &cl_init->init_flags)) + __set_bit(NFS_CS_PNFS, &clp->cl_flags); + if (test_bit(NFS_CS_NETUNREACH_FATAL, &cl_init->init_flags)) + __set_bit(NFS_CS_NETUNREACH_FATAL, &clp->cl_flags); + /* + * Set up the connection to the server before we add add to the + * global list. + */ + err = nfs_create_rpc_client(clp, cl_init, RPC_AUTH_GSS_KRB5I); + if (err == -EINVAL) + err = nfs_create_rpc_client(clp, cl_init, RPC_AUTH_UNIX); + if (err < 0) + goto error; + + /* If no clientaddr= option was specified, find a usable cb address */ + if (ip_addr == NULL) { + struct sockaddr_storage cb_addr; + struct sockaddr *sap = (struct sockaddr *)&cb_addr; + + err = rpc_localaddr(clp->cl_rpcclient, sap, sizeof(cb_addr)); + if (err < 0) + goto error; + err = rpc_ntop(sap, buf, sizeof(buf)); + if (err < 0) + goto error; + ip_addr = (const char *)buf; + } + strscpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr)); + + err = nfs_idmap_new(clp); + if (err < 0) { + dprintk("%s: failed to create idmapper. Error = %d\n", + __func__, err); + goto error; + } + __set_bit(NFS_CS_IDMAP, &clp->cl_res_state); + return clp; + +error: + nfs_free_client(clp); + return ERR_PTR(err); +} + +/* + * Destroy the NFS4 callback service + */ +static void nfs4_destroy_callback(struct nfs_client *clp) +{ + if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state)) { + struct rpc_xprt *xprt; + + xprt = rcu_dereference_raw(clp->cl_rpcclient->cl_xprt); + nfs_callback_down(clp->cl_mvops->minor_version, clp->cl_net, + xprt); + } +} + +static void nfs4_shutdown_client(struct nfs_client *clp) +{ + if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state)) + nfs4_kill_renewd(clp); + clp->cl_mvops->shutdown_client(clp); + nfs4_destroy_callback(clp); + if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state)) + nfs_idmap_delete(clp); + + rpc_destroy_wait_queue(&clp->cl_rpcwaitq); + kfree(clp->cl_serverowner); + kfree(clp->cl_serverscope); + kfree(clp->cl_implid); + kfree(clp->cl_owner_id); +} + +void nfs4_free_client(struct nfs_client *clp) +{ + nfs4_shutdown_client(clp); + nfs_free_client(clp); +} + +/* + * Initialize the NFS4 callback service + */ +static int nfs4_init_callback(struct nfs_client *clp) +{ + struct rpc_xprt *xprt; + int error; + + xprt = rcu_dereference_raw(clp->cl_rpcclient->cl_xprt); + + if (nfs4_has_session(clp)) { + error = xprt_setup_backchannel(xprt, NFS41_BC_MIN_CALLBACKS); + if (error < 0) + return error; + } + + error = nfs_callback_up(clp->cl_mvops->minor_version, xprt); + if (error < 0) { + dprintk("%s: failed to start callback. Error = %d\n", + __func__, error); + return error; + } + __set_bit(NFS_CS_CALLBACK, &clp->cl_res_state); + + return 0; +} + +/** + * nfs41_init_client - nfs_client initialization tasks for NFSv4.1+ + * @clp: nfs_client to initialize + * + * Returns zero on success, or a negative errno if some error occurred. + */ +int nfs41_init_client(struct nfs_client *clp) +{ + struct nfs4_session *session = NULL; + + /* + * Create the session and mark it expired. + * When a SEQUENCE operation encounters the expired session + * it will do session recovery to initialize it. + */ + session = nfs4_alloc_session(clp); + if (!session) + return -ENOMEM; + + clp->cl_session = session; + + /* + * The create session reply races with the server back + * channel probe. Mark the client NFS_CS_SESSION_INITING + * so that the client back channel can find the + * nfs_client struct + */ + nfs_mark_client_ready(clp, NFS_CS_SESSION_INITING); + return 0; +} + +/* + * Initialize the minor version specific parts of an NFS4 client record + */ +static int nfs4_init_client_minor_version(struct nfs_client *clp) +{ + int ret; + + ret = clp->cl_mvops->init_client(clp); + if (ret) + return ret; + return nfs4_init_callback(clp); +} + +static void nfs4_add_trunk(struct nfs_client *clp, struct nfs_client *old) +{ + struct sockaddr_storage clp_addr, old_addr; + struct sockaddr *clp_sap = (struct sockaddr *)&clp_addr; + struct sockaddr *old_sap = (struct sockaddr *)&old_addr; + size_t clp_salen; + struct xprt_create xprt_args = { + .ident = old->cl_proto, + .net = old->cl_net, + .servername = old->cl_hostname, + }; + int max_connect = test_bit(NFS_CS_PNFS, &clp->cl_flags) ? + clp->cl_max_connect : old->cl_max_connect; + + if (clp->cl_proto != old->cl_proto) + return; + clp_salen = rpc_peeraddr(clp->cl_rpcclient, clp_sap, sizeof(clp_addr)); + rpc_peeraddr(old->cl_rpcclient, old_sap, sizeof(old_addr)); + + if (clp_addr.ss_family != old_addr.ss_family) + return; + + xprt_args.dstaddr = clp_sap; + xprt_args.addrlen = clp_salen; + + rpc_clnt_add_xprt(old->cl_rpcclient, &xprt_args, + rpc_clnt_test_and_add_xprt, &max_connect); +} + +/** + * nfs4_init_client - Initialise an NFS4 client record + * + * @clp: nfs_client to initialise + * @cl_init: pointer to nfs_client_initdata + * + * Returns pointer to an NFS client, or an ERR_PTR value. + */ +struct nfs_client *nfs4_init_client(struct nfs_client *clp, + const struct nfs_client_initdata *cl_init) +{ + struct nfs_client *old; + int error; + + if (clp->cl_cons_state == NFS_CS_READY) + /* the client is initialised already */ + return clp; + + error = nfs4_init_client_minor_version(clp); + if (error < 0) + goto error; + + error = nfs4_discover_server_trunking(clp, &old); + if (error < 0) + goto error; + + if (clp != old) { + clp->cl_preserve_clid = true; + /* + * Mark the client as having failed initialization so other + * processes walking the nfs_client_list in nfs_match_client() + * won't try to use it. + */ + nfs_mark_client_ready(clp, -EPERM); + if (old->cl_mvops->session_trunk) + nfs4_add_trunk(clp, old); + } + clear_bit(NFS_CS_TSM_POSSIBLE, &clp->cl_flags); + nfs_put_client(clp); + return old; + +error: + nfs_mark_client_ready(clp, error); + nfs_put_client(clp); + return ERR_PTR(error); +} + +static bool nfs4_match_client_owner_id(const struct nfs_client *clp1, + const struct nfs_client *clp2) +{ + if (clp1->cl_owner_id == NULL || clp2->cl_owner_id == NULL) + return true; + return strcmp(clp1->cl_owner_id, clp2->cl_owner_id) == 0; +} + +int nfs4_match_client(struct nfs_client *pos, struct nfs_client *new, + struct nfs_client **prev, struct nfs_net *nn) +{ + int status; + + if (pos->rpc_ops != new->rpc_ops) + return 1; + + if (pos->cl_minorversion != new->cl_minorversion) + return 1; + + /* If "pos" isn't marked ready, we can't trust the + * remaining fields in "pos", especially the client + * ID and serverowner fields. Wait for CREATE_SESSION + * to finish. */ + if (pos->cl_cons_state > NFS_CS_READY) { + refcount_inc(&pos->cl_count); + spin_unlock(&nn->nfs_client_lock); + + nfs_put_client(*prev); + *prev = pos; + + status = nfs_wait_client_init_complete(pos); + spin_lock(&nn->nfs_client_lock); + + if (status < 0) + return status; + } + + if (pos->cl_cons_state != NFS_CS_READY) + return 1; + + if (pos->cl_clientid != new->cl_clientid) + return 1; + + /* NFSv4.1 always uses the uniform string, however someone + * might switch the uniquifier string on us. + */ + if (!nfs4_match_client_owner_id(pos, new)) + return 1; + + return 0; +} + +/* + * Returns true if the server major ids match + */ +bool +nfs4_check_serverowner_major_id(struct nfs41_server_owner *o1, + struct nfs41_server_owner *o2) +{ + if (o1->major_id_sz != o2->major_id_sz) + return false; + return memcmp(o1->major_id, o2->major_id, o1->major_id_sz) == 0; +} + +/* + * Returns true if the server scopes match + */ +static bool +nfs4_check_server_scope(struct nfs41_server_scope *s1, + struct nfs41_server_scope *s2) +{ + if (s1->server_scope_sz != s2->server_scope_sz) + return false; + return memcmp(s1->server_scope, s2->server_scope, + s1->server_scope_sz) == 0; +} + +/** + * nfs4_detect_session_trunking - Checks for session trunking. + * @clp: original mount nfs_client + * @res: result structure from an exchange_id using the original mount + * nfs_client with a new multi_addr transport + * @xprt: pointer to the transport to add. + * + * Called after a successful EXCHANGE_ID on a multi-addr connection. + * Upon success, add the transport. + * + * Returns zero on success, otherwise -EINVAL + * + * Note: since the exchange_id for the new multi_addr transport uses the + * same nfs_client from the original mount, the cl_owner_id is reused, + * so eir_clientowner is the same. + */ +int nfs4_detect_session_trunking(struct nfs_client *clp, + struct nfs41_exchange_id_res *res, + struct rpc_xprt *xprt) +{ + /* Check eir_clientid */ + if (clp->cl_clientid != res->clientid) + goto out_err; + + /* Check eir_server_owner so_major_id */ + if (!nfs4_check_serverowner_major_id(clp->cl_serverowner, + res->server_owner)) + goto out_err; + + /* Check eir_server_owner so_minor_id */ + if (clp->cl_serverowner->minor_id != res->server_owner->minor_id) + goto out_err; + + /* Check eir_server_scope */ + if (!nfs4_check_server_scope(clp->cl_serverscope, res->server_scope)) + goto out_err; + + pr_info("NFS: %s: Session trunking succeeded for %s\n", + clp->cl_hostname, + xprt->address_strings[RPC_DISPLAY_ADDR]); + + return 0; +out_err: + pr_info("NFS: %s: Session trunking failed for %s\n", clp->cl_hostname, + xprt->address_strings[RPC_DISPLAY_ADDR]); + + return -EINVAL; +} + +/** + * nfs41_walk_client_list - Find nfs_client that matches a client/server owner + * + * @new: nfs_client with client ID to test + * @result: OUT: found nfs_client, or new + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in "result." + * + * NB: nfs41_walk_client_list() relies on the new nfs_client being + * the last nfs_client on the list. + */ +int nfs41_walk_client_list(struct nfs_client *new, + struct nfs_client **result, + const struct cred *cred) +{ + struct nfs_net *nn = net_generic(new->cl_net, nfs_net_id); + struct nfs_client *pos, *prev = NULL; + int status = -NFS4ERR_STALE_CLIENTID; + + spin_lock(&nn->nfs_client_lock); + list_for_each_entry(pos, &nn->nfs_client_list, cl_share_link) { + + if (pos == new) + goto found; + + status = nfs4_match_client(pos, new, &prev, nn); + if (status < 0) + goto out; + if (status != 0) + continue; + + /* + * Note that session trunking is just a special subcase of + * client id trunking. In either case, we want to fall back + * to using the existing nfs_client. + */ + if (!nfs4_check_serverowner_major_id(pos->cl_serverowner, + new->cl_serverowner)) + continue; + +found: + refcount_inc(&pos->cl_count); + *result = pos; + status = 0; + break; + } + +out: + spin_unlock(&nn->nfs_client_lock); + nfs_put_client(prev); + return status; +} + +static void nfs4_destroy_server(struct nfs_server *server) +{ + LIST_HEAD(freeme); + + nfs_server_return_all_delegations(server); + unset_pnfs_layoutdriver(server); + nfs4_purge_state_owners(server, &freeme); + nfs4_free_state_owners(&freeme); + kfree(server->delegation_hash_table); +} + +/* + * NFSv4.0 callback thread helper + * + * Find a client by callback identifier + */ +struct nfs_client * +nfs4_find_client_ident(struct net *net, int cb_ident) +{ + struct nfs_client *clp; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + spin_lock(&nn->nfs_client_lock); + clp = idr_find(&nn->cb_ident_idr, cb_ident); + if (clp) + refcount_inc(&clp->cl_count); + spin_unlock(&nn->nfs_client_lock); + return clp; +} + +/* Common match routine for v4.0 and v4.1 callback services */ +static bool nfs4_cb_match_client(const struct sockaddr *addr, + struct nfs_client *clp, u32 minorversion) +{ + struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr; + + /* Don't match clients that failed to initialise */ + if (!(clp->cl_cons_state == NFS_CS_READY || + clp->cl_cons_state == NFS_CS_SESSION_INITING)) + return false; + + smp_rmb(); + + /* Match the version and minorversion */ + if (clp->rpc_ops->version != 4 || + clp->cl_minorversion != minorversion) + return false; + + /* Match only the IP address, not the port number */ + return rpc_cmp_addr(addr, clap); +} + +/* + * NFSv4.1 callback thread helper + * For CB_COMPOUND calls, find a client by IP address, protocol version, + * minorversion, and sessionID + * + * Returns NULL if no such client + */ +struct nfs_client * +nfs4_find_client_sessionid(struct net *net, const struct sockaddr *addr, + struct nfs4_sessionid *sid, u32 minorversion) +{ + struct nfs_client *clp; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + spin_lock(&nn->nfs_client_lock); + list_for_each_entry(clp, &nn->nfs_client_list, cl_share_link) { + if (!nfs4_cb_match_client(addr, clp, minorversion)) + continue; + + if (!nfs4_has_session(clp)) + continue; + + /* Match sessionid*/ + if (memcmp(clp->cl_session->sess_id.data, + sid->data, NFS4_MAX_SESSIONID_LEN) != 0) + continue; + + refcount_inc(&clp->cl_count); + spin_unlock(&nn->nfs_client_lock); + return clp; + } + spin_unlock(&nn->nfs_client_lock); + return NULL; +} + +/* + * Set up an NFS4 client + */ +static int nfs4_set_client(struct nfs_server *server, + struct nfs_client_initdata *cl_init) +{ + struct nfs_client *clp; + + cl_init->nfs_mod = &nfs_v4; + cl_init->cred = server->cred; + + if (cl_init->minorversion == 0) { + __set_bit(NFS_CS_REUSEPORT, &cl_init->init_flags); + cl_init->max_connect = 0; + } + + switch (cl_init->proto) { + case XPRT_TRANSPORT_RDMA: + case XPRT_TRANSPORT_TCP: + case XPRT_TRANSPORT_TCP_TLS: + break; + default: + cl_init->nconnect = 0; + } + + if (server->flags & NFS_MOUNT_NORESVPORT) + __set_bit(NFS_CS_NORESVPORT, &cl_init->init_flags); + if (server->options & NFS_OPTION_MIGRATION) + __set_bit(NFS_CS_MIGRATION, &cl_init->init_flags); + if (test_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status)) + __set_bit(NFS_CS_TSM_POSSIBLE, &cl_init->init_flags); + server->port = rpc_get_port((struct sockaddr *)cl_init->addr); + + if (server->flags & NFS_MOUNT_NETUNREACH_FATAL) + __set_bit(NFS_CS_NETUNREACH_FATAL, &cl_init->init_flags); + + /* Allocate or find a client reference we can use */ + clp = nfs_get_client(cl_init); + if (IS_ERR(clp)) + return PTR_ERR(clp); + + if (server->nfs_client == clp) { + nfs_put_client(clp); + return -ELOOP; + } + + /* + * Query for the lease time on clientid setup or renewal + * + * Note that this will be set on nfs_clients that were created + * only for the DS role and did not set this bit, but now will + * serve a dual role. + */ + set_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state); + + server->nfs_client = clp; + nfs_sysfs_add_server(server); + nfs_sysfs_link_rpc_client(server, clp->cl_rpcclient, "_state"); + + return 0; +} + +/* + * Set up a pNFS Data Server client. + * + * Return any existing nfs_client that matches server address,port,version + * and minorversion. + * + * For a new nfs_client, use a soft mount (default), a low retrans and a + * low timeout interval so that if a connection is lost, we retry through + * the MDS. + */ +struct nfs_client *nfs4_set_ds_client(struct nfs_server *mds_srv, + const struct sockaddr_storage *ds_addr, int ds_addrlen, + int ds_proto, unsigned int ds_timeo, unsigned int ds_retrans, + u32 minor_version) +{ + struct rpc_timeout ds_timeout; + struct nfs_client *mds_clp = mds_srv->nfs_client; + struct nfs_client_initdata cl_init = { + .addr = ds_addr, + .addrlen = ds_addrlen, + .nodename = mds_clp->cl_rpcclient->cl_nodename, + .ip_addr = mds_clp->cl_ipaddr, + .nfs_mod = &nfs_v4, + .proto = ds_proto, + .minorversion = minor_version, + .net = mds_clp->cl_net, + .timeparms = &ds_timeout, + .cred = mds_srv->cred, + .xprtsec = { + .policy = RPC_XPRTSEC_NONE, + .cert_serial = TLS_NO_CERT, + .privkey_serial = TLS_NO_PRIVKEY, + }, + }; + char buf[INET6_ADDRSTRLEN + 1]; + + if (rpc_ntop((struct sockaddr *)ds_addr, buf, sizeof(buf)) <= 0) + return ERR_PTR(-EINVAL); + cl_init.hostname = buf; + + switch (ds_proto) { + case XPRT_TRANSPORT_TCP_TLS: + if (mds_srv->nfs_client->cl_xprtsec.policy != RPC_XPRTSEC_NONE) + cl_init.xprtsec = mds_srv->nfs_client->cl_xprtsec; + else + ds_proto = XPRT_TRANSPORT_TCP; + fallthrough; + case XPRT_TRANSPORT_RDMA: + case XPRT_TRANSPORT_TCP: + if (mds_clp->cl_nconnect > 1) { + cl_init.nconnect = mds_clp->cl_nconnect; + cl_init.max_connect = NFS_MAX_TRANSPORTS; + } + } + + if (mds_srv->flags & NFS_MOUNT_NORESVPORT) + __set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); + if (test_bit(NFS_CS_NETUNREACH_FATAL, &mds_clp->cl_flags)) + __set_bit(NFS_CS_NETUNREACH_FATAL, &cl_init.init_flags); + + __set_bit(NFS_CS_PNFS, &cl_init.init_flags); + cl_init.max_connect = NFS_MAX_TRANSPORTS; + /* + * Set an authflavor equual to the MDS value. Use the MDS nfs_client + * cl_ipaddr so as to use the same EXCHANGE_ID co_ownerid as the MDS + * (section 13.1 RFC 5661). + */ + nfs_init_timeout_values(&ds_timeout, ds_proto, ds_timeo, ds_retrans); + return nfs_get_client(&cl_init); +} +EXPORT_SYMBOL_GPL(nfs4_set_ds_client); + +/* + * Session has been established, and the client marked ready. + * Limit the mount rsize, wsize and dtsize using negotiated fore + * channel attributes. + */ +void nfs4_session_limit_rwsize(struct nfs_server *server) +{ + struct nfs4_session *sess; + u32 server_resp_sz; + u32 server_rqst_sz; + + if (!nfs4_has_session(server->nfs_client)) + return; + sess = server->nfs_client->cl_session; + server_resp_sz = sess->fc_attrs.max_resp_sz - nfs41_maxread_overhead; + server_rqst_sz = sess->fc_attrs.max_rqst_sz - nfs41_maxwrite_overhead; + + if (server->dtsize > server_resp_sz) + server->dtsize = server_resp_sz; + if (server->rsize > server_resp_sz) + server->rsize = server_resp_sz; + if (server->wsize > server_rqst_sz) + server->wsize = server_rqst_sz; +} + +/* + * Limit xattr sizes using the channel attributes. + */ +void nfs4_session_limit_xasize(struct nfs_server *server) +{ +#ifdef CONFIG_NFS_V4_2 + struct nfs4_session *sess; + u32 server_gxa_sz; + u32 server_sxa_sz; + u32 server_lxa_sz; + + if (!nfs4_has_session(server->nfs_client)) + return; + + sess = server->nfs_client->cl_session; + + server_gxa_sz = sess->fc_attrs.max_resp_sz - nfs42_maxgetxattr_overhead; + server_sxa_sz = sess->fc_attrs.max_rqst_sz - nfs42_maxsetxattr_overhead; + server_lxa_sz = sess->fc_attrs.max_resp_sz - + nfs42_maxlistxattrs_overhead; + + if (server->gxasize > server_gxa_sz) + server->gxasize = server_gxa_sz; + if (server->sxasize > server_sxa_sz) + server->sxasize = server_sxa_sz; + if (server->lxasize > server_lxa_sz) + server->lxasize = server_lxa_sz; +#endif +} + +static int nfs4_server_common_setup(struct nfs_server *server, + struct nfs_fh *mntfh, bool auth_probe) +{ + int error; + + error = nfs4_delegation_hash_alloc(server); + if (error) + return error; + + /* data servers support only a subset of NFSv4.1 */ + if (is_ds_only_client(server->nfs_client)) + return -EPROTONOSUPPORT; + + /* We must ensure the session is initialised first */ + error = nfs4_init_session(server->nfs_client); + if (error < 0) + return error; + + nfs_server_set_init_caps(server); + + /* Probe the root fh to retrieve its FSID and filehandle */ + error = nfs4_get_rootfh(server, mntfh, auth_probe); + if (error < 0) + return error; + + dprintk("Server FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + nfs_display_fhandle(mntfh, "Pseudo-fs root FH"); + + error = nfs_probe_server(server, mntfh); + if (error < 0) + return error; + + nfs4_session_limit_rwsize(server); + nfs4_session_limit_xasize(server); + + if (server->namelen == 0 || server->namelen > NFS4_MAXNAMLEN) + server->namelen = NFS4_MAXNAMLEN; + + nfs_server_insert_lists(server); + server->mount_time = jiffies; + server->destroy = nfs4_destroy_server; + return 0; +} + +/* + * Create a version 4 volume record + */ +static int nfs4_init_server(struct nfs_server *server, struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct rpc_timeout timeparms; + struct nfs_client_initdata cl_init = { + .hostname = ctx->nfs_server.hostname, + .addr = &ctx->nfs_server._address, + .addrlen = ctx->nfs_server.addrlen, + .ip_addr = ctx->client_address, + .proto = ctx->nfs_server.protocol, + .minorversion = ctx->minorversion, + .net = fc->net_ns, + .timeparms = &timeparms, + .xprtsec = ctx->xprtsec, + .nconnect = ctx->nfs_server.nconnect, + .max_connect = ctx->nfs_server.max_connect, + }; + int error; + + nfs_init_timeout_values(&timeparms, ctx->nfs_server.protocol, + ctx->timeo, ctx->retrans); + + /* Initialise the client representation from the mount data */ + server->flags = ctx->flags; + server->options = ctx->options; + server->auth_info = ctx->auth_info; + + /* Use the first specified auth flavor. If this flavor isn't + * allowed by the server, use the SECINFO path to try the + * other specified flavors */ + if (ctx->auth_info.flavor_len >= 1) + ctx->selected_flavor = ctx->auth_info.flavors[0]; + else + ctx->selected_flavor = RPC_AUTH_UNIX; + + /* Get a client record */ + error = nfs4_set_client(server, &cl_init); + if (error < 0) + return error; + + if (ctx->bsize) { + server->bsize = ctx->bsize; + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_BSIZE; + } + if (ctx->rsize) { + server->rsize = + nfs_io_size(ctx->rsize, server->nfs_client->cl_proto); + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_RSIZE; + } + if (ctx->wsize) { + server->wsize = + nfs_io_size(ctx->wsize, server->nfs_client->cl_proto); + server->automount_inherit |= NFS_AUTOMOUNT_INHERIT_WSIZE; + } + + server->acregmin = ctx->acregmin * HZ; + server->acregmax = ctx->acregmax * HZ; + server->acdirmin = ctx->acdirmin * HZ; + server->acdirmax = ctx->acdirmax * HZ; + server->port = ctx->nfs_server.port; + + return nfs_init_server_rpcclient(server, &timeparms, + ctx->selected_flavor); +} + +/* + * Create a version 4 volume record + * - keyed on server and FSID + */ +struct nfs_server *nfs4_create_server(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_server *server; + bool auth_probe; + int error; + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + server->cred = get_cred(fc->cred); + + auth_probe = ctx->auth_info.flavor_len < 1; + + /* set up the general RPC client */ + error = nfs4_init_server(server, fc); + if (error < 0) + goto error; + + error = nfs4_server_common_setup(server, ctx->mntfh, auth_probe); + if (error < 0) + goto error; + + return server; + +error: + nfs_free_server(server); + return ERR_PTR(error); +} + +/* + * Create an NFS4 referral server record + */ +struct nfs_server *nfs4_create_referral_server(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_server *parent_server = NFS_SB(ctx->clone_data.sb); + struct nfs_client *parent_client = parent_server->nfs_client; + struct nfs_client_initdata cl_init = { + .hostname = ctx->nfs_server.hostname, + .addr = &ctx->nfs_server._address, + .addrlen = ctx->nfs_server.addrlen, + .ip_addr = parent_client->cl_ipaddr, + .minorversion = parent_client->cl_mvops->minor_version, + .net = parent_client->cl_net, + .timeparms = parent_server->client->cl_timeout, + .xprtsec = parent_client->cl_xprtsec, + .nconnect = parent_client->cl_nconnect, + .max_connect = parent_client->cl_max_connect, + }; + struct nfs_server *server; + bool auth_probe; + int error; + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + server->cred = get_cred(parent_server->cred); + + /* Initialise the client representation from the parent server */ + nfs_server_copy_userdata(server, parent_server); + + /* Get a client representation */ +#if IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA) + rpc_set_port(&ctx->nfs_server.address, NFS_RDMA_PORT); + cl_init.proto = XPRT_TRANSPORT_RDMA; + error = nfs4_set_client(server, &cl_init); + if (!error) + goto init_server; +#endif /* IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA) */ + + cl_init.proto = XPRT_TRANSPORT_TCP; + if (parent_client->cl_xprtsec.policy != RPC_XPRTSEC_NONE) + cl_init.proto = XPRT_TRANSPORT_TCP_TLS; + rpc_set_port(&ctx->nfs_server.address, NFS_PORT); + error = nfs4_set_client(server, &cl_init); + if (error < 0) + goto error; + +#if IS_ENABLED(CONFIG_SUNRPC_XPRT_RDMA) +init_server: +#endif + error = nfs_init_server_rpcclient(server, parent_server->client->cl_timeout, + ctx->selected_flavor); + if (error < 0) + goto error; + + auth_probe = parent_server->auth_info.flavor_len < 1; + + error = nfs4_server_common_setup(server, ctx->mntfh, auth_probe); + if (error < 0) + goto error; + + return server; + +error: + nfs_free_server(server); + return ERR_PTR(error); +} + +/** + * nfs4_update_server - Move an nfs_server to a different nfs_client + * + * @server: represents FSID to be moved + * @hostname: new end-point's hostname + * @sap: new end-point's socket address + * @salen: size of "sap" + * @net: net namespace + * + * The nfs_server must be quiescent before this function is invoked. + * Either its session is drained (NFSv4.1+), or its transport is + * plugged and drained (NFSv4.0). + * + * Returns zero on success, or a negative errno value. + */ +int nfs4_update_server(struct nfs_server *server, const char *hostname, + struct sockaddr_storage *sap, size_t salen, struct net *net) +{ + struct nfs_client *clp = server->nfs_client; + struct rpc_clnt *clnt = server->client; + struct xprt_create xargs = { + .ident = clp->cl_proto, + .net = net, + .dstaddr = (struct sockaddr *)sap, + .addrlen = salen, + .servername = hostname, + /* cel: bleh. We might need to pass TLS parameters here */ + }; + char buf[INET6_ADDRSTRLEN + 1]; + struct sockaddr_storage address; + struct sockaddr *localaddr = (struct sockaddr *)&address; + struct nfs_client_initdata cl_init = { + .hostname = hostname, + .addr = sap, + .addrlen = salen, + .ip_addr = buf, + .proto = clp->cl_proto, + .minorversion = clp->cl_minorversion, + .net = net, + .timeparms = clnt->cl_timeout, + .xprtsec = clp->cl_xprtsec, + .nconnect = clp->cl_nconnect, + .max_connect = clp->cl_max_connect, + }; + int error; + + error = rpc_switch_client_transport(clnt, &xargs, clnt->cl_timeout); + if (error != 0) + return error; + + error = rpc_localaddr(clnt, localaddr, sizeof(address)); + if (error != 0) + return error; + + if (rpc_ntop(localaddr, buf, sizeof(buf)) == 0) + return -EAFNOSUPPORT; + + nfs_server_remove_lists(server); + set_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); + error = nfs4_set_client(server, &cl_init); + clear_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); + if (error != 0) { + nfs_server_insert_lists(server); + return error; + } + nfs_put_client(clp); + + if (server->nfs_client->cl_hostname == NULL) { + server->nfs_client->cl_hostname = kstrdup(hostname, GFP_KERNEL); + if (server->nfs_client->cl_hostname == NULL) + return -ENOMEM; + } + nfs_server_insert_lists(server); + + return nfs_probe_server(server, NFS_FH(d_inode(server->super->s_root))); +} diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c new file mode 100644 index 000000000..be40e126c --- /dev/null +++ b/fs/nfs/nfs4file.c @@ -0,0 +1,459 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/file.c + * + * Copyright (C) 1992 Rick Sladkey + */ +#include +#include +#include +#include +#include +#include +#include +#include "delegation.h" +#include "internal.h" +#include "iostat.h" +#include "fscache.h" +#include "pnfs.h" + +#include "nfstrace.h" + +#ifdef CONFIG_NFS_V4_2 +#include "nfs42.h" +#endif + +#define NFSDBG_FACILITY NFSDBG_FILE + +static int +nfs4_file_open(struct inode *inode, struct file *filp) +{ + struct nfs_open_context *ctx; + struct dentry *dentry = file_dentry(filp); + struct dentry *parent = NULL; + struct inode *dir; + unsigned openflags = filp->f_flags; + struct iattr attr; + int err; + + /* + * If no cached dentry exists or if it's negative, NFSv4 handled the + * opens in ->lookup() or ->create(). + * + * We only get this far for a cached positive dentry. We skipped + * revalidation, so handle it here by dropping the dentry and returning + * -EOPENSTALE. The VFS will retry the lookup/create/open. + */ + + dprintk("NFS: open file(%pd2)\n", dentry); + + err = nfs_check_flags(openflags); + if (err) + return err; + + /* We can't create new files here */ + openflags &= ~(O_CREAT|O_EXCL); + + parent = dget_parent(dentry); + dir = d_inode(parent); + + ctx = alloc_nfs_open_context(file_dentry(filp), + flags_to_mode(openflags), filp); + err = PTR_ERR(ctx); + if (IS_ERR(ctx)) + goto out; + + attr.ia_valid = ATTR_OPEN; + if (openflags & O_TRUNC) { + attr.ia_valid |= ATTR_SIZE; + attr.ia_size = 0; + filemap_write_and_wait(inode->i_mapping); + } + + inode = NFS_PROTO(dir)->open_context(dir, ctx, openflags, &attr, NULL); + if (IS_ERR(inode)) { + err = PTR_ERR(inode); + switch (err) { + default: + goto out_put_ctx; + case -ENOENT: + case -ESTALE: + case -EISDIR: + case -ENOTDIR: + case -ELOOP: + goto out_drop; + } + } + if (inode != d_inode(dentry)) + goto out_drop; + + nfs_file_set_open_context(filp, ctx); + nfs_fscache_open_file(inode, filp); + err = 0; + filp->f_mode |= FMODE_CAN_ODIRECT; + +out_put_ctx: + put_nfs_open_context(ctx); +out: + dput(parent); + return err; + +out_drop: + d_drop(dentry); + err = -EOPENSTALE; + goto out_put_ctx; +} + +/* + * Flush all dirty pages, and check for write errors. + */ +static int +nfs4_file_flush(struct file *file, fl_owner_t id) +{ + struct inode *inode = file_inode(file); + errseq_t since; + + dprintk("NFS: flush(%pD2)\n", file); + + nfs_inc_stats(inode, NFSIOS_VFSFLUSH); + if ((file->f_mode & FMODE_WRITE) == 0) + return 0; + + /* + * If we're holding a write delegation, then check if we're required + * to flush the i/o on close. If not, then just start the i/o now. + */ + if (!nfs4_delegation_flush_on_close(inode)) + return filemap_fdatawrite(file->f_mapping); + + /* Flush writes to the server and return any errors */ + since = filemap_sample_wb_err(file->f_mapping); + nfs_wb_all(inode); + return filemap_check_wb_err(file->f_mapping, since); +} + +#ifdef CONFIG_NFS_V4_2 +static ssize_t __nfs4_copy_file_range(struct file *file_in, loff_t pos_in, + struct file *file_out, loff_t pos_out, + size_t count, unsigned int flags) +{ + struct nfs42_copy_notify_res *cn_resp = NULL; + struct nl4_server *nss = NULL; + nfs4_stateid *cnrs = NULL; + ssize_t ret; + bool sync = false; + + /* Only offload copy if superblock is the same */ + if (file_in->f_op != &nfs4_file_operations) + return -EXDEV; + if (!nfs_server_capable(file_inode(file_out), NFS_CAP_COPY) || + !nfs_server_capable(file_inode(file_in), NFS_CAP_COPY)) + return -EOPNOTSUPP; + if (file_inode(file_in) == file_inode(file_out)) + return -EOPNOTSUPP; + /* if the copy size if smaller than 2 RPC payloads, make it + * synchronous + */ + if (count <= 2 * NFS_SERVER(file_inode(file_in))->rsize) + sync = true; +retry: + if (!nfs42_files_from_same_server(file_in, file_out)) { + /* + * for inter copy, if copy size is too small + * then fallback to generic copy. + */ + if (sync) + return -EOPNOTSUPP; + cn_resp = kzalloc_obj(struct nfs42_copy_notify_res); + if (unlikely(cn_resp == NULL)) + return -ENOMEM; + + ret = nfs42_proc_copy_notify(file_in, file_out, cn_resp); + if (ret) { + ret = -EOPNOTSUPP; + goto out; + } + nss = &cn_resp->cnr_src; + cnrs = &cn_resp->cnr_stateid; + } + ret = nfs42_proc_copy(file_in, pos_in, file_out, pos_out, count, + nss, cnrs, sync); +out: + kfree(cn_resp); + + if (ret == -EAGAIN) + goto retry; + return ret; +} + +static ssize_t nfs4_copy_file_range(struct file *file_in, loff_t pos_in, + struct file *file_out, loff_t pos_out, + size_t count, unsigned int flags) +{ + ssize_t ret; + + ret = __nfs4_copy_file_range(file_in, pos_in, file_out, pos_out, count, + flags); + if (ret == -EOPNOTSUPP || ret == -EXDEV) + ret = splice_copy_file_range(file_in, pos_in, file_out, + pos_out, count); + return ret; +} + +static loff_t nfs4_file_llseek(struct file *filep, loff_t offset, int whence) +{ + loff_t ret; + + switch (whence) { + case SEEK_HOLE: + case SEEK_DATA: + ret = nfs42_proc_llseek(filep, offset, whence); + if (ret != -EOPNOTSUPP) + return ret; + fallthrough; + default: + return nfs_file_llseek(filep, offset, whence); + } +} + +static long nfs42_fallocate(struct file *filep, int mode, loff_t offset, loff_t len) +{ + struct inode *inode = file_inode(filep); + long ret; + + if (!S_ISREG(inode->i_mode)) + return -EOPNOTSUPP; + + switch (mode) { + case 0: + case FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE: + case FALLOC_FL_ZERO_RANGE: + break; + default: + return -EOPNOTSUPP; + } + + ret = inode_newsize_ok(inode, offset + len); + if (ret < 0) + return ret; + + if (mode & FALLOC_FL_PUNCH_HOLE) + return nfs42_proc_deallocate(filep, offset, len); + else if (mode & FALLOC_FL_ZERO_RANGE) + return nfs42_proc_zero_range(filep, offset ,len); + return nfs42_proc_allocate(filep, offset, len); +} + +static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off, + struct file *dst_file, loff_t dst_off, loff_t count, + unsigned int remap_flags) +{ + struct inode *dst_inode = file_inode(dst_file); + struct nfs_server *server = NFS_SERVER(dst_inode); + struct inode *src_inode = file_inode(src_file); + unsigned int bs = server->clone_blksize; + int ret; + + /* NFS does not support deduplication. */ + if (remap_flags & REMAP_FILE_DEDUP) + return -EOPNOTSUPP; + + if (remap_flags & ~REMAP_FILE_ADVISORY) + return -EINVAL; + + if (IS_SWAPFILE(dst_inode) || IS_SWAPFILE(src_inode)) + return -ETXTBSY; + + /* check alignment w.r.t. clone_blksize */ + ret = -EINVAL; + if (bs) { + if (!IS_ALIGNED(src_off, bs) || !IS_ALIGNED(dst_off, bs)) + goto out; + if (!IS_ALIGNED(count, bs) && i_size_read(src_inode) != (src_off + count)) + goto out; + } + + /* XXX: do we lock at all? what if server needs CB_RECALL_LAYOUT? */ + lock_two_nondirectories(src_inode, dst_inode); + /* flush all pending writes on both src and dst so that server + * has the latest data */ + nfs_file_block_o_direct(NFS_I(src_inode)); + ret = nfs_sync_inode(src_inode); + if (ret) + goto out_unlock; + nfs_file_block_o_direct(NFS_I(dst_inode)); + ret = nfs_sync_inode(dst_inode); + if (ret) + goto out_unlock; + + ret = nfs42_proc_clone(src_file, dst_file, src_off, dst_off, count); + + /* truncate inode page cache of the dst range so that future reads can fetch + * new data from server */ + if (!ret) + truncate_inode_pages_range(&dst_inode->i_data, dst_off, dst_off + count - 1); + +out_unlock: + unlock_two_nondirectories(src_inode, dst_inode); +out: + return ret < 0 ? ret : count; +} + +static int read_name_gen = 1; +#define SSC_READ_NAME_BODY "ssc_read_%d" + +static struct file *__nfs42_ssc_open(struct vfsmount *ss_mnt, + struct nfs_fh *src_fh, nfs4_stateid *stateid) +{ + struct nfs_fattr *fattr = nfs_alloc_fattr(); + struct file *filep, *res; + struct nfs_server *server; + struct inode *r_ino = NULL; + struct nfs_open_context *ctx; + struct nfs4_state_owner *sp; + char *read_name = NULL; + int len, status = 0; + + server = NFS_SB(ss_mnt->mnt_sb); + + if (!fattr) + return ERR_PTR(-ENOMEM); + + status = nfs4_proc_getattr(server, src_fh, fattr, NULL); + if (status < 0) { + res = ERR_PTR(status); + goto out; + } + + if (!S_ISREG(fattr->mode)) { + res = ERR_PTR(-EBADF); + goto out; + } + + res = ERR_PTR(-ENOMEM); + len = strlen(SSC_READ_NAME_BODY) + 16; + read_name = kzalloc(len, GFP_KERNEL); + if (read_name == NULL) + goto out; + snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++); + + r_ino = nfs_fhget(ss_mnt->mnt_sb, src_fh, fattr); + if (IS_ERR(r_ino)) { + res = ERR_CAST(r_ino); + goto out_free_name; + } + + filep = alloc_file_pseudo(r_ino, ss_mnt, read_name, O_RDONLY, + r_ino->i_fop); + if (IS_ERR(filep)) { + res = ERR_CAST(filep); + iput(r_ino); + goto out_free_name; + } + + ctx = alloc_nfs_open_context(filep->f_path.dentry, + flags_to_mode(filep->f_flags), filep); + if (IS_ERR(ctx)) { + res = ERR_CAST(ctx); + goto out_filep; + } + + res = ERR_PTR(-EINVAL); + sp = nfs4_get_state_owner(server, ctx->cred, GFP_KERNEL); + if (sp == NULL) + goto out_ctx; + + ctx->state = nfs4_get_open_state(r_ino, sp); + if (ctx->state == NULL) + goto out_stateowner; + + set_bit(NFS_SRV_SSC_COPY_STATE, &ctx->state->flags); + memcpy(&ctx->state->open_stateid.other, &stateid->other, + NFS4_STATEID_OTHER_SIZE); + update_open_stateid(ctx->state, stateid, NULL, filep->f_mode); + set_bit(NFS_OPEN_STATE, &ctx->state->flags); + + nfs_file_set_open_context(filep, ctx); + put_nfs_open_context(ctx); + + file_ra_state_init(&filep->f_ra, filep->f_mapping->host->i_mapping); + res = filep; +out_free_name: + kfree(read_name); +out: + nfs_free_fattr(fattr); + return res; +out_stateowner: + nfs4_put_state_owner(sp); +out_ctx: + put_nfs_open_context(ctx); +out_filep: + fput(filep); + goto out_free_name; +} + +static void __nfs42_ssc_close(struct file *filep) +{ + struct nfs_open_context *ctx = nfs_file_open_context(filep); + + ctx->state->flags = 0; +} + +static const struct nfs4_ssc_client_ops nfs4_ssc_clnt_ops_tbl = { + .sco_open = __nfs42_ssc_open, + .sco_close = __nfs42_ssc_close, +}; + +/** + * nfs42_ssc_register_ops - Wrapper to register NFS_V4 ops in nfs_common + * + * Return values: + * None + */ +void nfs42_ssc_register_ops(void) +{ + nfs42_ssc_register(&nfs4_ssc_clnt_ops_tbl); +} + +/** + * nfs42_ssc_unregister_ops - wrapper to un-register NFS_V4 ops in nfs_common + * + * Return values: + * None. + */ +void nfs42_ssc_unregister_ops(void) +{ + nfs42_ssc_unregister(&nfs4_ssc_clnt_ops_tbl); +} +#endif /* CONFIG_NFS_V4_2 */ + +static int nfs4_setlease(struct file *file, int arg, struct file_lease **lease, + void **priv) +{ + return nfs4_proc_setlease(file, arg, lease, priv); +} + +const struct file_operations nfs4_file_operations = { + .read_iter = nfs_file_read, + .write_iter = nfs_file_write, + .mmap_prepare = nfs_file_mmap_prepare, + .open = nfs4_file_open, + .flush = nfs4_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = iter_file_splice_write, + .check_flags = nfs_check_flags, + .setlease = nfs4_setlease, +#ifdef CONFIG_NFS_V4_2 + .copy_file_range = nfs4_copy_file_range, + .llseek = nfs4_file_llseek, + .fallocate = nfs42_fallocate, + .remap_file_range = nfs42_remap_file_range, +#else + .llseek = nfs_file_llseek, +#endif + .fop_flags = FOP_DONTCACHE, +}; diff --git a/fs/nfs/nfs4getroot.c b/fs/nfs/nfs4getroot.c new file mode 100644 index 000000000..e67ea345d --- /dev/null +++ b/fs/nfs/nfs4getroot.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0 +/* +* Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. +* Written by David Howells (dhowells@redhat.com) +*/ + +#include +#include "nfs4_fs.h" +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool auth_probe) +{ + struct nfs_fattr *fattr = nfs_alloc_fattr(); + int ret = -ENOMEM; + + if (fattr == NULL) + goto out; + + /* Start by getting the root filehandle from the server */ + ret = nfs4_proc_get_rootfh(server, mntfh, fattr, auth_probe); + if (ret < 0) { + dprintk("nfs4_get_rootfh: getroot error = %d\n", -ret); + goto out; + } + + if (!(fattr->valid & NFS_ATTR_FATTR_TYPE) || !S_ISDIR(fattr->mode)) { + printk(KERN_ERR "nfs4_get_rootfh:" + " getroot encountered non-directory\n"); + ret = -ENOTDIR; + goto out; + } + + memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid)); +out: + nfs_free_fattr(fattr); + return ret; +} diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c new file mode 100644 index 000000000..bc397110d --- /dev/null +++ b/fs/nfs/nfs4idmap.c @@ -0,0 +1,793 @@ +/* + * fs/nfs/idmap.c + * + * UID and GID to name mapping for clients. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Marius Aamodt Eriksen + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "netns.h" +#include "nfs4idmap.h" +#include "nfs4trace.h" + +#define NFS_UINT_MAXLEN 11 + +static const struct cred *id_resolver_cache; +static struct key_type key_type_id_resolver_legacy; + +struct idmap_legacy_upcalldata { + struct rpc_pipe_msg pipe_msg; + struct idmap_msg idmap_msg; + struct key *authkey; + struct idmap *idmap; +}; + +struct idmap { + struct rpc_pipe_dir_object idmap_pdo; + struct rpc_pipe *idmap_pipe; + struct idmap_legacy_upcalldata *idmap_upcall_data; + struct mutex idmap_mutex; + struct user_namespace *user_ns; +}; + +static struct user_namespace *idmap_userns(const struct idmap *idmap) +{ + if (idmap && idmap->user_ns) + return idmap->user_ns; + return &init_user_ns; +} + +/** + * nfs_fattr_init_names - initialise the nfs_fattr owner_name/group_name fields + * @fattr: fully initialised struct nfs_fattr + * @owner_name: owner name string cache + * @group_name: group name string cache + */ +void nfs_fattr_init_names(struct nfs_fattr *fattr, + struct nfs4_string *owner_name, + struct nfs4_string *group_name) +{ + fattr->owner_name = owner_name; + fattr->group_name = group_name; +} + +static void nfs_fattr_free_owner_name(struct nfs_fattr *fattr) +{ + fattr->valid &= ~NFS_ATTR_FATTR_OWNER_NAME; + kfree(fattr->owner_name->data); +} + +static void nfs_fattr_free_group_name(struct nfs_fattr *fattr) +{ + fattr->valid &= ~NFS_ATTR_FATTR_GROUP_NAME; + kfree(fattr->group_name->data); +} + +static bool nfs_fattr_map_owner_name(struct nfs_server *server, struct nfs_fattr *fattr) +{ + struct nfs4_string *owner = fattr->owner_name; + kuid_t uid; + + if (!(fattr->valid & NFS_ATTR_FATTR_OWNER_NAME)) + return false; + if (nfs_map_name_to_uid(server, owner->data, owner->len, &uid) == 0) { + fattr->uid = uid; + fattr->valid |= NFS_ATTR_FATTR_OWNER; + } + return true; +} + +static bool nfs_fattr_map_group_name(struct nfs_server *server, struct nfs_fattr *fattr) +{ + struct nfs4_string *group = fattr->group_name; + kgid_t gid; + + if (!(fattr->valid & NFS_ATTR_FATTR_GROUP_NAME)) + return false; + if (nfs_map_group_to_gid(server, group->data, group->len, &gid) == 0) { + fattr->gid = gid; + fattr->valid |= NFS_ATTR_FATTR_GROUP; + } + return true; +} + +/** + * nfs_fattr_free_names - free up the NFSv4 owner and group strings + * @fattr: a fully initialised nfs_fattr structure + */ +void nfs_fattr_free_names(struct nfs_fattr *fattr) +{ + if (fattr->valid & NFS_ATTR_FATTR_OWNER_NAME) + nfs_fattr_free_owner_name(fattr); + if (fattr->valid & NFS_ATTR_FATTR_GROUP_NAME) + nfs_fattr_free_group_name(fattr); +} + +/** + * nfs_fattr_map_and_free_names - map owner/group strings into uid/gid and free + * @server: pointer to the filesystem nfs_server structure + * @fattr: a fully initialised nfs_fattr structure + * + * This helper maps the cached NFSv4 owner/group strings in fattr into + * their numeric uid/gid equivalents, and then frees the cached strings. + */ +void nfs_fattr_map_and_free_names(struct nfs_server *server, struct nfs_fattr *fattr) +{ + if (nfs_fattr_map_owner_name(server, fattr)) + nfs_fattr_free_owner_name(fattr); + if (nfs_fattr_map_group_name(server, fattr)) + nfs_fattr_free_group_name(fattr); +} + +int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *res) +{ + unsigned long val; + char buf[16]; + + if (memchr(name, '@', namelen) != NULL || namelen >= sizeof(buf)) + return 0; + memcpy(buf, name, namelen); + buf[namelen] = '\0'; + if (kstrtoul(buf, 0, &val) != 0) + return 0; + *res = val; + return 1; +} +EXPORT_SYMBOL_GPL(nfs_map_string_to_numeric); + +static int nfs_map_numeric_to_string(__u32 id, char *buf, size_t buflen) +{ + return snprintf(buf, buflen, "%u", id); +} + +static struct key_type key_type_id_resolver = { + .name = "id_resolver", + .preparse = user_preparse, + .free_preparse = user_free_preparse, + .instantiate = generic_key_instantiate, + .revoke = user_revoke, + .destroy = user_destroy, + .describe = user_describe, + .read = user_read, +}; + +int nfs_idmap_init(void) +{ + struct cred *cred; + struct key *keyring; + int ret = 0; + + printk(KERN_NOTICE "NFS: Registering the %s key type\n", + key_type_id_resolver.name); + + cred = prepare_kernel_cred(&init_task); + if (!cred) + return -ENOMEM; + + keyring = keyring_alloc(".id_resolver", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, + (KEY_POS_ALL & ~KEY_POS_SETATTR) | + KEY_USR_VIEW | KEY_USR_READ, + KEY_ALLOC_NOT_IN_QUOTA, NULL, NULL); + if (IS_ERR(keyring)) { + ret = PTR_ERR(keyring); + goto failed_put_cred; + } + + ret = register_key_type(&key_type_id_resolver); + if (ret < 0) + goto failed_put_key; + + ret = register_key_type(&key_type_id_resolver_legacy); + if (ret < 0) + goto failed_reg_legacy; + + set_bit(KEY_FLAG_ROOT_CAN_CLEAR, &keyring->flags); + cred->thread_keyring = keyring; + cred->jit_keyring = KEY_REQKEY_DEFL_THREAD_KEYRING; + id_resolver_cache = cred; + return 0; + +failed_reg_legacy: + unregister_key_type(&key_type_id_resolver); +failed_put_key: + key_put(keyring); +failed_put_cred: + put_cred(cred); + return ret; +} + +void nfs_idmap_quit(void) +{ + key_revoke(id_resolver_cache->thread_keyring); + unregister_key_type(&key_type_id_resolver); + unregister_key_type(&key_type_id_resolver_legacy); + put_cred(id_resolver_cache); +} + +/* + * Assemble the description to pass to request_key() + * This function will allocate a new string and update dest to point + * at it. The caller is responsible for freeing dest. + * + * On error 0 is returned. Otherwise, the length of dest is returned. + */ +static ssize_t nfs_idmap_get_desc(const char *name, size_t namelen, + const char *type, size_t typelen, char **desc) +{ + char *cp; + size_t desclen = typelen + namelen + 2; + + *desc = kmalloc(desclen, GFP_KERNEL); + if (!*desc) + return -ENOMEM; + + cp = *desc; + memcpy(cp, type, typelen); + cp += typelen; + *cp++ = ':'; + + memcpy(cp, name, namelen); + cp += namelen; + *cp = '\0'; + return desclen; +} + +static struct key *nfs_idmap_request_key(const char *name, size_t namelen, + const char *type, struct idmap *idmap) +{ + char *desc; + struct key *rkey = ERR_PTR(-EAGAIN); + ssize_t ret; + + ret = nfs_idmap_get_desc(name, namelen, type, strlen(type), &desc); + if (ret < 0) + return ERR_PTR(ret); + + if (!idmap->user_ns || idmap->user_ns == &init_user_ns) + rkey = request_key(&key_type_id_resolver, desc, ""); + if (IS_ERR(rkey)) { + mutex_lock(&idmap->idmap_mutex); + rkey = request_key_with_auxdata(&key_type_id_resolver_legacy, + desc, NULL, "", 0, idmap); + mutex_unlock(&idmap->idmap_mutex); + } + if (!IS_ERR(rkey)) + set_bit(KEY_FLAG_ROOT_CAN_INVAL, &rkey->flags); + + kfree(desc); + return rkey; +} + +static ssize_t nfs_idmap_get_key(const char *name, size_t namelen, + const char *type, void *data, + size_t data_size, struct idmap *idmap) +{ + struct key *rkey; + const struct user_key_payload *payload; + ssize_t ret; + + scoped_with_creds(id_resolver_cache) + rkey = nfs_idmap_request_key(name, namelen, type, idmap); + if (IS_ERR(rkey)) { + ret = PTR_ERR(rkey); + goto out; + } + + rcu_read_lock(); + rkey->perm |= KEY_USR_VIEW; + + ret = key_validate(rkey); + if (ret < 0) + goto out_up; + + payload = user_key_payload_rcu(rkey); + if (IS_ERR_OR_NULL(payload)) { + ret = PTR_ERR(payload); + goto out_up; + } + + ret = payload->datalen; + if (ret > 0 && ret <= data_size) + memcpy(data, payload->data, ret); + else + ret = -EINVAL; + +out_up: + rcu_read_unlock(); + key_put(rkey); +out: + return ret; +} + +/* ID -> Name */ +static ssize_t nfs_idmap_lookup_name(__u32 id, const char *type, char *buf, + size_t buflen, struct idmap *idmap) +{ + char id_str[NFS_UINT_MAXLEN]; + int id_len; + ssize_t ret; + + id_len = nfs_map_numeric_to_string(id, id_str, sizeof(id_str)); + ret = nfs_idmap_get_key(id_str, id_len, type, buf, buflen, idmap); + if (ret < 0) + return -EINVAL; + return ret; +} + +/* Name -> ID */ +static int nfs_idmap_lookup_id(const char *name, size_t namelen, const char *type, + __u32 *id, struct idmap *idmap) +{ + char id_str[NFS_UINT_MAXLEN]; + long id_long; + ssize_t data_size; + int ret = 0; + + data_size = nfs_idmap_get_key(name, namelen, type, id_str, NFS_UINT_MAXLEN, idmap); + if (data_size <= 0) { + ret = -EINVAL; + } else { + ret = kstrtol(id_str, 10, &id_long); + if (!ret) + *id = (__u32)id_long; + } + return ret; +} + +/* idmap classic begins here */ + +enum { + Opt_find_uid, Opt_find_gid, Opt_find_user, Opt_find_group, Opt_find_err +}; + +static const match_table_t nfs_idmap_tokens = { + { Opt_find_uid, "uid:%s" }, + { Opt_find_gid, "gid:%s" }, + { Opt_find_user, "user:%s" }, + { Opt_find_group, "group:%s" }, + { Opt_find_err, NULL } +}; + +static int nfs_idmap_legacy_upcall(struct key *, void *); +static ssize_t idmap_pipe_downcall(struct file *, const char __user *, + size_t); +static void idmap_release_pipe(struct inode *); +static void idmap_pipe_destroy_msg(struct rpc_pipe_msg *); + +static const struct rpc_pipe_ops idmap_upcall_ops = { + .upcall = rpc_pipe_generic_upcall, + .downcall = idmap_pipe_downcall, + .release_pipe = idmap_release_pipe, + .destroy_msg = idmap_pipe_destroy_msg, +}; + +static struct key_type key_type_id_resolver_legacy = { + .name = "id_legacy", + .preparse = user_preparse, + .free_preparse = user_free_preparse, + .instantiate = generic_key_instantiate, + .revoke = user_revoke, + .destroy = user_destroy, + .describe = user_describe, + .read = user_read, + .request_key = nfs_idmap_legacy_upcall, +}; + +static void nfs_idmap_pipe_destroy(struct dentry *dir, + struct rpc_pipe_dir_object *pdo) +{ + struct idmap *idmap = pdo->pdo_data; + + rpc_unlink(idmap->idmap_pipe); +} + +static int nfs_idmap_pipe_create(struct dentry *dir, + struct rpc_pipe_dir_object *pdo) +{ + struct idmap *idmap = pdo->pdo_data; + + return rpc_mkpipe_dentry(dir, "idmap", idmap, idmap->idmap_pipe); +} + +static const struct rpc_pipe_dir_object_ops nfs_idmap_pipe_dir_object_ops = { + .create = nfs_idmap_pipe_create, + .destroy = nfs_idmap_pipe_destroy, +}; + +int +nfs_idmap_new(struct nfs_client *clp) +{ + struct idmap *idmap; + struct rpc_pipe *pipe; + int error; + + idmap = kzalloc_obj(*idmap); + if (idmap == NULL) + return -ENOMEM; + + mutex_init(&idmap->idmap_mutex); + idmap->user_ns = get_user_ns(clp->cl_rpcclient->cl_cred->user_ns); + + rpc_init_pipe_dir_object(&idmap->idmap_pdo, + &nfs_idmap_pipe_dir_object_ops, + idmap); + + pipe = rpc_mkpipe_data(&idmap_upcall_ops, 0); + if (IS_ERR(pipe)) { + error = PTR_ERR(pipe); + goto err; + } + idmap->idmap_pipe = pipe; + + error = rpc_add_pipe_dir_object(clp->cl_net, + &clp->cl_rpcclient->cl_pipedir_objects, + &idmap->idmap_pdo); + if (error) + goto err_destroy_pipe; + + clp->cl_idmap = idmap; + return 0; +err_destroy_pipe: + rpc_destroy_pipe_data(idmap->idmap_pipe); +err: + put_user_ns(idmap->user_ns); + kfree(idmap); + return error; +} + +void +nfs_idmap_delete(struct nfs_client *clp) +{ + struct idmap *idmap = clp->cl_idmap; + + if (!idmap) + return; + clp->cl_idmap = NULL; + rpc_remove_pipe_dir_object(clp->cl_net, + &clp->cl_rpcclient->cl_pipedir_objects, + &idmap->idmap_pdo); + rpc_destroy_pipe_data(idmap->idmap_pipe); + put_user_ns(idmap->user_ns); + kfree(idmap); +} + +static int nfs_idmap_prepare_message(char *desc, struct idmap *idmap, + struct idmap_msg *im, + struct rpc_pipe_msg *msg) +{ + substring_t substr; + int token, ret; + + im->im_type = IDMAP_TYPE_GROUP; + token = match_token(desc, nfs_idmap_tokens, &substr); + + switch (token) { + case Opt_find_uid: + im->im_type = IDMAP_TYPE_USER; + fallthrough; + case Opt_find_gid: + im->im_conv = IDMAP_CONV_NAMETOID; + ret = match_strlcpy(im->im_name, &substr, IDMAP_NAMESZ); + break; + + case Opt_find_user: + im->im_type = IDMAP_TYPE_USER; + fallthrough; + case Opt_find_group: + im->im_conv = IDMAP_CONV_IDTONAME; + ret = match_int(&substr, &im->im_id); + if (ret) + goto out; + break; + + default: + ret = -EINVAL; + goto out; + } + + msg->data = im; + msg->len = sizeof(struct idmap_msg); + +out: + return ret; +} + +static bool +nfs_idmap_prepare_pipe_upcall(struct idmap *idmap, + struct idmap_legacy_upcalldata *data) +{ + if (idmap->idmap_upcall_data != NULL) { + WARN_ON_ONCE(1); + return false; + } + idmap->idmap_upcall_data = data; + return true; +} + +static void nfs_idmap_complete_pipe_upcall(struct idmap_legacy_upcalldata *data, + int ret) +{ + complete_request_key(data->authkey, ret); + key_put(data->authkey); + kfree(data); +} + +static void nfs_idmap_abort_pipe_upcall(struct idmap *idmap, + struct idmap_legacy_upcalldata *data, + int ret) +{ + if (cmpxchg(&idmap->idmap_upcall_data, data, NULL) == data) + nfs_idmap_complete_pipe_upcall(data, ret); +} + +static int nfs_idmap_legacy_upcall(struct key *authkey, void *aux) +{ + struct idmap_legacy_upcalldata *data; + struct request_key_auth *rka = get_request_key_auth(authkey); + struct rpc_pipe_msg *msg; + struct idmap_msg *im; + struct idmap *idmap = aux; + struct key *key = rka->target_key; + int ret = -ENOKEY; + + if (!aux) + goto out1; + + /* msg and im are freed in idmap_pipe_destroy_msg */ + ret = -ENOMEM; + data = kzalloc_obj(*data); + if (!data) + goto out1; + + msg = &data->pipe_msg; + im = &data->idmap_msg; + data->idmap = idmap; + data->authkey = key_get(authkey); + + ret = nfs_idmap_prepare_message(key->description, idmap, im, msg); + if (ret < 0) + goto out2; + + ret = -EAGAIN; + if (!nfs_idmap_prepare_pipe_upcall(idmap, data)) + goto out2; + + ret = rpc_queue_upcall(idmap->idmap_pipe, msg); + if (ret < 0) + nfs_idmap_abort_pipe_upcall(idmap, data, ret); + + return ret; +out2: + kfree(data); +out1: + complete_request_key(authkey, ret); + return ret; +} + +static int nfs_idmap_instantiate(struct key *key, struct key *authkey, char *data, size_t datalen) +{ + return key_instantiate_and_link(key, data, datalen, + id_resolver_cache->thread_keyring, + authkey); +} + +static int nfs_idmap_read_and_verify_message(struct idmap_msg *im, + struct idmap_msg *upcall, + struct key *key, struct key *authkey) +{ + char id_str[NFS_UINT_MAXLEN]; + size_t len; + int ret = -ENOKEY; + + /* ret = -ENOKEY */ + if (upcall->im_type != im->im_type || upcall->im_conv != im->im_conv) + goto out; + switch (im->im_conv) { + case IDMAP_CONV_NAMETOID: + if (strcmp(upcall->im_name, im->im_name) != 0) + break; + /* Note: here we store the NUL terminator too */ + len = 1 + nfs_map_numeric_to_string(im->im_id, id_str, + sizeof(id_str)); + ret = nfs_idmap_instantiate(key, authkey, id_str, len); + break; + case IDMAP_CONV_IDTONAME: + if (upcall->im_id != im->im_id) + break; + len = strlen(im->im_name); + ret = nfs_idmap_instantiate(key, authkey, im->im_name, len); + break; + default: + ret = -EINVAL; + } +out: + return ret; +} + +static ssize_t +idmap_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) +{ + struct request_key_auth *rka; + struct rpc_inode *rpci = RPC_I(file_inode(filp)); + struct idmap *idmap = (struct idmap *)rpci->private; + struct idmap_legacy_upcalldata *data; + struct key *authkey; + struct idmap_msg im; + size_t namelen_in; + int ret = -ENOKEY; + + /* If instantiation is successful, anyone waiting for key construction + * will have been woken up and someone else may now have used + * idmap_key_cons - so after this point we may no longer touch it. + */ + data = xchg(&idmap->idmap_upcall_data, NULL); + if (data == NULL) + goto out_noupcall; + + authkey = data->authkey; + rka = get_request_key_auth(authkey); + + if (mlen != sizeof(im)) { + ret = -ENOSPC; + goto out; + } + + if (copy_from_user(&im, src, mlen) != 0) { + ret = -EFAULT; + goto out; + } + + if (!(im.im_status & IDMAP_STATUS_SUCCESS)) { + ret = -ENOKEY; + goto out; + } + + namelen_in = strnlen(im.im_name, IDMAP_NAMESZ); + if (namelen_in == 0 || namelen_in == IDMAP_NAMESZ) { + ret = -EINVAL; + goto out; + } + + ret = nfs_idmap_read_and_verify_message(&im, &data->idmap_msg, + rka->target_key, authkey); + if (ret >= 0) { + key_set_timeout(rka->target_key, nfs_idmap_cache_timeout); + ret = mlen; + } + +out: + nfs_idmap_complete_pipe_upcall(data, ret); +out_noupcall: + return ret; +} + +static void +idmap_pipe_destroy_msg(struct rpc_pipe_msg *msg) +{ + struct idmap_legacy_upcalldata *data = container_of(msg, + struct idmap_legacy_upcalldata, + pipe_msg); + struct idmap *idmap = data->idmap; + + if (msg->errno) + nfs_idmap_abort_pipe_upcall(idmap, data, msg->errno); +} + +static void +idmap_release_pipe(struct inode *inode) +{ + struct rpc_inode *rpci = RPC_I(inode); + struct idmap *idmap = (struct idmap *)rpci->private; + struct idmap_legacy_upcalldata *data; + + data = xchg(&idmap->idmap_upcall_data, NULL); + if (data) + nfs_idmap_complete_pipe_upcall(data, -EPIPE); +} + +int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_t namelen, kuid_t *uid) +{ + struct idmap *idmap = server->nfs_client->cl_idmap; + __u32 id = -1; + int ret = 0; + + if (!nfs_map_string_to_numeric(name, namelen, &id)) + ret = nfs_idmap_lookup_id(name, namelen, "uid", &id, idmap); + if (ret == 0) { + *uid = make_kuid(idmap_userns(idmap), id); + if (!uid_valid(*uid)) + ret = -ERANGE; + } + trace_nfs4_map_name_to_uid(name, namelen, id, ret); + return ret; +} + +int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size_t namelen, kgid_t *gid) +{ + struct idmap *idmap = server->nfs_client->cl_idmap; + __u32 id = -1; + int ret = 0; + + if (!nfs_map_string_to_numeric(name, namelen, &id)) + ret = nfs_idmap_lookup_id(name, namelen, "gid", &id, idmap); + if (ret == 0) { + *gid = make_kgid(idmap_userns(idmap), id); + if (!gid_valid(*gid)) + ret = -ERANGE; + } + trace_nfs4_map_group_to_gid(name, namelen, id, ret); + return ret; +} + +int nfs_map_uid_to_name(const struct nfs_server *server, kuid_t uid, char *buf, size_t buflen) +{ + struct idmap *idmap = server->nfs_client->cl_idmap; + int ret = -EINVAL; + __u32 id; + + id = from_kuid_munged(idmap_userns(idmap), uid); + if (!(server->caps & NFS_CAP_UIDGID_NOMAP)) + ret = nfs_idmap_lookup_name(id, "user", buf, buflen, idmap); + if (ret < 0) + ret = nfs_map_numeric_to_string(id, buf, buflen); + trace_nfs4_map_uid_to_name(buf, ret, id, ret); + return ret; +} +int nfs_map_gid_to_group(const struct nfs_server *server, kgid_t gid, char *buf, size_t buflen) +{ + struct idmap *idmap = server->nfs_client->cl_idmap; + int ret = -EINVAL; + __u32 id; + + id = from_kgid_munged(idmap_userns(idmap), gid); + if (!(server->caps & NFS_CAP_UIDGID_NOMAP)) + ret = nfs_idmap_lookup_name(id, "group", buf, buflen, idmap); + if (ret < 0) + ret = nfs_map_numeric_to_string(id, buf, buflen); + trace_nfs4_map_gid_to_group(buf, ret, id, ret); + return ret; +} diff --git a/fs/nfs/nfs4idmap.h b/fs/nfs/nfs4idmap.h new file mode 100644 index 000000000..de44d7330 --- /dev/null +++ b/fs/nfs/nfs4idmap.h @@ -0,0 +1,68 @@ +/* + * fs/nfs/nfs4idmap.h + * + * UID and GID to name mapping for clients. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Marius Aamodt Eriksen + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ +#ifndef NFS_IDMAP_H +#define NFS_IDMAP_H + +#include +#include + + +/* Forward declaration to make this header independent of others */ +struct nfs_client; +struct nfs_server; +struct nfs_fattr; +struct nfs4_string; + +int nfs_idmap_init(void); +void nfs_idmap_quit(void); +int nfs_idmap_new(struct nfs_client *); +void nfs_idmap_delete(struct nfs_client *); + +void nfs_fattr_init_names(struct nfs_fattr *fattr, + struct nfs4_string *owner_name, + struct nfs4_string *group_name); +void nfs_fattr_free_names(struct nfs_fattr *); +void nfs_fattr_map_and_free_names(struct nfs_server *, struct nfs_fattr *); + +int nfs_map_name_to_uid(const struct nfs_server *, const char *, size_t, kuid_t *); +int nfs_map_group_to_gid(const struct nfs_server *, const char *, size_t, kgid_t *); +int nfs_map_uid_to_name(const struct nfs_server *, kuid_t, char *, size_t); +int nfs_map_gid_to_group(const struct nfs_server *, kgid_t, char *, size_t); + +int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *res); + +extern unsigned int nfs_idmap_cache_timeout; +#endif /* NFS_IDMAP_H */ diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c new file mode 100644 index 000000000..2a03f02bb --- /dev/null +++ b/fs/nfs/nfs4namespace.c @@ -0,0 +1,564 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/nfs4namespace.c + * + * Copyright (C) 2005 Trond Myklebust + * - Modified by David Howells + * + * NFSv4 namespace + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" +#include "nfs4_fs.h" +#include "nfs.h" +#include "dns_resolve.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +/* + * Work out the length that an NFSv4 path would render to as a standard posix + * path, with a leading slash but no terminating slash. + */ +static ssize_t nfs4_pathname_len(const struct nfs4_pathname *pathname) +{ + ssize_t len = 0; + int i; + + for (i = 0; i < pathname->ncomponents; i++) { + const struct nfs4_string *component = &pathname->components[i]; + + if (component->len > NAME_MAX) + goto too_long; + len += 1 + component->len; /* Adding "/foo" */ + if (len > PATH_MAX) + goto too_long; + } + return len; + +too_long: + return -ENAMETOOLONG; +} + +/* + * Convert the NFSv4 pathname components into a standard posix path. + */ +static char *nfs4_pathname_string(const struct nfs4_pathname *pathname, + unsigned short *_len) +{ + ssize_t len; + char *buf, *p; + int i; + + len = nfs4_pathname_len(pathname); + if (len < 0) + return ERR_PTR(len); + *_len = len; + + p = buf = kmalloc(len + 1, GFP_KERNEL); + if (!buf) + return ERR_PTR(-ENOMEM); + + for (i = 0; i < pathname->ncomponents; i++) { + const struct nfs4_string *component = &pathname->components[i]; + + *p++ = '/'; + memcpy(p, component->data, component->len); + p += component->len; + } + + *p = 0; + return buf; +} + +/* + * return the path component of ":" + * nfspath - the ":" string + * end - one past the last char that could contain ":" + * returns NULL on failure + */ +static char *nfs_path_component(const char *nfspath, const char *end) +{ + char *p; + + if (*nfspath == '[') { + /* parse [] escaped IPv6 addrs */ + p = strchr(nfspath, ']'); + if (p != NULL && ++p < end && *p == ':') + return p + 1; + } else { + /* otherwise split on first colon */ + p = strchr(nfspath, ':'); + if (p != NULL && p < end) + return p + 1; + } + return NULL; +} + +/* + * Determine the mount path as a string + */ +static char *nfs4_path(struct dentry *dentry, char *buffer, ssize_t buflen) +{ + char *limit; + char *path = nfs_path(&limit, dentry, buffer, buflen, + NFS_PATH_CANONICAL); + if (!IS_ERR(path)) { + char *path_component = nfs_path_component(path, limit); + if (path_component) + return path_component; + } + return path; +} + +/* + * Check that fs_locations::fs_root [RFC3530 6.3] is a prefix for what we + * believe to be the server path to this dentry + */ +static int nfs4_validate_fspath(struct dentry *dentry, + const struct nfs4_fs_locations *locations, + struct nfs_fs_context *ctx) +{ + const char *path; + char *fs_path; + unsigned short len; + char *buf; + int n; + + buf = kmalloc(4096, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + path = nfs4_path(dentry, buf, 4096); + if (IS_ERR(path)) { + kfree(buf); + return PTR_ERR(path); + } + + fs_path = nfs4_pathname_string(&locations->fs_path, &len); + if (IS_ERR(fs_path)) { + kfree(buf); + return PTR_ERR(fs_path); + } + + n = strncmp(path, fs_path, len); + kfree(buf); + kfree(fs_path); + if (n != 0) { + dprintk("%s: path %s does not begin with fsroot %s\n", + __func__, path, ctx->nfs_server.export_path); + return -ENOENT; + } + + return 0; +} + +size_t nfs_parse_server_name(char *string, size_t len, struct sockaddr_storage *ss, + size_t salen, struct net *net, int port) +{ + struct sockaddr *sa = (struct sockaddr *)ss; + ssize_t ret; + + ret = rpc_pton(net, string, len, sa, salen); + if (ret == 0) { + ret = rpc_uaddr2sockaddr(net, string, len, sa, salen); + if (ret == 0) { + ret = nfs_dns_resolve_name(net, string, len, ss, salen); + if (ret < 0) + ret = 0; + } + } else if (port) { + rpc_set_port(sa, port); + } + return ret; +} + +/** + * nfs_find_best_sec - Find a security mechanism supported locally + * @clnt: pointer to rpc_clnt + * @server: NFS server struct + * @flavors: List of security tuples returned by SECINFO procedure + * + * Return an rpc client that uses the first security mechanism in + * "flavors" that is locally supported. The "flavors" array + * is searched in the order returned from the server, per RFC 3530 + * recommendation and each flavor is checked for membership in the + * sec= mount option list if it exists. + * + * Return -EPERM if no matching flavor is found in the array. + * + * Please call rpc_shutdown_client() when you are done with this rpc client. + * + */ +static struct rpc_clnt *nfs_find_best_sec(struct rpc_clnt *clnt, + struct nfs_server *server, + struct nfs4_secinfo_flavors *flavors) +{ + rpc_authflavor_t pflavor; + struct nfs4_secinfo4 *secinfo; + unsigned int i; + + for (i = 0; i < flavors->num_flavors; i++) { + secinfo = &flavors->flavors[i]; + + switch (secinfo->flavor) { + case RPC_AUTH_NULL: + case RPC_AUTH_UNIX: + case RPC_AUTH_GSS: + pflavor = rpcauth_get_pseudoflavor(secinfo->flavor, + &secinfo->flavor_info); + /* does the pseudoflavor match a sec= mount opt? */ + if (pflavor != RPC_AUTH_MAXFLAVOR && + nfs_auth_info_match(&server->auth_info, pflavor)) { + struct rpc_clnt *new; + struct rpc_cred *cred; + + /* Cloning creates an rpc_auth for the flavor */ + new = rpc_clone_client_set_auth(clnt, pflavor); + if (IS_ERR(new)) + continue; + /** + * Check that the user actually can use the + * flavor. This is mostly for RPC_AUTH_GSS + * where cr_init obtains a gss context + */ + cred = rpcauth_lookupcred(new->cl_auth, 0); + if (IS_ERR(cred)) { + rpc_shutdown_client(new); + continue; + } + put_rpccred(cred); + return new; + } + } + } + return ERR_PTR(-EPERM); +} + +/** + * nfs4_negotiate_security - in response to an NFS4ERR_WRONGSEC on lookup, + * return an rpc_clnt that uses the best available security flavor with + * respect to the secinfo flavor list and the sec= mount options. + * + * @clnt: RPC client to clone + * @inode: directory inode + * @name: lookup name + * + * Please call rpc_shutdown_client() when you are done with this rpc client. + */ +struct rpc_clnt * +nfs4_negotiate_security(struct rpc_clnt *clnt, struct inode *inode, + const struct qstr *name) +{ + struct page *page; + struct nfs4_secinfo_flavors *flavors; + struct rpc_clnt *new; + int err; + + page = alloc_page(GFP_KERNEL); + if (!page) + return ERR_PTR(-ENOMEM); + + flavors = page_address(page); + + err = nfs4_proc_secinfo(inode, name, flavors); + if (err < 0) { + new = ERR_PTR(err); + goto out; + } + + new = nfs_find_best_sec(clnt, NFS_SERVER(inode), flavors); + +out: + put_page(page); + return new; +} + +static int try_location(struct fs_context *fc, + const struct nfs4_fs_location *location) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + unsigned int len, s; + char *export_path, *source, *p; + int ret = -ENOENT; + + /* Allocate a buffer big enough to hold any of the hostnames plus a + * terminating char and also a buffer big enough to hold the hostname + * plus a colon plus the path. + */ + len = 0; + for (s = 0; s < location->nservers; s++) { + const struct nfs4_string *buf = &location->servers[s]; + if (buf->len > len) + len = buf->len; + } + + kfree(ctx->nfs_server.hostname); + ctx->nfs_server.hostname = kmalloc(len + 1, GFP_KERNEL); + if (!ctx->nfs_server.hostname) + return -ENOMEM; + + export_path = nfs4_pathname_string(&location->rootpath, + &ctx->nfs_server.export_path_len); + if (IS_ERR(export_path)) + return PTR_ERR(export_path); + + kfree(ctx->nfs_server.export_path); + ctx->nfs_server.export_path = export_path; + + source = kmalloc(len + 1 + ctx->nfs_server.export_path_len + 1, + GFP_KERNEL); + if (!source) + return -ENOMEM; + + kfree(fc->source); + fc->source = source; + for (s = 0; s < location->nservers; s++) { + const struct nfs4_string *buf = &location->servers[s]; + + if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len)) + continue; + + ctx->nfs_server.addrlen = + nfs_parse_server_name(buf->data, buf->len, + &ctx->nfs_server._address, + sizeof(ctx->nfs_server._address), + fc->net_ns, 0); + if (ctx->nfs_server.addrlen == 0) + continue; + + rpc_set_port(&ctx->nfs_server.address, NFS_PORT); + + memcpy(ctx->nfs_server.hostname, buf->data, buf->len); + ctx->nfs_server.hostname[buf->len] = '\0'; + + p = source; + memcpy(p, buf->data, buf->len); + p += buf->len; + *p++ = ':'; + memcpy(p, ctx->nfs_server.export_path, ctx->nfs_server.export_path_len); + p += ctx->nfs_server.export_path_len; + *p = 0; + + ret = nfs4_get_referral_tree(fc); + if (ret == 0) + return 0; + } + + return ret; +} + +/** + * nfs_follow_referral - set up mountpoint when hitting a referral on moved error + * @fc: pointer to struct nfs_fs_context + * @locations: array of NFSv4 server location information + * + */ +static int nfs_follow_referral(struct fs_context *fc, + const struct nfs4_fs_locations *locations) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + int loc, error; + + if (locations == NULL || locations->nlocations <= 0) + return -ENOENT; + + dprintk("%s: referral at %pd2\n", __func__, ctx->clone_data.dentry); + + /* Ensure fs path is a prefix of current dentry path */ + error = nfs4_validate_fspath(ctx->clone_data.dentry, locations, ctx); + if (error < 0) + return error; + + error = -ENOENT; + for (loc = 0; loc < locations->nlocations; loc++) { + const struct nfs4_fs_location *location = &locations->locations[loc]; + + if (location == NULL || location->nservers <= 0 || + location->rootpath.ncomponents == 0) + continue; + + error = try_location(fc, location); + if (error == 0) + return 0; + } + + return error; +} + +/* + * nfs_do_refmount - handle crossing a referral on server + * @dentry - dentry of referral + * + */ +static int nfs_do_refmount(struct fs_context *fc, struct rpc_clnt *client) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct dentry *dentry, *parent; + struct nfs4_fs_locations *fs_locations = NULL; + struct page *page; + int err = -ENOMEM; + + /* BUG_ON(IS_ROOT(dentry)); */ + page = alloc_page(GFP_KERNEL); + if (!page) + return -ENOMEM; + + fs_locations = kmalloc_obj(struct nfs4_fs_locations); + if (!fs_locations) + goto out_free; + fs_locations->fattr = nfs_alloc_fattr(); + if (!fs_locations->fattr) + goto out_free_2; + + /* Get locations */ + dentry = ctx->clone_data.dentry; + parent = dget_parent(dentry); + dprintk("%s: getting locations for %pd2\n", + __func__, dentry); + + err = nfs4_proc_fs_locations(client, d_inode(parent), &dentry->d_name, fs_locations, page); + dput(parent); + if (err != 0) + goto out_free_3; + + err = -ENOENT; + if (fs_locations->nlocations <= 0 || + fs_locations->fs_path.ncomponents <= 0) + goto out_free_3; + + err = nfs_follow_referral(fc, fs_locations); +out_free_3: + kfree(fs_locations->fattr); +out_free_2: + kfree(fs_locations); +out_free: + __free_page(page); + return err; +} + +int nfs4_submount(struct fs_context *fc, struct nfs_server *server) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct dentry *dentry = ctx->clone_data.dentry; + struct dentry *parent = dget_parent(dentry); + struct inode *dir = d_inode(parent); + struct rpc_clnt *client; + int ret; + + /* Look it up again to get its attributes and sec flavor */ + client = nfs4_proc_lookup_mountpoint(dir, dentry, ctx->mntfh, + ctx->clone_data.fattr); + dput(parent); + if (IS_ERR(client)) + return PTR_ERR(client); + + ctx->selected_flavor = client->cl_auth->au_flavor; + if (ctx->clone_data.fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) { + ret = nfs_do_refmount(fc, client); + } else { + ret = nfs_do_submount(fc); + } + + rpc_shutdown_client(client); + return ret; +} + +/* + * Try one location from the fs_locations array. + * + * Returns zero on success, or a negative errno value. + */ +static int nfs4_try_replacing_one_location(struct nfs_server *server, + const struct nfs4_fs_location *location) +{ + struct net *net = rpc_net_ns(server->client); + struct sockaddr_storage *sap; + unsigned int s; + size_t salen; + int error; + + sap = kmalloc_obj(*sap); + if (sap == NULL) + return -ENOMEM; + + error = -ENOENT; + for (s = 0; s < location->nservers; s++) { + const struct nfs4_string *buf = &location->servers[s]; + char *hostname; + + if (buf->len <= 0 || buf->len > PAGE_SIZE) + continue; + + if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len) != NULL) + continue; + + salen = nfs_parse_server_name(buf->data, buf->len, + sap, sizeof(*sap), net, 0); + if (salen == 0) + continue; + rpc_set_port((struct sockaddr *)sap, NFS_PORT); + + error = -ENOMEM; + hostname = kmemdup_nul(buf->data, buf->len, GFP_KERNEL); + if (hostname == NULL) + break; + + error = nfs4_update_server(server, hostname, sap, salen, net); + kfree(hostname); + if (error == 0) + break; + } + + kfree(sap); + return error; +} + +/** + * nfs4_replace_transport - set up transport to destination server + * + * @server: export being migrated + * @locations: fs_locations array + * + * Returns zero on success, or a negative errno value. + * + * The client tries all the entries in the "locations" array, in the + * order returned by the server, until one works or the end of the + * array is reached. + */ +int nfs4_replace_transport(struct nfs_server *server, + const struct nfs4_fs_locations *locations) +{ + int loc, error; + + error = -ENOENT; + if (locations == NULL || locations->nlocations <= 0) + goto out; + + for (loc = 0; loc < locations->nlocations; loc++) { + const struct nfs4_fs_location *location = + &locations->locations[loc]; + + if (location == NULL || location->nservers <= 0 || + location->rootpath.ncomponents == 0) + continue; + + error = nfs4_try_replacing_one_location(server, location); + if (error == 0) + break; + } + +out: + return error; +} diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c new file mode 100644 index 000000000..5709c6fea --- /dev/null +++ b/fs/nfs/nfs4proc.c @@ -0,0 +1,10797 @@ +/* + * fs/nfs/nfs4proc.c + * + * Client-side procedure declarations for NFSv4. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Kendrick Smith + * Andy Adamson + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "delegation.h" +#include "internal.h" +#include "iostat.h" +#include "callback.h" +#include "pnfs.h" +#include "netns.h" +#include "sysfs.h" +#include "nfs4idmap.h" +#include "nfs4session.h" +#include "fscache.h" +#include "nfs40.h" +#include "nfs42.h" + +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PROC + +#define NFS4_BITMASK_SZ 3 + +#define NFS4_POLL_RETRY_MIN (HZ/10) +#define NFS4_POLL_RETRY_MAX (15*HZ) + +/* file attributes which can be mapped to nfs attributes */ +#define NFS4_VALID_ATTRS (ATTR_MODE \ + | ATTR_UID \ + | ATTR_GID \ + | ATTR_SIZE \ + | ATTR_ATIME \ + | ATTR_MTIME \ + | ATTR_CTIME \ + | ATTR_ATIME_SET \ + | ATTR_MTIME_SET) + +struct nfs4_opendata; +static int _nfs4_recover_proc_open(struct nfs4_opendata *data); +static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); +static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr); +static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode); +static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, + struct nfs_fattr *fattr, struct iattr *sattr, + struct nfs_open_context *ctx, struct nfs4_label *ilabel); +static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, + const struct cred *cred, + struct nfs4_slot *slot, + bool is_privileged); +static int nfs41_test_stateid(struct nfs_server *, const nfs4_stateid *, + const struct cred *); +static int nfs41_free_stateid(struct nfs_server *, nfs4_stateid *, + const struct cred *, bool); + +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +static inline struct nfs4_label * +nfs4_label_init_security(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, struct nfs4_label *label) +{ + struct lsm_context shim; + int err; + + if (label == NULL) + return NULL; + + if (nfs_server_capable(dir, NFS_CAP_SECURITY_LABEL) == 0) + return NULL; + + label->lfs = 0; + label->pi = 0; + label->len = 0; + label->label = NULL; + + err = security_dentry_init_security(dentry, sattr->ia_mode, + &dentry->d_name, NULL, &shim); + if (err) + return NULL; + + label->lsmid = shim.id; + label->label = shim.context; + label->len = shim.len; + return label; +} +static inline void +nfs4_label_release_security(struct nfs4_label *label) +{ + struct lsm_context shim; + + if (label) { + shim.context = label->label; + shim.len = label->len; + shim.id = label->lsmid; + security_release_secctx(&shim); + } +} +static inline u32 *nfs4_bitmask(struct nfs_server *server, struct nfs4_label *label) +{ + if (label) + return server->attr_bitmask; + + return server->attr_bitmask_nl; +} +#else +static inline struct nfs4_label * +nfs4_label_init_security(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, struct nfs4_label *l) +{ return NULL; } +static inline void +nfs4_label_release_security(struct nfs4_label *label) +{ return; } +static inline u32 * +nfs4_bitmask(struct nfs_server *server, struct nfs4_label *label) +{ return server->attr_bitmask; } +#endif + +/* Prevent leaks of NFSv4 errors into userland */ +static int nfs4_map_errors(int err) +{ + if (err >= -1000) + return err; + switch (err) { + case -NFS4ERR_RESOURCE: + case -NFS4ERR_LAYOUTTRYLATER: + case -NFS4ERR_RECALLCONFLICT: + case -NFS4ERR_RETURNCONFLICT: + return -EREMOTEIO; + case -NFS4ERR_WRONGSEC: + case -NFS4ERR_WRONG_CRED: + return -EPERM; + case -NFS4ERR_BADOWNER: + case -NFS4ERR_BADNAME: + return -EINVAL; + case -NFS4ERR_SHARE_DENIED: + return -EACCES; + case -NFS4ERR_MINOR_VERS_MISMATCH: + return -EPROTONOSUPPORT; + case -NFS4ERR_FILE_OPEN: + return -EBUSY; + case -NFS4ERR_NOT_SAME: + return -ENOTSYNC; + case -ENETDOWN: + case -ENETUNREACH: + break; + default: + dprintk("%s could not handle NFSv4 error %d\n", + __func__, -err); + break; + } + return -EIO; +} + +/* + * This is our standard bitmap for GETATTR requests. + */ +const u32 nfs4_fattr_bitmap[3] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_CHANGE + | FATTR4_WORD0_SIZE + | FATTR4_WORD0_FSID + | FATTR4_WORD0_FILEID, + FATTR4_WORD1_MODE + | FATTR4_WORD1_NUMLINKS + | FATTR4_WORD1_OWNER + | FATTR4_WORD1_OWNER_GROUP + | FATTR4_WORD1_RAWDEV + | FATTR4_WORD1_SPACE_USED + | FATTR4_WORD1_TIME_ACCESS + | FATTR4_WORD1_TIME_CREATE + | FATTR4_WORD1_TIME_METADATA + | FATTR4_WORD1_TIME_MODIFY + | FATTR4_WORD1_MOUNTED_ON_FILEID, +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + FATTR4_WORD2_SECURITY_LABEL +#endif +}; + +static const u32 nfs4_pnfs_open_bitmap[3] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_CHANGE + | FATTR4_WORD0_SIZE + | FATTR4_WORD0_FSID + | FATTR4_WORD0_FILEID, + FATTR4_WORD1_MODE + | FATTR4_WORD1_NUMLINKS + | FATTR4_WORD1_OWNER + | FATTR4_WORD1_OWNER_GROUP + | FATTR4_WORD1_RAWDEV + | FATTR4_WORD1_SPACE_USED + | FATTR4_WORD1_TIME_ACCESS + | FATTR4_WORD1_TIME_CREATE + | FATTR4_WORD1_TIME_METADATA + | FATTR4_WORD1_TIME_MODIFY, + FATTR4_WORD2_MDSTHRESHOLD +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + | FATTR4_WORD2_SECURITY_LABEL +#endif +}; + +static const u32 nfs4_open_noattr_bitmap[3] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_FILEID, +}; + +const u32 nfs4_statfs_bitmap[3] = { + FATTR4_WORD0_FILES_AVAIL + | FATTR4_WORD0_FILES_FREE + | FATTR4_WORD0_FILES_TOTAL, + FATTR4_WORD1_SPACE_AVAIL + | FATTR4_WORD1_SPACE_FREE + | FATTR4_WORD1_SPACE_TOTAL +}; + +const u32 nfs4_pathconf_bitmap[3] = { + FATTR4_WORD0_MAXLINK + | FATTR4_WORD0_MAXNAME, + 0 +}; + +const u32 nfs4_fsinfo_bitmap[3] = { FATTR4_WORD0_MAXFILESIZE + | FATTR4_WORD0_MAXREAD + | FATTR4_WORD0_MAXWRITE + | FATTR4_WORD0_LEASE_TIME, + FATTR4_WORD1_TIME_DELTA + | FATTR4_WORD1_FS_LAYOUT_TYPES, + FATTR4_WORD2_LAYOUT_BLKSIZE + | FATTR4_WORD2_CLONE_BLKSIZE + | FATTR4_WORD2_CHANGE_ATTR_TYPE + | FATTR4_WORD2_XATTR_SUPPORT +}; + +const u32 nfs4_fs_locations_bitmap[3] = { + FATTR4_WORD0_CHANGE + | FATTR4_WORD0_SIZE + | FATTR4_WORD0_FSID + | FATTR4_WORD0_FILEID + | FATTR4_WORD0_FS_LOCATIONS, + FATTR4_WORD1_OWNER + | FATTR4_WORD1_OWNER_GROUP + | FATTR4_WORD1_RAWDEV + | FATTR4_WORD1_SPACE_USED + | FATTR4_WORD1_TIME_ACCESS + | FATTR4_WORD1_TIME_METADATA + | FATTR4_WORD1_TIME_MODIFY + | FATTR4_WORD1_MOUNTED_ON_FILEID, +}; + +static void nfs4_bitmap_copy_adjust(__u32 *dst, const __u32 *src, + struct inode *inode, unsigned long flags) +{ + unsigned long cache_validity; + + memcpy(dst, src, NFS4_BITMASK_SZ*sizeof(*dst)); + if (!inode || !nfs_have_read_or_write_delegation(inode)) + return; + + cache_validity = READ_ONCE(NFS_I(inode)->cache_validity) | flags; + + /* Remove the attributes over which we have full control */ + dst[1] &= ~FATTR4_WORD1_RAWDEV; + if (!(cache_validity & NFS_INO_INVALID_SIZE)) + dst[0] &= ~FATTR4_WORD0_SIZE; + + if (!(cache_validity & NFS_INO_INVALID_CHANGE)) + dst[0] &= ~FATTR4_WORD0_CHANGE; + + if (!(cache_validity & NFS_INO_INVALID_MODE)) + dst[1] &= ~FATTR4_WORD1_MODE; + if (!(cache_validity & NFS_INO_INVALID_OTHER)) + dst[1] &= ~(FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP); + + if (!(cache_validity & NFS_INO_INVALID_BTIME)) + dst[1] &= ~FATTR4_WORD1_TIME_CREATE; + + if (nfs_have_delegated_mtime(inode)) { + if (!(cache_validity & NFS_INO_INVALID_ATIME)) + dst[1] &= ~(FATTR4_WORD1_TIME_ACCESS|FATTR4_WORD1_TIME_ACCESS_SET); + if (!(cache_validity & NFS_INO_INVALID_MTIME)) + dst[1] &= ~(FATTR4_WORD1_TIME_MODIFY|FATTR4_WORD1_TIME_MODIFY_SET); + if (!(cache_validity & NFS_INO_INVALID_CTIME)) + dst[1] &= ~(FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY_SET); + } else if (nfs_have_delegated_atime(inode)) { + if (!(cache_validity & NFS_INO_INVALID_ATIME)) + dst[1] &= ~(FATTR4_WORD1_TIME_ACCESS|FATTR4_WORD1_TIME_ACCESS_SET); + } +} + +static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry, + struct nfs4_readdir_arg *readdir) +{ + unsigned int attrs = FATTR4_WORD0_FILEID | FATTR4_WORD0_TYPE; + __be32 *start, *p; + + if (cookie > 2) { + readdir->cookie = cookie; + memcpy(&readdir->verifier, verifier, sizeof(readdir->verifier)); + return; + } + + readdir->cookie = 0; + memset(&readdir->verifier, 0, sizeof(readdir->verifier)); + if (cookie == 2) + return; + + /* + * NFSv4 servers do not return entries for '.' and '..' + * Therefore, we fake these entries here. We let '.' + * have cookie 0 and '..' have cookie 1. Note that + * when talking to the server, we always send cookie 0 + * instead of 1 or 2. + */ + start = p = kmap_atomic(*readdir->pages); + + if (cookie == 0) { + *p++ = xdr_one; /* next */ + *p++ = xdr_zero; /* cookie, first word */ + *p++ = xdr_one; /* cookie, second word */ + *p++ = xdr_one; /* entry len */ + memcpy(p, ".\0\0\0", 4); /* entry */ + p++; + *p++ = xdr_one; /* bitmap length */ + *p++ = htonl(attrs); /* bitmap */ + *p++ = htonl(12); /* attribute buffer length */ + *p++ = htonl(NF4DIR); + p = xdr_encode_hyper(p, d_inode(dentry)->i_ino); + } + + *p++ = xdr_one; /* next */ + *p++ = xdr_zero; /* cookie, first word */ + *p++ = xdr_two; /* cookie, second word */ + *p++ = xdr_two; /* entry len */ + memcpy(p, "..\0\0", 4); /* entry */ + p++; + *p++ = xdr_one; /* bitmap length */ + *p++ = htonl(attrs); /* bitmap */ + *p++ = htonl(12); /* attribute buffer length */ + *p++ = htonl(NF4DIR); + spin_lock(&dentry->d_lock); + p = xdr_encode_hyper(p, d_inode(dentry->d_parent)->i_ino); + spin_unlock(&dentry->d_lock); + + readdir->pgbase = (char *)p - (char *)start; + readdir->count -= readdir->pgbase; + kunmap_atomic(start); +} + +static void nfs4_fattr_set_prechange(struct nfs_fattr *fattr, u64 version) +{ + if (!(fattr->valid & NFS_ATTR_FATTR_PRECHANGE)) { + fattr->pre_change_attr = version; + fattr->valid |= NFS_ATTR_FATTR_PRECHANGE; + } +} + +static void nfs4_test_and_free_stateid(struct nfs_server *server, + nfs4_stateid *stateid, + const struct cred *cred) +{ + const struct nfs4_minor_version_ops *ops = server->nfs_client->cl_mvops; + + ops->test_and_free_expired(server, stateid, cred); +} + +static void __nfs4_free_revoked_stateid(struct nfs_server *server, + nfs4_stateid *stateid, + const struct cred *cred) +{ + stateid->type = NFS4_REVOKED_STATEID_TYPE; + nfs4_test_and_free_stateid(server, stateid, cred); +} + +static void nfs4_free_revoked_stateid(struct nfs_server *server, + const nfs4_stateid *stateid, + const struct cred *cred) +{ + nfs4_stateid tmp; + + nfs4_stateid_copy(&tmp, stateid); + __nfs4_free_revoked_stateid(server, &tmp, cred); +} + +static long nfs4_update_delay(long *timeout) +{ + long ret; + if (!timeout) + return NFS4_POLL_RETRY_MAX; + if (*timeout <= 0) + *timeout = NFS4_POLL_RETRY_MIN; + if (*timeout > NFS4_POLL_RETRY_MAX) + *timeout = NFS4_POLL_RETRY_MAX; + ret = *timeout; + *timeout <<= 1; + return ret; +} + +static int nfs4_delay_killable(long *timeout) +{ + might_sleep(); + + if (unlikely(nfs_current_task_exiting())) + return -EINTR; + __set_current_state(TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(nfs4_update_delay(timeout)); + if (!__fatal_signal_pending(current)) + return 0; + return -EINTR; +} + +static int nfs4_delay_interruptible(long *timeout) +{ + might_sleep(); + + if (unlikely(nfs_current_task_exiting())) + return -EINTR; + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(nfs4_update_delay(timeout)); + if (!signal_pending(current)) + return 0; + return __fatal_signal_pending(current) ? -EINTR :-ERESTARTSYS; +} + +static int nfs4_delay(long *timeout, bool interruptible) +{ + if (interruptible) + return nfs4_delay_interruptible(timeout); + return nfs4_delay_killable(timeout); +} + +static const nfs4_stateid * +nfs4_recoverable_stateid(const nfs4_stateid *stateid) +{ + if (!stateid) + return NULL; + switch (stateid->type) { + case NFS4_OPEN_STATEID_TYPE: + case NFS4_LOCK_STATEID_TYPE: + case NFS4_DELEGATION_STATEID_TYPE: + return stateid; + default: + break; + } + return NULL; +} + +/* This is the error handling routine for processes that are allowed + * to sleep. + */ +static int nfs4_do_handle_exception(struct nfs_server *server, + int errorcode, struct nfs4_exception *exception) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state *state = exception->state; + const nfs4_stateid *stateid; + struct inode *inode = exception->inode; + int ret = errorcode; + + exception->delay = 0; + exception->recovering = 0; + exception->retry = 0; + + stateid = nfs4_recoverable_stateid(exception->stateid); + if (stateid == NULL && state != NULL) + stateid = nfs4_recoverable_stateid(&state->stateid); + + switch(errorcode) { + case 0: + return 0; + case -NFS4ERR_BADHANDLE: + case -ESTALE: + if (inode != NULL && S_ISREG(inode->i_mode)) + pnfs_destroy_layout(NFS_I(inode)); + break; + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_PARTNER_NO_AUTH: + if (inode != NULL && stateid != NULL) { + nfs_inode_find_state_and_recover(inode, + stateid); + goto wait_on_recovery; + } + fallthrough; + case -NFS4ERR_OPENMODE: + if (inode) { + int err; + + err = nfs_async_inode_return_delegation(inode, + stateid); + if (err == 0) + goto wait_on_recovery; + if (stateid != NULL && stateid->type == NFS4_DELEGATION_STATEID_TYPE) { + exception->retry = 1; + break; + } + } + if (state == NULL) + break; + ret = nfs4_schedule_stateid_recovery(server, state); + if (ret < 0) + break; + goto wait_on_recovery; + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_STALE_CLIENTID: + nfs4_schedule_lease_recovery(clp); + goto wait_on_recovery; + case -NFS4ERR_MOVED: + ret = nfs4_schedule_migration_recovery(server); + if (ret < 0) + break; + goto wait_on_recovery; + case -NFS4ERR_LEASE_MOVED: + nfs4_schedule_lease_moved_recovery(clp); + goto wait_on_recovery; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + /* Handled in nfs41_sequence_process() */ + goto wait_on_recovery; + case -NFS4ERR_FILE_OPEN: + if (exception->timeout > HZ) { + /* We have retried a decent amount, time to + * fail + */ + ret = -EBUSY; + break; + } + fallthrough; + case -NFS4ERR_DELAY: + nfs_inc_server_stats(server, NFSIOS_DELAY); + fallthrough; + case -NFS4ERR_GRACE: + case -NFS4ERR_LAYOUTTRYLATER: + case -NFS4ERR_RECALLCONFLICT: + case -NFS4ERR_RETURNCONFLICT: + exception->delay = 1; + return 0; + + case -NFS4ERR_RETRY_UNCACHED_REP: + case -NFS4ERR_OLD_STATEID: + exception->retry = 1; + break; + case -NFS4ERR_BADOWNER: + /* The following works around a Linux server bug! */ + case -NFS4ERR_BADNAME: + if (server->caps & NFS_CAP_UIDGID_NOMAP) { + server->caps &= ~NFS_CAP_UIDGID_NOMAP; + exception->retry = 1; + printk(KERN_WARNING "NFS: v4 server %s " + "does not accept raw " + "uid/gids. " + "Reenabling the idmapper.\n", + server->nfs_client->cl_hostname); + } + } + /* We failed to handle the error */ + return nfs4_map_errors(ret); +wait_on_recovery: + exception->recovering = 1; + return 0; +} + +/* + * Track the number of NFS4ERR_DELAY related retransmissions and return + * EAGAIN if the 'softerr' mount option is set, and we've exceeded the limit + * set by 'nfs_delay_retrans'. + */ +static int nfs4_exception_should_retrans(const struct nfs_server *server, + struct nfs4_exception *exception) +{ + if (server->flags & NFS_MOUNT_SOFTERR && nfs_delay_retrans >= 0) { + if (exception->retrans++ >= (unsigned short)nfs_delay_retrans) + return -EAGAIN; + } + return 0; +} + +/* This is the error handling routine for processes that are allowed + * to sleep. + */ +int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception) +{ + struct nfs_client *clp = server->nfs_client; + int ret; + + ret = nfs4_do_handle_exception(server, errorcode, exception); + if (exception->delay) { + int ret2 = nfs4_exception_should_retrans(server, exception); + if (ret2 < 0) { + exception->retry = 0; + return ret2; + } + ret = nfs4_delay(&exception->timeout, + exception->interruptible); + goto out_retry; + } + if (exception->recovering) { + if (exception->task_is_privileged) + return -EDEADLOCK; + ret = nfs4_wait_clnt_recover(clp); + if (test_bit(NFS_MIG_FAILED, &server->mig_status)) + return -EIO; + goto out_retry; + } + return ret; +out_retry: + if (ret == 0) + exception->retry = 1; + return ret; +} + +static int +nfs4_async_handle_exception(struct rpc_task *task, struct nfs_server *server, + int errorcode, struct nfs4_exception *exception) +{ + struct nfs_client *clp = server->nfs_client; + int ret; + + if ((task->tk_rpc_status == -ENETDOWN || + task->tk_rpc_status == -ENETUNREACH) && + task->tk_flags & RPC_TASK_NETUNREACH_FATAL) { + exception->delay = 0; + exception->recovering = 0; + exception->retry = 0; + return -EIO; + } + + ret = nfs4_do_handle_exception(server, errorcode, exception); + if (exception->delay) { + int ret2 = nfs4_exception_should_retrans(server, exception); + if (ret2 < 0) { + exception->retry = 0; + return ret2; + } + rpc_delay(task, nfs4_update_delay(&exception->timeout)); + goto out_retry; + } + if (exception->recovering) { + if (exception->task_is_privileged) + return -EDEADLOCK; + rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL); + if (test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) == 0) + rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task); + goto out_retry; + } + if (test_bit(NFS_MIG_FAILED, &server->mig_status)) + ret = -EIO; + return ret; +out_retry: + if (ret == 0) { + exception->retry = 1; + /* + * For NFS4ERR_MOVED, the client transport will need to + * be recomputed after migration recovery has completed. + */ + if (errorcode == -NFS4ERR_MOVED) + rpc_task_release_transport(task); + } + return ret; +} + +int +nfs4_async_handle_error(struct rpc_task *task, struct nfs_server *server, + struct nfs4_state *state, long *timeout) +{ + struct nfs4_exception exception = { + .state = state, + }; + + if (task->tk_status >= 0) + return 0; + if (timeout) + exception.timeout = *timeout; + task->tk_status = nfs4_async_handle_exception(task, server, + task->tk_status, + &exception); + if (exception.delay && timeout) + *timeout = exception.timeout; + if (exception.retry) + return -EAGAIN; + return 0; +} + +/* + * Return 'true' if 'clp' is using an rpc_client that is integrity protected + * or 'false' otherwise. + */ +static bool _nfs4_is_integrity_protected(struct nfs_client *clp) +{ + rpc_authflavor_t flavor = clp->cl_rpcclient->cl_auth->au_flavor; + return (flavor == RPC_AUTH_GSS_KRB5I) || (flavor == RPC_AUTH_GSS_KRB5P); +} + +void do_renew_lease(struct nfs_client *clp, unsigned long timestamp) +{ + spin_lock(&clp->cl_lock); + if (time_before(clp->cl_last_renewal,timestamp)) + clp->cl_last_renewal = timestamp; + spin_unlock(&clp->cl_lock); +} + +void renew_lease(const struct nfs_server *server, unsigned long timestamp) +{ + struct nfs_client *clp = server->nfs_client; + + if (!nfs4_has_session(clp)) + do_renew_lease(clp, timestamp); +} + +void nfs4_init_sequence(struct nfs_client *clp, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, int cache_reply, + int privileged) +{ + args->sa_slot = NULL; + args->sa_cache_this = cache_reply; + args->sa_privileged = privileged; + + res->sr_slot = NULL; + res->sr_slot_ops = clp->cl_mvops->sequence_slot_ops; +} + +static void nfs41_release_slot(struct nfs4_slot *slot) +{ + struct nfs4_session *session; + struct nfs4_slot_table *tbl; + bool send_new_highest_used_slotid = false; + + if (!slot) + return; + tbl = slot->table; + session = tbl->session; + + /* Bump the slot sequence number */ + if (slot->seq_done) + slot->seq_nr++; + slot->seq_done = 0; + + spin_lock(&tbl->slot_tbl_lock); + /* Be nice to the server: try to ensure that the last transmitted + * value for highest_user_slotid <= target_highest_slotid + */ + if (tbl->highest_used_slotid > tbl->target_highest_slotid) + send_new_highest_used_slotid = true; + + if (nfs41_wake_and_assign_slot(tbl, slot)) { + send_new_highest_used_slotid = false; + goto out_unlock; + } + nfs4_free_slot(tbl, slot); + + if (tbl->highest_used_slotid != NFS4_NO_SLOT) + send_new_highest_used_slotid = false; +out_unlock: + spin_unlock(&tbl->slot_tbl_lock); + if (send_new_highest_used_slotid) + nfs41_notify_server(session->clp); + if (waitqueue_active(&tbl->slot_waitq)) + wake_up_all(&tbl->slot_waitq); +} + +static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) +{ + nfs41_release_slot(res->sr_slot); + res->sr_slot = NULL; +} + +static void nfs4_slot_sequence_record_sent(struct nfs4_slot *slot, + u32 seqnr) +{ + if ((s32)(seqnr - slot->seq_nr_highest_sent) > 0) + slot->seq_nr_highest_sent = seqnr; +} +static void nfs4_slot_sequence_acked(struct nfs4_slot *slot, u32 seqnr) +{ + nfs4_slot_sequence_record_sent(slot, seqnr); + slot->seq_nr_last_acked = seqnr; +} + +static void nfs4_probe_sequence(struct nfs_client *client, const struct cred *cred, + struct nfs4_slot *slot) +{ + struct rpc_task *task = _nfs41_proc_sequence(client, cred, slot, true); + if (!IS_ERR(task)) + rpc_put_task_async(task); +} + +static int nfs41_sequence_process(struct rpc_task *task, + struct nfs4_sequence_res *res) +{ + struct nfs4_session *session; + struct nfs4_slot *slot = res->sr_slot; + struct nfs_client *clp; + int status; + int ret = 1; + + if (slot == NULL) + goto out_noaction; + /* don't increment the sequence number if the task wasn't sent */ + if (!RPC_WAS_SENT(task) || slot->seq_done) + goto out; + + session = slot->table->session; + clp = session->clp; + + trace_nfs4_sequence_done(session, res); + + status = res->sr_status; + if (task->tk_status == -NFS4ERR_DEADSESSION) + status = -NFS4ERR_DEADSESSION; + + /* Check the SEQUENCE operation status */ + switch (status) { + case 0: + /* Mark this sequence number as having been acked */ + nfs4_slot_sequence_acked(slot, slot->seq_nr); + /* Update the slot's sequence and clientid lease timer */ + slot->seq_done = 1; + do_renew_lease(clp, res->sr_timestamp); + /* Check sequence flags */ + nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags, + !!slot->privileged); + nfs41_update_target_slotid(slot->table, slot, res); + break; + case 1: + /* + * sr_status remains 1 if an RPC level error occurred. + * The server may or may not have processed the sequence + * operation.. + */ + nfs4_slot_sequence_record_sent(slot, slot->seq_nr); + slot->seq_done = 1; + goto out; + case -NFS4ERR_DELAY: + /* The server detected a resend of the RPC call and + * returned NFS4ERR_DELAY as per Section 2.10.6.2 + * of RFC5661. + */ + dprintk("%s: slot=%u seq=%u: Operation in progress\n", + __func__, + slot->slot_nr, + slot->seq_nr); + goto out_retry; + case -NFS4ERR_RETRY_UNCACHED_REP: + case -NFS4ERR_SEQ_FALSE_RETRY: + /* + * The server thinks we tried to replay a request. + * Retry the call after bumping the sequence ID. + */ + nfs4_slot_sequence_acked(slot, slot->seq_nr); + goto retry_new_seq; + case -NFS4ERR_BADSLOT: + /* + * The slot id we used was probably retired. Try again + * using a different slot id. + */ + if (slot->slot_nr < slot->table->target_highest_slotid) + goto session_recover; + goto retry_nowait; + case -NFS4ERR_SEQ_MISORDERED: + nfs4_slot_sequence_record_sent(slot, slot->seq_nr); + /* + * Were one or more calls using this slot interrupted? + * If the server never received the request, then our + * transmitted slot sequence number may be too high. However, + * if the server did receive the request then it might + * accidentally give us a reply with a mismatched operation. + * We can sort this out by sending a lone sequence operation + * to the server on the same slot. + */ + if ((s32)(slot->seq_nr - slot->seq_nr_last_acked) > 1) { + slot->seq_nr--; + if (task->tk_msg.rpc_proc != &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE]) { + nfs4_probe_sequence(clp, task->tk_msg.rpc_cred, slot); + res->sr_slot = NULL; + } + goto retry_nowait; + } + /* + * RFC5661: + * A retry might be sent while the original request is + * still in progress on the replier. The replier SHOULD + * deal with the issue by returning NFS4ERR_DELAY as the + * reply to SEQUENCE or CB_SEQUENCE operation, but + * implementations MAY return NFS4ERR_SEQ_MISORDERED. + * + * Restart the search after a delay. + */ + slot->seq_nr = slot->seq_nr_highest_sent; + goto out_retry; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + goto session_recover; + default: + /* Just update the slot sequence no. */ + slot->seq_done = 1; + } +out: + /* The session may be reset by one of the error handlers. */ + dprintk("%s: Error %d free the slot \n", __func__, res->sr_status); +out_noaction: + return ret; +session_recover: + set_bit(NFS4_SLOT_TBL_DRAINING, &session->fc_slot_table.slot_tbl_state); + nfs4_schedule_session_recovery(session, status); + dprintk("%s ERROR: %d Reset session\n", __func__, status); + nfs41_sequence_free_slot(res); + goto out; +retry_new_seq: + ++slot->seq_nr; +retry_nowait: + if (rpc_restart_call_prepare(task)) { + nfs41_sequence_free_slot(res); + task->tk_status = 0; + ret = 0; + } + goto out; +out_retry: + if (!rpc_restart_call(task)) + goto out; + rpc_delay(task, NFS4_POLL_RETRY_MAX); + return 0; +} + +int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) +{ + if (!nfs41_sequence_process(task, res)) + return 0; + if (res->sr_slot != NULL) + nfs41_sequence_free_slot(res); + return 1; + +} +EXPORT_SYMBOL_GPL(nfs41_sequence_done); + +static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + + dprintk("--> %s data->seq_server %p\n", __func__, data->seq_server); + + nfs4_setup_sequence(data->seq_server->nfs_client, + data->seq_args, data->seq_res, task); +} + +static void nfs41_call_sync_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + + nfs41_sequence_done(task, data->seq_res); +} + +static const struct rpc_call_ops nfs41_call_sync_ops = { + .rpc_call_prepare = nfs41_call_sync_prepare, + .rpc_call_done = nfs41_call_sync_done, +}; + + +static void nfs41_sequence_res_init(struct nfs4_sequence_res *res) +{ + res->sr_timestamp = jiffies; + res->sr_status_flags = 0; + res->sr_status = 1; +} + +static +void nfs4_sequence_attach_slot(struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct nfs4_slot *slot) +{ + if (!slot) + return; + slot->privileged = args->sa_privileged ? 1 : 0; + args->sa_slot = slot; + + res->sr_slot = slot; +} + +static void nfs4_sequence_free_slot(struct nfs4_sequence_res *res) +{ + if (res->sr_slot != NULL) + res->sr_slot_ops->free_slot(res); +} + +static int nfs4_sequence_process(struct rpc_task *task, struct nfs4_sequence_res *res) +{ + if (res->sr_slot == NULL) + return 1; + return res->sr_slot_ops->process(task, res); +} + +int nfs4_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) +{ + if (res->sr_slot == NULL) + return 1; + return res->sr_slot_ops->done(task, res); +} +EXPORT_SYMBOL_GPL(nfs4_sequence_done); + + +int nfs4_setup_sequence(struct nfs_client *client, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + struct rpc_task *task) +{ + struct nfs4_session *session = nfs4_get_session(client); + struct nfs4_slot_table *tbl = client->cl_slot_tbl; + struct nfs4_slot *slot; + + /* slot already allocated? */ + if (res->sr_slot != NULL) + goto out_start; + + if (session) + tbl = &session->fc_slot_table; + + spin_lock(&tbl->slot_tbl_lock); + /* The state manager will wait until the slot table is empty */ + if (nfs4_slot_tbl_draining(tbl) && !args->sa_privileged) + goto out_sleep; + + slot = nfs4_alloc_slot(tbl); + if (IS_ERR(slot)) { + if (slot == ERR_PTR(-ENOMEM)) + goto out_sleep_timeout; + goto out_sleep; + } + spin_unlock(&tbl->slot_tbl_lock); + + nfs4_sequence_attach_slot(args, res, slot); + + trace_nfs4_setup_sequence(session, args); +out_start: + nfs41_sequence_res_init(res); + rpc_call_start(task); + return 0; +out_sleep_timeout: + /* Try again in 1/4 second */ + if (args->sa_privileged) + rpc_sleep_on_priority_timeout(&tbl->slot_tbl_waitq, task, + jiffies + (HZ >> 2), RPC_PRIORITY_PRIVILEGED); + else + rpc_sleep_on_timeout(&tbl->slot_tbl_waitq, task, + NULL, jiffies + (HZ >> 2)); + spin_unlock(&tbl->slot_tbl_lock); + return -EAGAIN; +out_sleep: + if (args->sa_privileged) + rpc_sleep_on_priority(&tbl->slot_tbl_waitq, task, + RPC_PRIORITY_PRIVILEGED); + else + rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL); + spin_unlock(&tbl->slot_tbl_lock); + return -EAGAIN; +} +EXPORT_SYMBOL_GPL(nfs4_setup_sequence); + +static int nfs4_call_sync_custom(struct rpc_task_setup *task_setup) +{ + int ret; + struct rpc_task *task; + + task = rpc_run_task(task_setup); + if (IS_ERR(task)) + return PTR_ERR(task); + + ret = task->tk_status; + rpc_put_task(task); + return ret; +} + +static int nfs4_do_call_sync(struct rpc_clnt *clnt, + struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + unsigned short task_flags) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_call_sync_data data = { + .seq_server = server, + .seq_args = args, + .seq_res = res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = clnt, + .rpc_message = msg, + .callback_ops = clp->cl_mvops->call_sync_ops, + .callback_data = &data, + .flags = task_flags, + }; + + return nfs4_call_sync_custom(&task_setup); +} + +int nfs4_call_sync_sequence(struct rpc_clnt *clnt, + struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res) +{ + unsigned short task_flags = 0; + + if (server->caps & NFS_CAP_MOVEABLE) + task_flags = RPC_TASK_MOVEABLE; + return nfs4_do_call_sync(clnt, server, msg, args, res, task_flags); +} + + +int nfs4_call_sync(struct rpc_clnt *clnt, + struct nfs_server *server, + struct rpc_message *msg, + struct nfs4_sequence_args *args, + struct nfs4_sequence_res *res, + int cache_reply) +{ + nfs4_init_sequence(server->nfs_client, args, res, cache_reply, 0); + return nfs4_call_sync_sequence(clnt, server, msg, args, res); +} + +static void +nfs4_inc_nlink_locked(struct inode *inode) +{ + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_NLINK); + inc_nlink(inode); +} + +static void +nfs4_inc_nlink(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs4_inc_nlink_locked(inode); + spin_unlock(&inode->i_lock); +} + +static void +nfs4_dec_nlink_locked(struct inode *inode) +{ + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_NLINK); + drop_nlink(inode); +} + +static void +nfs4_update_changeattr_locked(struct inode *inode, + struct nfs4_change_info *cinfo, + unsigned long timestamp, unsigned long cache_validity) +{ + struct nfs_inode *nfsi = NFS_I(inode); + u64 change_attr = inode_peek_iversion_raw(inode); + + if (!nfs_have_delegated_mtime(inode)) + cache_validity |= NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME; + if (S_ISDIR(inode->i_mode)) + cache_validity |= NFS_INO_INVALID_DATA; + + switch (NFS_SERVER(inode)->change_attr_type) { + case NFS4_CHANGE_TYPE_IS_UNDEFINED: + if (cinfo->after == change_attr) + goto out; + break; + default: + if ((s64)(change_attr - cinfo->after) >= 0) + goto out; + } + + inode_set_iversion_raw(inode, cinfo->after); + if (!cinfo->atomic || cinfo->before != change_attr) { + if (S_ISDIR(inode->i_mode)) + nfs_force_lookup_revalidate(inode); + + if (!nfs_have_delegated_attributes(inode)) + cache_validity |= + NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL | + NFS_INO_INVALID_SIZE | NFS_INO_INVALID_OTHER | + NFS_INO_INVALID_BLOCKS | NFS_INO_INVALID_NLINK | + NFS_INO_INVALID_MODE | NFS_INO_INVALID_BTIME | + NFS_INO_INVALID_XATTR; + nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); + } + nfsi->attrtimeo_timestamp = jiffies; + nfsi->read_cache_jiffies = timestamp; + nfsi->attr_gencount = nfs_inc_attr_generation_counter(); + nfsi->cache_validity &= ~NFS_INO_INVALID_CHANGE; +out: + nfs_set_cache_invalid(inode, cache_validity); +} + +void +nfs4_update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo, + unsigned long timestamp, unsigned long cache_validity) +{ + spin_lock(&dir->i_lock); + nfs4_update_changeattr_locked(dir, cinfo, timestamp, cache_validity); + spin_unlock(&dir->i_lock); +} + +struct nfs4_open_createattrs { + struct nfs4_label *label; + struct iattr *sattr; + const __u32 verf[2]; +}; + +static bool nfs4_clear_cap_atomic_open_v1(struct nfs_server *server, + int err, struct nfs4_exception *exception) +{ + if (err != -EINVAL) + return false; + if (!(server->caps & NFS_CAP_ATOMIC_OPEN_V1)) + return false; + server->caps &= ~NFS_CAP_ATOMIC_OPEN_V1; + exception->retry = 1; + return true; +} + +static fmode_t _nfs4_ctx_to_accessmode(const struct nfs_open_context *ctx) +{ + return ctx->mode & (FMODE_READ|FMODE_WRITE|FMODE_EXEC); +} + +static fmode_t _nfs4_ctx_to_openmode(const struct nfs_open_context *ctx) +{ + fmode_t ret = ctx->mode & (FMODE_READ|FMODE_WRITE); + + return (ctx->mode & FMODE_EXEC) ? FMODE_READ | ret : ret; +} + +static u32 +nfs4_fmode_to_share_access(fmode_t fmode) +{ + u32 res = 0; + + switch (fmode & (FMODE_READ | FMODE_WRITE)) { + case FMODE_READ: + res = NFS4_SHARE_ACCESS_READ; + break; + case FMODE_WRITE: + res = NFS4_SHARE_ACCESS_WRITE; + break; + case FMODE_READ|FMODE_WRITE: + res = NFS4_SHARE_ACCESS_BOTH; + } + return res; +} + +static u32 +nfs4_map_atomic_open_share(struct nfs_server *server, + fmode_t fmode, int openflags) +{ + u32 res = nfs4_fmode_to_share_access(fmode); + + if (!(server->caps & NFS_CAP_ATOMIC_OPEN_V1)) + goto out; + /* Want no delegation if we're using O_DIRECT */ + if (openflags & O_DIRECT) { + res |= NFS4_SHARE_WANT_NO_DELEG; + goto out; + } + /* res |= NFS4_SHARE_WANT_NO_PREFERENCE; */ + if (server->caps & NFS_CAP_DELEGTIME) + res |= NFS4_SHARE_WANT_DELEG_TIMESTAMPS; + if (server->caps & NFS_CAP_OPEN_XOR) + res |= NFS4_SHARE_WANT_OPEN_XOR_DELEGATION; +out: + return res; +} + +static enum open_claim_type4 +nfs4_map_atomic_open_claim(struct nfs_server *server, + enum open_claim_type4 claim) +{ + if (server->caps & NFS_CAP_ATOMIC_OPEN_V1) + return claim; + switch (claim) { + default: + return claim; + case NFS4_OPEN_CLAIM_FH: + return NFS4_OPEN_CLAIM_NULL; + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: + return NFS4_OPEN_CLAIM_DELEGATE_CUR; + case NFS4_OPEN_CLAIM_DELEG_PREV_FH: + return NFS4_OPEN_CLAIM_DELEGATE_PREV; + } +} + +static void nfs4_init_opendata_res(struct nfs4_opendata *p) +{ + p->o_res.f_attr = &p->f_attr; + p->o_res.seqid = p->o_arg.seqid; + p->c_res.seqid = p->c_arg.seqid; + p->o_res.server = p->o_arg.server; + p->o_res.access_request = p->o_arg.access; + nfs_fattr_init(&p->f_attr); + nfs_fattr_init_names(&p->f_attr, &p->owner_name, &p->group_name); +} + +static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, + struct nfs4_state_owner *sp, fmode_t fmode, int flags, + const struct nfs4_open_createattrs *c, + enum open_claim_type4 claim, + gfp_t gfp_mask) +{ + struct dentry *parent = dget_parent(dentry); + struct inode *dir = d_inode(parent); + struct nfs_server *server = NFS_SERVER(dir); + struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t); + struct nfs4_label *label = (c != NULL) ? c->label : NULL; + struct nfs4_opendata *p; + + p = kzalloc_obj(*p, gfp_mask); + if (p == NULL) + goto err; + + p->f_attr.label = nfs4_label_alloc(server, gfp_mask); + if (IS_ERR(p->f_attr.label)) + goto err_free_p; + + p->a_label = nfs4_label_alloc(server, gfp_mask); + if (IS_ERR(p->a_label)) + goto err_free_f; + + alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid; + p->o_arg.seqid = alloc_seqid(&sp->so_seqid, gfp_mask); + if (IS_ERR(p->o_arg.seqid)) + goto err_free_label; + nfs_sb_active(dentry->d_sb); + p->dentry = dget(dentry); + p->dir = parent; + p->owner = sp; + atomic_inc(&sp->so_count); + p->o_arg.open_flags = flags; + p->o_arg.fmode = fmode & (FMODE_READ|FMODE_WRITE); + p->o_arg.claim = nfs4_map_atomic_open_claim(server, claim); + p->o_arg.share_access = nfs4_map_atomic_open_share(server, + fmode, flags); + if (flags & O_CREAT) { + p->o_arg.umask = current_umask(); + p->o_arg.label = nfs4_label_copy(p->a_label, label); + if (c->sattr != NULL && c->sattr->ia_valid != 0) { + p->o_arg.u.attrs = &p->attrs; + memcpy(&p->attrs, c->sattr, sizeof(p->attrs)); + + memcpy(p->o_arg.u.verifier.data, c->verf, + sizeof(p->o_arg.u.verifier.data)); + } + } + /* ask server to check for all possible rights as results + * are cached */ + switch (p->o_arg.claim) { + default: + break; + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_FH: + p->o_arg.access = NFS4_ACCESS_READ | NFS4_ACCESS_MODIFY | + NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE | + NFS4_ACCESS_EXECUTE | + nfs_access_xattr_mask(server); + } + p->o_arg.clientid = server->nfs_client->cl_clientid; + p->o_arg.id.create_time = ktime_to_ns(sp->so_seqid.create_time); + p->o_arg.id.uniquifier = sp->so_seqid.owner_id; + p->o_arg.name = &dentry->d_name; + p->o_arg.server = server; + p->o_arg.bitmask = nfs4_bitmask(server, label); + p->o_arg.open_bitmap = &nfs4_fattr_bitmap[0]; + switch (p->o_arg.claim) { + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + case NFS4_OPEN_CLAIM_DELEGATE_PREV: + p->o_arg.fh = NFS_FH(dir); + break; + case NFS4_OPEN_CLAIM_PREVIOUS: + case NFS4_OPEN_CLAIM_FH: + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: + case NFS4_OPEN_CLAIM_DELEG_PREV_FH: + p->o_arg.fh = NFS_FH(d_inode(dentry)); + } + p->c_arg.fh = &p->o_res.fh; + p->c_arg.stateid = &p->o_res.stateid; + p->c_arg.seqid = p->o_arg.seqid; + nfs4_init_opendata_res(p); + kref_init(&p->kref); + return p; + +err_free_label: + nfs4_label_free(p->a_label); +err_free_f: + nfs4_label_free(p->f_attr.label); +err_free_p: + kfree(p); +err: + dput(parent); + return NULL; +} + +static void nfs4_opendata_free(struct kref *kref) +{ + struct nfs4_opendata *p = container_of(kref, + struct nfs4_opendata, kref); + struct super_block *sb = p->dentry->d_sb; + + nfs4_lgopen_release(p->lgp); + nfs_free_seqid(p->o_arg.seqid); + nfs4_sequence_free_slot(&p->o_res.seq_res); + if (p->state != NULL) + nfs4_put_open_state(p->state); + nfs4_put_state_owner(p->owner); + + nfs4_label_free(p->a_label); + nfs4_label_free(p->f_attr.label); + + dput(p->dir); + dput(p->dentry); + nfs_sb_deactive(sb); + nfs_fattr_free_names(&p->f_attr); + kfree(p->f_attr.mdsthreshold); + kfree(p); +} + +static void nfs4_opendata_put(struct nfs4_opendata *p) +{ + if (p != NULL) + kref_put(&p->kref, nfs4_opendata_free); +} + +static bool nfs4_mode_match_open_stateid(struct nfs4_state *state, + fmode_t fmode) +{ + switch(fmode & (FMODE_READ|FMODE_WRITE)) { + case FMODE_READ|FMODE_WRITE: + return state->n_rdwr != 0; + case FMODE_WRITE: + return state->n_wronly != 0; + case FMODE_READ: + return state->n_rdonly != 0; + } + WARN_ON_ONCE(1); + return false; +} + +static int can_open_cached(struct nfs4_state *state, fmode_t mode, + int open_mode, enum open_claim_type4 claim) +{ + int ret = 0; + + if (open_mode & (O_EXCL|O_TRUNC)) + goto out; + switch (claim) { + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_FH: + goto out; + default: + break; + } + switch (mode & (FMODE_READ|FMODE_WRITE)) { + case FMODE_READ: + ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0 + && state->n_rdonly != 0; + break; + case FMODE_WRITE: + ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0 + && state->n_wronly != 0; + break; + case FMODE_READ|FMODE_WRITE: + ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0 + && state->n_rdwr != 0; + } +out: + return ret; +} + +static bool can_open_delegated(const struct inode *inode, fmode_t fmode, + enum open_claim_type4 claim, nfs4_stateid *stateid) +{ + struct nfs_delegation *delegation; + bool ret = false; + + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return false; + if ((delegation->type & fmode) != fmode) + goto out_put_delegation; + + switch (claim) { + case NFS4_OPEN_CLAIM_PREVIOUS: + if (test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags)) + break; + fallthrough; + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_FH: + nfs_mark_delegation_referenced(delegation); + /* Save the delegation stateid */ + if (stateid) + nfs4_stateid_copy(stateid, &delegation->stateid); + ret = true; + break; + default: + break; + } + +out_put_delegation: + nfs_put_delegation(delegation); + return ret; +} + +static void update_open_stateflags(struct nfs4_state *state, fmode_t fmode) +{ + switch (fmode) { + case FMODE_WRITE: + state->n_wronly++; + break; + case FMODE_READ: + state->n_rdonly++; + break; + case FMODE_READ|FMODE_WRITE: + state->n_rdwr++; + } + nfs4_state_set_mode_locked(state, state->state | fmode); +} + +static bool nfs_open_stateid_recover_openmode(struct nfs4_state *state) +{ + if (state->n_rdonly && !test_bit(NFS_O_RDONLY_STATE, &state->flags)) + return true; + if (state->n_wronly && !test_bit(NFS_O_WRONLY_STATE, &state->flags)) + return true; + if (state->n_rdwr && !test_bit(NFS_O_RDWR_STATE, &state->flags)) + return true; + return false; +} + +static void nfs_state_log_update_open_stateid(struct nfs4_state *state) +{ + if (test_and_clear_bit(NFS_STATE_CHANGE_WAIT, &state->flags)) + wake_up_all(&state->waitq); +} + +static void nfs_test_and_clear_all_open_stateid(struct nfs4_state *state) +{ + struct nfs_client *clp = state->owner->so_server->nfs_client; + bool need_recover = false; + + if (test_and_clear_bit(NFS_O_RDONLY_STATE, &state->flags) && state->n_rdonly) + need_recover = true; + if (test_and_clear_bit(NFS_O_WRONLY_STATE, &state->flags) && state->n_wronly) + need_recover = true; + if (test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags) && state->n_rdwr) + need_recover = true; + if (need_recover) + nfs4_state_mark_reclaim_nograce(clp, state); +} + +/* + * Check for whether or not the caller may update the open stateid + * to the value passed in by stateid. + * + * Note: This function relies heavily on the server implementing + * RFC7530 Section 9.1.4.2, and RFC5661 Section 8.2.2 + * correctly. + * i.e. The stateid seqids have to be initialised to 1, and + * are then incremented on every state transition. + */ +static bool nfs_stateid_is_sequential(struct nfs4_state *state, + const nfs4_stateid *stateid) +{ + if (test_bit(NFS_OPEN_STATE, &state->flags)) { + /* The common case - we're updating to a new sequence number */ + if (nfs4_stateid_match_other(stateid, &state->open_stateid)) { + if (nfs4_stateid_is_next(&state->open_stateid, stateid)) + return true; + return false; + } + /* The server returned a new stateid */ + } + /* This is the first OPEN in this generation */ + if (stateid->seqid == cpu_to_be32(1)) + return true; + return false; +} + +static void nfs_resync_open_stateid_locked(struct nfs4_state *state) +{ + if (!(state->n_wronly || state->n_rdonly || state->n_rdwr)) + return; + if (state->n_wronly) + set_bit(NFS_O_WRONLY_STATE, &state->flags); + if (state->n_rdonly) + set_bit(NFS_O_RDONLY_STATE, &state->flags); + if (state->n_rdwr) + set_bit(NFS_O_RDWR_STATE, &state->flags); + set_bit(NFS_OPEN_STATE, &state->flags); +} + +static void nfs_clear_open_stateid_locked(struct nfs4_state *state, + nfs4_stateid *stateid, fmode_t fmode) +{ + clear_bit(NFS_O_RDWR_STATE, &state->flags); + switch (fmode & (FMODE_READ|FMODE_WRITE)) { + case FMODE_WRITE: + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + break; + case FMODE_READ: + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + break; + case 0: + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + clear_bit(NFS_OPEN_STATE, &state->flags); + } + if (stateid == NULL) + return; + /* Handle OPEN+OPEN_DOWNGRADE races */ + if (nfs4_stateid_match_other(stateid, &state->open_stateid) && + !nfs4_stateid_is_newer(stateid, &state->open_stateid)) { + nfs_resync_open_stateid_locked(state); + goto out; + } + if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0) + nfs4_stateid_copy(&state->stateid, stateid); + nfs4_stateid_copy(&state->open_stateid, stateid); + trace_nfs4_open_stateid_update(state->inode, stateid, 0); +out: + nfs_state_log_update_open_stateid(state); +} + +static void nfs_clear_open_stateid(struct nfs4_state *state, + nfs4_stateid *arg_stateid, + nfs4_stateid *stateid, fmode_t fmode) +{ + write_seqlock(&state->seqlock); + /* Ignore, if the CLOSE argment doesn't match the current stateid */ + if (nfs4_state_match_open_stateid_other(state, arg_stateid)) + nfs_clear_open_stateid_locked(state, stateid, fmode); + write_sequnlock(&state->seqlock); + if (test_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags)) + nfs4_schedule_state_manager(state->owner->so_server->nfs_client); +} + +static void nfs_set_open_stateid_locked(struct nfs4_state *state, + const nfs4_stateid *stateid, nfs4_stateid *freeme) + __must_hold(&state->owner->so_lock) + __must_hold(&state->seqlock) + __must_hold(RCU) + +{ + DEFINE_WAIT(wait); + int status = 0; + for (;;) { + + if (nfs_stateid_is_sequential(state, stateid)) + break; + + if (status) { + if (nfs4_stateid_match_other(stateid, &state->open_stateid) && + !nfs4_stateid_is_newer(stateid, &state->open_stateid)) { + trace_nfs4_open_stateid_update_skip(state->inode, + stateid, status); + return; + } else { + break; + } + } + + /* Rely on seqids for serialisation with NFSv4.0 */ + if (!nfs4_has_session(NFS_SERVER(state->inode)->nfs_client)) + break; + + set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); + prepare_to_wait(&state->waitq, &wait, TASK_KILLABLE); + /* + * Ensure we process the state changes in the same order + * in which the server processed them by delaying the + * update of the stateid until we are in sequence. + */ + write_sequnlock(&state->seqlock); + spin_unlock(&state->owner->so_lock); + rcu_read_unlock(); + trace_nfs4_open_stateid_update_wait(state->inode, stateid, 0); + + if (!fatal_signal_pending(current) && + !nfs_current_task_exiting()) { + if (schedule_timeout(5*HZ) == 0) + status = -EAGAIN; + else + status = 0; + } else + status = -EINTR; + finish_wait(&state->waitq, &wait); + rcu_read_lock(); + spin_lock(&state->owner->so_lock); + write_seqlock(&state->seqlock); + } + + if (test_bit(NFS_OPEN_STATE, &state->flags) && + !nfs4_stateid_match_other(stateid, &state->open_stateid)) { + nfs4_stateid_copy(freeme, &state->open_stateid); + nfs_test_and_clear_all_open_stateid(state); + } + + if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0) + nfs4_stateid_copy(&state->stateid, stateid); + nfs4_stateid_copy(&state->open_stateid, stateid); + trace_nfs4_open_stateid_update(state->inode, stateid, status); + nfs_state_log_update_open_stateid(state); +} + +static void nfs_state_set_open_stateid(struct nfs4_state *state, + const nfs4_stateid *open_stateid, + fmode_t fmode, + nfs4_stateid *freeme) +{ + /* + * Protect the call to nfs4_state_set_mode_locked and + * serialise the stateid update + */ + write_seqlock(&state->seqlock); + nfs_set_open_stateid_locked(state, open_stateid, freeme); + switch (fmode) { + case FMODE_READ: + set_bit(NFS_O_RDONLY_STATE, &state->flags); + break; + case FMODE_WRITE: + set_bit(NFS_O_WRONLY_STATE, &state->flags); + break; + case FMODE_READ|FMODE_WRITE: + set_bit(NFS_O_RDWR_STATE, &state->flags); + } + set_bit(NFS_OPEN_STATE, &state->flags); + write_sequnlock(&state->seqlock); +} + +void nfs_state_clear_open_state_flags(struct nfs4_state *state) +{ + clear_bit(NFS_O_RDWR_STATE, &state->flags); + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + clear_bit(NFS_OPEN_STATE, &state->flags); +} + +static void nfs_state_set_delegation(struct nfs4_state *state, + const nfs4_stateid *deleg_stateid, + fmode_t fmode) +{ + /* + * Protect the call to nfs4_state_set_mode_locked and + * serialise the stateid update + */ + write_seqlock(&state->seqlock); + nfs4_stateid_copy(&state->stateid, deleg_stateid); + set_bit(NFS_DELEGATED_STATE, &state->flags); + write_sequnlock(&state->seqlock); +} + +void nfs_state_clear_delegation(struct nfs4_state *state) +{ + write_seqlock(&state->seqlock); + nfs4_stateid_copy(&state->stateid, &state->open_stateid); + clear_bit(NFS_DELEGATED_STATE, &state->flags); + write_sequnlock(&state->seqlock); +} + +int update_open_stateid(struct nfs4_state *state, + const nfs4_stateid *open_stateid, + const nfs4_stateid *delegation, + fmode_t fmode) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs_client *clp = server->nfs_client; + struct nfs_delegation *deleg_cur; + nfs4_stateid freeme = { }; + int ret = 0; + + fmode &= (FMODE_READ|FMODE_WRITE); + + spin_lock(&state->owner->so_lock); + if (open_stateid != NULL) { + rcu_read_lock(); + nfs_state_set_open_stateid(state, open_stateid, fmode, &freeme); + rcu_read_unlock(); + ret = 1; + } + + deleg_cur = nfs4_get_valid_delegation(state->inode); + if (deleg_cur == NULL) + goto no_delegation; + + spin_lock(&deleg_cur->lock); + if (!deleg_cur->inode || + test_bit(NFS_DELEGATION_RETURNING, &deleg_cur->flags) || + (deleg_cur->type & fmode) != fmode) + goto no_delegation_unlock; + + if (delegation == NULL) + delegation = &deleg_cur->stateid; + else if (!nfs4_stateid_match_other(&deleg_cur->stateid, delegation)) + goto no_delegation_unlock; + + nfs_mark_delegation_referenced(deleg_cur); + nfs_state_set_delegation(state, &deleg_cur->stateid, fmode); + ret = 1; +no_delegation_unlock: + spin_unlock(&deleg_cur->lock); + nfs_put_delegation(deleg_cur); +no_delegation: + if (ret) + update_open_stateflags(state, fmode); + spin_unlock(&state->owner->so_lock); + + if (test_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags)) + nfs4_schedule_state_manager(clp); + if (freeme.type != 0) + nfs4_test_and_free_stateid(server, &freeme, + state->owner->so_cred); + + return ret; +} + +static bool nfs4_update_lock_stateid(struct nfs4_lock_state *lsp, + const nfs4_stateid *stateid) +{ + struct nfs4_state *state = lsp->ls_state; + bool ret = false; + + spin_lock(&state->state_lock); + if (!nfs4_stateid_match_other(stateid, &lsp->ls_stateid)) + goto out_noupdate; + if (!nfs4_stateid_is_newer(stateid, &lsp->ls_stateid)) + goto out_noupdate; + nfs4_stateid_copy(&lsp->ls_stateid, stateid); + ret = true; +out_noupdate: + spin_unlock(&state->state_lock); + return ret; +} + +static void nfs4_return_incompatible_delegation(struct inode *inode, fmode_t fmode) +{ + struct nfs_delegation *delegation; + + fmode &= FMODE_READ|FMODE_WRITE; + delegation = nfs4_get_valid_delegation(inode); + if (!delegation) + return; + if ((delegation->type & fmode) != fmode) + nfs4_inode_return_delegation(inode); + nfs_put_delegation(delegation); +} + +static struct nfs4_state *nfs4_try_open_cached(struct nfs4_opendata *opendata) +{ + struct nfs4_state *state = opendata->state; + int open_mode = opendata->o_arg.open_flags; + fmode_t fmode = opendata->o_arg.fmode; + enum open_claim_type4 claim = opendata->o_arg.claim; + nfs4_stateid stateid; + int ret = -EAGAIN; + + for (;;) { + spin_lock(&state->owner->so_lock); + if (can_open_cached(state, fmode, open_mode, claim)) { + update_open_stateflags(state, fmode); + spin_unlock(&state->owner->so_lock); + goto out_return_state; + } + spin_unlock(&state->owner->so_lock); + + if (!can_open_delegated(state->inode, fmode, claim, &stateid)) + break; + + nfs_release_seqid(opendata->o_arg.seqid); + if (!opendata->is_recover) { + ret = nfs_may_open(state->inode, state->owner->so_cred, open_mode); + if (ret != 0) + goto out; + } + ret = -EAGAIN; + + /* Try to update the stateid using the delegation */ + if (update_open_stateid(state, NULL, &stateid, fmode)) + goto out_return_state; + } +out: + return ERR_PTR(ret); +out_return_state: + refcount_inc(&state->count); + return state; +} + +static void +nfs4_process_delegation(struct inode *inode, const struct cred *cred, + enum open_claim_type4 claim, + const struct nfs4_open_delegation *delegation) +{ + switch (delegation->open_delegation_type) { + case NFS4_OPEN_DELEGATE_READ: + case NFS4_OPEN_DELEGATE_WRITE: + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + break; + default: + return; + } + switch (claim) { + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: + pr_err_ratelimited("NFS: Broken NFSv4 server %s is " + "returning a delegation for " + "OPEN(CLAIM_DELEGATE_CUR)\n", + NFS_SERVER(inode)->nfs_client->cl_hostname); + break; + case NFS4_OPEN_CLAIM_PREVIOUS: + nfs_inode_reclaim_delegation(inode, cred, delegation->type, + &delegation->stateid, + delegation->pagemod_limit, + delegation->open_delegation_type); + break; + default: + nfs_inode_set_delegation(inode, cred, delegation->type, + &delegation->stateid, + delegation->pagemod_limit, + delegation->open_delegation_type); + } + if (delegation->do_recall) + nfs_async_inode_return_delegation(inode, &delegation->stateid); +} + +/* + * Check the inode attributes against the CLAIM_PREVIOUS returned attributes + * and update the nfs4_state. + */ +static struct nfs4_state * +_nfs4_opendata_reclaim_to_nfs4_state(struct nfs4_opendata *data) +{ + struct inode *inode = data->state->inode; + struct nfs4_state *state = data->state; + int ret; + + if (!data->rpc_done) { + if (data->rpc_status) + return ERR_PTR(data->rpc_status); + return nfs4_try_open_cached(data); + } + + ret = nfs_refresh_inode(inode, &data->f_attr); + if (ret) + return ERR_PTR(ret); + + nfs4_process_delegation(state->inode, + data->owner->so_cred, + data->o_arg.claim, + &data->o_res.delegation); + + if (!(data->o_res.rflags & NFS4_OPEN_RESULT_NO_OPEN_STATEID)) { + if (!update_open_stateid(state, &data->o_res.stateid, + NULL, data->o_arg.fmode)) + return ERR_PTR(-EAGAIN); + } else if (!update_open_stateid(state, NULL, NULL, data->o_arg.fmode)) + return ERR_PTR(-EAGAIN); + refcount_inc(&state->count); + + return state; +} + +static struct inode * +nfs4_opendata_get_inode(struct nfs4_opendata *data) +{ + struct inode *inode; + + switch (data->o_arg.claim) { + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + case NFS4_OPEN_CLAIM_DELEGATE_PREV: + if (!(data->f_attr.valid & NFS_ATTR_FATTR)) + return ERR_PTR(-EAGAIN); + inode = nfs_fhget(data->dir->d_sb, &data->o_res.fh, + &data->f_attr); + break; + default: + inode = d_inode(data->dentry); + ihold(inode); + nfs_refresh_inode(inode, &data->f_attr); + } + return inode; +} + +static struct nfs4_state * +nfs4_opendata_find_nfs4_state(struct nfs4_opendata *data) +{ + struct nfs4_state *state; + struct inode *inode; + + inode = nfs4_opendata_get_inode(data); + if (IS_ERR(inode)) + return ERR_CAST(inode); + if (data->state != NULL && data->state->inode == inode) { + state = data->state; + refcount_inc(&state->count); + } else + state = nfs4_get_open_state(inode, data->owner); + iput(inode); + if (state == NULL) + state = ERR_PTR(-ENOMEM); + return state; +} + +static struct nfs4_state * +_nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data) +{ + struct nfs4_state *state; + + if (!data->rpc_done) { + state = nfs4_try_open_cached(data); + trace_nfs4_cached_open(data->state); + goto out; + } + + state = nfs4_opendata_find_nfs4_state(data); + if (IS_ERR(state)) + goto out; + + nfs4_process_delegation(state->inode, + data->owner->so_cred, + data->o_arg.claim, + &data->o_res.delegation); + + if (!(data->o_res.rflags & NFS4_OPEN_RESULT_NO_OPEN_STATEID)) { + if (!update_open_stateid(state, &data->o_res.stateid, + NULL, data->o_arg.fmode)) { + nfs4_put_open_state(state); + state = ERR_PTR(-EAGAIN); + } + } else if (!update_open_stateid(state, NULL, NULL, data->o_arg.fmode)) { + nfs4_put_open_state(state); + state = ERR_PTR(-EAGAIN); + } +out: + nfs_release_seqid(data->o_arg.seqid); + return state; +} + +static struct nfs4_state * +nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data) +{ + struct nfs4_state *ret; + + if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) + ret =_nfs4_opendata_reclaim_to_nfs4_state(data); + else + ret = _nfs4_opendata_to_nfs4_state(data); + nfs4_sequence_free_slot(&data->o_res.seq_res); + return ret; +} + +static struct nfs_open_context * +nfs4_state_find_open_context_mode(struct nfs4_state *state, fmode_t mode) +{ + struct nfs_inode *nfsi = NFS_I(state->inode); + struct nfs_open_context *ctx; + + rcu_read_lock(); + list_for_each_entry_rcu(ctx, &nfsi->open_files, list) { + if (ctx->state != state) + continue; + if ((ctx->mode & mode) != mode) + continue; + if (!get_nfs_open_context(ctx)) + continue; + rcu_read_unlock(); + return ctx; + } + rcu_read_unlock(); + return ERR_PTR(-ENOENT); +} + +static struct nfs_open_context * +nfs4_state_find_open_context(struct nfs4_state *state) +{ + struct nfs_open_context *ctx; + + ctx = nfs4_state_find_open_context_mode(state, FMODE_READ|FMODE_WRITE); + if (!IS_ERR(ctx)) + return ctx; + ctx = nfs4_state_find_open_context_mode(state, FMODE_WRITE); + if (!IS_ERR(ctx)) + return ctx; + return nfs4_state_find_open_context_mode(state, FMODE_READ); +} + +static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context *ctx, + struct nfs4_state *state, enum open_claim_type4 claim) +{ + struct nfs4_opendata *opendata; + + opendata = nfs4_opendata_alloc(ctx->dentry, state->owner, 0, 0, + NULL, claim, GFP_NOFS); + if (opendata == NULL) + return ERR_PTR(-ENOMEM); + opendata->state = state; + refcount_inc(&state->count); + return opendata; +} + +static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, + fmode_t fmode) +{ + struct nfs4_state *newstate; + struct nfs_server *server = NFS_SB(opendata->dentry->d_sb); + int openflags = opendata->o_arg.open_flags; + int ret; + + if (!nfs4_mode_match_open_stateid(opendata->state, fmode)) + return 0; + opendata->o_arg.fmode = fmode; + opendata->o_arg.share_access = + nfs4_map_atomic_open_share(server, fmode, openflags); + memset(&opendata->o_res, 0, sizeof(opendata->o_res)); + memset(&opendata->c_res, 0, sizeof(opendata->c_res)); + nfs4_init_opendata_res(opendata); + ret = _nfs4_recover_proc_open(opendata); + if (ret != 0) + return ret; + newstate = nfs4_opendata_to_nfs4_state(opendata); + if (IS_ERR(newstate)) + return PTR_ERR(newstate); + if (newstate != opendata->state) + ret = -ESTALE; + nfs4_close_state(newstate, fmode); + return ret; +} + +static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state *state) +{ + int ret; + + /* memory barrier prior to reading state->n_* */ + smp_rmb(); + ret = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE); + if (ret != 0) + return ret; + ret = nfs4_open_recover_helper(opendata, FMODE_WRITE); + if (ret != 0) + return ret; + ret = nfs4_open_recover_helper(opendata, FMODE_READ); + if (ret != 0) + return ret; + /* + * We may have performed cached opens for all three recoveries. + * Check if we need to update the current stateid. + */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0 && + !nfs4_stateid_match(&state->stateid, &state->open_stateid)) { + write_seqlock(&state->seqlock); + if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0) + nfs4_stateid_copy(&state->stateid, &state->open_stateid); + write_sequnlock(&state->seqlock); + } + return 0; +} + +/* + * OPEN_RECLAIM: + * reclaim state on the server after a reboot. + */ +static int _nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct nfs_delegation *delegation; + struct nfs4_opendata *opendata; + u32 delegation_type = NFS4_OPEN_DELEGATE_NONE; + int status; + + opendata = nfs4_open_recoverdata_alloc(ctx, state, + NFS4_OPEN_CLAIM_PREVIOUS); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(state->inode)->delegation); + if (delegation != NULL && test_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags) != 0) { + switch(delegation->type) { + case FMODE_READ: + delegation_type = NFS4_OPEN_DELEGATE_READ; + if (test_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags)) + delegation_type = NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG; + break; + case FMODE_WRITE: + case FMODE_READ|FMODE_WRITE: + delegation_type = NFS4_OPEN_DELEGATE_WRITE; + if (test_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags)) + delegation_type = NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG; + } + } + rcu_read_unlock(); + opendata->o_arg.u.delegation_type = delegation_type; + status = nfs4_open_recover(opendata, state); + nfs4_opendata_put(opendata); + return status; +} + +static int nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_exception exception = { }; + int err; + do { + err = _nfs4_do_open_reclaim(ctx, state); + trace_nfs4_open_reclaim(ctx, 0, err); + if (nfs4_clear_cap_atomic_open_v1(server, err, &exception)) + continue; + if (err != -NFS4ERR_DELAY) + break; + nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + return err; +} + +int nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state) +{ + struct nfs_open_context *ctx; + int ret; + + ctx = nfs4_state_find_open_context(state); + if (IS_ERR(ctx)) + return -EAGAIN; + clear_bit(NFS_DELEGATED_STATE, &state->flags); + nfs_state_clear_open_state_flags(state); + ret = nfs4_do_open_reclaim(ctx, state); + put_nfs_open_context(ctx); + return ret; +} + +static int nfs4_handle_delegation_recall_error(struct nfs_server *server, struct nfs4_state *state, const nfs4_stateid *stateid, struct file_lock *fl, int err) +{ + switch (err) { + default: + printk(KERN_ERR "NFS: %s: unhandled error " + "%d.\n", __func__, err); + fallthrough; + case 0: + case -ENOENT: + case -EAGAIN: + case -ESTALE: + case -ETIMEDOUT: + break; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_DEADSESSION: + return -EAGAIN; + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_STALE_STATEID: + /* Don't recall a delegation if it was lost */ + nfs4_schedule_lease_recovery(server->nfs_client); + return -EAGAIN; + case -NFS4ERR_MOVED: + nfs4_schedule_migration_recovery(server); + return -EAGAIN; + case -NFS4ERR_LEASE_MOVED: + nfs4_schedule_lease_moved_recovery(server->nfs_client); + return -EAGAIN; + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_OPENMODE: + nfs_inode_find_state_and_recover(state->inode, + stateid); + nfs4_schedule_stateid_recovery(server, state); + return -EAGAIN; + case -NFS4ERR_DELAY: + case -NFS4ERR_GRACE: + ssleep(1); + return -EAGAIN; + case -ENOMEM: + case -NFS4ERR_DENIED: + if (fl) { + struct nfs4_lock_state *lsp = fl->fl_u.nfs4_fl.owner; + if (lsp) + set_bit(NFS_LOCK_LOST, &lsp->ls_flags); + } + return 0; + } + return err; +} + +int nfs4_open_delegation_recall(struct nfs_open_context *ctx, + struct nfs4_state *state, const nfs4_stateid *stateid) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_opendata *opendata; + int err = 0; + + opendata = nfs4_open_recoverdata_alloc(ctx, state, + NFS4_OPEN_CLAIM_DELEG_CUR_FH); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); + nfs4_stateid_copy(&opendata->o_arg.u.delegation, stateid); + if (!test_bit(NFS_O_RDWR_STATE, &state->flags)) { + err = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE); + if (err) + goto out; + } + if (!test_bit(NFS_O_WRONLY_STATE, &state->flags)) { + err = nfs4_open_recover_helper(opendata, FMODE_WRITE); + if (err) + goto out; + } + if (!test_bit(NFS_O_RDONLY_STATE, &state->flags)) { + err = nfs4_open_recover_helper(opendata, FMODE_READ); + if (err) + goto out; + } + nfs_state_clear_delegation(state); +out: + nfs4_opendata_put(opendata); + return nfs4_handle_delegation_recall_error(server, state, stateid, NULL, err); +} + +static void nfs4_open_confirm_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_opendata *data = calldata; + + nfs4_setup_sequence(data->o_arg.server->nfs_client, + &data->c_arg.seq_args, &data->c_res.seq_res, task); +} + +static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_opendata *data = calldata; + + data->c_res.seq_res.sr_slot_ops->done(task, &data->c_res.seq_res); + + data->rpc_status = task->tk_status; + if (data->rpc_status == 0) { + nfs4_stateid_copy(&data->o_res.stateid, &data->c_res.stateid); + nfs_confirm_seqid(&data->owner->so_seqid, 0); + renew_lease(data->o_res.server, data->timestamp); + data->rpc_done = true; + } +} + +static void nfs4_open_confirm_release(void *calldata) +{ + struct nfs4_opendata *data = calldata; + struct nfs4_state *state = NULL; + + /* If this request hasn't been cancelled, do nothing */ + if (!data->cancelled) + goto out_free; + /* In case of error, no cleanup! */ + if (!data->rpc_done) + goto out_free; + state = nfs4_opendata_to_nfs4_state(data); + if (!IS_ERR(state)) + nfs4_close_state(state, data->o_arg.fmode); +out_free: + nfs4_opendata_put(data); +} + +static const struct rpc_call_ops nfs4_open_confirm_ops = { + .rpc_call_prepare = nfs4_open_confirm_prepare, + .rpc_call_done = nfs4_open_confirm_done, + .rpc_release = nfs4_open_confirm_release, +}; + +/* + * Note: On error, nfs4_proc_open_confirm will free the struct nfs4_opendata + */ +static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) +{ + struct nfs_server *server = NFS_SERVER(d_inode(data->dir)); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_CONFIRM], + .rpc_argp = &data->c_arg, + .rpc_resp = &data->c_res, + .rpc_cred = data->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_open_confirm_ops, + .callback_data = data, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + int status; + + nfs4_init_sequence(server->nfs_client, &data->c_arg.seq_args, + &data->c_res.seq_res, 1, data->is_recover); + kref_get(&data->kref); + data->rpc_done = false; + data->rpc_status = 0; + data->timestamp = jiffies; + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + status = rpc_wait_for_completion_task(task); + if (status != 0) { + data->cancelled = true; + smp_wmb(); + } else + status = data->rpc_status; + rpc_put_task(task); + return status; +} + +static void nfs4_open_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_opendata *data = calldata; + struct nfs4_state_owner *sp = data->owner; + struct nfs_client *clp = sp->so_server->nfs_client; + enum open_claim_type4 claim = data->o_arg.claim; + + if (nfs_wait_on_sequence(data->o_arg.seqid, task) != 0) + goto out_wait; + /* + * Check if we still need to send an OPEN call, or if we can use + * a delegation instead. + */ + if (data->state != NULL) { + if (can_open_cached(data->state, data->o_arg.fmode, + data->o_arg.open_flags, claim)) + goto out_no_action; + if (can_open_delegated(data->state->inode, data->o_arg.fmode, + claim, NULL)) { + trace_nfs4_cached_open(data->state); + goto out_no_action; + } + } + /* Update client id. */ + data->o_arg.clientid = clp->cl_clientid; + switch (claim) { + default: + break; + case NFS4_OPEN_CLAIM_PREVIOUS: + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: + case NFS4_OPEN_CLAIM_DELEG_PREV_FH: + data->o_arg.open_bitmap = &nfs4_open_noattr_bitmap[0]; + fallthrough; + case NFS4_OPEN_CLAIM_FH: + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR]; + } + data->timestamp = jiffies; + if (nfs4_setup_sequence(data->o_arg.server->nfs_client, + &data->o_arg.seq_args, + &data->o_res.seq_res, + task) != 0) + nfs_release_seqid(data->o_arg.seqid); + + /* Set the create mode (note dependency on the session type) */ + data->o_arg.createmode = NFS4_CREATE_UNCHECKED; + if (data->o_arg.open_flags & O_EXCL) { + data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE4_1; + if (clp->cl_mvops->minor_version == 0) { + data->o_arg.createmode = NFS4_CREATE_EXCLUSIVE; + /* don't put an ACCESS op in OPEN compound if O_EXCL, + * because ACCESS will return permission denied for + * all bits until close */ + data->o_res.access_request = data->o_arg.access = 0; + } else if (nfs4_has_persistent_session(clp)) + data->o_arg.createmode = NFS4_CREATE_GUARDED; + } + return; + +out_no_action: + task->tk_action = NULL; +out_wait: + nfs4_sequence_done(task, &data->o_res.seq_res); +} + +static void nfs4_open_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_opendata *data = calldata; + + data->rpc_status = task->tk_status; + + if (!nfs4_sequence_process(task, &data->o_res.seq_res)) + return; + + if (task->tk_status == 0) { + if (data->o_res.f_attr->valid & NFS_ATTR_FATTR_TYPE) { + switch (data->o_res.f_attr->mode & S_IFMT) { + case S_IFREG: + break; + case S_IFLNK: + data->rpc_status = -ELOOP; + break; + case S_IFDIR: + data->rpc_status = -EISDIR; + break; + default: + data->rpc_status = -ENOTDIR; + } + } + renew_lease(data->o_res.server, data->timestamp); + if (!(data->o_res.rflags & NFS4_OPEN_RESULT_CONFIRM)) + nfs_confirm_seqid(&data->owner->so_seqid, 0); + } + data->rpc_done = true; +} + +static void nfs4_open_release(void *calldata) +{ + struct nfs4_opendata *data = calldata; + struct nfs4_state *state = NULL; + + /* In case of error, no cleanup! */ + if (data->rpc_status != 0 || !data->rpc_done) { + nfs_release_seqid(data->o_arg.seqid); + goto out_free; + } + /* If this request hasn't been cancelled, do nothing */ + if (!data->cancelled) + goto out_free; + /* In case we need an open_confirm, no cleanup! */ + if (data->o_res.rflags & NFS4_OPEN_RESULT_CONFIRM) + goto out_free; + state = nfs4_opendata_to_nfs4_state(data); + if (!IS_ERR(state)) + nfs4_close_state(state, data->o_arg.fmode); +out_free: + nfs4_opendata_put(data); +} + +static const struct rpc_call_ops nfs4_open_ops = { + .rpc_call_prepare = nfs4_open_prepare, + .rpc_call_done = nfs4_open_done, + .rpc_release = nfs4_open_release, +}; + +static int nfs4_run_open_task(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct inode *dir = d_inode(data->dir); + struct nfs_server *server = NFS_SERVER(dir); + struct nfs_client *clp = server->nfs_client; + struct nfs_openargs *o_arg = &data->o_arg; + struct nfs_openres *o_res = &data->o_res; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN], + .rpc_argp = o_arg, + .rpc_resp = o_res, + .rpc_cred = data->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_open_ops, + .callback_data = data, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + int status; + + if (nfs_server_capable(dir, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + kref_get(&data->kref); + data->rpc_done = false; + data->rpc_status = 0; + data->cancelled = false; + data->is_recover = false; + if (!ctx) { + nfs4_init_sequence(clp, &o_arg->seq_args, &o_res->seq_res, 1, 1); + data->is_recover = true; + task_setup_data.flags |= RPC_TASK_TIMEOUT; + } else { + nfs4_init_sequence(clp, &o_arg->seq_args, &o_res->seq_res, 1, 0); + pnfs_lgopen_prepare(data, ctx); + } + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + status = rpc_wait_for_completion_task(task); + if (status != 0) { + data->cancelled = true; + smp_wmb(); + } else + status = data->rpc_status; + rpc_put_task(task); + + return status; +} + +static int _nfs4_recover_proc_open(struct nfs4_opendata *data) +{ + struct inode *dir = d_inode(data->dir); + struct nfs_openres *o_res = &data->o_res; + int status; + + status = nfs4_run_open_task(data, NULL); + if (status != 0 || !data->rpc_done) + return status; + + nfs_fattr_map_and_free_names(NFS_SERVER(dir), &data->f_attr); + + if (o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) + status = _nfs4_proc_open_confirm(data); + + return status; +} + +/* + * Additional permission checks in order to distinguish between an + * open for read, and an open for execute. This works around the + * fact that NFSv4 OPEN treats read and execute permissions as being + * the same. + * Note that in the non-execute case, we want to turn off permission + * checking if we just created a new file (POSIX open() semantics). + */ +static int nfs4_opendata_access(const struct cred *cred, + struct nfs4_opendata *opendata, + struct nfs4_state *state, fmode_t fmode) +{ + struct nfs_access_entry cache; + u32 mask, flags; + + /* access call failed or for some reason the server doesn't + * support any access modes -- defer access call until later */ + if (opendata->o_res.access_supported == 0) + return 0; + + mask = 0; + if (fmode & FMODE_EXEC) { + /* ONLY check for exec rights */ + if (S_ISDIR(state->inode->i_mode)) + mask = NFS4_ACCESS_LOOKUP; + else + mask = NFS4_ACCESS_EXECUTE; + } else if ((fmode & FMODE_READ) && !opendata->file_created) + mask = NFS4_ACCESS_READ; + + nfs_access_set_mask(&cache, opendata->o_res.access_result); + nfs_access_add_cache(state->inode, &cache, cred); + + flags = NFS4_ACCESS_READ | NFS4_ACCESS_EXECUTE | NFS4_ACCESS_LOOKUP; + if ((mask & ~cache.mask & flags) == 0) + return 0; + + return -EACCES; +} + +/* + * Note: On error, nfs4_proc_open will free the struct nfs4_opendata + */ +static int _nfs4_proc_open(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct inode *dir = d_inode(data->dir); + struct nfs_server *server = NFS_SERVER(dir); + struct nfs_openargs *o_arg = &data->o_arg; + struct nfs_openres *o_res = &data->o_res; + int status; + + status = nfs4_run_open_task(data, ctx); + if (!data->rpc_done) + return status; + if (status != 0) { + if (status == -NFS4ERR_BADNAME && + !(o_arg->open_flags & O_CREAT)) + return -ENOENT; + return status; + } + + nfs_fattr_map_and_free_names(server, &data->f_attr); + + if (o_arg->open_flags & O_CREAT) { + if (o_arg->open_flags & O_EXCL) + data->file_created = true; + else if (o_res->cinfo.before != o_res->cinfo.after) + data->file_created = true; + if (data->file_created || + inode_peek_iversion_raw(dir) != o_res->cinfo.after) + nfs4_update_changeattr(dir, &o_res->cinfo, + o_res->f_attr->time_start, + NFS_INO_INVALID_DATA); + } + if ((o_res->rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) == 0) + server->caps &= ~NFS_CAP_POSIX_LOCK; + if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) { + status = _nfs4_proc_open_confirm(data); + if (status != 0) + return status; + } + if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) { + struct nfs_fh *fh = &o_res->fh; + + nfs4_sequence_free_slot(&o_res->seq_res); + if (o_arg->claim == NFS4_OPEN_CLAIM_FH) + fh = NFS_FH(d_inode(data->dentry)); + nfs4_proc_getattr(server, fh, o_res->f_attr, NULL); + } + return 0; +} + +/* + * OPEN_EXPIRED: + * reclaim state on the server after a network partition. + * Assumes caller holds the appropriate lock + */ +static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct nfs4_opendata *opendata; + int ret; + + opendata = nfs4_open_recoverdata_alloc(ctx, state, NFS4_OPEN_CLAIM_FH); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); + /* + * We're not recovering a delegation, so ask for no delegation. + * Otherwise the recovery thread could deadlock with an outstanding + * delegation return. + */ + opendata->o_arg.open_flags = O_DIRECT; + ret = nfs4_open_recover(opendata, state); + if (ret == -ESTALE) + d_drop(ctx->dentry); + nfs4_opendata_put(opendata); + return ret; +} + +static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_exception exception = { }; + int err; + + do { + err = _nfs4_open_expired(ctx, state); + trace_nfs4_open_expired(ctx, 0, err); + if (nfs4_clear_cap_atomic_open_v1(server, err, &exception)) + continue; + switch (err) { + default: + goto out; + case -NFS4ERR_GRACE: + case -NFS4ERR_DELAY: + nfs4_handle_exception(server, err, &exception); + err = 0; + } + } while (exception.retry); +out: + return err; +} + +int nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state) +{ + struct nfs_open_context *ctx; + int ret; + + ctx = nfs4_state_find_open_context(state); + if (IS_ERR(ctx)) + return -EAGAIN; + ret = nfs4_do_open_expired(ctx, state); + put_nfs_open_context(ctx); + return ret; +} + +void nfs_finish_clear_delegation_stateid(struct nfs4_state *state, + const nfs4_stateid *stateid) +{ + nfs_remove_bad_delegation(state->inode, stateid); + nfs_state_clear_delegation(state); +} + +static int nfs41_test_and_free_expired_stateid(struct nfs_server *server, + nfs4_stateid *stateid, const struct cred *cred) +{ + int status; + + switch (stateid->type) { + default: + break; + case NFS4_INVALID_STATEID_TYPE: + case NFS4_SPECIAL_STATEID_TYPE: + case NFS4_FREED_STATEID_TYPE: + return -NFS4ERR_BAD_STATEID; + case NFS4_REVOKED_STATEID_TYPE: + goto out_free; + } + + status = nfs41_test_stateid(server, stateid, cred); + switch (status) { + case -NFS4ERR_EXPIRED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + break; + default: + return status; + } +out_free: + /* Ack the revoked state to the server */ + nfs41_free_stateid(server, stateid, cred, true); + return -NFS4ERR_EXPIRED; +} + +static int nfs41_check_delegation_stateid(struct nfs4_state *state) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + nfs4_stateid stateid; + struct nfs_delegation *delegation; + const struct cred *cred = NULL; + int status, ret = NFS_OK; + + /* Get the delegation credential for use by test/free_stateid */ + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(state->inode)->delegation); + if (delegation == NULL) { + rcu_read_unlock(); + nfs_state_clear_delegation(state); + return NFS_OK; + } + + spin_lock(&delegation->lock); + nfs4_stateid_copy(&stateid, &delegation->stateid); + + if (!test_and_clear_bit(NFS_DELEGATION_TEST_EXPIRED, + &delegation->flags)) { + spin_unlock(&delegation->lock); + rcu_read_unlock(); + return NFS_OK; + } + + if (delegation->cred) + cred = get_cred(delegation->cred); + spin_unlock(&delegation->lock); + rcu_read_unlock(); + status = nfs41_test_and_free_expired_stateid(server, &stateid, cred); + trace_nfs4_test_delegation_stateid(state, NULL, status); + if (status == -NFS4ERR_EXPIRED || status == -NFS4ERR_BAD_STATEID) + nfs_finish_clear_delegation_stateid(state, &stateid); + else + ret = status; + + put_cred(cred); + return ret; +} + +static void nfs41_delegation_recover_stateid(struct nfs4_state *state) +{ + nfs4_stateid tmp; + + if (test_bit(NFS_DELEGATED_STATE, &state->flags) && + nfs4_copy_delegation_stateid(state->inode, state->state, + &tmp, NULL) && + nfs4_stateid_match_other(&state->stateid, &tmp)) + nfs_state_set_delegation(state, &tmp, state->state); + else + nfs_state_clear_delegation(state); +} + +/** + * nfs41_check_expired_locks - possibly free a lock stateid + * + * @state: NFSv4 state for an inode + * + * Returns NFS_OK if recovery for this stateid is now finished. + * Otherwise a negative NFS4ERR value is returned. + */ +static int nfs41_check_expired_locks(struct nfs4_state *state) +{ + int status, ret = NFS_OK; + struct nfs4_lock_state *lsp, *prev = NULL; + struct nfs_server *server = NFS_SERVER(state->inode); + + if (!test_bit(LK_STATE_IN_USE, &state->flags)) + goto out; + + spin_lock(&state->state_lock); + list_for_each_entry(lsp, &state->lock_states, ls_locks) { + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags)) { + const struct cred *cred = lsp->ls_state->owner->so_cred; + + refcount_inc(&lsp->ls_count); + spin_unlock(&state->state_lock); + + nfs4_put_lock_state(prev); + prev = lsp; + + status = nfs41_test_and_free_expired_stateid(server, + &lsp->ls_stateid, + cred); + trace_nfs4_test_lock_stateid(state, lsp, status); + if (status == -NFS4ERR_EXPIRED || + status == -NFS4ERR_BAD_STATEID) { + clear_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags); + lsp->ls_stateid.type = NFS4_INVALID_STATEID_TYPE; + if (!recover_lost_locks) + set_bit(NFS_LOCK_LOST, &lsp->ls_flags); + } else if (status != NFS_OK) { + ret = status; + nfs4_put_lock_state(prev); + goto out; + } + spin_lock(&state->state_lock); + } + } + spin_unlock(&state->state_lock); + nfs4_put_lock_state(prev); +out: + return ret; +} + +/** + * nfs41_check_open_stateid - possibly free an open stateid + * + * @state: NFSv4 state for an inode + * + * Returns NFS_OK if recovery for this stateid is now finished. + * Otherwise a negative NFS4ERR value is returned. + */ +static int nfs41_check_open_stateid(struct nfs4_state *state) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + nfs4_stateid *stateid = &state->open_stateid; + const struct cred *cred = state->owner->so_cred; + int status; + + if (test_bit(NFS_OPEN_STATE, &state->flags) == 0) + return -NFS4ERR_BAD_STATEID; + status = nfs41_test_and_free_expired_stateid(server, stateid, cred); + trace_nfs4_test_open_stateid(state, NULL, status); + if (status == -NFS4ERR_EXPIRED || status == -NFS4ERR_BAD_STATEID) { + nfs_state_clear_open_state_flags(state); + stateid->type = NFS4_INVALID_STATEID_TYPE; + return status; + } + if (nfs_open_stateid_recover_openmode(state)) + return -NFS4ERR_OPENMODE; + return NFS_OK; +} + +static int nfs41_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state) +{ + int status; + + status = nfs41_check_delegation_stateid(state); + if (status != NFS_OK) + return status; + nfs41_delegation_recover_stateid(state); + + status = nfs41_check_expired_locks(state); + if (status != NFS_OK) + return status; + status = nfs41_check_open_stateid(state); + if (status != NFS_OK) + status = nfs4_open_expired(sp, state); + return status; +} + +/* + * on an EXCLUSIVE create, the server should send back a bitmask with FATTR4-* + * fields corresponding to attributes that were used to store the verifier. + * Make sure we clobber those fields in the later setattr call + */ +static unsigned nfs4_exclusive_attrset(struct nfs4_opendata *opendata, + struct iattr *sattr, struct nfs4_label **label) +{ + const __u32 *bitmask = opendata->o_arg.server->exclcreat_bitmask; + __u32 attrset[3]; + unsigned ret; + unsigned i; + + for (i = 0; i < ARRAY_SIZE(attrset); i++) { + attrset[i] = opendata->o_res.attrset[i]; + if (opendata->o_arg.createmode == NFS4_CREATE_EXCLUSIVE4_1) + attrset[i] &= ~bitmask[i]; + } + + ret = (opendata->o_arg.createmode == NFS4_CREATE_EXCLUSIVE) ? + sattr->ia_valid : 0; + + if ((attrset[1] & (FATTR4_WORD1_TIME_ACCESS|FATTR4_WORD1_TIME_ACCESS_SET))) { + if (sattr->ia_valid & ATTR_ATIME_SET) + ret |= ATTR_ATIME_SET; + else + ret |= ATTR_ATIME; + } + + if ((attrset[1] & (FATTR4_WORD1_TIME_MODIFY|FATTR4_WORD1_TIME_MODIFY_SET))) { + if (sattr->ia_valid & ATTR_MTIME_SET) + ret |= ATTR_MTIME_SET; + else + ret |= ATTR_MTIME; + } + + if (!(attrset[2] & FATTR4_WORD2_SECURITY_LABEL)) + *label = NULL; + return ret; +} + +static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, + struct nfs_open_context *ctx) +{ + struct nfs4_state_owner *sp = opendata->owner; + struct nfs_server *server = sp->so_server; + struct dentry *dentry; + struct nfs4_state *state; + fmode_t acc_mode = _nfs4_ctx_to_accessmode(ctx); + struct inode *dir = d_inode(opendata->dir); + unsigned long dir_verifier; + int ret; + + dir_verifier = nfs_save_change_attribute(dir); + + ret = _nfs4_proc_open(opendata, ctx); + if (ret != 0) + goto out; + + state = _nfs4_opendata_to_nfs4_state(opendata); + ret = PTR_ERR(state); + if (IS_ERR(state)) + goto out; + ctx->state = state; + if (server->caps & NFS_CAP_POSIX_LOCK) + set_bit(NFS_STATE_POSIX_LOCKS, &state->flags); + if (opendata->o_res.rflags & NFS4_OPEN_RESULT_MAY_NOTIFY_LOCK) + set_bit(NFS_STATE_MAY_NOTIFY_LOCK, &state->flags); + if (opendata->o_res.rflags & NFS4_OPEN_RESULT_PRESERVE_UNLINKED) + set_bit(NFS_INO_PRESERVE_UNLINKED, &NFS_I(state->inode)->flags); + + switch(opendata->o_arg.claim) { + default: + break; + case NFS4_OPEN_CLAIM_NULL: + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + case NFS4_OPEN_CLAIM_DELEGATE_PREV: + if (!opendata->rpc_done) + break; + if (opendata->o_res.delegation.type != 0) + dir_verifier = nfs_save_change_attribute(dir); + } + + dentry = opendata->dentry; + nfs_set_verifier(dentry, dir_verifier); + if (d_really_is_negative(dentry)) { + struct dentry *alias; + d_drop(dentry); + alias = d_splice_alias(igrab(state->inode), dentry); + /* d_splice_alias() can't fail here - it's a non-directory */ + if (alias) { + dput(ctx->dentry); + nfs_set_verifier(alias, dir_verifier); + ctx->dentry = dentry = alias; + } + } + + /* Parse layoutget results before we check for access */ + pnfs_parse_lgopen(state->inode, opendata->lgp, ctx); + + ret = nfs4_opendata_access(sp->so_cred, opendata, state, acc_mode); + if (ret != 0) + goto out; + + if (d_inode(dentry) == state->inode) + nfs_inode_attach_open_context(ctx); + +out: + if (!opendata->cancelled) { + if (opendata->lgp) { + nfs4_lgopen_release(opendata->lgp); + opendata->lgp = NULL; + } + nfs4_sequence_free_slot(&opendata->o_res.seq_res); + } + return ret; +} + +/* + * Returns a referenced nfs4_state + */ +static int _nfs4_do_open(struct inode *dir, + struct nfs_open_context *ctx, + int flags, + const struct nfs4_open_createattrs *c, + int *opened) +{ + struct nfs4_state_owner *sp; + struct nfs4_state *state = NULL; + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_opendata *opendata; + struct dentry *dentry = ctx->dentry; + const struct cred *cred = ctx->cred; + struct nfs4_threshold **ctx_th = &ctx->mdsthreshold; + fmode_t fmode = _nfs4_ctx_to_openmode(ctx); + enum open_claim_type4 claim = NFS4_OPEN_CLAIM_NULL; + struct iattr *sattr = c->sattr; + struct nfs4_label *label = c->label; + int status; + + /* Protect against reboot recovery conflicts */ + status = -ENOMEM; + sp = nfs4_get_state_owner(server, cred, GFP_KERNEL); + if (sp == NULL) { + dprintk("nfs4_do_open: nfs4_get_state_owner failed!\n"); + goto out_err; + } + status = nfs4_client_recover_expired_lease(server->nfs_client); + if (status != 0) + goto err_put_state_owner; + if (d_really_is_positive(dentry)) + nfs4_return_incompatible_delegation(d_inode(dentry), fmode); + status = -ENOMEM; + if (d_really_is_positive(dentry)) + claim = NFS4_OPEN_CLAIM_FH; + opendata = nfs4_opendata_alloc(dentry, sp, fmode, flags, + c, claim, GFP_KERNEL); + if (opendata == NULL) + goto err_put_state_owner; + + if (server->attr_bitmask[2] & FATTR4_WORD2_MDSTHRESHOLD) { + if (!opendata->f_attr.mdsthreshold) { + opendata->f_attr.mdsthreshold = pnfs_mdsthreshold_alloc(); + if (!opendata->f_attr.mdsthreshold) + goto err_opendata_put; + } + opendata->o_arg.open_bitmap = &nfs4_pnfs_open_bitmap[0]; + } + if (d_really_is_positive(dentry)) + opendata->state = nfs4_get_open_state(d_inode(dentry), sp); + + status = _nfs4_open_and_get_state(opendata, ctx); + if (status != 0) + goto err_opendata_put; + state = ctx->state; + + if ((opendata->o_arg.open_flags & (O_CREAT|O_EXCL)) == (O_CREAT|O_EXCL) && + (opendata->o_arg.createmode != NFS4_CREATE_GUARDED)) { + unsigned attrs = nfs4_exclusive_attrset(opendata, sattr, &label); + /* + * send create attributes which was not set by open + * with an extra setattr. + */ + if (attrs || label) { + unsigned ia_old = sattr->ia_valid; + + sattr->ia_valid = attrs; + nfs_fattr_init(opendata->o_res.f_attr); + status = nfs4_do_setattr(state->inode, cred, + opendata->o_res.f_attr, sattr, + ctx, label); + if (status == 0) { + nfs_setattr_update_inode(state->inode, sattr, + opendata->o_res.f_attr); + nfs_setsecurity(state->inode, opendata->o_res.f_attr); + } + sattr->ia_valid = ia_old; + } + } + if (opened && opendata->file_created) + *opened = 1; + + if (pnfs_use_threshold(ctx_th, opendata->f_attr.mdsthreshold, server)) { + *ctx_th = opendata->f_attr.mdsthreshold; + opendata->f_attr.mdsthreshold = NULL; + } + + nfs4_opendata_put(opendata); + nfs4_put_state_owner(sp); + return 0; +err_opendata_put: + nfs4_opendata_put(opendata); +err_put_state_owner: + nfs4_put_state_owner(sp); +out_err: + return status; +} + + +static struct nfs4_state *nfs4_do_open(struct inode *dir, + struct nfs_open_context *ctx, + int flags, + struct iattr *sattr, + struct nfs4_label *label, + int *opened) +{ + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_exception exception = { + .interruptible = true, + }; + struct nfs4_state *res; + struct nfs4_open_createattrs c = { + .label = label, + .sattr = sattr, + .verf = { + [0] = (__u32)jiffies, + [1] = (__u32)current->pid, + }, + }; + int status; + + do { + status = _nfs4_do_open(dir, ctx, flags, &c, opened); + res = ctx->state; + trace_nfs4_open_file(ctx, flags, status); + if (status == 0) + break; + /* NOTE: BAD_SEQID means the server and client disagree about the + * book-keeping w.r.t. state-changing operations + * (OPEN/CLOSE/LOCK/LOCKU...) + * It is actually a sign of a bug on the client or on the server. + * + * If we receive a BAD_SEQID error in the particular case of + * doing an OPEN, we assume that nfs_increment_open_seqid() will + * have unhashed the old state_owner for us, and that we can + * therefore safely retry using a new one. We should still warn + * the user though... + */ + if (status == -NFS4ERR_BAD_SEQID) { + pr_warn_ratelimited("NFS: v4 server %s " + " returned a bad sequence-id error!\n", + NFS_SERVER(dir)->nfs_client->cl_hostname); + exception.retry = 1; + continue; + } + /* + * BAD_STATEID on OPEN means that the server cancelled our + * state before it received the OPEN_CONFIRM. + * Recover by retrying the request as per the discussion + * on Page 181 of RFC3530. + */ + if (status == -NFS4ERR_BAD_STATEID) { + exception.retry = 1; + continue; + } + if (status == -NFS4ERR_EXPIRED) { + nfs4_schedule_lease_recovery(server->nfs_client); + exception.retry = 1; + continue; + } + if (status == -EAGAIN) { + /* We must have found a delegation */ + exception.retry = 1; + continue; + } + if (nfs4_clear_cap_atomic_open_v1(server, status, &exception)) + continue; + res = ERR_PTR(nfs4_handle_exception(server, + status, &exception)); + } while (exception.retry); + return res; +} + +static int _nfs4_do_setattr(struct inode *inode, + struct nfs_setattrargs *arg, + struct nfs_setattrres *res, + const struct cred *cred, + struct nfs_open_context *ctx) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETATTR], + .rpc_argp = arg, + .rpc_resp = res, + .rpc_cred = cred, + }; + const struct cred *delegation_cred = NULL; + unsigned long timestamp = jiffies; + bool truncate; + int status; + + nfs_fattr_init(res->fattr); + + /* Servers should only apply open mode checks for file size changes */ + truncate = (arg->iap->ia_valid & ATTR_SIZE) ? true : false; + if (!truncate) { + nfs4_inode_make_writeable(inode); + goto zero_stateid; + } + + if (nfs4_copy_delegation_stateid(inode, FMODE_WRITE, &arg->stateid, &delegation_cred)) { + /* Use that stateid */ + } else if (ctx != NULL && ctx->state) { + struct nfs_lock_context *l_ctx; + if (!nfs4_valid_open_stateid(ctx->state)) + return -EBADF; + l_ctx = nfs_get_lock_context(ctx); + if (IS_ERR(l_ctx)) + return PTR_ERR(l_ctx); + status = nfs4_select_rw_stateid(ctx->state, FMODE_WRITE, l_ctx, + &arg->stateid, &delegation_cred); + nfs_put_lock_context(l_ctx); + if (status == -EIO) + return -EBADF; + else if (status == -EAGAIN) + goto zero_stateid; + } else { +zero_stateid: + nfs4_stateid_copy(&arg->stateid, &zero_stateid); + } + if (delegation_cred) + msg.rpc_cred = delegation_cred; + + status = nfs4_call_sync(server->client, server, &msg, &arg->seq_args, &res->seq_res, 1); + + put_cred(delegation_cred); + if (status == 0 && ctx != NULL) + renew_lease(server, timestamp); + trace_nfs4_setattr(inode, &arg->stateid, status); + return status; +} + +static int nfs4_do_setattr(struct inode *inode, const struct cred *cred, + struct nfs_fattr *fattr, struct iattr *sattr, + struct nfs_open_context *ctx, struct nfs4_label *ilabel) +{ + struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS4_BITMASK_SZ]; + struct nfs4_state *state = ctx ? ctx->state : NULL; + struct nfs_setattrargs arg = { + .fh = NFS_FH(inode), + .iap = sattr, + .server = server, + .bitmask = bitmask, + .label = ilabel, + }; + struct nfs_setattrres res = { + .fattr = fattr, + .server = server, + }; + struct nfs4_exception exception = { + .state = state, + .inode = inode, + .stateid = &arg.stateid, + }; + unsigned long adjust_flags = NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME; + int err; + + if (sattr->ia_valid & (ATTR_MODE | ATTR_KILL_SUID | ATTR_KILL_SGID)) + adjust_flags |= NFS_INO_INVALID_MODE; + if (sattr->ia_valid & (ATTR_UID | ATTR_GID)) + adjust_flags |= NFS_INO_INVALID_OTHER; + if (sattr->ia_valid & ATTR_ATIME) + adjust_flags |= NFS_INO_INVALID_ATIME; + if (sattr->ia_valid & ATTR_MTIME) + adjust_flags |= NFS_INO_INVALID_MTIME; + + do { + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, fattr->label), + inode, adjust_flags); + + err = _nfs4_do_setattr(inode, &arg, &res, cred, ctx); + switch (err) { + case -NFS4ERR_OPENMODE: + if (!(sattr->ia_valid & ATTR_SIZE)) { + pr_warn_once("NFSv4: server %s is incorrectly " + "applying open mode checks to " + "a SETATTR that is not " + "changing file size.\n", + server->nfs_client->cl_hostname); + } + if (state && !(state->state & FMODE_WRITE)) { + err = -EBADF; + if (sattr->ia_valid & ATTR_OPEN) + err = -EACCES; + goto out; + } + } + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); +out: + return err; +} + +static bool +nfs4_wait_on_layoutreturn(struct inode *inode, struct rpc_task *task) +{ + if (inode == NULL || !nfs_have_layout(inode)) + return false; + + return pnfs_wait_on_layoutreturn(inode, task); +} + +/* + * Update the seqid of an open stateid + */ +static void nfs4_sync_open_stateid(nfs4_stateid *dst, + struct nfs4_state *state) +{ + __be32 seqid_open; + u32 dst_seqid; + int seq; + + for (;;) { + if (!nfs4_valid_open_stateid(state)) + break; + seq = read_seqbegin(&state->seqlock); + if (!nfs4_state_match_open_stateid_other(state, dst)) { + nfs4_stateid_copy(dst, &state->open_stateid); + if (read_seqretry(&state->seqlock, seq)) + continue; + break; + } + seqid_open = state->open_stateid.seqid; + if (read_seqretry(&state->seqlock, seq)) + continue; + + dst_seqid = be32_to_cpu(dst->seqid); + if ((s32)(dst_seqid - be32_to_cpu(seqid_open)) < 0) + dst->seqid = seqid_open; + break; + } +} + +/* + * Update the seqid of an open stateid after receiving + * NFS4ERR_OLD_STATEID + */ +static bool nfs4_refresh_open_old_stateid(nfs4_stateid *dst, + struct nfs4_state *state) +{ + __be32 seqid_open; + u32 dst_seqid; + bool ret; + int seq, status = -EAGAIN; + DEFINE_WAIT(wait); + + for (;;) { + ret = false; + if (!nfs4_valid_open_stateid(state)) + break; + seq = read_seqbegin(&state->seqlock); + if (!nfs4_state_match_open_stateid_other(state, dst)) { + if (read_seqretry(&state->seqlock, seq)) + continue; + break; + } + + write_seqlock(&state->seqlock); + seqid_open = state->open_stateid.seqid; + + dst_seqid = be32_to_cpu(dst->seqid); + + /* Did another OPEN bump the state's seqid? try again: */ + if ((s32)(be32_to_cpu(seqid_open) - dst_seqid) > 0) { + dst->seqid = seqid_open; + write_sequnlock(&state->seqlock); + ret = true; + break; + } + + /* server says we're behind but we haven't seen the update yet */ + set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); + prepare_to_wait(&state->waitq, &wait, TASK_KILLABLE); + write_sequnlock(&state->seqlock); + trace_nfs4_close_stateid_update_wait(state->inode, dst, 0); + + if (fatal_signal_pending(current) || nfs_current_task_exiting()) + status = -EINTR; + else + if (schedule_timeout(5*HZ) != 0) + status = 0; + + finish_wait(&state->waitq, &wait); + + if (!status) + continue; + if (status == -EINTR) + break; + + /* we slept the whole 5 seconds, we must have lost a seqid */ + dst->seqid = cpu_to_be32(dst_seqid + 1); + ret = true; + break; + } + + return ret; +} + +struct nfs4_closedata { + struct inode *inode; + struct nfs4_state *state; + struct nfs_closeargs arg; + struct nfs_closeres res; + struct { + struct nfs4_layoutreturn_args arg; + struct nfs4_layoutreturn_res res; + struct nfs4_xdr_opaque_data ld_private; + u32 roc_barrier; + bool roc; + } lr; + struct nfs_fattr fattr; + unsigned long timestamp; + unsigned short retrans; +}; + +static void nfs4_free_closedata(void *data) +{ + struct nfs4_closedata *calldata = data; + struct nfs4_state_owner *sp = calldata->state->owner; + struct super_block *sb = calldata->state->inode->i_sb; + + if (calldata->lr.roc) + pnfs_roc_release(&calldata->lr.arg, &calldata->lr.res, + calldata->res.lr_ret); + nfs4_put_open_state(calldata->state); + nfs_free_seqid(calldata->arg.seqid); + nfs4_put_state_owner(sp); + nfs_sb_deactive(sb); + kfree(calldata); +} + +static void nfs4_close_done(struct rpc_task *task, void *data) +{ + struct nfs4_closedata *calldata = data; + struct nfs4_state *state = calldata->state; + struct nfs_server *server = NFS_SERVER(calldata->inode); + nfs4_stateid *res_stateid = NULL; + struct nfs4_exception exception = { + .state = state, + .inode = calldata->inode, + .stateid = &calldata->arg.stateid, + .retrans = calldata->retrans, + }; + + if (!nfs4_sequence_done(task, &calldata->res.seq_res)) + return; + trace_nfs4_close(state, &calldata->arg, &calldata->res, task->tk_status); + + /* Handle Layoutreturn errors */ + if (pnfs_roc_done(task, &calldata->arg.lr_args, &calldata->res.lr_res, + &calldata->res.lr_ret) == -EAGAIN) + goto out_restart; + + /* hmm. we are done with the inode, and in the process of freeing + * the state_owner. we keep this around to process errors + */ + switch (task->tk_status) { + case 0: + res_stateid = &calldata->res.stateid; + renew_lease(server, calldata->timestamp); + break; + case -NFS4ERR_ACCESS: + if (calldata->arg.bitmask != NULL) { + calldata->arg.bitmask = NULL; + calldata->res.fattr = NULL; + goto out_restart; + + } + break; + case -NFS4ERR_OLD_STATEID: + /* Did we race with OPEN? */ + if (nfs4_refresh_open_old_stateid(&calldata->arg.stateid, + state)) + goto out_restart; + goto out_release; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_EXPIRED: + nfs4_free_revoked_stateid(server, + &calldata->arg.stateid, + task->tk_msg.rpc_cred); + fallthrough; + case -NFS4ERR_BAD_STATEID: + if (calldata->arg.fmode == 0) + break; + fallthrough; + default: + task->tk_status = nfs4_async_handle_exception(task, + server, task->tk_status, &exception); + calldata->retrans = exception.retrans; + if (exception.retry) + goto out_restart; + } + nfs_clear_open_stateid(state, &calldata->arg.stateid, + res_stateid, calldata->arg.fmode); +out_release: + task->tk_status = 0; + nfs_release_seqid(calldata->arg.seqid); + nfs_refresh_inode(calldata->inode, &calldata->fattr); + dprintk("%s: ret = %d\n", __func__, task->tk_status); + return; +out_restart: + task->tk_status = 0; + rpc_restart_call_prepare(task); + goto out_release; +} + +static void nfs4_close_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_closedata *calldata = data; + struct nfs4_state *state = calldata->state; + struct inode *inode = calldata->inode; + struct nfs_server *server = NFS_SERVER(inode); + struct pnfs_layout_hdr *lo; + bool is_rdonly, is_wronly, is_rdwr; + int call_close = 0; + + if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) + goto out_wait; + + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; + spin_lock(&state->owner->so_lock); + is_rdwr = test_bit(NFS_O_RDWR_STATE, &state->flags); + is_rdonly = test_bit(NFS_O_RDONLY_STATE, &state->flags); + is_wronly = test_bit(NFS_O_WRONLY_STATE, &state->flags); + /* Calculate the change in open mode */ + calldata->arg.fmode = 0; + if (state->n_rdwr == 0) { + if (state->n_rdonly == 0) + call_close |= is_rdonly; + else if (is_rdonly) + calldata->arg.fmode |= FMODE_READ; + if (state->n_wronly == 0) + call_close |= is_wronly; + else if (is_wronly) + calldata->arg.fmode |= FMODE_WRITE; + if (calldata->arg.fmode != (FMODE_READ|FMODE_WRITE)) + call_close |= is_rdwr; + } else if (is_rdwr) + calldata->arg.fmode |= FMODE_READ|FMODE_WRITE; + + nfs4_sync_open_stateid(&calldata->arg.stateid, state); + if (!nfs4_valid_open_stateid(state)) + call_close = 0; + spin_unlock(&state->owner->so_lock); + + if (!call_close) { + /* Note: exit _without_ calling nfs4_close_done */ + goto out_no_action; + } + + if (!calldata->lr.roc && nfs4_wait_on_layoutreturn(inode, task)) { + nfs_release_seqid(calldata->arg.seqid); + goto out_wait; + } + + lo = calldata->arg.lr_args ? calldata->arg.lr_args->layout : NULL; + if (lo && !pnfs_layout_is_valid(lo)) { + calldata->arg.lr_args = NULL; + calldata->res.lr_res = NULL; + } + + if (calldata->arg.fmode == 0) + task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE]; + + if (calldata->arg.fmode == 0 || calldata->arg.fmode == FMODE_READ) { + /* Close-to-open cache consistency revalidation */ + if (!nfs4_have_delegation(inode, FMODE_READ, 0)) { + nfs4_bitmask_set(calldata->arg.bitmask_store, + server->cache_consistency_bitmask, + inode, 0); + calldata->arg.bitmask = calldata->arg.bitmask_store; + } else + calldata->arg.bitmask = NULL; + } + + calldata->arg.share_access = + nfs4_fmode_to_share_access(calldata->arg.fmode); + + if (calldata->res.fattr == NULL) + calldata->arg.bitmask = NULL; + else if (calldata->arg.bitmask == NULL) + calldata->res.fattr = NULL; + calldata->timestamp = jiffies; + if (nfs4_setup_sequence(NFS_SERVER(inode)->nfs_client, + &calldata->arg.seq_args, + &calldata->res.seq_res, + task) != 0) + nfs_release_seqid(calldata->arg.seqid); + return; +out_no_action: + task->tk_action = NULL; +out_wait: + nfs4_sequence_done(task, &calldata->res.seq_res); +} + +static const struct rpc_call_ops nfs4_close_ops = { + .rpc_call_prepare = nfs4_close_prepare, + .rpc_call_done = nfs4_close_done, + .rpc_release = nfs4_free_closedata, +}; + +/* + * It is possible for data to be read/written from a mem-mapped file + * after the sys_close call (which hits the vfs layer as a flush). + * This means that we can't safely call nfsv4 close on a file until + * the inode is cleared. This in turn means that we are not good + * NFSv4 citizens - we do not indicate to the server to update the file's + * share state even when we are done with one of the three share + * stateid's in the inode. + * + * NOTE: Caller must be holding the sp->so_owner semaphore! + */ +int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs_client *clp = server->nfs_client; + struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t); + struct nfs4_closedata *calldata; + struct nfs4_state_owner *sp = state->owner; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE], + .rpc_cred = state->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_close_ops, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + int status = -ENOMEM; + + if (nfs_server_capable(state->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_CLEANUP, + &task_setup_data.rpc_client, &msg); + + calldata = kzalloc_obj(*calldata, gfp_mask); + if (calldata == NULL) + goto out; + nfs4_init_sequence(clp, &calldata->arg.seq_args, + &calldata->res.seq_res, 1, 0); + calldata->inode = state->inode; + calldata->state = state; + calldata->arg.fh = NFS_FH(state->inode); + if (!nfs4_copy_open_stateid(&calldata->arg.stateid, state)) + goto out_free_calldata; + /* Serialization for the sequence id */ + alloc_seqid = clp->cl_mvops->alloc_seqid; + calldata->arg.seqid = alloc_seqid(&state->owner->so_seqid, gfp_mask); + if (IS_ERR(calldata->arg.seqid)) + goto out_free_calldata; + nfs_fattr_init(&calldata->fattr); + calldata->arg.fmode = 0; + calldata->lr.arg.ld_private = &calldata->lr.ld_private; + calldata->res.fattr = &calldata->fattr; + calldata->res.seqid = calldata->arg.seqid; + calldata->res.server = server; + calldata->res.lr_ret = -NFS4ERR_NOMATCHING_LAYOUT; + calldata->lr.roc = pnfs_roc(state->inode, &calldata->lr.arg, + &calldata->lr.res, msg.rpc_cred, wait); + if (calldata->lr.roc) { + calldata->arg.lr_args = &calldata->lr.arg; + calldata->res.lr_res = &calldata->lr.res; + } + nfs_sb_active(calldata->inode->i_sb); + + msg.rpc_argp = &calldata->arg; + msg.rpc_resp = &calldata->res; + task_setup_data.callback_data = calldata; + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + status = 0; + if (wait) + status = rpc_wait_for_completion_task(task); + rpc_put_task(task); + return status; +out_free_calldata: + kfree(calldata); +out: + nfs4_put_open_state(state); + nfs4_put_state_owner(sp); + return status; +} + +static struct inode * +nfs4_atomic_open(struct inode *dir, struct nfs_open_context *ctx, + int open_flags, struct iattr *attr, int *opened) +{ + struct nfs4_state *state; + struct nfs4_label l, *label; + + label = nfs4_label_init_security(dir, ctx->dentry, attr, &l); + + /* Protect against concurrent sillydeletes */ + state = nfs4_do_open(dir, ctx, open_flags, attr, label, opened); + + nfs4_label_release_security(label); + + if (IS_ERR(state)) + return ERR_CAST(state); + return state->inode; +} + +static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) +{ + struct dentry *dentry = ctx->dentry; + if (ctx->state == NULL) + return; + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + nfs4_inode_set_return_delegation_on_close(d_inode(dentry)); + if (is_sync) + nfs4_close_sync(ctx->state, _nfs4_ctx_to_openmode(ctx)); + else + nfs4_close_state(ctx->state, _nfs4_ctx_to_openmode(ctx)); +} + +#define FATTR4_WORD1_NFS40_MASK (2*FATTR4_WORD1_MOUNTED_ON_FILEID - 1UL) +#define FATTR4_WORD2_NFS41_MASK (2*FATTR4_WORD2_SUPPATTR_EXCLCREAT - 1UL) +#define FATTR4_WORD2_NFS42_MASK (2*FATTR4_WORD2_OPEN_ARGUMENTS - 1UL) + +#define FATTR4_WORD2_NFS42_TIME_DELEG_MASK \ + (FATTR4_WORD2_TIME_DELEG_MODIFY|FATTR4_WORD2_TIME_DELEG_ACCESS) +static bool nfs4_server_delegtime_capable(struct nfs4_server_caps_res *res) +{ + u32 share_access_want = res->open_caps.oa_share_access_want[0]; + u32 attr_bitmask = res->attr_bitmask[2]; + + return (share_access_want & NFS4_SHARE_WANT_DELEG_TIMESTAMPS) && + ((attr_bitmask & FATTR4_WORD2_NFS42_TIME_DELEG_MASK) == + FATTR4_WORD2_NFS42_TIME_DELEG_MASK); +} + +static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) +{ + u32 minorversion = server->nfs_client->cl_minorversion; + u32 bitmask[3] = { + [0] = FATTR4_WORD0_SUPPORTED_ATTRS, + }; + struct nfs4_server_caps_arg args = { + .fhandle = fhandle, + .bitmask = bitmask, + }; + struct nfs4_server_caps_res res = {}; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS], + .rpc_argp = &args, + .rpc_resp = &res, + }; + int status; + int i; + + bitmask[0] = FATTR4_WORD0_SUPPORTED_ATTRS | + FATTR4_WORD0_FH_EXPIRE_TYPE | + FATTR4_WORD0_LINK_SUPPORT | + FATTR4_WORD0_SYMLINK_SUPPORT | + FATTR4_WORD0_ACLSUPPORT | + FATTR4_WORD0_CASE_INSENSITIVE | + FATTR4_WORD0_CASE_PRESERVING; + if (minorversion) + bitmask[2] = FATTR4_WORD2_SUPPATTR_EXCLCREAT; + if (minorversion > 1) + bitmask[2] |= FATTR4_WORD2_OPEN_ARGUMENTS; + + status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); + if (status == 0) { + bitmask[0] = (FATTR4_WORD0_SUPPORTED_ATTRS | + FATTR4_WORD0_FH_EXPIRE_TYPE | + FATTR4_WORD0_LINK_SUPPORT | + FATTR4_WORD0_SYMLINK_SUPPORT | + FATTR4_WORD0_ACLSUPPORT | + FATTR4_WORD0_CASE_INSENSITIVE | + FATTR4_WORD0_CASE_PRESERVING) & + res.attr_bitmask[0]; + /* Sanity check the server answers */ + switch (minorversion) { + case 0: + res.attr_bitmask[1] &= FATTR4_WORD1_NFS40_MASK; + res.attr_bitmask[2] = 0; + break; + case 1: + res.attr_bitmask[2] &= FATTR4_WORD2_NFS41_MASK; + bitmask[2] = FATTR4_WORD2_SUPPATTR_EXCLCREAT & + res.attr_bitmask[2]; + break; + case 2: + res.attr_bitmask[2] &= FATTR4_WORD2_NFS42_MASK; + bitmask[2] = (FATTR4_WORD2_SUPPATTR_EXCLCREAT | + FATTR4_WORD2_OPEN_ARGUMENTS) & + res.attr_bitmask[2]; + } + memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask)); + server->caps &= + ~(NFS_CAP_ACLS | NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS | + NFS_CAP_SECURITY_LABEL | NFS_CAP_FS_LOCATIONS | + NFS_CAP_OPEN_XOR | NFS_CAP_DELEGTIME | + NFS_CAP_CASE_INSENSITIVE | NFS_CAP_CASE_NONPRESERVING); + server->fattr_valid = NFS_ATTR_FATTR_V4; + if (res.attr_bitmask[0] & FATTR4_WORD0_ACL && + res.acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) + server->caps |= NFS_CAP_ACLS; + if (res.has_links != 0) + server->caps |= NFS_CAP_HARDLINKS; + if (res.has_symlinks != 0) + server->caps |= NFS_CAP_SYMLINKS; + if (res.case_insensitive) + server->caps |= NFS_CAP_CASE_INSENSITIVE; + if ((res.attr_bitmask[0] & FATTR4_WORD0_CASE_PRESERVING) && + !res.case_preserving) + server->caps |= NFS_CAP_CASE_NONPRESERVING; +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + if (res.attr_bitmask[2] & FATTR4_WORD2_SECURITY_LABEL) + server->caps |= NFS_CAP_SECURITY_LABEL; +#endif + if (res.attr_bitmask[0] & FATTR4_WORD0_FS_LOCATIONS) + server->caps |= NFS_CAP_FS_LOCATIONS; + if (!(res.attr_bitmask[0] & FATTR4_WORD0_FILEID)) + server->fattr_valid &= ~NFS_ATTR_FATTR_FILEID; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_MODE)) + server->fattr_valid &= ~NFS_ATTR_FATTR_MODE; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_NUMLINKS)) + server->fattr_valid &= ~NFS_ATTR_FATTR_NLINK; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_OWNER)) + server->fattr_valid &= ~(NFS_ATTR_FATTR_OWNER | + NFS_ATTR_FATTR_OWNER_NAME); + if (!(res.attr_bitmask[1] & FATTR4_WORD1_OWNER_GROUP)) + server->fattr_valid &= ~(NFS_ATTR_FATTR_GROUP | + NFS_ATTR_FATTR_GROUP_NAME); + if (!(res.attr_bitmask[1] & FATTR4_WORD1_SPACE_USED)) + server->fattr_valid &= ~NFS_ATTR_FATTR_SPACE_USED; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_TIME_ACCESS)) + server->fattr_valid &= ~NFS_ATTR_FATTR_ATIME; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_TIME_METADATA)) + server->fattr_valid &= ~NFS_ATTR_FATTR_CTIME; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_TIME_MODIFY)) + server->fattr_valid &= ~NFS_ATTR_FATTR_MTIME; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_TIME_MODIFY)) + server->fattr_valid &= ~NFS_ATTR_FATTR_MTIME; + if (!(res.attr_bitmask[1] & FATTR4_WORD1_TIME_CREATE)) + server->fattr_valid &= ~NFS_ATTR_FATTR_BTIME; + memcpy(server->attr_bitmask_nl, res.attr_bitmask, + sizeof(server->attr_bitmask)); + server->attr_bitmask_nl[2] &= ~FATTR4_WORD2_SECURITY_LABEL; + + if (res.open_caps.oa_share_access_want[0] & + NFS4_SHARE_WANT_OPEN_XOR_DELEGATION) + server->caps |= NFS_CAP_OPEN_XOR; + if (nfs4_server_delegtime_capable(&res)) + server->caps |= NFS_CAP_DELEGTIME; + + memcpy(server->cache_consistency_bitmask, res.attr_bitmask, sizeof(server->cache_consistency_bitmask)); + server->cache_consistency_bitmask[0] &= FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE; + server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY; + server->cache_consistency_bitmask[2] = 0; + + /* Avoid a regression due to buggy server */ + for (i = 0; i < ARRAY_SIZE(res.exclcreat_bitmask); i++) + res.exclcreat_bitmask[i] &= res.attr_bitmask[i]; + memcpy(server->exclcreat_bitmask, res.exclcreat_bitmask, + sizeof(server->exclcreat_bitmask)); + + server->acl_bitmask = res.acl_bitmask; + server->fh_expire_type = res.fh_expire_type; + } + + return status; +} + +int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + do { + err = nfs4_handle_exception(server, + _nfs4_server_capabilities(server, fhandle), + &exception); + } while (exception.retry); + return err; +} + +static void test_fs_location_for_trunking(struct nfs4_fs_location *location, + struct nfs_client *clp, + struct nfs_server *server) +{ + int i; + + for (i = 0; i < location->nservers; i++) { + struct nfs4_string *srv_loc = &location->servers[i]; + struct sockaddr_storage addr; + size_t addrlen; + struct xprt_create xprt_args = { + .ident = 0, + .net = clp->cl_net, + }; + struct nfs4_add_xprt_data xprtdata = { + .clp = clp, + }; + struct rpc_add_xprt_test rpcdata = { + .add_xprt_test = clp->cl_mvops->session_trunk, + .data = &xprtdata, + }; + char *servername = NULL; + + if (!srv_loc->len) + continue; + + addrlen = nfs_parse_server_name(srv_loc->data, srv_loc->len, + &addr, sizeof(addr), + clp->cl_net, server->port); + if (!addrlen) + return; + xprt_args.dstaddr = (struct sockaddr *)&addr; + xprt_args.addrlen = addrlen; + servername = kmalloc(srv_loc->len + 1, GFP_KERNEL); + if (!servername) + return; + memcpy(servername, srv_loc->data, srv_loc->len); + servername[srv_loc->len] = '\0'; + xprt_args.servername = servername; + + xprtdata.cred = nfs4_get_clid_cred(clp); + rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args, + rpc_clnt_setup_test_and_add_xprt, + &rpcdata); + if (xprtdata.cred) + put_cred(xprtdata.cred); + kfree(servername); + } +} + +static bool _is_same_nfs4_pathname(struct nfs4_pathname *path1, + struct nfs4_pathname *path2) +{ + int i; + + if (path1->ncomponents != path2->ncomponents) + return false; + for (i = 0; i < path1->ncomponents; i++) { + if (path1->components[i].len != path2->components[i].len) + return false; + if (memcmp(path1->components[i].data, path2->components[i].data, + path1->components[i].len)) + return false; + } + return true; +} + +static int _nfs4_discover_trunking(struct nfs_server *server, + struct nfs_fh *fhandle) +{ + struct nfs4_fs_locations *locations = NULL; + struct page *page; + const struct cred *cred; + struct nfs_client *clp = server->nfs_client; + const struct nfs4_state_maintenance_ops *ops = + clp->cl_mvops->state_renewal_ops; + int status = -ENOMEM, i; + + cred = ops->get_state_renewal_cred(clp); + if (cred == NULL) { + cred = nfs4_get_clid_cred(clp); + if (cred == NULL) + return -ENOKEY; + } + + page = alloc_page(GFP_KERNEL); + if (!page) + goto out_put_cred; + locations = kmalloc_obj(struct nfs4_fs_locations); + if (!locations) + goto out_free; + locations->fattr = nfs_alloc_fattr(); + if (!locations->fattr) + goto out_free_2; + + status = nfs4_proc_get_locations(server, fhandle, locations, page, + cred); + if (status) + goto out_free_3; + + for (i = 0; i < locations->nlocations; i++) { + if (!_is_same_nfs4_pathname(&locations->fs_path, + &locations->locations[i].rootpath)) + continue; + test_fs_location_for_trunking(&locations->locations[i], clp, + server); + } +out_free_3: + kfree(locations->fattr); +out_free_2: + kfree(locations); +out_free: + __free_page(page); +out_put_cred: + put_cred(cred); + return status; +} + +static int nfs4_discover_trunking(struct nfs_server *server, + struct nfs_fh *fhandle) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + struct nfs_client *clp = server->nfs_client; + int err = 0; + + if (!nfs4_has_session(clp)) + goto out; + do { + err = nfs4_handle_exception(server, + _nfs4_discover_trunking(server, fhandle), + &exception); + } while (exception.retry); +out: + return err; +} + +static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + u32 bitmask[3] = { + [0] = FATTR4_WORD0_TYPE | FATTR4_WORD0_CHANGE | + FATTR4_WORD0_SIZE | FATTR4_WORD0_FSID, + }; + struct nfs4_lookup_root_arg args = { + .bitmask = bitmask, + }; + struct nfs4_lookup_res res = { + .server = server, + .fattr = fattr, + .fh = fhandle, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP_ROOT], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + nfs_fattr_init(fattr); + return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); +} + +static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_lookup_root(server, fhandle, fattr); + trace_nfs4_lookup_root(server, fhandle, fattr, err); + switch (err) { + case 0: + case -NFS4ERR_WRONGSEC: + goto out; + default: + err = nfs4_handle_exception(server, err, &exception); + } + } while (exception.retry); +out: + return err; +} + +static int nfs4_lookup_root_sec(struct nfs_server *server, + struct nfs_fh *fhandle, struct nfs_fattr *fattr, + rpc_authflavor_t flavor) +{ + struct rpc_auth_create_args auth_args = { + .pseudoflavor = flavor, + }; + struct rpc_auth *auth; + + auth = rpcauth_create(&auth_args, server->client); + if (IS_ERR(auth)) + return -EACCES; + return nfs4_lookup_root(server, fhandle, fattr); +} + +/* + * Retry pseudoroot lookup with various security flavors. We do this when: + * + * NFSv4.0: the PUTROOTFH operation returns NFS4ERR_WRONGSEC + * NFSv4.1: the server does not support the SECINFO_NO_NAME operation + * + * Returns zero on success, or a negative NFS4ERR value, or a + * negative errno value. + */ +int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + /* Per 3530bis 15.33.5 */ + static const rpc_authflavor_t flav_array[] = { + RPC_AUTH_GSS_KRB5P, + RPC_AUTH_GSS_KRB5I, + RPC_AUTH_GSS_KRB5, + RPC_AUTH_UNIX, /* courtesy */ + RPC_AUTH_NULL, + }; + int status = -EPERM; + size_t i; + + if (server->auth_info.flavor_len > 0) { + /* try each flavor specified by user */ + for (i = 0; i < server->auth_info.flavor_len; i++) { + status = nfs4_lookup_root_sec( + server, fhandle, fattr, + server->auth_info.flavors[i]); + if (status == -NFS4ERR_WRONGSEC || status == -EACCES) + continue; + break; + } + } else { + /* no flavors specified by user, try default list */ + for (i = 0; i < ARRAY_SIZE(flav_array); i++) { + status = nfs4_lookup_root_sec(server, fhandle, fattr, + flav_array[i]); + if (status == -NFS4ERR_WRONGSEC || status == -EACCES) + continue; + break; + } + } + + /* + * -EACCES could mean that the user doesn't have correct permissions + * to access the mount. It could also mean that we tried to mount + * with a gss auth flavor, but rpc.gssd isn't running. Either way, + * existing mount programs don't handle -EACCES very well so it should + * be mapped to -EPERM instead. + */ + if (status == -EACCES) + status = -EPERM; + return status; +} + +/** + * nfs4_proc_get_rootfh - get file handle for server's pseudoroot + * @server: initialized nfs_server handle + * @fhandle: we fill in the pseudo-fs root file handle + * @fattr: we fill in a bare bones struct fattr + * @auth_probe: probe the auth flavours + * + * Returns zero on success, or a negative errno. + */ +int nfs4_proc_get_rootfh(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, bool auth_probe) +{ + int status = 0; + + if (!auth_probe) + status = nfs4_lookup_root(server, fhandle, fattr); + + if (auth_probe || status == NFS4ERR_WRONGSEC) + status = server->nfs_client->cl_mvops->find_root_sec( + server, fhandle, fattr); + + return nfs4_map_errors(status); +} + +static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, + struct nfs_fsinfo *info) +{ + int error; + struct nfs_fattr *fattr = info->fattr; + + error = nfs4_server_capabilities(server, mntfh); + if (error < 0) { + dprintk("nfs4_get_root: getcaps error = %d\n", -error); + return error; + } + + error = nfs4_proc_getattr(server, mntfh, fattr, NULL); + if (error < 0) { + dprintk("nfs4_get_root: getattr error = %d\n", -error); + goto out; + } + + if (fattr->valid & NFS_ATTR_FATTR_FSID && + !nfs_fsid_equal(&server->fsid, &fattr->fsid)) + memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid)); + +out: + return error; +} + +/* + * Get locations and (maybe) other attributes of a referral. + * Note that we'll actually follow the referral later when + * we detect fsid mismatch in inode revalidation + */ +static int nfs4_get_referral(struct rpc_clnt *client, struct inode *dir, + const struct qstr *name, struct nfs_fattr *fattr, + struct nfs_fh *fhandle) +{ + int status = -ENOMEM; + struct page *page = NULL; + struct nfs4_fs_locations *locations = NULL; + + page = alloc_page(GFP_KERNEL); + if (page == NULL) + goto out; + locations = kmalloc_obj(struct nfs4_fs_locations); + if (locations == NULL) + goto out; + + locations->fattr = fattr; + + status = nfs4_proc_fs_locations(client, dir, name, locations, page); + if (status != 0) + goto out; + + /* + * If the fsid didn't change, this is a migration event, not a + * referral. Cause us to drop into the exception handler, which + * will kick off migration recovery. + */ + if (nfs_fsid_equal(&NFS_SERVER(dir)->fsid, &fattr->fsid)) { + dprintk("%s: server did not return a different fsid for" + " a referral at %s\n", __func__, name->name); + status = -NFS4ERR_MOVED; + goto out; + } + /* Fixup attributes for the nfs_lookup() call to nfs_fhget() */ + nfs_fixup_referral_attributes(fattr); + memset(fhandle, 0, sizeof(struct nfs_fh)); +out: + if (page) + __free_page(page); + kfree(locations); + return status; +} + +static bool should_request_dir_deleg(struct inode *inode) +{ + if (!directory_delegations) + return false; + if (!inode) + return false; + if (!S_ISDIR(inode->i_mode)) + return false; + if (!nfs_server_capable(inode, NFS_CAP_DIR_DELEG)) + return false; + if (!test_and_clear_bit(NFS_INO_REQ_DIR_DELEG, &(NFS_I(inode)->flags))) + return false; + if (nfs4_have_delegation(inode, FMODE_READ, 0)) + return false; + return true; +} + +static void nfs4_call_getattr_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + nfs4_setup_sequence(data->seq_server->nfs_client, data->seq_args, + data->seq_res, task); +} + +static void nfs4_call_getattr_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_call_sync_data *data = calldata; + + nfs4_sequence_process(task, data->seq_res); +} + +static const struct rpc_call_ops nfs4_call_getattr_ops = { + .rpc_call_prepare = nfs4_call_getattr_prepare, + .rpc_call_done = nfs4_call_getattr_done, +}; + +static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode) +{ + __u32 bitmask[NFS4_BITMASK_SZ]; + struct nfs4_getattr_arg args = { + .fh = fhandle, + .bitmask = bitmask, + }; + struct nfs4_getattr_res res = { + .fattr = fattr, + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETATTR], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct nfs4_call_sync_data data = { + .seq_server = server, + .seq_args = &args.seq_args, + .seq_res = &res.seq_res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_call_getattr_ops, + .callback_data = &data, + }; + struct nfs_client *clp = server->nfs_client; + struct nfs4_gdd_res gdd_res; + int status; + + if (nfs4_has_session(clp)) + task_setup.flags = RPC_TASK_MOVEABLE; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (inode && (server->flags & NFS_MOUNT_SOFTREVAL)) + task_setup.flags |= RPC_TASK_TIMEOUT; + + args.get_dir_deleg = should_request_dir_deleg(inode); + if (args.get_dir_deleg) + res.gdd_res = &gdd_res; + + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, fattr->label), inode, 0); + nfs_fattr_init(fattr); + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 0); + + status = nfs4_call_sync_custom(&task_setup); + + if (args.get_dir_deleg) { + switch (status) { + case 0: + if (gdd_res.status != GDD4_OK) + break; + nfs_inode_set_delegation(inode, current_cred(), + FMODE_READ, &gdd_res.deleg, 0, + NFS4_OPEN_DELEGATE_READ); + break; + case -ENOTSUPP: + case -EOPNOTSUPP: + server->caps &= ~NFS_CAP_DIR_DELEG; + break; + case -NFS4ERR_INVAL: + case -NFS4ERR_IO: + case -NFS4ERR_DIRDELEG_UNAVAIL: + case -NFS4ERR_NOTDIR: + clear_bit(NFS_INO_REQ_DIR_DELEG, &(NFS_I(inode)->flags)); + status = -EAGAIN; + } + } + + nfs4_sequence_free_slot(&res.seq_res); + return status; +} + +int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_getattr(server, fhandle, fattr, inode); + trace_nfs4_getattr(server, fhandle, fattr, err); + switch (err) { + default: + err = nfs4_handle_exception(server, err, &exception); + break; + case -EAGAIN: + case -ENOTSUPP: + case -EOPNOTSUPP: + exception.retry = true; + } + } while (exception.retry); + return err; +} + +/* + * The file is not closed if it is opened due to the a request to change + * the size of the file. The open call will not be needed once the + * VFS layer lookup-intents are implemented. + * + * Close is called when the inode is destroyed. + * If we haven't opened the file for O_WRONLY, we + * need to in the size_change case to obtain a stateid. + * + * Got race? + * Because OPEN is always done by name in nfsv4, it is + * possible that we opened a different file by the same + * name. We can recognize this race condition, but we + * can't do anything about it besides returning an error. + * + * This will be fixed with VFS changes (lookup-intent). + */ +static int +nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, + struct iattr *sattr) +{ + struct inode *inode = d_inode(dentry); + const struct cred *cred = NULL; + struct nfs_open_context *ctx = NULL; + int status; + + if (pnfs_ld_layoutret_on_setattr(inode) && + sattr->ia_valid & ATTR_SIZE && + sattr->ia_size < i_size_read(inode)) + pnfs_commit_and_return_layout(inode); + + nfs_fattr_init(fattr); + + /* Deal with open(O_TRUNC) */ + if (sattr->ia_valid & ATTR_OPEN) + sattr->ia_valid &= ~(ATTR_MTIME|ATTR_CTIME); + + /* Optimization: if the end result is no change, don't RPC */ + if ((sattr->ia_valid & ~(ATTR_FILE|ATTR_OPEN)) == 0) + return 0; + + /* Search for an existing open(O_WRITE) file */ + if (sattr->ia_valid & ATTR_FILE) { + + ctx = nfs_file_open_context(sattr->ia_file); + if (ctx) + cred = ctx->cred; + } + + /* Return any delegations if we're going to change ACLs */ + if ((sattr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) + nfs4_inode_make_writeable(inode); + + status = nfs4_do_setattr(inode, cred, fattr, sattr, ctx, NULL); + if (status == 0) { + nfs_setattr_update_inode(inode, sattr, fattr); + nfs_setsecurity(inode, fattr); + } + return status; +} + +static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, + struct dentry *dentry, const struct qstr *name, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + struct nfs_server *server = NFS_SERVER(dir); + int status; + struct nfs4_lookup_arg args = { + .bitmask = server->attr_bitmask, + .dir_fh = NFS_FH(dir), + .name = name, + }; + struct nfs4_lookup_res res = { + .server = server, + .fattr = fattr, + .fh = fhandle, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP], + .rpc_argp = &args, + .rpc_resp = &res, + }; + unsigned short task_flags = 0; + + if (nfs_server_capable(dir, NFS_CAP_MOVEABLE)) + task_flags = RPC_TASK_MOVEABLE; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (nfs_lookup_is_soft_revalidate(dentry)) + task_flags |= RPC_TASK_TIMEOUT; + + args.bitmask = nfs4_bitmask(server, fattr->label); + + nfs_fattr_init(fattr); + + dprintk("NFS call lookup %pd2\n", dentry); + nfs4_init_sequence(server->nfs_client, &args.seq_args, &res.seq_res, 0, 0); + status = nfs4_do_call_sync(clnt, server, &msg, + &args.seq_args, &res.seq_res, task_flags); + dprintk("NFS reply lookup: %d\n", status); + return status; +} + +static void nfs_fixup_secinfo_attributes(struct nfs_fattr *fattr) +{ + fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE | + NFS_ATTR_FATTR_NLINK | NFS_ATTR_FATTR_MOUNTPOINT; + fattr->mode = S_IFDIR | S_IRUGO | S_IXUGO; + fattr->nlink = 2; +} + +static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, + struct dentry *dentry, const struct qstr *name, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + struct rpc_clnt *client = *clnt; + int err; + do { + err = _nfs4_proc_lookup(client, dir, dentry, name, fhandle, fattr); + trace_nfs4_lookup(dir, name, err); + switch (err) { + case -NFS4ERR_BADNAME: + err = -ENOENT; + goto out; + case -NFS4ERR_MOVED: + err = nfs4_get_referral(client, dir, name, fattr, fhandle); + if (err == -NFS4ERR_MOVED) + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); + goto out; + case -NFS4ERR_WRONGSEC: + err = -EPERM; + if (client != *clnt) + goto out; + client = nfs4_negotiate_security(client, dir, name); + if (IS_ERR(client)) + return PTR_ERR(client); + + exception.retry = 1; + break; + default: + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); + } + } while (exception.retry); + +out: + if (err == 0) + *clnt = client; + else if (client != *clnt) + rpc_shutdown_client(client); + + return err; +} + +static int nfs4_proc_lookup(struct inode *dir, struct dentry *dentry, const struct qstr *name, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + int status; + struct rpc_clnt *client = NFS_CLIENT(dir); + + status = nfs4_proc_lookup_common(&client, dir, dentry, name, fhandle, fattr); + if (client != NFS_CLIENT(dir)) { + rpc_shutdown_client(client); + nfs_fixup_secinfo_attributes(fattr); + } + return status; +} + +struct rpc_clnt * +nfs4_proc_lookup_mountpoint(struct inode *dir, struct dentry *dentry, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + struct rpc_clnt *client = NFS_CLIENT(dir); + int status; + + status = nfs4_proc_lookup_common(&client, dir, dentry, &dentry->d_name, + fhandle, fattr); + if (status < 0) + return ERR_PTR(status); + return (client == NFS_CLIENT(dir)) ? rpc_clone_client(client) : client; +} + +static int _nfs4_proc_lookupp(struct inode *inode, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + struct rpc_clnt *clnt = NFS_CLIENT(inode); + struct nfs_server *server = NFS_SERVER(inode); + int status; + struct nfs4_lookupp_arg args = { + .bitmask = server->attr_bitmask, + .fh = NFS_FH(inode), + }; + struct nfs4_lookupp_res res = { + .server = server, + .fattr = fattr, + .fh = fhandle, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUPP], + .rpc_argp = &args, + .rpc_resp = &res, + }; + unsigned short task_flags = 0; + + if (server->flags & NFS_MOUNT_SOFTREVAL) + task_flags |= RPC_TASK_TIMEOUT; + if (server->caps & NFS_CAP_MOVEABLE) + task_flags |= RPC_TASK_MOVEABLE; + + args.bitmask = nfs4_bitmask(server, fattr->label); + + nfs_fattr_init(fattr); + nfs4_init_sequence(server->nfs_client, &args.seq_args, &res.seq_res, 0, 0); + + dprintk("NFS call lookupp ino=0x%llx\n", inode->i_ino); + status = nfs4_do_call_sync(clnt, server, &msg, &args.seq_args, + &res.seq_res, task_flags); + dprintk("NFS reply lookupp: %d\n", status); + return status; +} + +static int nfs4_proc_lookupp(struct inode *inode, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_lookupp(inode, fhandle, fattr); + trace_nfs4_lookupp(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry, + const struct cred *cred) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs4_accessargs args = { + .fh = NFS_FH(inode), + .access = entry->mask, + }; + struct nfs4_accessres res = { + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ACCESS], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + int status = 0; + + if (!nfs4_have_delegation(inode, FMODE_READ, 0)) { + nfs_request_directory_delegation(inode); + res.fattr = nfs_alloc_fattr(); + if (res.fattr == NULL) + return -ENOMEM; + args.bitmask = server->cache_consistency_bitmask; + } + status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); + if (!status) { + nfs_access_set_mask(entry, res.access); + if (res.fattr) + nfs_refresh_inode(inode, res.fattr); + } + nfs_free_fattr(res.fattr); + return status; +} + +static int nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry, + const struct cred *cred) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_access(inode, entry, cred); + trace_nfs4_access(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +/* + * TODO: For the time being, we don't try to get any attributes + * along with any of the zero-copy operations READ, READDIR, + * READLINK, WRITE. + * + * In the case of the first three, we want to put the GETATTR + * after the read-type operation -- this is because it is hard + * to predict the length of a GETATTR response in v4, and thus + * align the READ data correctly. This means that the GETATTR + * may end up partially falling into the page cache, and we should + * shift it into the 'tail' of the xdr_buf before processing. + * To do this efficiently, we need to know the total length + * of data received, which doesn't seem to be available outside + * of the RPC layer. + * + * In the case of WRITE, we also want to put the GETATTR after + * the operation -- in this case because we want to make sure + * we get the post-operation mtime and size. + * + * Both of these changes to the XDR layer would in fact be quite + * minor, but I decided to leave them for a subsequent patch. + */ +static int _nfs4_proc_readlink(struct inode *inode, struct page *page, + unsigned int pgbase, unsigned int pglen) +{ + struct nfs4_readlink args = { + .fh = NFS_FH(inode), + .pgbase = pgbase, + .pglen = pglen, + .pages = &page, + }; + struct nfs4_readlink_res res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READLINK], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + return nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), &msg, &args.seq_args, &res.seq_res, 0); +} + +static int nfs4_proc_readlink(struct inode *inode, struct page *page, + unsigned int pgbase, unsigned int pglen) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_readlink(inode, page, pgbase, pglen); + trace_nfs4_readlink(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +/* + * This is just for mknod. open(O_CREAT) will always do ->open_context(). + */ +static int +nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, + int flags) +{ + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_label l, *ilabel; + struct nfs_open_context *ctx; + struct nfs4_state *state; + int status = 0; + + ctx = alloc_nfs_open_context(dentry, FMODE_READ, NULL); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + ilabel = nfs4_label_init_security(dir, dentry, sattr, &l); + + nfs_request_directory_delegation(dir); + + if (!(server->attr_bitmask[2] & FATTR4_WORD2_MODE_UMASK)) + sattr->ia_mode &= ~current_umask(); + state = nfs4_do_open(dir, ctx, flags, sattr, ilabel, NULL); + if (IS_ERR(state)) { + status = PTR_ERR(state); + goto out; + } +out: + nfs4_label_release_security(ilabel); + put_nfs_open_context(ctx); + return status; +} + +static int +_nfs4_proc_remove(struct inode *dir, const struct qstr *name, u32 ftype) +{ + struct nfs_server *server = NFS_SERVER(dir); + struct nfs_removeargs args = { + .fh = NFS_FH(dir), + .name = *name, + }; + struct nfs_removeres res = { + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE], + .rpc_argp = &args, + .rpc_resp = &res, + }; + unsigned long timestamp = jiffies; + int status; + + status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 1); + if (status == 0) { + spin_lock(&dir->i_lock); + /* Removing a directory decrements nlink in the parent */ + if (ftype == NF4DIR && dir->i_nlink > 2) + nfs4_dec_nlink_locked(dir); + nfs4_update_changeattr_locked(dir, &res.cinfo, timestamp, + NFS_INO_INVALID_DATA); + spin_unlock(&dir->i_lock); + } + return status; +} + +static int nfs4_proc_remove(struct inode *dir, struct dentry *dentry) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + struct inode *inode = d_inode(dentry); + int err; + + if (inode) { + if (inode->i_nlink == 1) + nfs4_inode_return_delegation(inode); + else + nfs4_inode_make_writeable(inode); + } + do { + err = _nfs4_proc_remove(dir, &dentry->d_name, NF4REG); + trace_nfs4_remove(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + return err; +} + +static int nfs4_proc_rmdir(struct inode *dir, const struct qstr *name) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + do { + err = _nfs4_proc_remove(dir, name, NF4DIR); + trace_nfs4_remove(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + return err; +} + +static void nfs4_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) +{ + struct nfs_removeargs *args = msg->rpc_argp; + struct nfs_removeres *res = msg->rpc_resp; + struct nfs_server *server = NFS_SB(dentry->d_sb); + + res->server = server; + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; + nfs4_init_sequence(server->nfs_client, &args->seq_args, + &res->seq_res, 1, 0); + + nfs_fattr_init(res->dir_attr); + nfs_request_directory_delegation(d_inode(dentry->d_parent)); + + if (inode) { + nfs4_inode_return_delegation(inode); + nfs_d_prune_case_insensitive_aliases(inode); + } +} + +static void nfs4_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data) +{ + nfs4_setup_sequence(NFS_SB(data->dentry->d_sb)->nfs_client, + &data->args.seq_args, + &data->res.seq_res, + task); +} + +static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) +{ + struct nfs_unlinkdata *data = task->tk_calldata; + struct nfs_removeres *res = &data->res; + + if (!nfs4_sequence_done(task, &res->seq_res)) + return 0; + if (nfs4_async_handle_error(task, res->server, NULL, + &data->timeout) == -EAGAIN) + return 0; + if (task->tk_status == 0) + nfs4_update_changeattr(dir, &res->cinfo, + res->dir_attr->time_start, + NFS_INO_INVALID_DATA); + return 1; +} + +static void nfs4_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry, + struct inode *same_parent) +{ + struct nfs_server *server = NFS_SB(old_dentry->d_sb); + struct nfs_renameargs *arg = msg->rpc_argp; + struct nfs_renameres *res = msg->rpc_resp; + struct inode *old_inode = d_inode(old_dentry); + struct inode *new_inode = d_inode(new_dentry); + + if (old_inode) + nfs4_inode_make_writeable(old_inode); + if (new_inode) + nfs4_inode_return_delegation(new_inode); + if (same_parent) + nfs_request_directory_delegation(same_parent); + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME]; + res->server = server; + nfs4_init_sequence(server->nfs_client, &arg->seq_args, + &res->seq_res, 1, 0); +} + +static void nfs4_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data) +{ + nfs4_setup_sequence(NFS_SERVER(data->old_dir)->nfs_client, + &data->args.seq_args, + &data->res.seq_res, + task); +} + +static int nfs4_proc_rename_done(struct rpc_task *task, struct inode *old_dir, + struct inode *new_dir) +{ + struct nfs_renamedata *data = task->tk_calldata; + struct nfs_renameres *res = &data->res; + + if (!nfs4_sequence_done(task, &res->seq_res)) + return 0; + if (nfs4_async_handle_error(task, res->server, NULL, &data->timeout) == -EAGAIN) + return 0; + + if (task->tk_status == 0) { + nfs_d_prune_case_insensitive_aliases(d_inode(data->old_dentry)); + if (new_dir != old_dir) { + /* Note: If we moved a directory, nlink will change */ + nfs4_update_changeattr(old_dir, &res->old_cinfo, + res->old_fattr->time_start, + NFS_INO_INVALID_NLINK | + NFS_INO_INVALID_DATA); + nfs4_update_changeattr(new_dir, &res->new_cinfo, + res->new_fattr->time_start, + NFS_INO_INVALID_NLINK | + NFS_INO_INVALID_DATA); + nfs_update_delegated_mtime(new_dir); + } else + nfs4_update_changeattr(old_dir, &res->old_cinfo, + res->old_fattr->time_start, + NFS_INO_INVALID_DATA); + } + return 1; +} + +static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) +{ + struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS4_BITMASK_SZ]; + struct nfs4_link_arg arg = { + .fh = NFS_FH(inode), + .dir_fh = NFS_FH(dir), + .name = name, + .bitmask = bitmask, + }; + struct nfs4_link_res res = { + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status = -ENOMEM; + + res.fattr = nfs_alloc_fattr_with_label(server); + if (res.fattr == NULL) + goto out; + + nfs4_inode_make_writeable(inode); + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, res.fattr->label), + inode, + NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME); + status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); + if (!status) { + nfs4_update_changeattr(dir, &res.cinfo, res.fattr->time_start, + NFS_INO_INVALID_DATA); + nfs4_inc_nlink(inode); + status = nfs_post_op_update_inode(inode, res.fattr); + if (!status) + nfs_setsecurity(inode, res.fattr); + } + +out: + nfs_free_fattr(res.fattr); + return status; +} + +static int nfs4_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), + _nfs4_proc_link(inode, dir, name), + &exception); + } while (exception.retry); + return err; +} + +struct nfs4_createdata { + struct rpc_message msg; + struct nfs4_create_arg arg; + struct nfs4_create_res res; + struct nfs_fh fh; + struct nfs_fattr fattr; +}; + +static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir, + const struct qstr *name, struct iattr *sattr, u32 ftype) +{ + struct nfs4_createdata *data; + + data = kzalloc_obj(*data); + if (data != NULL) { + struct nfs_server *server = NFS_SERVER(dir); + + data->fattr.label = nfs4_label_alloc(server, GFP_KERNEL); + if (IS_ERR(data->fattr.label)) + goto out_free; + + data->msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE]; + data->msg.rpc_argp = &data->arg; + data->msg.rpc_resp = &data->res; + data->arg.dir_fh = NFS_FH(dir); + data->arg.server = server; + data->arg.name = name; + data->arg.attrs = sattr; + data->arg.ftype = ftype; + data->arg.bitmask = nfs4_bitmask(server, data->fattr.label); + data->arg.umask = current_umask(); + data->res.server = server; + data->res.fh = &data->fh; + data->res.fattr = &data->fattr; + nfs_fattr_init(data->res.fattr); + } + return data; +out_free: + kfree(data); + return NULL; +} + +static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_createdata *data) +{ + int status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &data->msg, + &data->arg.seq_args, &data->res.seq_res, 1); + if (status == 0) { + spin_lock(&dir->i_lock); + nfs4_update_changeattr_locked(dir, &data->res.dir_cinfo, + data->res.fattr->time_start, + NFS_INO_INVALID_DATA); + spin_unlock(&dir->i_lock); + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); + } + return status; +} + +static struct dentry *nfs4_do_mkdir(struct inode *dir, struct dentry *dentry, + struct nfs4_createdata *data, int *statusp) +{ + struct dentry *ret; + + *statusp = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &data->msg, + &data->arg.seq_args, &data->res.seq_res, 1); + + if (*statusp) + return NULL; + + spin_lock(&dir->i_lock); + /* Creating a directory bumps nlink in the parent */ + nfs4_inc_nlink_locked(dir); + nfs4_update_changeattr_locked(dir, &data->res.dir_cinfo, + data->res.fattr->time_start, + NFS_INO_INVALID_DATA); + spin_unlock(&dir->i_lock); + ret = nfs_add_or_obtain(dentry, data->res.fh, data->res.fattr); + if (!IS_ERR(ret)) + return ret; + *statusp = PTR_ERR(ret); + return NULL; +} + +static void nfs4_free_createdata(struct nfs4_createdata *data) +{ + nfs4_label_free(data->fattr.label); + kfree(data); +} + +static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, + struct folio *folio, unsigned int len, struct iattr *sattr, + struct nfs4_label *label) +{ + struct page *page = &folio->page; + struct nfs4_createdata *data; + int status = -ENAMETOOLONG; + + if (len > NFS4_MAXPATHLEN) + goto out; + + status = -ENOMEM; + data = nfs4_alloc_createdata(dir, &dentry->d_name, sattr, NF4LNK); + if (data == NULL) + goto out; + + data->msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SYMLINK]; + data->arg.u.symlink.pages = &page; + data->arg.u.symlink.len = len; + data->arg.label = label; + + status = nfs4_do_create(dir, dentry, data); + + nfs4_free_createdata(data); +out: + return status; +} + +static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, + struct folio *folio, unsigned int len, struct iattr *sattr) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + struct nfs4_label l, *label; + int err; + + label = nfs4_label_init_security(dir, dentry, sattr, &l); + + do { + err = _nfs4_proc_symlink(dir, dentry, folio, len, sattr, label); + trace_nfs4_symlink(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + + nfs4_label_release_security(label); + return err; +} + +static struct dentry *_nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, + struct nfs4_label *label, int *statusp) +{ + struct nfs4_createdata *data; + struct dentry *ret = NULL; + + *statusp = -ENOMEM; + data = nfs4_alloc_createdata(dir, &dentry->d_name, sattr, NF4DIR); + if (data == NULL) + goto out; + + data->arg.label = label; + ret = nfs4_do_mkdir(dir, dentry, data, statusp); + + nfs4_free_createdata(data); +out: + return ret; +} + +static struct dentry *nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, + struct iattr *sattr) +{ + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_exception exception = { + .interruptible = true, + }; + struct nfs4_label l, *label; + struct dentry *alias; + int err; + + label = nfs4_label_init_security(dir, dentry, sattr, &l); + + if (!(server->attr_bitmask[2] & FATTR4_WORD2_MODE_UMASK)) + sattr->ia_mode &= ~current_umask(); + do { + alias = _nfs4_proc_mkdir(dir, dentry, sattr, label, &err); + trace_nfs4_mkdir(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, &exception); + if (err) + alias = ERR_PTR(err); + } while (exception.retry); + nfs4_label_release_security(label); + + return alias; +} + +static int _nfs4_proc_readdir(struct nfs_readdir_arg *nr_arg, + struct nfs_readdir_res *nr_res) +{ + struct inode *dir = d_inode(nr_arg->dentry); + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_readdir_arg args = { + .fh = NFS_FH(dir), + .pages = nr_arg->pages, + .pgbase = 0, + .count = nr_arg->page_len, + .plus = nr_arg->plus, + }; + struct nfs4_readdir_res res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READDIR], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = nr_arg->cred, + }; + int status; + + dprintk("%s: dentry = %pd2, cookie = %llu\n", __func__, + nr_arg->dentry, (unsigned long long)nr_arg->cookie); + if (!(server->caps & NFS_CAP_SECURITY_LABEL)) + args.bitmask = server->attr_bitmask_nl; + else + args.bitmask = server->attr_bitmask; + + nfs4_setup_readdir(nr_arg->cookie, nr_arg->verf, nr_arg->dentry, &args); + res.pgbase = args.pgbase; + status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, + &res.seq_res, 0); + if (status >= 0) { + memcpy(nr_res->verf, res.verifier.data, NFS4_VERIFIER_SIZE); + status += args.pgbase; + } + + nfs_invalidate_atime(dir); + + dprintk("%s: returns %d\n", __func__, status); + return status; +} + +static int nfs4_proc_readdir(struct nfs_readdir_arg *arg, + struct nfs_readdir_res *res) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_readdir(arg, res); + trace_nfs4_readdir(d_inode(arg->dentry), err); + err = nfs4_handle_exception(NFS_SERVER(d_inode(arg->dentry)), + err, &exception); + } while (exception.retry); + return err; +} + +static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, struct nfs4_label *label, dev_t rdev) +{ + struct nfs4_createdata *data; + int mode = sattr->ia_mode; + int status = -ENOMEM; + + data = nfs4_alloc_createdata(dir, &dentry->d_name, sattr, NF4SOCK); + if (data == NULL) + goto out; + + if (S_ISFIFO(mode)) + data->arg.ftype = NF4FIFO; + else if (S_ISBLK(mode)) { + data->arg.ftype = NF4BLK; + data->arg.u.device.specdata1 = MAJOR(rdev); + data->arg.u.device.specdata2 = MINOR(rdev); + } + else if (S_ISCHR(mode)) { + data->arg.ftype = NF4CHR; + data->arg.u.device.specdata1 = MAJOR(rdev); + data->arg.u.device.specdata2 = MINOR(rdev); + } else if (!S_ISSOCK(mode)) { + status = -EINVAL; + goto out_free; + } + + data->arg.label = label; + status = nfs4_do_create(dir, dentry, data); +out_free: + nfs4_free_createdata(data); +out: + return status; +} + +static int nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, dev_t rdev) +{ + struct nfs_server *server = NFS_SERVER(dir); + struct nfs4_exception exception = { + .interruptible = true, + }; + struct nfs4_label l, *label; + int err; + + label = nfs4_label_init_security(dir, dentry, sattr, &l); + + if (!(server->attr_bitmask[2] & FATTR4_WORD2_MODE_UMASK)) + sattr->ia_mode &= ~current_umask(); + do { + err = _nfs4_proc_mknod(dir, dentry, sattr, label, rdev); + trace_nfs4_mknod(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + + nfs4_label_release_security(label); + + return err; +} + +static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsstat *fsstat) +{ + struct nfs4_statfs_arg args = { + .fh = fhandle, + .bitmask = server->attr_bitmask, + }; + struct nfs4_statfs_res res = { + .fsstat = fsstat, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_STATFS], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + nfs_fattr_init(fsstat->fattr); + return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); +} + +static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = nfs4_handle_exception(server, + _nfs4_proc_statfs(server, fhandle, fsstat), + &exception); + } while (exception.retry); + return err; +} + +static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *fsinfo) +{ + struct nfs4_fsinfo_arg args = { + .fh = fhandle, + .bitmask = server->attr_bitmask, + }; + struct nfs4_fsinfo_res res = { + .fsinfo = fsinfo, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSINFO], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); +} + +static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + do { + err = _nfs4_do_fsinfo(server, fhandle, fsinfo); + trace_nfs4_fsinfo(server, fhandle, fsinfo->fattr, err); + if (err == 0) { + nfs4_set_lease_period(server->nfs_client, fsinfo->lease_time); + break; + } + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + return err; +} + +static int nfs4_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) +{ + int error; + + nfs_fattr_init(fsinfo->fattr); + error = nfs4_do_fsinfo(server, fhandle, fsinfo); + if (error == 0) { + /* block layout checks this! */ + server->pnfs_blksize = fsinfo->blksize; + set_pnfs_layoutdriver(server, fhandle, fsinfo); + } + + return error; +} + +static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_pathconf *pathconf) +{ + struct nfs4_pathconf_arg args = { + .fh = fhandle, + .bitmask = server->attr_bitmask, + }; + struct nfs4_pathconf_res res = { + .pathconf = pathconf, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_PATHCONF], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + /* None of the pathconf attributes are mandatory to implement */ + if ((args.bitmask[0] & nfs4_pathconf_bitmap[0]) == 0) { + memset(pathconf, 0, sizeof(*pathconf)); + return 0; + } + + nfs_fattr_init(pathconf->fattr); + return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); +} + +static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_pathconf *pathconf) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + do { + err = nfs4_handle_exception(server, + _nfs4_proc_pathconf(server, fhandle, pathconf), + &exception); + } while (exception.retry); + return err; +} + +int nfs4_set_rw_stateid(nfs4_stateid *stateid, + const struct nfs_open_context *ctx, + const struct nfs_lock_context *l_ctx, + fmode_t fmode) +{ + return nfs4_select_rw_stateid(ctx->state, fmode, l_ctx, stateid, NULL); +} +EXPORT_SYMBOL_GPL(nfs4_set_rw_stateid); + +static bool nfs4_stateid_is_current(nfs4_stateid *stateid, + const struct nfs_open_context *ctx, + const struct nfs_lock_context *l_ctx, + fmode_t fmode) +{ + nfs4_stateid _current_stateid; + + /* If the current stateid represents a lost lock, then exit */ + if (nfs4_set_rw_stateid(&_current_stateid, ctx, l_ctx, fmode) == -EIO) + return true; + return nfs4_stateid_match(stateid, &_current_stateid); +} + +static bool nfs4_error_stateid_expired(int err) +{ + switch (err) { + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_OLD_STATEID: + case -NFS4ERR_OPENMODE: + case -NFS4ERR_EXPIRED: + return true; + } + return false; +} + +static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + struct nfs_server *server = NFS_SERVER(hdr->inode); + + trace_nfs4_read(hdr, task->tk_status); + if (task->tk_status < 0) { + struct nfs4_exception exception = { + .inode = hdr->inode, + .state = hdr->args.context->state, + .stateid = &hdr->args.stateid, + .retrans = hdr->retrans, + }; + task->tk_status = nfs4_async_handle_exception(task, + server, task->tk_status, &exception); + hdr->retrans = exception.retrans; + if (exception.retry) { + rpc_restart_call_prepare(task); + return -EAGAIN; + } + } + + if (task->tk_status > 0) + renew_lease(server, hdr->timestamp); + return 0; +} + +static bool nfs4_read_stateid_changed(struct rpc_task *task, + struct nfs_pgio_args *args) +{ + + if (!nfs4_error_stateid_expired(task->tk_status) || + nfs4_stateid_is_current(&args->stateid, + args->context, + args->lock_context, + FMODE_READ)) + return false; + rpc_restart_call_prepare(task); + return true; +} + +static bool nfs4_read_plus_not_supported(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs_server *server = NFS_SERVER(hdr->inode); + struct rpc_message *msg = &task->tk_msg; + + if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS] && + task->tk_status == -ENOTSUPP) { + server->caps &= ~NFS_CAP_READ_PLUS; + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; + rpc_restart_call_prepare(task); + return true; + } + return false; +} + +static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + if (!nfs4_sequence_done(task, &hdr->res.seq_res)) + return -EAGAIN; + if (nfs4_read_stateid_changed(task, &hdr->args)) + return -EAGAIN; + if (nfs4_read_plus_not_supported(task, hdr)) + return -EAGAIN; + if (task->tk_status > 0) + nfs_invalidate_atime(hdr->inode); + return hdr->pgio_done_cb ? hdr->pgio_done_cb(task, hdr) : + nfs4_read_done_cb(task, hdr); +} + +#if defined CONFIG_NFS_V4_2 && defined CONFIG_NFS_V4_2_READ_PLUS +static bool nfs42_read_plus_support(struct nfs_pgio_header *hdr, + struct rpc_message *msg) +{ + /* Note: We don't use READ_PLUS with pNFS yet */ + if (nfs_server_capable(hdr->inode, NFS_CAP_READ_PLUS) && !hdr->ds_clp) { + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS]; + return nfs_read_alloc_scratch(hdr, READ_PLUS_SCRATCH_SIZE); + } + return false; +} +#else +static bool nfs42_read_plus_support(struct nfs_pgio_header *hdr, + struct rpc_message *msg) +{ + return false; +} +#endif /* CONFIG_NFS_V4_2 */ + +static void nfs4_proc_read_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg) +{ + hdr->timestamp = jiffies; + if (!hdr->pgio_done_cb) + hdr->pgio_done_cb = nfs4_read_done_cb; + if (!nfs42_read_plus_support(hdr, msg)) + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; + nfs4_init_sequence(NFS_SERVER(hdr->inode)->nfs_client, + &hdr->args.seq_args, &hdr->res.seq_res, 0, 0); +} + +static int nfs4_proc_pgio_rpc_prepare(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + if (nfs4_setup_sequence(NFS_SERVER(hdr->inode)->nfs_client, + &hdr->args.seq_args, + &hdr->res.seq_res, + task)) + return 0; + if (nfs4_set_rw_stateid(&hdr->args.stateid, hdr->args.context, + hdr->args.lock_context, + hdr->rw_mode) == -EIO) + return -EIO; + if (unlikely(test_bit(NFS_CONTEXT_BAD, &hdr->args.context->flags))) + return -EIO; + return 0; +} + +static int nfs4_write_done_cb(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct inode *inode = hdr->inode; + + trace_nfs4_write(hdr, task->tk_status); + if (task->tk_status < 0) { + struct nfs4_exception exception = { + .inode = hdr->inode, + .state = hdr->args.context->state, + .stateid = &hdr->args.stateid, + .retrans = hdr->retrans, + }; + task->tk_status = nfs4_async_handle_exception(task, + NFS_SERVER(inode), task->tk_status, + &exception); + hdr->retrans = exception.retrans; + if (exception.retry) { + rpc_restart_call_prepare(task); + return -EAGAIN; + } + } + if (task->tk_status >= 0) { + renew_lease(NFS_SERVER(inode), hdr->timestamp); + nfs_writeback_update_inode(hdr); + } + return 0; +} + +static bool nfs4_write_stateid_changed(struct rpc_task *task, + struct nfs_pgio_args *args) +{ + + if (!nfs4_error_stateid_expired(task->tk_status) || + nfs4_stateid_is_current(&args->stateid, + args->context, + args->lock_context, + FMODE_WRITE)) + return false; + rpc_restart_call_prepare(task); + return true; +} + +static int nfs4_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + if (!nfs4_sequence_done(task, &hdr->res.seq_res)) + return -EAGAIN; + if (nfs4_write_stateid_changed(task, &hdr->args)) + return -EAGAIN; + return hdr->pgio_done_cb ? hdr->pgio_done_cb(task, hdr) : + nfs4_write_done_cb(task, hdr); +} + +static +bool nfs4_write_need_cache_consistency_data(struct nfs_pgio_header *hdr) +{ + /* Don't request attributes for pNFS or O_DIRECT writes */ + if (hdr->ds_clp != NULL || hdr->dreq != NULL) + return false; + /* Otherwise, request attributes if and only if we don't hold + * a delegation + */ + return nfs4_have_delegation(hdr->inode, FMODE_READ, 0) == 0; +} + +void nfs4_bitmask_set(__u32 bitmask[], const __u32 src[], + struct inode *inode, unsigned long cache_validity) +{ + struct nfs_server *server = NFS_SERVER(inode); + unsigned int i; + + memcpy(bitmask, src, sizeof(*bitmask) * NFS4_BITMASK_SZ); + cache_validity |= READ_ONCE(NFS_I(inode)->cache_validity); + + if (cache_validity & NFS_INO_INVALID_CHANGE) + bitmask[0] |= FATTR4_WORD0_CHANGE; + if (cache_validity & NFS_INO_INVALID_ATIME) + bitmask[1] |= FATTR4_WORD1_TIME_ACCESS; + if (cache_validity & NFS_INO_INVALID_MODE) + bitmask[1] |= FATTR4_WORD1_MODE; + if (cache_validity & NFS_INO_INVALID_OTHER) + bitmask[1] |= FATTR4_WORD1_OWNER | FATTR4_WORD1_OWNER_GROUP; + if (cache_validity & NFS_INO_INVALID_NLINK) + bitmask[1] |= FATTR4_WORD1_NUMLINKS; + if (cache_validity & NFS_INO_INVALID_CTIME) + bitmask[1] |= FATTR4_WORD1_TIME_METADATA; + if (cache_validity & NFS_INO_INVALID_MTIME) + bitmask[1] |= FATTR4_WORD1_TIME_MODIFY; + if (cache_validity & NFS_INO_INVALID_BLOCKS) + bitmask[1] |= FATTR4_WORD1_SPACE_USED; + if (cache_validity & NFS_INO_INVALID_BTIME) + bitmask[1] |= FATTR4_WORD1_TIME_CREATE; + + if (cache_validity & NFS_INO_INVALID_SIZE) + bitmask[0] |= FATTR4_WORD0_SIZE; + + for (i = 0; i < NFS4_BITMASK_SZ; i++) + bitmask[i] &= server->attr_bitmask[i]; +} + +static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + struct nfs_server *server = NFS_SERVER(hdr->inode); + + if (!nfs4_write_need_cache_consistency_data(hdr)) { + hdr->args.bitmask = NULL; + hdr->res.fattr = NULL; + } else { + nfs4_bitmask_set(hdr->args.bitmask_store, + server->cache_consistency_bitmask, + hdr->inode, NFS_INO_INVALID_BLOCKS); + hdr->args.bitmask = hdr->args.bitmask_store; + } + + if (!hdr->pgio_done_cb) + hdr->pgio_done_cb = nfs4_write_done_cb; + hdr->res.server = server; + hdr->timestamp = jiffies; + + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE]; + nfs4_init_sequence(server->nfs_client, &hdr->args.seq_args, + &hdr->res.seq_res, 0, 0); + nfs4_state_protect_write(hdr->ds_clp ? hdr->ds_clp : server->nfs_client, clnt, msg, hdr); +} + +static void nfs4_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) +{ + nfs4_setup_sequence(NFS_SERVER(data->inode)->nfs_client, + &data->args.seq_args, + &data->res.seq_res, + task); +} + +static int nfs4_commit_done_cb(struct rpc_task *task, struct nfs_commit_data *data) +{ + struct inode *inode = data->inode; + + trace_nfs4_commit(data, task->tk_status); + if (nfs4_async_handle_error(task, NFS_SERVER(inode), + NULL, NULL) == -EAGAIN) { + rpc_restart_call_prepare(task); + return -EAGAIN; + } + return 0; +} + +static int nfs4_commit_done(struct rpc_task *task, struct nfs_commit_data *data) +{ + if (!nfs4_sequence_done(task, &data->res.seq_res)) + return -EAGAIN; + return data->commit_done_cb(task, data); +} + +static void nfs4_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + struct nfs_server *server = NFS_SERVER(data->inode); + + if (data->commit_done_cb == NULL) + data->commit_done_cb = nfs4_commit_done_cb; + data->res.server = server; + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT]; + nfs4_init_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, 1, 0); + nfs4_state_protect(data->ds_clp ? data->ds_clp : server->nfs_client, + NFS_SP4_MACH_CRED_COMMIT, clnt, msg); +} + +static int _nfs4_proc_commit(struct file *dst, struct nfs_commitargs *args, + struct nfs_commitres *res) +{ + struct inode *dst_inode = file_inode(dst); + struct nfs_server *server = NFS_SERVER(dst_inode); + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT], + .rpc_argp = args, + .rpc_resp = res, + }; + + args->fh = NFS_FH(dst_inode); + return nfs4_call_sync(server->client, server, &msg, + &args->seq_args, &res->seq_res, 1); +} + +int nfs4_proc_commit(struct file *dst, __u64 offset, __u32 count, struct nfs_commitres *res) +{ + struct nfs_commitargs args = { + .offset = offset, + .count = count, + }; + struct nfs_server *dst_server = NFS_SERVER(file_inode(dst)); + struct nfs4_exception exception = { }; + int status; + + do { + status = _nfs4_proc_commit(dst, &args, res); + status = nfs4_handle_exception(dst_server, status, &exception); + } while (exception.retry); + + return status; +} + +static bool nfs4_server_supports_acls(const struct nfs_server *server, + enum nfs4_acl_type type) +{ + switch (type) { + default: + return server->attr_bitmask[0] & FATTR4_WORD0_ACL; + case NFS4ACL_DACL: + return server->attr_bitmask[1] & FATTR4_WORD1_DACL; + case NFS4ACL_SACL: + return server->attr_bitmask[1] & FATTR4_WORD1_SACL; + } +} + +/* Assuming that XATTR_SIZE_MAX is a multiple of PAGE_SIZE, and that + * it's OK to put sizeof(void) * (XATTR_SIZE_MAX/PAGE_SIZE) bytes on + * the stack. + */ +#define NFS4ACL_MAXPAGES DIV_ROUND_UP(XATTR_SIZE_MAX, PAGE_SIZE) + +int nfs4_buf_to_pages_noslab(const void *buf, size_t buflen, + struct page **pages) +{ + struct page *newpage, **spages; + int rc = 0; + size_t len; + spages = pages; + + do { + len = min_t(size_t, PAGE_SIZE, buflen); + newpage = alloc_page(GFP_KERNEL); + + if (newpage == NULL) + goto unwind; + memcpy(page_address(newpage), buf, len); + buf += len; + buflen -= len; + *pages++ = newpage; + rc++; + } while (buflen != 0); + + return rc; + +unwind: + for(; rc > 0; rc--) + __free_page(spages[rc-1]); + return -ENOMEM; +} + +struct nfs4_cached_acl { + enum nfs4_acl_type type; + int cached; + size_t len; + char data[]; +}; + +static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + spin_lock(&inode->i_lock); + kfree(nfsi->nfs4_acl); + nfsi->nfs4_acl = acl; + spin_unlock(&inode->i_lock); +} + +static void nfs4_zap_acl_attr(struct inode *inode) +{ + nfs4_set_cached_acl(inode, NULL); +} + +static ssize_t nfs4_read_cached_acl(struct inode *inode, char *buf, + size_t buflen, enum nfs4_acl_type type) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs4_cached_acl *acl; + int ret = -ENOENT; + + spin_lock(&inode->i_lock); + acl = nfsi->nfs4_acl; + if (acl == NULL) + goto out; + if (acl->type != type) + goto out; + if (buf == NULL) /* user is just asking for length */ + goto out_len; + if (acl->cached == 0) + goto out; + ret = -ERANGE; /* see getxattr(2) man page */ + if (acl->len > buflen) + goto out; + memcpy(buf, acl->data, acl->len); +out_len: + ret = acl->len; +out: + spin_unlock(&inode->i_lock); + return ret; +} + +static void nfs4_write_cached_acl(struct inode *inode, struct page **pages, + size_t pgbase, size_t acl_len, + enum nfs4_acl_type type) +{ + struct nfs4_cached_acl *acl; + size_t buflen = sizeof(*acl) + acl_len; + + if (buflen <= PAGE_SIZE) { + acl = kmalloc(buflen, GFP_KERNEL); + if (acl == NULL) + goto out; + acl->cached = 1; + _copy_from_pages(acl->data, pages, pgbase, acl_len); + } else { + acl = kmalloc_obj(*acl); + if (acl == NULL) + goto out; + acl->cached = 0; + } + acl->type = type; + acl->len = acl_len; +out: + nfs4_set_cached_acl(inode, acl); +} + +/* + * The getxattr API returns the required buffer length when called with a + * NULL buf. The NFSv4 acl tool then calls getxattr again after allocating + * the required buf. On a NULL buf, we send a page of data to the server + * guessing that the ACL request can be serviced by a page. If so, we cache + * up to the page of ACL data, and the 2nd call to getxattr is serviced by + * the cache. If not so, we throw away the page, and cache the required + * length. The next getxattr call will then produce another round trip to + * the server, this time with the input buf of the required size. + */ +static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, + size_t buflen, enum nfs4_acl_type type) +{ + struct page **pages; + struct nfs_getaclargs args = { + .fh = NFS_FH(inode), + .acl_type = type, + .acl_len = buflen, + }; + struct nfs_getaclres res = { + .acl_type = type, + .acl_len = buflen, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL], + .rpc_argp = &args, + .rpc_resp = &res, + }; + unsigned int npages; + int ret = -ENOMEM, i; + struct nfs_server *server = NFS_SERVER(inode); + + if (buflen == 0) + buflen = server->rsize; + + npages = DIV_ROUND_UP(buflen, PAGE_SIZE) + 1; + pages = kmalloc_objs(struct page *, npages); + if (!pages) + return -ENOMEM; + + args.acl_pages = pages; + + for (i = 0; i < npages; i++) { + pages[i] = alloc_page(GFP_KERNEL); + if (!pages[i]) + goto out_free; + } + + /* for decoding across pages */ + res.acl_scratch = folio_alloc(GFP_KERNEL, 0); + if (!res.acl_scratch) + goto out_free; + + args.acl_len = npages * PAGE_SIZE; + + dprintk("%s buf %p buflen %zu npages %d args.acl_len %zu\n", + __func__, buf, buflen, npages, args.acl_len); + ret = nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), + &msg, &args.seq_args, &res.seq_res, 0); + if (ret) + goto out_free; + + /* Handle the case where the passed-in buffer is too short */ + if (res.acl_flags & NFS4_ACL_TRUNC) { + /* Did the user only issue a request for the acl length? */ + if (buf == NULL) + goto out_ok; + ret = -ERANGE; + goto out_free; + } + nfs4_write_cached_acl(inode, pages, res.acl_data_offset, res.acl_len, + type); + if (buf) { + if (res.acl_len > buflen) { + ret = -ERANGE; + goto out_free; + } + _copy_from_pages(buf, pages, res.acl_data_offset, res.acl_len); + } +out_ok: + ret = res.acl_len; +out_free: + while (--i >= 0) + __free_page(pages[i]); + if (res.acl_scratch) + folio_put(res.acl_scratch); + kfree(pages); + return ret; +} + +static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, + size_t buflen, enum nfs4_acl_type type) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + ssize_t ret; + do { + ret = __nfs4_get_acl_uncached(inode, buf, buflen, type); + trace_nfs4_get_acl(inode, ret); + if (ret >= 0) + break; + ret = nfs4_handle_exception(NFS_SERVER(inode), ret, &exception); + } while (exception.retry); + return ret; +} + +static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen, + enum nfs4_acl_type type) +{ + struct nfs_server *server = NFS_SERVER(inode); + int ret; + + if (unlikely(NFS_FH(inode)->size == 0)) + return -ENODATA; + if (!nfs4_server_supports_acls(server, type)) + return -EOPNOTSUPP; + ret = nfs_revalidate_inode(inode, NFS_INO_INVALID_CHANGE); + if (ret < 0) + return ret; + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); + ret = nfs4_read_cached_acl(inode, buf, buflen, type); + if (ret != -ENOENT) + /* -ENOENT is returned if there is no ACL or if there is an ACL + * but no cached acl data, just the acl length */ + return ret; + return nfs4_get_acl_uncached(inode, buf, buflen, type); +} + +static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, + size_t buflen, enum nfs4_acl_type type) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct page *pages[NFS4ACL_MAXPAGES]; + struct nfs_setaclargs arg = { + .fh = NFS_FH(inode), + .acl_type = type, + .acl_len = buflen, + .acl_pages = pages, + }; + struct nfs_setaclres res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETACL], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + unsigned int npages = DIV_ROUND_UP(buflen, PAGE_SIZE); + int ret, i; + + /* You can't remove system.nfs4_acl: */ + if (buflen == 0) + return -EINVAL; + if (!nfs4_server_supports_acls(server, type)) + return -EOPNOTSUPP; + if (npages > ARRAY_SIZE(pages)) + return -ERANGE; + i = nfs4_buf_to_pages_noslab(buf, buflen, arg.acl_pages); + if (i < 0) + return i; + nfs4_inode_make_writeable(inode); + ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); + + /* + * Free each page after tx, so the only ref left is + * held by the network stack + */ + for (; i > 0; i--) + put_page(pages[i-1]); + + /* + * Acl update can result in inode attribute update. + * so mark the attribute cache invalid. + */ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); + spin_unlock(&inode->i_lock); + nfs_access_zap_cache(inode); + nfs_zap_acl_cache(inode); + return ret; +} + +static int nfs4_proc_set_acl(struct inode *inode, const void *buf, + size_t buflen, enum nfs4_acl_type type) +{ + struct nfs4_exception exception = { }; + int err; + + if (unlikely(NFS_FH(inode)->size == 0)) + return -ENODATA; + do { + err = __nfs4_proc_set_acl(inode, buf, buflen, type); + trace_nfs4_set_acl(inode, err); + if (err == -NFS4ERR_BADOWNER || err == -NFS4ERR_BADNAME) { + /* + * no need to retry since the kernel + * isn't involved in encoding the ACEs. + */ + err = -EINVAL; + break; + } + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +static int _nfs4_get_security_label(struct inode *inode, void *buf, + size_t buflen) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs4_label label = {0, 0, 0, buflen, buf}; + + u32 bitmask[3] = { 0, 0, FATTR4_WORD2_SECURITY_LABEL }; + struct nfs_fattr fattr = { + .label = &label, + }; + struct nfs4_getattr_arg arg = { + .fh = NFS_FH(inode), + .bitmask = bitmask, + }; + struct nfs4_getattr_res res = { + .fattr = &fattr, + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETATTR], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int ret; + + nfs_fattr_init(&fattr); + + ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 0); + if (ret) + return ret; + if (!(fattr.valid & NFS_ATTR_FATTR_V4_SECURITY_LABEL)) + return -ENOENT; + return label.len; +} + +static int nfs4_get_security_label(struct inode *inode, void *buf, + size_t buflen) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + if (!nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) + return -EOPNOTSUPP; + + do { + err = _nfs4_get_security_label(inode, buf, buflen); + trace_nfs4_get_security_label(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +static int _nfs4_do_set_security_label(struct inode *inode, + struct nfs4_label *ilabel, + struct nfs_fattr *fattr) +{ + + struct iattr sattr = {0}; + struct nfs_server *server = NFS_SERVER(inode); + const u32 bitmask[3] = { 0, 0, FATTR4_WORD2_SECURITY_LABEL }; + struct nfs_setattrargs arg = { + .fh = NFS_FH(inode), + .iap = &sattr, + .server = server, + .bitmask = bitmask, + .label = ilabel, + }; + struct nfs_setattrres res = { + .fattr = fattr, + .server = server, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETATTR], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status; + + nfs4_stateid_copy(&arg.stateid, &zero_stateid); + + status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); + if (status) + dprintk("%s failed: %d\n", __func__, status); + + return status; +} + +static int nfs4_do_set_security_label(struct inode *inode, + struct nfs4_label *ilabel, + struct nfs_fattr *fattr) +{ + struct nfs4_exception exception = { }; + int err; + + do { + err = _nfs4_do_set_security_label(inode, ilabel, fattr); + trace_nfs4_set_security_label(inode, err); + err = nfs4_handle_exception(NFS_SERVER(inode), err, + &exception); + } while (exception.retry); + return err; +} + +static int +nfs4_set_security_label(struct inode *inode, const void *buf, size_t buflen) +{ + struct nfs4_label ilabel = {0, 0, 0, buflen, (char *)buf }; + struct nfs_fattr *fattr; + int status; + + if (!nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) + return -EOPNOTSUPP; + + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fattr == NULL) + return -ENOMEM; + + status = nfs4_do_set_security_label(inode, &ilabel, fattr); + if (status == 0) + nfs_setsecurity(inode, fattr); + + nfs_free_fattr(fattr); + return status; +} +#endif /* CONFIG_NFS_V4_SECURITY_LABEL */ + + +static void nfs4_init_boot_verifier(const struct nfs_client *clp, + nfs4_verifier *bootverf) +{ + __be32 verf[2]; + + if (test_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) { + /* An impossible timestamp guarantees this value + * will never match a generated boot time. */ + verf[0] = cpu_to_be32(U32_MAX); + verf[1] = cpu_to_be32(U32_MAX); + } else { + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + u64 ns = ktime_to_ns(nn->boot_time); + + verf[0] = cpu_to_be32(ns >> 32); + verf[1] = cpu_to_be32(ns); + } + memcpy(bootverf->data, verf, sizeof(bootverf->data)); +} + +static size_t +nfs4_get_uniquifier(struct nfs_client *clp, char *buf, size_t buflen) +{ + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + struct nfs_netns_client *nn_clp = nn->nfs_client; + const char *id; + + buf[0] = '\0'; + + if (nn_clp) { + rcu_read_lock(); + id = rcu_dereference(nn_clp->identifier); + if (id) + strscpy(buf, id, buflen); + rcu_read_unlock(); + } + + if (nfs4_client_id_uniquifier[0] != '\0' && buf[0] == '\0') + strscpy(buf, nfs4_client_id_uniquifier, buflen); + + return strlen(buf); +} + +static int +nfs4_init_nonuniform_client_string(struct nfs_client *clp) +{ + char buf[NFS4_CLIENT_ID_UNIQ_LEN]; + size_t buflen; + size_t len; + char *str; + + if (clp->cl_owner_id != NULL) + return 0; + + rcu_read_lock(); + len = 14 + + strlen(clp->cl_rpcclient->cl_nodename) + + 1 + + strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)) + + 1; + rcu_read_unlock(); + + buflen = nfs4_get_uniquifier(clp, buf, sizeof(buf)); + if (buflen) + len += buflen + 1; + + if (len > NFS4_OPAQUE_LIMIT + 1) + return -EINVAL; + + /* + * Since this string is allocated at mount time, and held until the + * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying + * about a memory-reclaim deadlock. + */ + str = kmalloc(len, GFP_KERNEL); + if (!str) + return -ENOMEM; + + rcu_read_lock(); + if (buflen) + scnprintf(str, len, "Linux NFSv4.0 %s/%s/%s", + clp->cl_rpcclient->cl_nodename, buf, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); + else + scnprintf(str, len, "Linux NFSv4.0 %s/%s", + clp->cl_rpcclient->cl_nodename, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); + rcu_read_unlock(); + + clp->cl_owner_id = str; + return 0; +} + +static int +nfs4_init_uniform_client_string(struct nfs_client *clp) +{ + char buf[NFS4_CLIENT_ID_UNIQ_LEN]; + size_t buflen; + size_t len; + char *str; + + if (clp->cl_owner_id != NULL) + return 0; + + len = 10 + 10 + 1 + 10 + 1 + + strlen(clp->cl_rpcclient->cl_nodename) + 1; + + buflen = nfs4_get_uniquifier(clp, buf, sizeof(buf)); + if (buflen) + len += buflen + 1; + + if (len > NFS4_OPAQUE_LIMIT + 1) + return -EINVAL; + + /* + * Since this string is allocated at mount time, and held until the + * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying + * about a memory-reclaim deadlock. + */ + str = kmalloc(len, GFP_KERNEL); + if (!str) + return -ENOMEM; + + if (buflen) + scnprintf(str, len, "Linux NFSv%u.%u %s/%s", + clp->rpc_ops->version, clp->cl_minorversion, + buf, clp->cl_rpcclient->cl_nodename); + else + scnprintf(str, len, "Linux NFSv%u.%u %s", + clp->rpc_ops->version, clp->cl_minorversion, + clp->cl_rpcclient->cl_nodename); + clp->cl_owner_id = str; + return 0; +} + +/* + * nfs4_callback_up_net() starts only "tcp" and "tcp6" callback + * services. Advertise one based on the address family of the + * clientaddr. + */ +static unsigned int +nfs4_init_callback_netid(const struct nfs_client *clp, char *buf, size_t len) +{ + if (strchr(clp->cl_ipaddr, ':') != NULL) + return scnprintf(buf, len, "tcp6"); + else + return scnprintf(buf, len, "tcp"); +} + +static void nfs4_setclientid_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_setclientid *sc = calldata; + + if (task->tk_status == 0) + sc->sc_cred = get_rpccred(task->tk_rqstp->rq_cred); +} + +static const struct rpc_call_ops nfs4_setclientid_ops = { + .rpc_call_done = nfs4_setclientid_done, +}; + +/** + * nfs4_proc_setclientid - Negotiate client ID + * @clp: state data structure + * @program: RPC program for NFSv4 callback service + * @port: IP port number for NFS4 callback service + * @cred: credential to use for this call + * @res: where to place the result + * + * Returns zero, a negative errno, or a negative NFS4ERR status code. + */ +int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, + unsigned short port, const struct cred *cred, + struct nfs4_setclientid_res *res) +{ + nfs4_verifier sc_verifier; + struct nfs4_setclientid setclientid = { + .sc_verifier = &sc_verifier, + .sc_prog = program, + .sc_clnt = clp, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID], + .rpc_argp = &setclientid, + .rpc_resp = res, + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs4_setclientid_ops, + .callback_data = &setclientid, + .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN, + }; + unsigned long now = jiffies; + int status; + + /* nfs_client_id4 */ + nfs4_init_boot_verifier(clp, &sc_verifier); + + if (test_bit(NFS_CS_MIGRATION, &clp->cl_flags)) + status = nfs4_init_uniform_client_string(clp); + else + status = nfs4_init_nonuniform_client_string(clp); + + if (status) + goto out; + + /* cb_client4 */ + setclientid.sc_netid_len = + nfs4_init_callback_netid(clp, + setclientid.sc_netid, + sizeof(setclientid.sc_netid)); + setclientid.sc_uaddr_len = scnprintf(setclientid.sc_uaddr, + sizeof(setclientid.sc_uaddr), "%s.%u.%u", + clp->cl_ipaddr, port >> 8, port & 255); + + dprintk("NFS call setclientid auth=%s, '%s'\n", + clp->cl_rpcclient->cl_auth->au_ops->au_name, + clp->cl_owner_id); + + status = nfs4_call_sync_custom(&task_setup_data); + if (setclientid.sc_cred) { + kfree(clp->cl_acceptor); + clp->cl_acceptor = rpcauth_stringify_acceptor(setclientid.sc_cred); + put_rpccred(setclientid.sc_cred); + } + + if (status == 0) + do_renew_lease(clp, now); +out: + trace_nfs4_setclientid(clp, status); + dprintk("NFS reply setclientid: %d\n", status); + return status; +} + +/** + * nfs4_proc_setclientid_confirm - Confirm client ID + * @clp: state data structure + * @arg: result of a previous SETCLIENTID + * @cred: credential to use for this call + * + * Returns zero, a negative errno, or a negative NFS4ERR status code. + */ +int nfs4_proc_setclientid_confirm(struct nfs_client *clp, + struct nfs4_setclientid_res *arg, + const struct cred *cred) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID_CONFIRM], + .rpc_argp = arg, + .rpc_cred = cred, + }; + int status; + + dprintk("NFS call setclientid_confirm auth=%s, (client ID %llx)\n", + clp->cl_rpcclient->cl_auth->au_ops->au_name, + clp->cl_clientid); + status = rpc_call_sync(clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); + trace_nfs4_setclientid_confirm(clp, status); + dprintk("NFS reply setclientid_confirm: %d\n", status); + return status; +} + +struct nfs4_delegreturndata { + struct nfs4_delegreturnargs args; + struct nfs4_delegreturnres res; + struct nfs_fh fh; + nfs4_stateid stateid; + unsigned long timestamp; + unsigned short retrans; + struct { + struct nfs4_layoutreturn_args arg; + struct nfs4_layoutreturn_res res; + struct nfs4_xdr_opaque_data ld_private; + u32 roc_barrier; + bool roc; + } lr; + struct nfs4_delegattr sattr; + struct nfs_fattr fattr; + int rpc_status; + struct inode *inode; +}; + +static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_delegreturndata *data = calldata; + struct nfs4_exception exception = { + .inode = data->inode, + .stateid = &data->stateid, + .task_is_privileged = data->args.seq_args.sa_privileged, + .retrans = data->retrans, + }; + + if (!nfs4_sequence_done(task, &data->res.seq_res)) + return; + + trace_nfs4_delegreturn_exit(&data->args, &data->res, task->tk_status); + + /* Handle Layoutreturn errors */ + if (pnfs_roc_done(task, &data->args.lr_args, &data->res.lr_res, + &data->res.lr_ret) == -EAGAIN) + goto out_restart; + + if (data->args.sattr_args && task->tk_status != 0) { + switch(data->res.sattr_ret) { + case 0: + data->args.sattr_args = NULL; + data->res.sattr_res = false; + break; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + /* Let the main handler below do stateid recovery */ + break; + case -NFS4ERR_OLD_STATEID: + if (nfs4_refresh_delegation_stateid(&data->stateid, + data->inode)) + goto out_restart; + fallthrough; + default: + data->args.sattr_args = NULL; + data->res.sattr_res = false; + goto out_restart; + } + } + + switch (task->tk_status) { + case 0: + renew_lease(data->res.server, data->timestamp); + break; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_EXPIRED: + nfs4_free_revoked_stateid(data->res.server, + data->args.stateid, + task->tk_msg.rpc_cred); + fallthrough; + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_STALE_STATEID: + case -ETIMEDOUT: + task->tk_status = 0; + break; + case -NFS4ERR_OLD_STATEID: + if (!nfs4_refresh_delegation_stateid(&data->stateid, data->inode)) + nfs4_stateid_seqid_inc(&data->stateid); + if (data->args.bitmask) { + data->args.bitmask = NULL; + data->res.fattr = NULL; + } + goto out_restart; + case -NFS4ERR_ACCESS: + if (data->args.bitmask) { + data->args.bitmask = NULL; + data->res.fattr = NULL; + goto out_restart; + } + fallthrough; + default: + task->tk_status = nfs4_async_handle_exception(task, + data->res.server, task->tk_status, + &exception); + data->retrans = exception.retrans; + if (exception.retry) + goto out_restart; + } + nfs_delegation_mark_returned(data->inode, data->args.stateid); + data->rpc_status = task->tk_status; + return; +out_restart: + task->tk_status = 0; + rpc_restart_call_prepare(task); +} + +static void nfs4_delegreturn_release(void *calldata) +{ + struct nfs4_delegreturndata *data = calldata; + struct inode *inode = data->inode; + + if (data->lr.roc) + pnfs_roc_release(&data->lr.arg, &data->lr.res, + data->res.lr_ret); + if (inode) { + nfs4_fattr_set_prechange(&data->fattr, + inode_peek_iversion_raw(inode)); + nfs_refresh_inode(inode, &data->fattr); + nfs_iput_and_deactive(inode); + } + kfree(calldata); +} + +static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_delegreturndata *d_data; + struct pnfs_layout_hdr *lo; + + d_data = data; + + if (!d_data->lr.roc && nfs4_wait_on_layoutreturn(d_data->inode, task)) { + nfs4_sequence_done(task, &d_data->res.seq_res); + return; + } + + lo = d_data->args.lr_args ? d_data->args.lr_args->layout : NULL; + if (lo && !pnfs_layout_is_valid(lo)) { + d_data->args.lr_args = NULL; + d_data->res.lr_res = NULL; + } + + nfs4_setup_sequence(d_data->res.server->nfs_client, + &d_data->args.seq_args, + &d_data->res.seq_res, + task); +} + +static const struct rpc_call_ops nfs4_delegreturn_ops = { + .rpc_call_prepare = nfs4_delegreturn_prepare, + .rpc_call_done = nfs4_delegreturn_done, + .rpc_release = nfs4_delegreturn_release, +}; + +static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred, + const nfs4_stateid *stateid, + struct nfs_delegation *delegation, + int issync) +{ + struct nfs4_delegreturndata *data; + struct nfs_server *server = NFS_SERVER(inode); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DELEGRETURN], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_delegreturn_ops, + .flags = RPC_TASK_ASYNC | RPC_TASK_TIMEOUT, + }; + int status = 0; + + if (nfs_server_capable(inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + data = kzalloc_obj(*data); + if (data == NULL) + return -ENOMEM; + + nfs4_state_protect(server->nfs_client, + NFS_SP4_MACH_CRED_CLEANUP, + &task_setup_data.rpc_client, &msg); + + data->args.fhandle = &data->fh; + data->args.stateid = &data->stateid; + nfs4_bitmask_set(data->args.bitmask_store, + server->cache_consistency_bitmask, inode, 0); + data->args.bitmask = data->args.bitmask_store; + nfs_copy_fh(&data->fh, NFS_FH(inode)); + nfs4_stateid_copy(&data->stateid, stateid); + data->res.fattr = &data->fattr; + data->res.server = server; + data->res.lr_ret = -NFS4ERR_NOMATCHING_LAYOUT; + data->lr.arg.ld_private = &data->lr.ld_private; + nfs_fattr_init(data->res.fattr); + data->timestamp = jiffies; + data->rpc_status = 0; + data->inode = nfs_igrab_and_active(inode); + if (data->inode || issync) { + data->lr.roc = pnfs_roc(inode, &data->lr.arg, &data->lr.res, + cred, issync); + if (data->lr.roc) { + data->args.lr_args = &data->lr.arg; + data->res.lr_res = &data->lr.res; + } + } + + if (delegation && + test_bit(NFS_DELEGATION_DELEGTIME, &delegation->flags)) { + if (delegation->type & FMODE_READ) { + data->sattr.atime = inode_get_atime(inode); + data->sattr.atime_set = true; + } + if (delegation->type & FMODE_WRITE) { + data->sattr.mtime = inode_get_mtime(inode); + data->sattr.mtime_set = true; + } + data->args.sattr_args = &data->sattr; + data->res.sattr_res = true; + } + + nfs4_init_sequence(server->nfs_client, &data->args.seq_args, + &data->res.seq_res, 1, !data->inode ? 1 : 0); + + task_setup_data.callback_data = data; + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + if (!issync) + goto out; + status = rpc_wait_for_completion_task(task); + if (status != 0) + goto out; + status = data->rpc_status; +out: + rpc_put_task(task); + return status; +} + +int nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred, + const nfs4_stateid *stateid, + struct nfs_delegation *delegation, int issync) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs4_exception exception = { }; + int err; + do { + err = _nfs4_proc_delegreturn(inode, cred, stateid, + delegation, issync); + trace_nfs4_delegreturn(inode, stateid, err); + switch (err) { + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_EXPIRED: + case 0: + return 0; + } + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + return err; +} + +static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct inode *inode = state->inode; + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_client *clp = server->nfs_client; + struct nfs_lockt_args arg = { + .fh = NFS_FH(inode), + .fl = request, + }; + struct nfs_lockt_res res = { + .denied = request, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCKT], + .rpc_argp = &arg, + .rpc_resp = &res, + .rpc_cred = state->owner->so_cred, + }; + struct nfs4_lock_state *lsp; + int status; + + arg.lock_owner.clientid = clp->cl_clientid; + status = nfs4_set_lock_state(state, request); + if (status != 0) + goto out; + lsp = request->fl_u.nfs4_fl.owner; + arg.lock_owner.id = lsp->ls_seqid.owner_id; + arg.lock_owner.s_dev = server->s_dev; + status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); + switch (status) { + case 0: + request->c.flc_type = F_UNLCK; + break; + case -NFS4ERR_DENIED: + status = 0; + } + request->fl_ops->fl_release_private(request); + request->fl_ops = NULL; +out: + return status; +} + +static int nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + + do { + err = _nfs4_proc_getlk(state, cmd, request); + trace_nfs4_get_lock(request, state, cmd, err); + err = nfs4_handle_exception(NFS_SERVER(state->inode), err, + &exception); + } while (exception.retry); + return err; +} + +/* + * Update the seqid of a lock stateid after receiving + * NFS4ERR_OLD_STATEID + */ +static bool nfs4_refresh_lock_old_stateid(nfs4_stateid *dst, + struct nfs4_lock_state *lsp) +{ + struct nfs4_state *state = lsp->ls_state; + bool ret = false; + + spin_lock(&state->state_lock); + if (!nfs4_stateid_match_other(dst, &lsp->ls_stateid)) + goto out; + if (!nfs4_stateid_is_newer(&lsp->ls_stateid, dst)) + nfs4_stateid_seqid_inc(dst); + else + dst->seqid = lsp->ls_stateid.seqid; + ret = true; +out: + spin_unlock(&state->state_lock); + return ret; +} + +static bool nfs4_sync_lock_stateid(nfs4_stateid *dst, + struct nfs4_lock_state *lsp) +{ + struct nfs4_state *state = lsp->ls_state; + bool ret; + + spin_lock(&state->state_lock); + ret = !nfs4_stateid_match_other(dst, &lsp->ls_stateid); + nfs4_stateid_copy(dst, &lsp->ls_stateid); + spin_unlock(&state->state_lock); + return ret; +} + +struct nfs4_unlockdata { + struct nfs_locku_args arg; + struct nfs_locku_res res; + struct nfs4_lock_state *lsp; + struct nfs_open_context *ctx; + struct nfs_lock_context *l_ctx; + struct file_lock fl; + struct nfs_server *server; + unsigned long timestamp; + unsigned short retrans; +}; + +static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl, + struct nfs_open_context *ctx, + struct nfs4_lock_state *lsp, + struct nfs_seqid *seqid) +{ + struct nfs4_unlockdata *p; + struct nfs4_state *state = lsp->ls_state; + struct inode *inode = state->inode; + struct nfs_lock_context *l_ctx; + + p = kzalloc_obj(*p); + if (p == NULL) + return NULL; + l_ctx = nfs_get_lock_context(ctx); + if (!IS_ERR(l_ctx)) { + p->l_ctx = l_ctx; + } else { + kfree(p); + return NULL; + } + p->arg.fh = NFS_FH(inode); + p->arg.fl = &p->fl; + p->arg.seqid = seqid; + p->res.seqid = seqid; + p->lsp = lsp; + /* Ensure we don't close file until we're done freeing locks! */ + p->ctx = get_nfs_open_context(ctx); + locks_init_lock(&p->fl); + locks_copy_lock(&p->fl, fl); + p->server = NFS_SERVER(inode); + spin_lock(&state->state_lock); + nfs4_stateid_copy(&p->arg.stateid, &lsp->ls_stateid); + spin_unlock(&state->state_lock); + return p; +} + +static void nfs4_locku_release_calldata(void *data) +{ + struct nfs4_unlockdata *calldata = data; + nfs_free_seqid(calldata->arg.seqid); + nfs4_put_lock_state(calldata->lsp); + nfs_put_lock_context(calldata->l_ctx); + put_nfs_open_context(calldata->ctx); + kfree(calldata); +} + +static void nfs4_locku_done(struct rpc_task *task, void *data) +{ + struct nfs4_unlockdata *calldata = data; + struct nfs4_exception exception = { + .inode = calldata->lsp->ls_state->inode, + .stateid = &calldata->arg.stateid, + .retrans = calldata->retrans, + }; + + if (!nfs4_sequence_done(task, &calldata->res.seq_res)) + return; + switch (task->tk_status) { + case 0: + renew_lease(calldata->server, calldata->timestamp); + if (nfs4_update_lock_stateid(calldata->lsp, + &calldata->res.stateid)) + break; + fallthrough; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_EXPIRED: + nfs4_free_revoked_stateid(calldata->server, + &calldata->arg.stateid, + task->tk_msg.rpc_cred); + fallthrough; + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_STALE_STATEID: + if (nfs4_sync_lock_stateid(&calldata->arg.stateid, + calldata->lsp)) + rpc_restart_call_prepare(task); + break; + case -NFS4ERR_OLD_STATEID: + if (nfs4_refresh_lock_old_stateid(&calldata->arg.stateid, + calldata->lsp)) + rpc_restart_call_prepare(task); + break; + default: + task->tk_status = nfs4_async_handle_exception(task, + calldata->server, task->tk_status, + &exception); + calldata->retrans = exception.retrans; + if (exception.retry) + rpc_restart_call_prepare(task); + } + nfs_release_seqid(calldata->arg.seqid); +} + +static void nfs4_locku_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_unlockdata *calldata = data; + + if (test_bit(NFS_CONTEXT_UNLOCK, &calldata->l_ctx->open_context->flags) && + nfs_async_iocounter_wait(task, calldata->l_ctx)) + return; + + if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) + goto out_wait; + if (test_bit(NFS_LOCK_INITIALIZED, &calldata->lsp->ls_flags) == 0) { + /* Note: exit _without_ running nfs4_locku_done */ + goto out_no_action; + } + calldata->timestamp = jiffies; + if (nfs4_setup_sequence(calldata->server->nfs_client, + &calldata->arg.seq_args, + &calldata->res.seq_res, + task) != 0) + nfs_release_seqid(calldata->arg.seqid); + return; +out_no_action: + task->tk_action = NULL; +out_wait: + nfs4_sequence_done(task, &calldata->res.seq_res); +} + +static const struct rpc_call_ops nfs4_locku_ops = { + .rpc_call_prepare = nfs4_locku_prepare, + .rpc_call_done = nfs4_locku_done, + .rpc_release = nfs4_locku_release_calldata, +}; + +static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, + struct nfs_open_context *ctx, + struct nfs4_lock_state *lsp, + struct nfs_seqid *seqid) +{ + struct nfs4_unlockdata *data; + struct nfs_client *clp = NFS_SERVER(lsp->ls_state->inode)->nfs_client; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCKU], + .rpc_cred = ctx->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(lsp->ls_state->inode), + .rpc_message = &msg, + .callback_ops = &nfs4_locku_ops, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC, + }; + + if (nfs_server_capable(lsp->ls_state->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_CLEANUP, + &task_setup_data.rpc_client, &msg); + + /* Ensure this is an unlock - when canceling a lock, the + * canceled lock is passed in, and it won't be an unlock. + */ + fl->c.flc_type = F_UNLCK; + if (fl->c.flc_flags & FL_CLOSE) + set_bit(NFS_CONTEXT_UNLOCK, &ctx->flags); + + data = nfs4_alloc_unlockdata(fl, ctx, lsp, seqid); + if (data == NULL) { + nfs_free_seqid(seqid); + return ERR_PTR(-ENOMEM); + } + + nfs4_init_sequence(clp, &data->arg.seq_args, &data->res.seq_res, 1, 0); + msg.rpc_argp = &data->arg; + msg.rpc_resp = &data->res; + task_setup_data.callback_data = data; + return rpc_run_task(&task_setup_data); +} + +static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct inode *inode = state->inode; + struct nfs4_state_owner *sp = state->owner; + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_seqid *seqid; + struct nfs4_lock_state *lsp; + struct rpc_task *task; + struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t); + int status = 0; + unsigned char saved_flags = request->c.flc_flags; + + status = nfs4_set_lock_state(state, request); + /* Unlock _before_ we do the RPC call */ + request->c.flc_flags |= FL_EXISTS; + /* Exclude nfs_delegation_claim_locks() */ + mutex_lock(&sp->so_delegreturn_mutex); + /* Exclude nfs4_reclaim_open_stateid() - note nesting! */ + down_read(&nfsi->rwsem); + if (locks_lock_inode_wait(inode, request) == -ENOENT) { + up_read(&nfsi->rwsem); + mutex_unlock(&sp->so_delegreturn_mutex); + goto out; + } + lsp = request->fl_u.nfs4_fl.owner; + set_bit(NFS_LOCK_UNLOCKING, &lsp->ls_flags); + up_read(&nfsi->rwsem); + mutex_unlock(&sp->so_delegreturn_mutex); + if (status != 0) + goto out; + /* Is this a delegated lock? */ + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) == 0) + goto out; + alloc_seqid = NFS_SERVER(inode)->nfs_client->cl_mvops->alloc_seqid; + seqid = alloc_seqid(&lsp->ls_seqid, GFP_KERNEL); + status = -ENOMEM; + if (IS_ERR(seqid)) + goto out; + task = nfs4_do_unlck(request, + nfs_file_open_context(request->c.flc_file), + lsp, seqid); + status = PTR_ERR(task); + if (IS_ERR(task)) + goto out; + status = rpc_wait_for_completion_task(task); + rpc_put_task(task); +out: + request->c.flc_flags = saved_flags; + trace_nfs4_unlock(request, state, F_SETLK, status); + return status; +} + +struct nfs4_lockdata { + struct nfs_lock_args arg; + struct nfs_lock_res res; + struct nfs4_lock_state *lsp; + struct nfs_open_context *ctx; + struct file_lock fl; + unsigned long timestamp; + int rpc_status; + int cancelled; + struct nfs_server *server; +}; + +static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, + struct nfs_open_context *ctx, struct nfs4_lock_state *lsp, + gfp_t gfp_mask) +{ + struct nfs4_lockdata *p; + struct inode *inode = lsp->ls_state->inode; + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_seqid *(*alloc_seqid)(struct nfs_seqid_counter *, gfp_t); + + p = kzalloc_obj(*p, gfp_mask); + if (p == NULL) + return NULL; + + p->arg.fh = NFS_FH(inode); + p->arg.fl = &p->fl; + p->arg.open_seqid = nfs_alloc_seqid(&lsp->ls_state->owner->so_seqid, gfp_mask); + if (IS_ERR(p->arg.open_seqid)) + goto out_free; + alloc_seqid = server->nfs_client->cl_mvops->alloc_seqid; + p->arg.lock_seqid = alloc_seqid(&lsp->ls_seqid, gfp_mask); + if (IS_ERR(p->arg.lock_seqid)) + goto out_free_seqid; + p->arg.lock_owner.clientid = server->nfs_client->cl_clientid; + p->arg.lock_owner.id = lsp->ls_seqid.owner_id; + p->arg.lock_owner.s_dev = server->s_dev; + p->res.lock_seqid = p->arg.lock_seqid; + p->lsp = lsp; + p->server = server; + p->ctx = get_nfs_open_context(ctx); + locks_init_lock(&p->fl); + locks_copy_lock(&p->fl, fl); + return p; +out_free_seqid: + nfs_free_seqid(p->arg.open_seqid); +out_free: + kfree(p); + return NULL; +} + +static void nfs4_lock_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_lockdata *data = calldata; + struct nfs4_state *state = data->lsp->ls_state; + + if (nfs_wait_on_sequence(data->arg.lock_seqid, task) != 0) + goto out_wait; + /* Do we need to do an open_to_lock_owner? */ + if (!test_bit(NFS_LOCK_INITIALIZED, &data->lsp->ls_flags)) { + if (nfs_wait_on_sequence(data->arg.open_seqid, task) != 0) { + goto out_release_lock_seqid; + } + nfs4_stateid_copy(&data->arg.open_stateid, + &state->open_stateid); + data->arg.new_lock_owner = 1; + data->res.open_seqid = data->arg.open_seqid; + } else { + data->arg.new_lock_owner = 0; + nfs4_stateid_copy(&data->arg.lock_stateid, + &data->lsp->ls_stateid); + } + if (!nfs4_valid_open_stateid(state)) { + data->rpc_status = -EBADF; + task->tk_action = NULL; + goto out_release_open_seqid; + } + data->timestamp = jiffies; + if (nfs4_setup_sequence(data->server->nfs_client, + &data->arg.seq_args, + &data->res.seq_res, + task) == 0) + return; +out_release_open_seqid: + nfs_release_seqid(data->arg.open_seqid); +out_release_lock_seqid: + nfs_release_seqid(data->arg.lock_seqid); +out_wait: + nfs4_sequence_done(task, &data->res.seq_res); + dprintk("%s: ret = %d\n", __func__, data->rpc_status); +} + +static void nfs4_lock_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_lockdata *data = calldata; + struct nfs4_lock_state *lsp = data->lsp; + + if (!nfs4_sequence_done(task, &data->res.seq_res)) + return; + + data->rpc_status = task->tk_status; + switch (task->tk_status) { + case 0: + renew_lease(NFS_SERVER(d_inode(data->ctx->dentry)), + data->timestamp); + if (data->arg.new_lock_owner != 0) { + nfs_confirm_seqid(&lsp->ls_seqid, 0); + nfs4_stateid_copy(&lsp->ls_stateid, &data->res.stateid); + set_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags); + } else if (!nfs4_update_lock_stateid(lsp, &data->res.stateid)) + goto out_restart; + break; + case -NFS4ERR_OLD_STATEID: + if (data->arg.new_lock_owner != 0 && + nfs4_refresh_open_old_stateid(&data->arg.open_stateid, + lsp->ls_state)) + goto out_restart; + if (nfs4_refresh_lock_old_stateid(&data->arg.lock_stateid, lsp)) + goto out_restart; + fallthrough; + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_EXPIRED: + if (data->arg.new_lock_owner != 0) { + if (!nfs4_stateid_match(&data->arg.open_stateid, + &lsp->ls_state->open_stateid)) + goto out_restart; + } else if (!nfs4_stateid_match(&data->arg.lock_stateid, + &lsp->ls_stateid)) + goto out_restart; + } +out_done: + dprintk("%s: ret = %d!\n", __func__, data->rpc_status); + return; +out_restart: + if (!data->cancelled) + rpc_restart_call_prepare(task); + goto out_done; +} + +static void nfs4_lock_release(void *calldata) +{ + struct nfs4_lockdata *data = calldata; + + nfs_free_seqid(data->arg.open_seqid); + if (data->cancelled && data->rpc_status == 0) { + struct rpc_task *task; + task = nfs4_do_unlck(&data->fl, data->ctx, data->lsp, + data->arg.lock_seqid); + if (!IS_ERR(task)) + rpc_put_task_async(task); + dprintk("%s: cancelling lock!\n", __func__); + } else + nfs_free_seqid(data->arg.lock_seqid); + nfs4_put_lock_state(data->lsp); + put_nfs_open_context(data->ctx); + kfree(data); +} + +static const struct rpc_call_ops nfs4_lock_ops = { + .rpc_call_prepare = nfs4_lock_prepare, + .rpc_call_done = nfs4_lock_done, + .rpc_release = nfs4_lock_release, +}; + +static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_state *lsp, int new_lock_owner, int error) +{ + switch (error) { + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED; + if (new_lock_owner != 0 || + test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) + nfs4_schedule_stateid_recovery(server, lsp->ls_state); + break; + case -NFS4ERR_STALE_STATEID: + lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED; + nfs4_schedule_lease_recovery(server->nfs_client); + } +} + +static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *fl, int recovery_type) +{ + struct nfs4_lockdata *data; + struct rpc_task *task; + struct nfs_client *clp = NFS_SERVER(state->inode)->nfs_client; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCK], + .rpc_cred = state->owner->so_cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_CLIENT(state->inode), + .rpc_message = &msg, + .callback_ops = &nfs4_lock_ops, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + int ret; + + if (nfs_server_capable(state->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + data = nfs4_alloc_lockdata(fl, + nfs_file_open_context(fl->c.flc_file), + fl->fl_u.nfs4_fl.owner, GFP_KERNEL); + if (data == NULL) + return -ENOMEM; + if (IS_SETLKW(cmd)) + data->arg.block = 1; + nfs4_init_sequence(clp, &data->arg.seq_args, &data->res.seq_res, 1, + recovery_type > NFS_LOCK_NEW); + msg.rpc_argp = &data->arg; + msg.rpc_resp = &data->res; + task_setup_data.callback_data = data; + + if (recovery_type == NFS_LOCK_RECLAIM) + data->arg.reclaim = NFS_LOCK_RECLAIM; + + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + ret = rpc_wait_for_completion_task(task); + if (ret == 0) { + ret = data->rpc_status; + if (ret) + nfs4_handle_setlk_error(data->server, data->lsp, + data->arg.new_lock_owner, ret); + } else + data->cancelled = true; + trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret); + rpc_put_task(task); + dprintk("%s: ret = %d\n", __func__, ret); + return ret; +} + +int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_exception exception = { + .inode = state->inode, + }; + int err; + + do { + /* Cache the lock if possible... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; + err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_RECLAIM); + if (err != -NFS4ERR_DELAY) + break; + nfs4_handle_exception(server, err, &exception); + } while (exception.retry); + return err; +} + +int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_exception exception = { + .inode = state->inode, + }; + int err; + + err = nfs4_set_lock_state(state, request); + if (err != 0) + return err; + if (!recover_lost_locks) { + set_bit(NFS_LOCK_LOST, &request->fl_u.nfs4_fl.owner->ls_flags); + return 0; + } + do { + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; + err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_EXPIRED); + switch (err) { + default: + goto out; + case -NFS4ERR_GRACE: + case -NFS4ERR_DELAY: + nfs4_handle_exception(server, err, &exception); + err = 0; + } + } while (exception.retry); +out: + return err; +} + +static int nfs41_lock_expired(struct nfs4_state *state, struct file_lock *request) +{ + struct nfs4_lock_state *lsp; + int status; + + status = nfs4_set_lock_state(state, request); + if (status != 0) + return status; + lsp = request->fl_u.nfs4_fl.owner; + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) || + test_bit(NFS_LOCK_LOST, &lsp->ls_flags)) + return 0; + return nfs4_lock_expired(state, request); +} + +static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct nfs_inode *nfsi = NFS_I(state->inode); + struct nfs4_state_owner *sp = state->owner; + unsigned char flags = request->c.flc_flags; + int status; + + request->c.flc_flags |= FL_ACCESS; + status = locks_lock_inode_wait(state->inode, request); + if (status < 0) + goto out; + mutex_lock(&sp->so_delegreturn_mutex); + down_read(&nfsi->rwsem); + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + /* Yes: cache locks! */ + /* ...but avoid races with delegation recall... */ + request->c.flc_flags = flags & ~FL_SLEEP; + status = locks_lock_inode_wait(state->inode, request); + up_read(&nfsi->rwsem); + mutex_unlock(&sp->so_delegreturn_mutex); + goto out; + } + up_read(&nfsi->rwsem); + mutex_unlock(&sp->so_delegreturn_mutex); + status = _nfs4_do_setlk(state, cmd, request, NFS_LOCK_NEW); + if (status) + goto out; + + down_read(&nfsi->rwsem); + request->c.flc_flags &= ~(FL_SLEEP | FL_ACCESS); + status = locks_lock_inode_wait(state->inode, request); + up_read(&nfsi->rwsem); +out: + request->c.flc_flags = flags; + return status; +} + +static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct nfs4_exception exception = { + .state = state, + .inode = state->inode, + .interruptible = true, + }; + int err; + + do { + err = _nfs4_proc_setlk(state, cmd, request); + if (err == -NFS4ERR_DENIED) + err = -EAGAIN; + err = nfs4_handle_exception(NFS_SERVER(state->inode), + err, &exception); + } while (exception.retry); + return err; +} + +#define NFS4_LOCK_MINTIMEOUT (1 * HZ) +#define NFS4_LOCK_MAXTIMEOUT (30 * HZ) + +static int +nfs4_retry_setlk_simple(struct nfs4_state *state, int cmd, + struct file_lock *request) +{ + int status = -ERESTARTSYS; + unsigned long timeout = NFS4_LOCK_MINTIMEOUT; + + while(!signalled()) { + status = nfs4_proc_setlk(state, cmd, request); + if ((status != -EAGAIN) || IS_SETLK(cmd)) + break; + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); + schedule_timeout(timeout); + timeout *= 2; + timeout = min_t(unsigned long, NFS4_LOCK_MAXTIMEOUT, timeout); + status = -ERESTARTSYS; + } + return status; +} + +struct nfs4_lock_waiter { + struct inode *inode; + struct nfs_lowner owner; + wait_queue_entry_t wait; +}; + +static int +nfs4_wake_lock_waiter(wait_queue_entry_t *wait, unsigned int mode, int flags, void *key) +{ + struct nfs4_lock_waiter *waiter = + container_of(wait, struct nfs4_lock_waiter, wait); + + /* NULL key means to wake up everyone */ + if (key) { + struct cb_notify_lock_args *cbnl = key; + struct nfs_lowner *lowner = &cbnl->cbnl_owner, + *wowner = &waiter->owner; + + /* Only wake if the callback was for the same owner. */ + if (lowner->id != wowner->id || lowner->s_dev != wowner->s_dev) + return 0; + + /* Make sure it's for the right inode */ + if (nfs_compare_fh(NFS_FH(waiter->inode), &cbnl->cbnl_fh)) + return 0; + } + + return woken_wake_function(wait, mode, flags, key); +} + +static int +nfs4_retry_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner; + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs_client *clp = server->nfs_client; + wait_queue_head_t *q = &clp->cl_lock_waitq; + struct nfs4_lock_waiter waiter = { + .inode = state->inode, + .owner = { .clientid = clp->cl_clientid, + .id = lsp->ls_seqid.owner_id, + .s_dev = server->s_dev }, + }; + int status; + + /* Don't bother with waitqueue if we don't expect a callback */ + if (!test_bit(NFS_STATE_MAY_NOTIFY_LOCK, &state->flags)) + return nfs4_retry_setlk_simple(state, cmd, request); + + init_wait(&waiter.wait); + waiter.wait.func = nfs4_wake_lock_waiter; + add_wait_queue(q, &waiter.wait); + + do { + status = nfs4_proc_setlk(state, cmd, request); + if (status != -EAGAIN || IS_SETLK(cmd)) + break; + + status = -ERESTARTSYS; + wait_woken(&waiter.wait, TASK_INTERRUPTIBLE|TASK_FREEZABLE, + NFS4_LOCK_MAXTIMEOUT); + } while (!signalled()); + + remove_wait_queue(q, &waiter.wait); + + return status; +} + +static int +nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request) +{ + struct nfs_open_context *ctx; + struct nfs4_state *state; + int status; + + /* verify open state */ + ctx = nfs_file_open_context(filp); + state = ctx->state; + + if (IS_GETLK(cmd)) { + if (state != NULL) + return nfs4_proc_getlk(state, F_GETLK, request); + return 0; + } + + if (!(IS_SETLK(cmd) || IS_SETLKW(cmd))) + return -EINVAL; + + if (lock_is_unlock(request)) { + if (state != NULL) + return nfs4_proc_unlck(state, cmd, request); + return 0; + } + + if (state == NULL) + return -ENOLCK; + + if ((request->c.flc_flags & FL_POSIX) && + !test_bit(NFS_STATE_POSIX_LOCKS, &state->flags)) + return -ENOLCK; + + /* + * Don't rely on the VFS having checked the file open mode, + * since it won't do this for flock() locks. + */ + switch (request->c.flc_type) { + case F_RDLCK: + if (!(filp->f_mode & FMODE_READ)) + return -EBADF; + break; + case F_WRLCK: + if (!(filp->f_mode & FMODE_WRITE)) + return -EBADF; + } + + status = nfs4_set_lock_state(state, request); + if (status != 0) + return status; + + return nfs4_retry_setlk(state, cmd, request); +} + +static int nfs4_delete_lease(struct file *file, void **priv) +{ + return generic_setlease(file, F_UNLCK, NULL, priv); +} + +static int nfs4_add_lease(struct file *file, int arg, struct file_lease **lease, + void **priv) +{ + struct inode *inode = file_inode(file); + fmode_t type = arg == F_RDLCK ? FMODE_READ : FMODE_WRITE; + int ret; + + /* No delegation, no lease */ + if (!nfs4_have_delegation(inode, type, 0)) + return -EAGAIN; + ret = generic_setlease(file, arg, lease, priv); + if (ret || nfs4_have_delegation(inode, type, 0)) + return ret; + /* We raced with a delegation return */ + nfs4_delete_lease(file, priv); + return -EAGAIN; +} + +int nfs4_proc_setlease(struct file *file, int arg, struct file_lease **lease, + void **priv) +{ + switch (arg) { + case F_RDLCK: + case F_WRLCK: + return nfs4_add_lease(file, arg, lease, priv); + case F_UNLCK: + return nfs4_delete_lease(file, priv); + default: + return -EINVAL; + } +} + +int nfs4_lock_delegation_recall(struct file_lock *fl, struct nfs4_state *state, const nfs4_stateid *stateid) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + int err; + + err = nfs4_set_lock_state(state, fl); + if (err != 0) + return err; + do { + err = _nfs4_do_setlk(state, F_SETLK, fl, NFS_LOCK_NEW); + if (err != -NFS4ERR_DELAY && err != -NFS4ERR_GRACE) + break; + ssleep(1); + } while (err == -NFS4ERR_DELAY || err == -NFSERR_GRACE); + return nfs4_handle_delegation_recall_error(server, state, stateid, fl, err); +} + +#define XATTR_NAME_NFSV4_ACL "system.nfs4_acl" + +static int nfs4_xattr_set_nfs4_acl(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + return nfs4_proc_set_acl(inode, buf, buflen, NFS4ACL_ACL); +} + +static int nfs4_xattr_get_nfs4_acl(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + return nfs4_proc_get_acl(inode, buf, buflen, NFS4ACL_ACL); +} + +static bool nfs4_xattr_list_nfs4_acl(struct dentry *dentry) +{ + return nfs4_server_supports_acls(NFS_SB(dentry->d_sb), NFS4ACL_ACL); +} + +#define XATTR_NAME_NFSV4_DACL "system.nfs4_dacl" + +static int nfs4_xattr_set_nfs4_dacl(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + return nfs4_proc_set_acl(inode, buf, buflen, NFS4ACL_DACL); +} + +static int nfs4_xattr_get_nfs4_dacl(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + return nfs4_proc_get_acl(inode, buf, buflen, NFS4ACL_DACL); +} + +static bool nfs4_xattr_list_nfs4_dacl(struct dentry *dentry) +{ + return nfs4_server_supports_acls(NFS_SB(dentry->d_sb), NFS4ACL_DACL); +} + +#define XATTR_NAME_NFSV4_SACL "system.nfs4_sacl" + +static int nfs4_xattr_set_nfs4_sacl(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + return nfs4_proc_set_acl(inode, buf, buflen, NFS4ACL_SACL); +} + +static int nfs4_xattr_get_nfs4_sacl(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + return nfs4_proc_get_acl(inode, buf, buflen, NFS4ACL_SACL); +} + +static bool nfs4_xattr_list_nfs4_sacl(struct dentry *dentry) +{ + return nfs4_server_supports_acls(NFS_SB(dentry->d_sb), NFS4ACL_SACL); +} + +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + +static int nfs4_xattr_set_nfs4_label(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + if (security_ismaclabel(key)) + return nfs4_set_security_label(inode, buf, buflen); + + return -EOPNOTSUPP; +} + +static int nfs4_xattr_get_nfs4_label(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + if (security_ismaclabel(key)) + return nfs4_get_security_label(inode, buf, buflen); + return -EOPNOTSUPP; +} + +static const struct xattr_handler nfs4_xattr_nfs4_label_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .get = nfs4_xattr_get_nfs4_label, + .set = nfs4_xattr_set_nfs4_label, +}; + +#endif + +#ifdef CONFIG_NFS_V4_2 +static int nfs4_xattr_set_nfs4_user(const struct xattr_handler *handler, + struct mnt_idmap *idmap, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + u32 mask; + int ret; + + if (!nfs_server_capable(inode, NFS_CAP_XATTR)) + return -EOPNOTSUPP; + + /* + * There is no mapping from the MAY_* flags to the NFS_ACCESS_XA* + * flags right now. Handling of xattr operations use the normal + * file read/write permissions. + * + * Just in case the server has other ideas (which RFC 8276 allows), + * do a cached access check for the XA* flags to possibly avoid + * doing an RPC and getting EACCES back. + */ + if (!nfs_access_get_cached(inode, current_cred(), &mask, true)) { + if (!(mask & NFS_ACCESS_XAWRITE)) + return -EACCES; + } + + if (buf == NULL) { + ret = nfs42_proc_removexattr(inode, key); + if (!ret) + nfs4_xattr_cache_remove(inode, key); + } else { + ret = nfs42_proc_setxattr(inode, key, buf, buflen, flags); + if (!ret) + nfs4_xattr_cache_add(inode, key, buf, NULL, buflen); + } + + return ret; +} + +static int nfs4_xattr_get_nfs4_user(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + u32 mask; + ssize_t ret; + + if (!nfs_server_capable(inode, NFS_CAP_XATTR)) + return -EOPNOTSUPP; + + if (!nfs_access_get_cached(inode, current_cred(), &mask, true)) { + if (!(mask & NFS_ACCESS_XAREAD)) + return -EACCES; + } + + ret = nfs_revalidate_inode(inode, NFS_INO_INVALID_CHANGE); + if (ret) + return ret; + + ret = nfs4_xattr_cache_get(inode, key, buf, buflen); + if (ret >= 0 || (ret < 0 && ret != -ENOENT)) + return ret; + + ret = nfs42_proc_getxattr(inode, key, buf, buflen); + + return ret; +} + +static ssize_t +nfs4_listxattr_nfs4_user(struct inode *inode, char *list, size_t list_len) +{ + u64 cookie; + bool eof; + ssize_t ret, size; + char *buf; + size_t buflen; + u32 mask; + + if (!nfs_server_capable(inode, NFS_CAP_XATTR)) + return 0; + + if (!nfs_access_get_cached(inode, current_cred(), &mask, true)) { + if (!(mask & NFS_ACCESS_XALIST)) + return 0; + } + + ret = nfs_revalidate_inode(inode, NFS_INO_INVALID_CHANGE); + if (ret) + return ret; + + ret = nfs4_xattr_cache_list(inode, list, list_len); + if (ret >= 0 || (ret < 0 && ret != -ENOENT)) + return ret; + + cookie = 0; + eof = false; + buflen = list_len ? list_len : XATTR_LIST_MAX; + buf = list_len ? list : NULL; + size = 0; + + while (!eof) { + ret = nfs42_proc_listxattrs(inode, buf, buflen, + &cookie, &eof); + if (ret < 0) + return ret; + + if (list_len) { + buf += ret; + buflen -= ret; + } + size += ret; + } + + if (list_len) + nfs4_xattr_cache_set_list(inode, list, size); + + return size; +} + +#else + +static ssize_t +nfs4_listxattr_nfs4_user(struct inode *inode, char *list, size_t list_len) +{ + return 0; +} +#endif /* CONFIG_NFS_V4_2 */ + +/* + * nfs_fhget will use either the mounted_on_fileid or the fileid + */ +static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr) +{ + if (!(((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) || + (fattr->valid & NFS_ATTR_FATTR_FILEID)) && + (fattr->valid & NFS_ATTR_FATTR_FSID) && + (fattr->valid & NFS_ATTR_FATTR_V4_LOCATIONS))) + return; + + fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE | + NFS_ATTR_FATTR_NLINK | NFS_ATTR_FATTR_V4_REFERRAL; + fattr->mode = S_IFDIR | S_IRUGO | S_IXUGO; + fattr->nlink = 2; +} + +static int _nfs4_proc_fs_locations(struct rpc_clnt *client, struct inode *dir, + const struct qstr *name, + struct nfs4_fs_locations *fs_locations, + struct page *page) +{ + struct nfs_server *server = NFS_SERVER(dir); + u32 bitmask[3]; + struct nfs4_fs_locations_arg args = { + .dir_fh = NFS_FH(dir), + .name = name, + .page = page, + .bitmask = bitmask, + }; + struct nfs4_fs_locations_res res = { + .fs_locations = fs_locations, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS], + .rpc_argp = &args, + .rpc_resp = &res, + }; + int status; + + dprintk("%s: start\n", __func__); + + bitmask[0] = nfs4_fattr_bitmap[0] | FATTR4_WORD0_FS_LOCATIONS; + bitmask[1] = nfs4_fattr_bitmap[1]; + + /* Ask for the fileid of the absent filesystem if mounted_on_fileid + * is not supported */ + if (NFS_SERVER(dir)->attr_bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID) + bitmask[0] &= ~FATTR4_WORD0_FILEID; + else + bitmask[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; + + nfs_fattr_init(fs_locations->fattr); + fs_locations->server = server; + fs_locations->nlocations = 0; + status = nfs4_call_sync(client, server, &msg, &args.seq_args, &res.seq_res, 0); + dprintk("%s: returned status = %d\n", __func__, status); + return status; +} + +int nfs4_proc_fs_locations(struct rpc_clnt *client, struct inode *dir, + const struct qstr *name, + struct nfs4_fs_locations *fs_locations, + struct page *page) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs4_proc_fs_locations(client, dir, name, + fs_locations, page); + trace_nfs4_get_fs_locations(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + return err; +} + +/* + * This operation also signals the server that this client is + * performing migration recovery. The server can stop asserting + * SEQ4_STATUS_LEASE_MOVED for this client. The client ID + * performing this operation is identified in the SEQUENCE + * operation in this compound. + * + * When the client supports GETATTR(fs_locations_info), it can + * be plumbed in here. + */ +static int _nfs41_proc_get_locations(struct nfs_server *server, + struct nfs_fh *fhandle, + struct nfs4_fs_locations *locations, + struct page *page, const struct cred *cred) +{ + struct rpc_clnt *clnt = server->client; + struct nfs_client *clp = server->nfs_client; + u32 bitmask[2] = { + [0] = FATTR4_WORD0_FSID | FATTR4_WORD0_FS_LOCATIONS, + }; + struct nfs4_fs_locations_arg args = { + .fh = fhandle, + .page = page, + .bitmask = bitmask, + .migration = 1, /* skip LOOKUP */ + }; + struct nfs4_fs_locations_res res = { + .fs_locations = locations, + .migration = 1, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + struct nfs4_call_sync_data data = { + .seq_server = server, + .seq_args = &args.seq_args, + .seq_res = &res.seq_res, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clnt, + .rpc_message = &msg, + .callback_ops = clp->cl_mvops->call_sync_ops, + .callback_data = &data, + .flags = RPC_TASK_NO_ROUND_ROBIN, + }; + int status; + + nfs_fattr_init(locations->fattr); + locations->server = server; + locations->nlocations = 0; + + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 1); + status = nfs4_call_sync_custom(&task_setup_data); + if (status == NFS4_OK && + res.seq_res.sr_status_flags & SEQ4_STATUS_LEASE_MOVED) + status = -NFS4ERR_LEASE_MOVED; + return status; +} + +/** + * nfs4_proc_get_locations - discover locations for a migrated FSID + * @server: pointer to nfs_server to process + * @fhandle: pointer to the kernel NFS client file handle + * @locations: result of query + * @page: buffer + * @cred: credential to use for this operation + * + * Returns NFS4_OK on success, a negative NFS4ERR status code if the + * operation failed, or a negative errno if a local error occurred. + * + * On success, "locations" is filled in, but if the server has + * no locations information, NFS_ATTR_FATTR_V4_LOCATIONS is not + * asserted. + * + * -NFS4ERR_LEASE_MOVED is returned if the server still has leases + * from this client that require migration recovery. + */ +int nfs4_proc_get_locations(struct nfs_server *server, + struct nfs_fh *fhandle, + struct nfs4_fs_locations *locations, + struct page *page, const struct cred *cred) +{ + struct nfs_client *clp = server->nfs_client; + const struct nfs4_mig_recovery_ops *ops = + clp->cl_mvops->mig_recovery_ops; + struct nfs4_exception exception = { + .interruptible = true, + }; + int status; + + dprintk("%s: FSID %llx:%llx on \"%s\"\n", __func__, + (unsigned long long)server->fsid.major, + (unsigned long long)server->fsid.minor, + clp->cl_hostname); + nfs_display_fhandle(fhandle, __func__); + + do { + status = ops->get_locations(server, fhandle, locations, page, + cred); + if (status != -NFS4ERR_DELAY) + break; + nfs4_handle_exception(server, status, &exception); + } while (exception.retry); + return status; +} + +/* + * This operation also signals the server that this client is + * performing "lease moved" recovery. The server can stop asserting + * SEQ4_STATUS_LEASE_MOVED for this client. The client ID performing + * this operation is identified in the SEQUENCE operation in this + * compound. + */ +static int _nfs41_proc_fsid_present(struct inode *inode, const struct cred *cred) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct rpc_clnt *clnt = server->client; + struct nfs4_fsid_present_arg args = { + .fh = NFS_FH(inode), + }; + struct nfs4_fsid_present_res res = { + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSID_PRESENT], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + int status; + + res.fh = nfs_alloc_fhandle(); + if (res.fh == NULL) + return -ENOMEM; + + nfs4_init_sequence(server->nfs_client, &args.seq_args, &res.seq_res, 0, 1); + status = nfs4_call_sync_sequence(clnt, server, &msg, + &args.seq_args, &res.seq_res); + nfs_free_fhandle(res.fh); + if (status == NFS4_OK && + res.seq_res.sr_status_flags & SEQ4_STATUS_LEASE_MOVED) + status = -NFS4ERR_LEASE_MOVED; + return status; +} + +/** + * nfs4_proc_fsid_present - Is this FSID present or absent on server? + * @inode: inode on FSID to check + * @cred: credential to use for this operation + * + * Server indicates whether the FSID is present, moved, or not + * recognized. This operation is necessary to clear a LEASE_MOVED + * condition for this client ID. + * + * Returns NFS4_OK if the FSID is present on this server, + * -NFS4ERR_MOVED if the FSID is no longer present, a negative + * NFS4ERR code if some error occurred on the server, or a + * negative errno if a local failure occurred. + */ +int nfs4_proc_fsid_present(struct inode *inode, const struct cred *cred) +{ + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_client *clp = server->nfs_client; + const struct nfs4_mig_recovery_ops *ops = + clp->cl_mvops->mig_recovery_ops; + struct nfs4_exception exception = { + .interruptible = true, + }; + int status; + + dprintk("%s: FSID %llx:%llx on \"%s\"\n", __func__, + (unsigned long long)server->fsid.major, + (unsigned long long)server->fsid.minor, + clp->cl_hostname); + nfs_display_fhandle(NFS_FH(inode), __func__); + + do { + status = ops->fsid_present(inode, cred); + if (status != -NFS4ERR_DELAY) + break; + nfs4_handle_exception(server, status, &exception); + } while (exception.retry); + return status; +} + +/* + * If 'use_integrity' is true and the state managment nfs_client + * cl_rpcclient is using krb5i/p, use the integrity protected cl_rpcclient + * and the machine credential as per RFC3530bis and RFC5661 Security + * Considerations sections. Otherwise, just use the user cred with the + * filesystem's rpc_client. + */ +static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors, bool use_integrity) +{ + int status; + struct rpc_clnt *clnt = NFS_SERVER(dir)->client; + struct nfs_client *clp = NFS_SERVER(dir)->nfs_client; + struct nfs4_secinfo_arg args = { + .dir_fh = NFS_FH(dir), + .name = name, + }; + struct nfs4_secinfo_res res = { + .flavors = flavors, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SECINFO], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct nfs4_call_sync_data data = { + .seq_server = NFS_SERVER(dir), + .seq_args = &args.seq_args, + .seq_res = &res.seq_res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = clnt, + .rpc_message = &msg, + .callback_ops = clp->cl_mvops->call_sync_ops, + .callback_data = &data, + .flags = RPC_TASK_NO_ROUND_ROBIN, + }; + const struct cred *cred = NULL; + + if (use_integrity) { + clnt = clp->cl_rpcclient; + task_setup.rpc_client = clnt; + + cred = nfs4_get_clid_cred(clp); + msg.rpc_cred = cred; + } + + dprintk("NFS call secinfo %s\n", name->name); + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_SECINFO, &clnt, &msg); + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 0); + status = nfs4_call_sync_custom(&task_setup); + + dprintk("NFS reply secinfo: %d\n", status); + + put_cred(cred); + return status; +} + +int nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, + struct nfs4_secinfo_flavors *flavors) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = -NFS4ERR_WRONGSEC; + + /* try to use integrity protection with machine cred */ + if (_nfs4_is_integrity_protected(NFS_SERVER(dir)->nfs_client)) + err = _nfs4_proc_secinfo(dir, name, flavors, true); + + /* + * if unable to use integrity protection, or SECINFO with + * integrity protection returns NFS4ERR_WRONGSEC (which is + * disallowed by spec, but exists in deployed servers) use + * the current filesystem's rpc_client and the user cred. + */ + if (err == -NFS4ERR_WRONGSEC) + err = _nfs4_proc_secinfo(dir, name, flavors, false); + + trace_nfs4_secinfo(dir, name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + return err; +} + +/* + * Check the exchange flags returned by the server for invalid flags, having + * both PNFS and NON_PNFS flags set, and not having one of NON_PNFS, PNFS, or + * DS flags set. + */ +static int nfs4_check_cl_exchange_flags(u32 flags, u32 version) +{ + if (version >= 2 && (flags & ~EXCHGID4_2_FLAG_MASK_R)) + goto out_inval; + else if (version < 2 && (flags & ~EXCHGID4_FLAG_MASK_R)) + goto out_inval; + if ((flags & EXCHGID4_FLAG_USE_PNFS_MDS) && + (flags & EXCHGID4_FLAG_USE_NON_PNFS)) + goto out_inval; + if (!(flags & (EXCHGID4_FLAG_MASK_PNFS))) + goto out_inval; + return NFS_OK; +out_inval: + return -NFS4ERR_INVAL; +} + +static bool +nfs41_same_server_scope(struct nfs41_server_scope *a, + struct nfs41_server_scope *b) +{ + if (a->server_scope_sz != b->server_scope_sz) + return false; + return memcmp(a->server_scope, b->server_scope, a->server_scope_sz) == 0; +} + +static void +nfs4_bind_one_conn_to_session_done(struct rpc_task *task, void *calldata) +{ + struct nfs41_bind_conn_to_session_args *args = task->tk_msg.rpc_argp; + struct nfs41_bind_conn_to_session_res *res = task->tk_msg.rpc_resp; + struct nfs_client *clp = args->client; + + switch (task->tk_status) { + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + nfs4_schedule_session_recovery(clp->cl_session, + task->tk_status); + return; + } + if (args->dir == NFS4_CDFC4_FORE_OR_BOTH && + res->dir != NFS4_CDFS4_BOTH) { + rpc_task_close_connection(task); + if (args->retries++ < MAX_BIND_CONN_TO_SESSION_RETRIES) + rpc_restart_call(task); + } +} + +static const struct rpc_call_ops nfs4_bind_one_conn_to_session_ops = { + .rpc_call_done = nfs4_bind_one_conn_to_session_done, +}; + +/* + * nfs4_proc_bind_one_conn_to_session() + * + * The 4.1 client currently uses the same TCP connection for the + * fore and backchannel. + */ +static +int nfs4_proc_bind_one_conn_to_session(struct rpc_clnt *clnt, + struct rpc_xprt *xprt, + struct nfs_client *clp, + const struct cred *cred) +{ + int status; + struct nfs41_bind_conn_to_session_args args = { + .client = clp, + .dir = NFS4_CDFC4_FORE_OR_BOTH, + .retries = 0, + }; + struct nfs41_bind_conn_to_session_res res; + struct rpc_message msg = { + .rpc_proc = + &nfs4_procedures[NFSPROC4_CLNT_BIND_CONN_TO_SESSION], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clnt, + .rpc_xprt = xprt, + .callback_ops = &nfs4_bind_one_conn_to_session_ops, + .rpc_message = &msg, + .flags = RPC_TASK_TIMEOUT, + }; + struct rpc_task *task; + + nfs4_copy_sessionid(&args.sessionid, &clp->cl_session->sess_id); + if (!(clp->cl_session->flags & SESSION4_BACK_CHAN)) + args.dir = NFS4_CDFC4_FORE; + + /* Do not set the backchannel flag unless this is clnt->cl_xprt */ + if (xprt != rcu_access_pointer(clnt->cl_xprt)) + args.dir = NFS4_CDFC4_FORE; + + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) { + status = task->tk_status; + rpc_put_task(task); + } else + status = PTR_ERR(task); + trace_nfs4_bind_conn_to_session(clp, status); + if (status == 0) { + if (memcmp(res.sessionid.data, + clp->cl_session->sess_id.data, NFS4_MAX_SESSIONID_LEN)) { + dprintk("NFS: %s: Session ID mismatch\n", __func__); + return -EIO; + } + if ((res.dir & args.dir) != res.dir || res.dir == 0) { + dprintk("NFS: %s: Unexpected direction from server\n", + __func__); + return -EIO; + } + if (res.use_conn_in_rdma_mode != args.use_conn_in_rdma_mode) { + dprintk("NFS: %s: Server returned RDMA mode = true\n", + __func__); + return -EIO; + } + } + + return status; +} + +struct rpc_bind_conn_calldata { + struct nfs_client *clp; + const struct cred *cred; +}; + +static int +nfs4_proc_bind_conn_to_session_callback(struct rpc_clnt *clnt, + struct rpc_xprt *xprt, + void *calldata) +{ + struct rpc_bind_conn_calldata *p = calldata; + + return nfs4_proc_bind_one_conn_to_session(clnt, xprt, p->clp, p->cred); +} + +int nfs4_proc_bind_conn_to_session(struct nfs_client *clp, const struct cred *cred) +{ + struct rpc_bind_conn_calldata data = { + .clp = clp, + .cred = cred, + }; + return rpc_clnt_iterate_for_each_xprt(clp->cl_rpcclient, + nfs4_proc_bind_conn_to_session_callback, &data); +} + +/* + * Minimum set of SP4_MACH_CRED operations from RFC 5661 in the enforce map + * and operations we'd like to see to enable certain features in the allow map + */ +static const struct nfs41_state_protection nfs4_sp4_mach_cred_request = { + .how = SP4_MACH_CRED, + .enforce.u.words = { + [1] = 1 << (OP_BIND_CONN_TO_SESSION - 32) | + 1 << (OP_EXCHANGE_ID - 32) | + 1 << (OP_CREATE_SESSION - 32) | + 1 << (OP_DESTROY_SESSION - 32) | + 1 << (OP_DESTROY_CLIENTID - 32) + }, + .allow.u.words = { + [0] = 1 << (OP_CLOSE) | + 1 << (OP_OPEN_DOWNGRADE) | + 1 << (OP_LOCKU) | + 1 << (OP_DELEGRETURN) | + 1 << (OP_COMMIT), + [1] = 1 << (OP_SECINFO - 32) | + 1 << (OP_SECINFO_NO_NAME - 32) | + 1 << (OP_LAYOUTRETURN - 32) | + 1 << (OP_TEST_STATEID - 32) | + 1 << (OP_FREE_STATEID - 32) | + 1 << (OP_WRITE - 32) + } +}; + +/* + * Select the state protection mode for client `clp' given the server results + * from exchange_id in `sp'. + * + * Returns 0 on success, negative errno otherwise. + */ +static int nfs4_sp4_select_mode(struct nfs_client *clp, + struct nfs41_state_protection *sp) +{ + static const u32 supported_enforce[NFS4_OP_MAP_NUM_WORDS] = { + [1] = 1 << (OP_BIND_CONN_TO_SESSION - 32) | + 1 << (OP_EXCHANGE_ID - 32) | + 1 << (OP_CREATE_SESSION - 32) | + 1 << (OP_DESTROY_SESSION - 32) | + 1 << (OP_DESTROY_CLIENTID - 32) + }; + unsigned long flags = 0; + unsigned int i; + int ret = 0; + + if (sp->how == SP4_MACH_CRED) { + /* Print state protect result */ + dfprintk(MOUNT, "Server SP4_MACH_CRED support:\n"); + for (i = 0; i <= LAST_NFS4_OP; i++) { + if (test_bit(i, sp->enforce.u.longs)) + dfprintk(MOUNT, " enforce op %d\n", i); + if (test_bit(i, sp->allow.u.longs)) + dfprintk(MOUNT, " allow op %d\n", i); + } + + /* make sure nothing is on enforce list that isn't supported */ + for (i = 0; i < NFS4_OP_MAP_NUM_WORDS; i++) { + if (sp->enforce.u.words[i] & ~supported_enforce[i]) { + dfprintk(MOUNT, "sp4_mach_cred: disabled\n"); + ret = -EINVAL; + goto out; + } + } + + /* + * Minimal mode - state operations are allowed to use machine + * credential. Note this already happens by default, so the + * client doesn't have to do anything more than the negotiation. + * + * NOTE: we don't care if EXCHANGE_ID is in the list - + * we're already using the machine cred for exchange_id + * and will never use a different cred. + */ + if (test_bit(OP_BIND_CONN_TO_SESSION, sp->enforce.u.longs) && + test_bit(OP_CREATE_SESSION, sp->enforce.u.longs) && + test_bit(OP_DESTROY_SESSION, sp->enforce.u.longs) && + test_bit(OP_DESTROY_CLIENTID, sp->enforce.u.longs)) { + dfprintk(MOUNT, "sp4_mach_cred:\n"); + dfprintk(MOUNT, " minimal mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_MINIMAL, &flags); + } else { + dfprintk(MOUNT, "sp4_mach_cred: disabled\n"); + ret = -EINVAL; + goto out; + } + + if (test_bit(OP_CLOSE, sp->allow.u.longs) && + test_bit(OP_OPEN_DOWNGRADE, sp->allow.u.longs) && + test_bit(OP_DELEGRETURN, sp->allow.u.longs) && + test_bit(OP_LOCKU, sp->allow.u.longs)) { + dfprintk(MOUNT, " cleanup mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_CLEANUP, &flags); + } + + if (test_bit(OP_LAYOUTRETURN, sp->allow.u.longs)) { + dfprintk(MOUNT, " pnfs cleanup mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_PNFS_CLEANUP, &flags); + } + + if (test_bit(OP_SECINFO, sp->allow.u.longs) && + test_bit(OP_SECINFO_NO_NAME, sp->allow.u.longs)) { + dfprintk(MOUNT, " secinfo mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_SECINFO, &flags); + } + + if (test_bit(OP_TEST_STATEID, sp->allow.u.longs) && + test_bit(OP_FREE_STATEID, sp->allow.u.longs)) { + dfprintk(MOUNT, " stateid mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_STATEID, &flags); + } + + if (test_bit(OP_WRITE, sp->allow.u.longs)) { + dfprintk(MOUNT, " write mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_WRITE, &flags); + } + + if (test_bit(OP_COMMIT, sp->allow.u.longs)) { + dfprintk(MOUNT, " commit mode enabled\n"); + __set_bit(NFS_SP4_MACH_CRED_COMMIT, &flags); + } + } +out: + clp->cl_sp4_flags = flags; + return ret; +} + +struct nfs41_exchange_id_data { + struct nfs41_exchange_id_res res; + struct nfs41_exchange_id_args args; +}; + +static void nfs4_exchange_id_release(void *data) +{ + struct nfs41_exchange_id_data *cdata = + (struct nfs41_exchange_id_data *)data; + + nfs_put_client(cdata->args.client); + kfree(cdata->res.impl_id); + kfree(cdata->res.server_scope); + kfree(cdata->res.server_owner); + kfree(cdata); +} + +static const struct rpc_call_ops nfs4_exchange_id_call_ops = { + .rpc_release = nfs4_exchange_id_release, +}; + +/* + * _nfs4_proc_exchange_id() + * + * Wrapper for EXCHANGE_ID operation. + */ +static struct rpc_task * +nfs4_run_exchange_id(struct nfs_client *clp, const struct cred *cred, + u32 sp4_how, struct rpc_xprt *xprt) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_EXCHANGE_ID], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clp->cl_rpcclient, + .callback_ops = &nfs4_exchange_id_call_ops, + .rpc_message = &msg, + .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN, + }; + struct nfs41_exchange_id_data *calldata; + int status; + + if (!refcount_inc_not_zero(&clp->cl_count)) + return ERR_PTR(-EIO); + + status = -ENOMEM; + calldata = kzalloc_obj(*calldata, GFP_NOFS); + if (!calldata) + goto out; + + nfs4_init_boot_verifier(clp, &calldata->args.verifier); + + status = nfs4_init_uniform_client_string(clp); + if (status) + goto out_calldata; + + calldata->res.server_owner = kzalloc_obj(struct nfs41_server_owner, + GFP_NOFS); + status = -ENOMEM; + if (unlikely(calldata->res.server_owner == NULL)) + goto out_calldata; + + calldata->res.server_scope = kzalloc_obj(struct nfs41_server_scope, + GFP_NOFS); + if (unlikely(calldata->res.server_scope == NULL)) + goto out_server_owner; + + calldata->res.impl_id = kzalloc_obj(struct nfs41_impl_id, GFP_NOFS); + if (unlikely(calldata->res.impl_id == NULL)) + goto out_server_scope; + + switch (sp4_how) { + case SP4_NONE: + calldata->args.state_protect.how = SP4_NONE; + break; + + case SP4_MACH_CRED: + calldata->args.state_protect = nfs4_sp4_mach_cred_request; + break; + + default: + /* unsupported! */ + WARN_ON_ONCE(1); + status = -EINVAL; + goto out_impl_id; + } + if (xprt) { + task_setup_data.rpc_xprt = xprt; + task_setup_data.flags |= RPC_TASK_SOFTCONN; + memcpy(calldata->args.verifier.data, clp->cl_confirm.data, + sizeof(calldata->args.verifier.data)); + } + calldata->args.client = clp; + calldata->args.flags = EXCHGID4_FLAG_SUPP_MOVED_REFER | + EXCHGID4_FLAG_BIND_PRINC_STATEID; +#ifdef CONFIG_NFS_V4_1_MIGRATION + calldata->args.flags |= EXCHGID4_FLAG_SUPP_MOVED_MIGR; +#endif + if (test_bit(NFS_CS_PNFS, &clp->cl_flags)) + calldata->args.flags |= EXCHGID4_FLAG_USE_PNFS_DS; + msg.rpc_argp = &calldata->args; + msg.rpc_resp = &calldata->res; + task_setup_data.callback_data = calldata; + + return rpc_run_task(&task_setup_data); + +out_impl_id: + kfree(calldata->res.impl_id); +out_server_scope: + kfree(calldata->res.server_scope); +out_server_owner: + kfree(calldata->res.server_owner); +out_calldata: + kfree(calldata); +out: + nfs_put_client(clp); + return ERR_PTR(status); +} + +/* + * _nfs4_proc_exchange_id() + * + * Wrapper for EXCHANGE_ID operation. + */ +static int _nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cred, + u32 sp4_how) +{ + struct rpc_task *task; + struct nfs41_exchange_id_args *argp; + struct nfs41_exchange_id_res *resp; + unsigned long now = jiffies; + int status; + + task = nfs4_run_exchange_id(clp, cred, sp4_how, NULL); + if (IS_ERR(task)) + return PTR_ERR(task); + + argp = task->tk_msg.rpc_argp; + resp = task->tk_msg.rpc_resp; + status = task->tk_status; + if (status != 0) + goto out; + + status = nfs4_check_cl_exchange_flags(resp->flags, + clp->cl_mvops->minor_version); + if (status != 0) + goto out; + + status = nfs4_sp4_select_mode(clp, &resp->state_protect); + if (status != 0) + goto out; + + do_renew_lease(clp, now); + + clp->cl_clientid = resp->clientid; + clp->cl_exchange_flags = resp->flags; + clp->cl_seqid = resp->seqid; + /* Client ID is not confirmed */ + if (!(resp->flags & EXCHGID4_FLAG_CONFIRMED_R)) + clear_bit(NFS4_SESSION_ESTABLISHED, + &clp->cl_session->session_state); + + if (clp->cl_serverscope != NULL && + !nfs41_same_server_scope(clp->cl_serverscope, + resp->server_scope)) { + dprintk("%s: server_scope mismatch detected\n", + __func__); + set_bit(NFS4CLNT_SERVER_SCOPE_MISMATCH, &clp->cl_state); + } + + swap(clp->cl_serverowner, resp->server_owner); + swap(clp->cl_serverscope, resp->server_scope); + swap(clp->cl_implid, resp->impl_id); + + /* Save the EXCHANGE_ID verifier session trunk tests */ + memcpy(clp->cl_confirm.data, argp->verifier.data, + sizeof(clp->cl_confirm.data)); +out: + trace_nfs4_exchange_id(clp, status); + rpc_put_task(task); + return status; +} + +/* + * nfs4_proc_exchange_id() + * + * Returns zero, a negative errno, or a negative NFS4ERR status code. + * + * Since the clientid has expired, all compounds using sessions + * associated with the stale clientid will be returning + * NFS4ERR_BADSESSION in the sequence operation, and will therefore + * be in some phase of session reset. + * + * Will attempt to negotiate SP4_MACH_CRED if krb5i / krb5p auth is used. + */ +int nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cred) +{ + rpc_authflavor_t authflavor = clp->cl_rpcclient->cl_auth->au_flavor; + int status; + + /* try SP4_MACH_CRED if krb5i/p */ + if (authflavor == RPC_AUTH_GSS_KRB5I || + authflavor == RPC_AUTH_GSS_KRB5P) { + status = _nfs4_proc_exchange_id(clp, cred, SP4_MACH_CRED); + if (!status) + return 0; + } + + /* try SP4_NONE */ + return _nfs4_proc_exchange_id(clp, cred, SP4_NONE); +} + +/** + * nfs4_test_session_trunk + * + * This is an add_xprt_test() test function called from + * rpc_clnt_setup_test_and_add_xprt. + * + * The rpc_xprt_switch is referrenced by rpc_clnt_setup_test_and_add_xprt + * and is dereferrenced in nfs4_exchange_id_release + * + * Upon success, add the new transport to the rpc_clnt + * + * @clnt: struct rpc_clnt to get new transport + * @xprt: the rpc_xprt to test + * @data: call data for _nfs4_proc_exchange_id. + */ +void nfs4_test_session_trunk(struct rpc_clnt *clnt, struct rpc_xprt *xprt, + void *data) +{ + struct nfs4_add_xprt_data *adata = data; + struct rpc_task *task; + int status; + + u32 sp4_how; + + dprintk("--> %s try %s\n", __func__, + xprt->address_strings[RPC_DISPLAY_ADDR]); + + sp4_how = (adata->clp->cl_sp4_flags == 0 ? SP4_NONE : SP4_MACH_CRED); + +try_again: + /* Test connection for session trunking. Async exchange_id call */ + task = nfs4_run_exchange_id(adata->clp, adata->cred, sp4_how, xprt); + if (IS_ERR(task)) + return; + + status = task->tk_status; + if (status == 0) { + status = nfs4_detect_session_trunking(adata->clp, + task->tk_msg.rpc_resp, xprt); + trace_nfs4_trunked_exchange_id(adata->clp, + xprt->address_strings[RPC_DISPLAY_ADDR], status); + } + if (status == 0) + rpc_clnt_xprt_switch_add_xprt(clnt, xprt); + else if (status != -NFS4ERR_DELAY && rpc_clnt_xprt_switch_has_addr(clnt, + (struct sockaddr *)&xprt->addr)) + rpc_clnt_xprt_switch_remove_xprt(clnt, xprt); + + rpc_put_task(task); + if (status == -NFS4ERR_DELAY) { + ssleep(1); + goto try_again; + } +} +EXPORT_SYMBOL_GPL(nfs4_test_session_trunk); + +static int _nfs4_proc_destroy_clientid(struct nfs_client *clp, + const struct cred *cred) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DESTROY_CLIENTID], + .rpc_argp = clp, + .rpc_cred = cred, + }; + int status; + + status = rpc_call_sync(clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); + trace_nfs4_destroy_clientid(clp, status); + if (status) + dprintk("NFS: Got error %d from the server %s on " + "DESTROY_CLIENTID.", status, clp->cl_hostname); + return status; +} + +static int nfs4_proc_destroy_clientid(struct nfs_client *clp, + const struct cred *cred) +{ + unsigned int loop; + int ret; + + for (loop = NFS4_MAX_LOOP_ON_RECOVER; loop != 0; loop--) { + ret = _nfs4_proc_destroy_clientid(clp, cred); + switch (ret) { + case -NFS4ERR_DELAY: + case -NFS4ERR_CLIENTID_BUSY: + ssleep(1); + break; + default: + return ret; + } + } + return 0; +} + +int nfs4_destroy_clientid(struct nfs_client *clp) +{ + const struct cred *cred; + int ret = 0; + + if (clp->cl_mvops->minor_version < 1) + goto out; + if (clp->cl_exchange_flags == 0) + goto out; + if (clp->cl_preserve_clid) + goto out; + cred = nfs4_get_clid_cred(clp); + ret = nfs4_proc_destroy_clientid(clp, cred); + put_cred(cred); + switch (ret) { + case 0: + case -NFS4ERR_STALE_CLIENTID: + clp->cl_exchange_flags = 0; + } +out: + return ret; +} + +struct nfs4_get_lease_time_data { + struct nfs4_get_lease_time_args *args; + struct nfs4_get_lease_time_res *res; + struct nfs_client *clp; +}; + +static void nfs4_get_lease_time_prepare(struct rpc_task *task, + void *calldata) +{ + struct nfs4_get_lease_time_data *data = + (struct nfs4_get_lease_time_data *)calldata; + + /* just setup sequence, do not trigger session recovery + since we're invoked within one */ + nfs4_setup_sequence(data->clp, + &data->args->la_seq_args, + &data->res->lr_seq_res, + task); +} + +/* + * Called from nfs4_state_manager thread for session setup, so don't recover + * from sequence operation or clientid errors. + */ +static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_get_lease_time_data *data = + (struct nfs4_get_lease_time_data *)calldata; + + if (!nfs4_sequence_done(task, &data->res->lr_seq_res)) + return; + switch (task->tk_status) { + case -NFS4ERR_DELAY: + case -NFS4ERR_GRACE: + rpc_delay(task, NFS4_POLL_RETRY_MIN); + task->tk_status = 0; + fallthrough; + case -NFS4ERR_RETRY_UNCACHED_REP: + rpc_restart_call_prepare(task); + return; + } +} + +static const struct rpc_call_ops nfs4_get_lease_time_ops = { + .rpc_call_prepare = nfs4_get_lease_time_prepare, + .rpc_call_done = nfs4_get_lease_time_done, +}; + +int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) +{ + struct nfs4_get_lease_time_args args; + struct nfs4_get_lease_time_res res = { + .lr_fsinfo = fsinfo, + }; + struct nfs4_get_lease_time_data data = { + .args = &args, + .res = &res, + .clp = clp, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GET_LEASE_TIME], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs4_get_lease_time_ops, + .callback_data = &data, + .flags = RPC_TASK_TIMEOUT, + }; + + nfs4_init_sequence(clp, &args.la_seq_args, &res.lr_seq_res, 0, 1); + return nfs4_call_sync_custom(&task_setup); +} + +/* + * Initialize the values to be used by the client in CREATE_SESSION + * If nfs4_init_session set the fore channel request and response sizes, + * use them. + * + * Set the back channel max_resp_sz_cached to zero to force the client to + * always set csa_cachethis to FALSE because the current implementation + * of the back channel DRC only supports caching the CB_SEQUENCE operation. + */ +static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args, + struct rpc_clnt *clnt) +{ + unsigned int max_rqst_sz, max_resp_sz; + unsigned int max_bc_payload = rpc_max_bc_payload(clnt); + unsigned int max_bc_slots = rpc_num_bc_slots(clnt); + + max_rqst_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxwrite_overhead; + max_resp_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxread_overhead; + + /* Fore channel attributes */ + args->fc_attrs.max_rqst_sz = max_rqst_sz; + args->fc_attrs.max_resp_sz = max_resp_sz; + args->fc_attrs.max_ops = NFS4_MAX_OPS; + args->fc_attrs.max_reqs = max_session_slots; + + dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u " + "max_ops=%u max_reqs=%u\n", + __func__, + args->fc_attrs.max_rqst_sz, args->fc_attrs.max_resp_sz, + args->fc_attrs.max_ops, args->fc_attrs.max_reqs); + + /* Back channel attributes */ + args->bc_attrs.max_rqst_sz = max_bc_payload; + args->bc_attrs.max_resp_sz = max_bc_payload; + args->bc_attrs.max_resp_sz_cached = 0; + args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS; + args->bc_attrs.max_reqs = max_t(unsigned short, max_session_cb_slots, 1); + if (args->bc_attrs.max_reqs > max_bc_slots) + args->bc_attrs.max_reqs = max_bc_slots; + + dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u " + "max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n", + __func__, + args->bc_attrs.max_rqst_sz, args->bc_attrs.max_resp_sz, + args->bc_attrs.max_resp_sz_cached, args->bc_attrs.max_ops, + args->bc_attrs.max_reqs); +} + +static int nfs4_verify_fore_channel_attrs(struct nfs41_create_session_args *args, + struct nfs41_create_session_res *res) +{ + struct nfs4_channel_attrs *sent = &args->fc_attrs; + struct nfs4_channel_attrs *rcvd = &res->fc_attrs; + + if (rcvd->max_resp_sz > sent->max_resp_sz) + return -EINVAL; + /* + * Our requested max_ops is the minimum we need; we're not + * prepared to break up compounds into smaller pieces than that. + * So, no point even trying to continue if the server won't + * cooperate: + */ + if (rcvd->max_ops < sent->max_ops) + return -EINVAL; + if (rcvd->max_reqs == 0) + return -EINVAL; + if (rcvd->max_reqs > NFS4_MAX_SLOT_TABLE) + rcvd->max_reqs = NFS4_MAX_SLOT_TABLE; + return 0; +} + +static int nfs4_verify_back_channel_attrs(struct nfs41_create_session_args *args, + struct nfs41_create_session_res *res) +{ + struct nfs4_channel_attrs *sent = &args->bc_attrs; + struct nfs4_channel_attrs *rcvd = &res->bc_attrs; + + if (!(res->flags & SESSION4_BACK_CHAN)) + goto out; + if (rcvd->max_rqst_sz > sent->max_rqst_sz) + return -EINVAL; + if (rcvd->max_resp_sz > sent->max_resp_sz) + return -EINVAL; + if (rcvd->max_resp_sz_cached > sent->max_resp_sz_cached) + return -EINVAL; + if (rcvd->max_ops > sent->max_ops) + return -EINVAL; + if (rcvd->max_reqs > sent->max_reqs) + return -EINVAL; +out: + return 0; +} + +static int nfs4_verify_channel_attrs(struct nfs41_create_session_args *args, + struct nfs41_create_session_res *res) +{ + int ret; + + ret = nfs4_verify_fore_channel_attrs(args, res); + if (ret) + return ret; + return nfs4_verify_back_channel_attrs(args, res); +} + +static void nfs4_update_session(struct nfs4_session *session, + struct nfs41_create_session_res *res) +{ + nfs4_copy_sessionid(&session->sess_id, &res->sessionid); + /* Mark client id and session as being confirmed */ + session->clp->cl_exchange_flags |= EXCHGID4_FLAG_CONFIRMED_R; + set_bit(NFS4_SESSION_ESTABLISHED, &session->session_state); + session->flags = res->flags; + memcpy(&session->fc_attrs, &res->fc_attrs, sizeof(session->fc_attrs)); + if (res->flags & SESSION4_BACK_CHAN) + memcpy(&session->bc_attrs, &res->bc_attrs, + sizeof(session->bc_attrs)); +} + +static int _nfs4_proc_create_session(struct nfs_client *clp, + const struct cred *cred) +{ + struct nfs4_session *session = clp->cl_session; + struct nfs41_create_session_args args = { + .client = clp, + .clientid = clp->cl_clientid, + .seqid = clp->cl_seqid, + .cb_program = NFS4_CALLBACK, + }; + struct nfs41_create_session_res res; + + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE_SESSION], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + int status; + + nfs4_init_channel_attrs(&args, clp->cl_rpcclient); + args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN); + + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); + trace_nfs4_create_session(clp, status); + + switch (status) { + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_DELAY: + case -ETIMEDOUT: + case -EACCES: + case -EAGAIN: + goto out; + } + + clp->cl_seqid++; + if (!status) { + /* Verify the session's negotiated channel_attrs values */ + status = nfs4_verify_channel_attrs(&args, &res); + /* Increment the clientid slot sequence id */ + if (status) + goto out; + nfs4_update_session(session, &res); + } +out: + return status; +} + +/* + * Issues a CREATE_SESSION operation to the server. + * It is the responsibility of the caller to verify the session is + * expired before calling this routine. + */ +int nfs4_proc_create_session(struct nfs_client *clp, const struct cred *cred) +{ + int status; + unsigned *ptr; + struct nfs4_session *session = clp->cl_session; + struct nfs4_add_xprt_data xprtdata = { + .clp = clp, + }; + struct rpc_add_xprt_test rpcdata = { + .add_xprt_test = clp->cl_mvops->session_trunk, + .data = &xprtdata, + }; + + dprintk("--> %s clp=%p session=%p\n", __func__, clp, session); + + status = _nfs4_proc_create_session(clp, cred); + if (status) + goto out; + + /* Init or reset the session slot tables */ + status = nfs4_setup_session_slot_tables(session); + dprintk("slot table setup returned %d\n", status); + if (status) + goto out; + + ptr = (unsigned *)&session->sess_id.data[0]; + dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__, + clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]); + rpc_clnt_probe_trunked_xprts(clp->cl_rpcclient, &rpcdata); +out: + return status; +} + +/* + * Issue the over-the-wire RPC DESTROY_SESSION. + * The caller must serialize access to this routine. + */ +int nfs4_proc_destroy_session(struct nfs4_session *session, + const struct cred *cred) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DESTROY_SESSION], + .rpc_argp = session, + .rpc_cred = cred, + }; + int status = 0; + + /* session is still being setup */ + if (!test_and_clear_bit(NFS4_SESSION_ESTABLISHED, &session->session_state)) + return 0; + + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); + trace_nfs4_destroy_session(session->clp, status); + + if (status) + dprintk("NFS: Got error %d from the server on DESTROY_SESSION. " + "Session has been destroyed regardless...\n", status); + rpc_clnt_manage_trunked_xprts(session->clp->cl_rpcclient); + return status; +} + +/* + * Renew the cl_session lease. + */ +struct nfs4_sequence_data { + struct nfs_client *clp; + struct nfs4_sequence_args args; + struct nfs4_sequence_res res; +}; + +static void nfs41_sequence_release(void *data) +{ + struct nfs4_sequence_data *calldata = data; + struct nfs_client *clp = calldata->clp; + + if (refcount_read(&clp->cl_count) > 1) + nfs4_schedule_state_renewal(clp); + nfs_put_client(clp); + kfree(calldata); +} + +static int nfs41_sequence_handle_errors(struct rpc_task *task, struct nfs_client *clp) +{ + switch(task->tk_status) { + case -NFS4ERR_DELAY: + rpc_delay(task, NFS4_POLL_RETRY_MAX); + return -EAGAIN; + default: + nfs4_schedule_lease_recovery(clp); + } + return 0; +} + +static void nfs41_sequence_call_done(struct rpc_task *task, void *data) +{ + struct nfs4_sequence_data *calldata = data; + struct nfs_client *clp = calldata->clp; + + if (!nfs41_sequence_done(task, task->tk_msg.rpc_resp)) + return; + + trace_nfs4_sequence(clp, task->tk_status); + if (task->tk_status < 0 && clp->cl_cons_state >= 0) { + dprintk("%s ERROR %d\n", __func__, task->tk_status); + if (refcount_read(&clp->cl_count) == 1) + return; + + if (nfs41_sequence_handle_errors(task, clp) == -EAGAIN) { + rpc_restart_call_prepare(task); + return; + } + } + dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred); +} + +static void nfs41_sequence_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_sequence_data *calldata = data; + struct nfs_client *clp = calldata->clp; + struct nfs4_sequence_args *args; + struct nfs4_sequence_res *res; + + args = task->tk_msg.rpc_argp; + res = task->tk_msg.rpc_resp; + + nfs4_setup_sequence(clp, args, res, task); +} + +static const struct rpc_call_ops nfs41_sequence_ops = { + .rpc_call_done = nfs41_sequence_call_done, + .rpc_call_prepare = nfs41_sequence_prepare, + .rpc_release = nfs41_sequence_release, +}; + +static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, + const struct cred *cred, + struct nfs4_slot *slot, + bool is_privileged) +{ + struct nfs4_sequence_data *calldata; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs41_sequence_ops, + .flags = RPC_TASK_ASYNC | RPC_TASK_TIMEOUT | RPC_TASK_MOVEABLE, + }; + struct rpc_task *ret; + + ret = ERR_PTR(-EIO); + if (!refcount_inc_not_zero(&clp->cl_count)) + goto out_err; + + ret = ERR_PTR(-ENOMEM); + calldata = kzalloc_obj(*calldata); + if (calldata == NULL) + goto out_put_clp; + nfs4_init_sequence(clp, &calldata->args, &calldata->res, 0, is_privileged); + nfs4_sequence_attach_slot(&calldata->args, &calldata->res, slot); + msg.rpc_argp = &calldata->args; + msg.rpc_resp = &calldata->res; + calldata->clp = clp; + task_setup_data.callback_data = calldata; + + ret = rpc_run_task(&task_setup_data); + if (IS_ERR(ret)) + goto out_err; + return ret; +out_put_clp: + nfs_put_client(clp); +out_err: + nfs41_release_slot(slot); + return ret; +} + +static int nfs41_proc_async_sequence(struct nfs_client *clp, const struct cred *cred, unsigned renew_flags) +{ + struct rpc_task *task; + int ret = 0; + + if ((renew_flags & NFS4_RENEW_TIMEOUT) == 0) + return -EAGAIN; + task = _nfs41_proc_sequence(clp, cred, NULL, false); + if (IS_ERR(task)) + ret = PTR_ERR(task); + else + rpc_put_task_async(task); + dprintk("<-- %s status=%d\n", __func__, ret); + return ret; +} + +static int nfs4_proc_sequence(struct nfs_client *clp, const struct cred *cred) +{ + struct rpc_task *task; + int ret; + + task = _nfs41_proc_sequence(clp, cred, NULL, true); + if (IS_ERR(task)) { + ret = PTR_ERR(task); + goto out; + } + ret = rpc_wait_for_completion_task(task); + if (!ret) + ret = task->tk_status; + rpc_put_task(task); +out: + dprintk("<-- %s status=%d\n", __func__, ret); + return ret; +} + +struct nfs4_reclaim_complete_data { + struct nfs_client *clp; + struct nfs41_reclaim_complete_args arg; + struct nfs41_reclaim_complete_res res; +}; + +static void nfs4_reclaim_complete_prepare(struct rpc_task *task, void *data) +{ + struct nfs4_reclaim_complete_data *calldata = data; + + nfs4_setup_sequence(calldata->clp, + &calldata->arg.seq_args, + &calldata->res.seq_res, + task); +} + +static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nfs_client *clp) +{ + switch(task->tk_status) { + case 0: + wake_up_all(&clp->cl_lock_waitq); + fallthrough; + case -NFS4ERR_COMPLETE_ALREADY: + case -NFS4ERR_WRONG_CRED: /* What to do here? */ + break; + case -NFS4ERR_DELAY: + rpc_delay(task, NFS4_POLL_RETRY_MAX); + fallthrough; + case -NFS4ERR_RETRY_UNCACHED_REP: + case -EACCES: + dprintk("%s: failed to reclaim complete error %d for server %s, retrying\n", + __func__, task->tk_status, clp->cl_hostname); + return -EAGAIN; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + break; + default: + nfs4_schedule_lease_recovery(clp); + } + return 0; +} + +static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data) +{ + struct nfs4_reclaim_complete_data *calldata = data; + struct nfs_client *clp = calldata->clp; + struct nfs4_sequence_res *res = &calldata->res.seq_res; + + if (!nfs41_sequence_done(task, res)) + return; + + trace_nfs4_reclaim_complete(clp, task->tk_status); + if (nfs41_reclaim_complete_handle_errors(task, clp) == -EAGAIN) { + rpc_restart_call_prepare(task); + return; + } +} + +static void nfs4_free_reclaim_complete_data(void *data) +{ + struct nfs4_reclaim_complete_data *calldata = data; + + kfree(calldata); +} + +static const struct rpc_call_ops nfs4_reclaim_complete_call_ops = { + .rpc_call_prepare = nfs4_reclaim_complete_prepare, + .rpc_call_done = nfs4_reclaim_complete_done, + .rpc_release = nfs4_free_reclaim_complete_data, +}; + +/* + * Issue a global reclaim complete. + */ +static int nfs41_proc_reclaim_complete(struct nfs_client *clp, + const struct cred *cred) +{ + struct nfs4_reclaim_complete_data *calldata; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RECLAIM_COMPLETE], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs4_reclaim_complete_call_ops, + .flags = RPC_TASK_NO_ROUND_ROBIN, + }; + int status = -ENOMEM; + + calldata = kzalloc_obj(*calldata, GFP_NOFS); + if (calldata == NULL) + goto out; + calldata->clp = clp; + calldata->arg.one_fs = 0; + + nfs4_init_sequence(clp, &calldata->arg.seq_args, &calldata->res.seq_res, 0, 1); + msg.rpc_argp = &calldata->arg; + msg.rpc_resp = &calldata->res; + task_setup_data.callback_data = calldata; + status = nfs4_call_sync_custom(&task_setup_data); +out: + dprintk("<-- %s status=%d\n", __func__, status); + return status; +} + +static void +nfs4_layoutget_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutget *lgp = calldata; + struct nfs_server *server = NFS_SERVER(lgp->args.inode); + + nfs4_setup_sequence(server->nfs_client, &lgp->args.seq_args, + &lgp->res.seq_res, task); +} + +static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutget *lgp = calldata; + + nfs41_sequence_process(task, &lgp->res.seq_res); +} + +static int +nfs4_layoutget_handle_exception(struct rpc_task *task, + struct nfs4_layoutget *lgp, struct nfs4_exception *exception) +{ + struct inode *inode = lgp->args.inode; + struct nfs_server *server = NFS_SERVER(inode); + struct pnfs_layout_hdr *lo = lgp->lo; + int nfs4err = task->tk_status; + int err, status = 0; + LIST_HEAD(head); + + dprintk("--> %s tk_status => %d\n", __func__, -task->tk_status); + + nfs4_sequence_free_slot(&lgp->res.seq_res); + + exception->state = NULL; + exception->stateid = NULL; + + switch (nfs4err) { + case 0: + goto out; + + /* + * NFS4ERR_LAYOUTUNAVAILABLE means we are not supposed to use pnfs + * on the file. set tk_status to -ENODATA to tell upper layer to + * retry go inband. + */ + case -NFS4ERR_LAYOUTUNAVAILABLE: + status = -ENODATA; + goto out; + /* + * NFS4ERR_BADLAYOUT means the MDS cannot return a layout of + * length lgp->args.minlength != 0 (see RFC5661 section 18.43.3). + */ + case -NFS4ERR_BADLAYOUT: + status = -EOVERFLOW; + goto out; + /* + * NFS4ERR_LAYOUTTRYLATER is a conflict with another client + * (or clients) writing to the same RAID stripe except when + * the minlength argument is 0 (see RFC5661 section 18.43.3). + * + * Treat it like we would RECALLCONFLICT -- we retry for a little + * while, and then eventually give up. + */ + case -NFS4ERR_LAYOUTTRYLATER: + if (lgp->args.minlength == 0) { + status = -EOVERFLOW; + goto out; + } + status = -EBUSY; + break; + case -NFS4ERR_RECALLCONFLICT: + case -NFS4ERR_RETURNCONFLICT: + status = -ERECALLCONFLICT; + break; + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + exception->timeout = 0; + spin_lock(&inode->i_lock); + /* If the open stateid was bad, then recover it. */ + if (!lo || test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags) || + !nfs4_stateid_match_other(&lgp->args.stateid, &lo->plh_stateid)) { + spin_unlock(&inode->i_lock); + exception->state = lgp->args.ctx->state; + exception->stateid = &lgp->args.stateid; + break; + } + + /* + * Mark the bad layout state as invalid, then retry + */ + pnfs_mark_layout_stateid_invalid(lo, &head); + spin_unlock(&inode->i_lock); + nfs_commit_inode(inode, 0); + pnfs_free_lseg_list(&head); + status = -EAGAIN; + goto out; + } + + err = nfs4_handle_exception(server, nfs4err, exception); + if (!status) { + if (exception->retry) + status = -EAGAIN; + else + status = err; + } +out: + return status; +} + +size_t max_response_pages(struct nfs_server *server) +{ + u32 max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz; + return nfs_page_array_len(0, max_resp_sz); +} + +static void nfs4_layoutget_release(void *calldata) +{ + struct nfs4_layoutget *lgp = calldata; + + nfs4_sequence_free_slot(&lgp->res.seq_res); + pnfs_layoutget_free(lgp); +} + +static const struct rpc_call_ops nfs4_layoutget_call_ops = { + .rpc_call_prepare = nfs4_layoutget_prepare, + .rpc_call_done = nfs4_layoutget_done, + .rpc_release = nfs4_layoutget_release, +}; + +struct pnfs_layout_segment * +nfs4_proc_layoutget(struct nfs4_layoutget *lgp, + struct nfs4_exception *exception) +{ + struct inode *inode = lgp->args.inode; + struct nfs_server *server = NFS_SERVER(inode); + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTGET], + .rpc_argp = &lgp->args, + .rpc_resp = &lgp->res, + .rpc_cred = lgp->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs4_layoutget_call_ops, + .callback_data = lgp, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF | + RPC_TASK_MOVEABLE, + }; + struct pnfs_layout_segment *lseg = NULL; + int status = 0; + + nfs4_init_sequence(server->nfs_client, &lgp->args.seq_args, + &lgp->res.seq_res, 0, 0); + exception->retry = 0; + + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return ERR_CAST(task); + + status = rpc_wait_for_completion_task(task); + if (status != 0) + goto out; + + if (task->tk_status < 0) { + exception->retry = 1; + status = nfs4_layoutget_handle_exception(task, lgp, exception); + } else if (lgp->res.layoutp->len == 0) { + exception->retry = 1; + status = -EAGAIN; + nfs4_update_delay(&exception->timeout); + } else + lseg = pnfs_layout_process(lgp); +out: + trace_nfs4_layoutget(lgp->args.ctx, + &lgp->args.range, + &lgp->res.range, + &lgp->res.stateid, + status); + + rpc_put_task(task); + dprintk("<-- %s status=%d\n", __func__, status); + if (status) + return ERR_PTR(status); + return lseg; +} + +static void +nfs4_layoutreturn_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + + nfs4_setup_sequence(lrp->clp, + &lrp->args.seq_args, + &lrp->res.seq_res, + task); + if (!pnfs_layout_is_valid(lrp->args.layout)) + rpc_exit(task, 0); +} + +static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + struct nfs_server *server; + + if (!nfs41_sequence_process(task, &lrp->res.seq_res)) + return; + + if (task->tk_rpc_status < 0) { + switch (task->tk_rpc_status) { + case -EACCES: + case -EIO: + case -EKEYEXPIRED: + case -ERESTARTSYS: + case -EINTR: + lrp->rpc_status = 0; + break; + case -ENETDOWN: + case -ENETUNREACH: + if (task->tk_flags & RPC_TASK_NETUNREACH_FATAL) + lrp->rpc_status = 0; + else + lrp->rpc_status = -EAGAIN; + break; + default: + lrp->rpc_status = -EAGAIN; + break; + } + lrp->res.lrs_present = 0; + return; + } + + server = NFS_SERVER(lrp->args.inode); + switch (task->tk_status) { + case -NFS4ERR_OLD_STATEID: + if (nfs4_layout_refresh_old_stateid(&lrp->args.stateid, + &lrp->args.range, + lrp->args.inode)) + goto out_restart; + fallthrough; + default: + task->tk_status = 0; + lrp->res.lrs_present = 0; + fallthrough; + case 0: + break; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + nfs4_schedule_session_recovery(server->nfs_client->cl_session, + task->tk_status); + lrp->res.lrs_present = 0; + lrp->rpc_status = -EAGAIN; + task->tk_status = 0; + break; + case -NFS4ERR_DELAY: + if (nfs4_async_handle_error(task, server, NULL, NULL) == + -EAGAIN) + goto out_restart; + lrp->res.lrs_present = 0; + break; + } + return; +out_restart: + task->tk_status = 0; + nfs4_sequence_free_slot(&lrp->res.seq_res); + rpc_restart_call_prepare(task); +} + +static void nfs4_layoutreturn_release(void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + struct pnfs_layout_hdr *lo = lrp->args.layout; + + if (lrp->rpc_status == 0 || !lrp->inode) + pnfs_layoutreturn_free_lsegs( + lo, &lrp->args.stateid, &lrp->args.range, + lrp->res.lrs_present ? &lrp->res.stateid : NULL); + else + pnfs_layoutreturn_retry_later(lo, &lrp->args.stateid, + &lrp->args.range); + nfs4_sequence_free_slot(&lrp->res.seq_res); + if (lrp->ld_private.ops && lrp->ld_private.ops->free) + lrp->ld_private.ops->free(&lrp->ld_private); + pnfs_put_layout_hdr(lrp->args.layout); + nfs_iput_and_deactive(lrp->inode); + put_cred(lrp->cred); + kfree(calldata); +} + +static const struct rpc_call_ops nfs4_layoutreturn_call_ops = { + .rpc_call_prepare = nfs4_layoutreturn_prepare, + .rpc_call_done = nfs4_layoutreturn_done, + .rpc_release = nfs4_layoutreturn_release, +}; + +int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, unsigned int flags) +{ + struct nfs_client *clp = NFS_SERVER(lrp->args.inode)->nfs_client; + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTRETURN], + .rpc_argp = &lrp->args, + .rpc_resp = &lrp->res, + .rpc_cred = lrp->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = NFS_SERVER(lrp->args.inode)->client, + .rpc_message = &msg, + .callback_ops = &nfs4_layoutreturn_call_ops, + .callback_data = lrp, + .flags = RPC_TASK_MOVEABLE, + }; + int status = 0; + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_PNFS_CLEANUP, + &task_setup_data.rpc_client, &msg); + + lrp->inode = nfs_igrab_and_active(lrp->args.inode); + if (flags & PNFS_FL_LAYOUTRETURN_ASYNC) { + if (!lrp->inode) { + nfs4_layoutreturn_release(lrp); + return -EAGAIN; + } + task_setup_data.flags |= RPC_TASK_ASYNC; + } + if (!lrp->inode) + flags |= PNFS_FL_LAYOUTRETURN_PRIVILEGED; + + nfs4_init_sequence(clp, &lrp->args.seq_args, &lrp->res.seq_res, 1, + flags & PNFS_FL_LAYOUTRETURN_PRIVILEGED ? 1 : 0); + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + if (!(flags & PNFS_FL_LAYOUTRETURN_ASYNC)) + status = task->tk_status; + trace_nfs4_layoutreturn(lrp->args.inode, &lrp->args.stateid, status); + dprintk("<-- %s status=%d\n", __func__, status); + rpc_put_task(task); + return status; +} + +static int +_nfs4_proc_getdeviceinfo(struct nfs_server *server, + struct pnfs_device *pdev, + const struct cred *cred) +{ + struct nfs4_getdeviceinfo_args args = { + .pdev = pdev, + .notify_types = NOTIFY_DEVICEID4_CHANGE | + NOTIFY_DEVICEID4_DELETE, + }; + struct nfs4_getdeviceinfo_res res = { + .pdev = pdev, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETDEVICEINFO], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + int status; + + status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); + if (res.notification & ~args.notify_types) + dprintk("%s: unsupported notification\n", __func__); + if (res.notification != args.notify_types) + pdev->nocache = 1; + + trace_nfs4_getdeviceinfo(server, &pdev->dev_id, status); + + dprintk("<-- %s status=%d\n", __func__, status); + + return status; +} + +int nfs4_proc_getdeviceinfo(struct nfs_server *server, + struct pnfs_device *pdev, + const struct cred *cred) +{ + struct nfs4_exception exception = { }; + int err; + + do { + err = nfs4_handle_exception(server, + _nfs4_proc_getdeviceinfo(server, pdev, cred), + &exception); + } while (exception.retry); + return err; +} +EXPORT_SYMBOL_GPL(nfs4_proc_getdeviceinfo); + +static void nfs4_layoutcommit_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutcommit_data *data = calldata; + struct nfs_server *server = NFS_SERVER(data->args.inode); + + nfs4_setup_sequence(server->nfs_client, + &data->args.seq_args, + &data->res.seq_res, + task); +} + +static void +nfs4_layoutcommit_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutcommit_data *data = calldata; + struct nfs_server *server = NFS_SERVER(data->args.inode); + + if (!nfs41_sequence_done(task, &data->res.seq_res)) + return; + + switch (task->tk_status) { /* Just ignore these failures */ + case -NFS4ERR_DELEG_REVOKED: /* layout was recalled */ + case -NFS4ERR_BADIOMODE: /* no IOMODE_RW layout for range */ + case -NFS4ERR_BADLAYOUT: /* no layout */ + case -NFS4ERR_GRACE: /* loca_recalim always false */ + task->tk_status = 0; + break; + case -NFS4ERR_OLD_STATEID: { + u32 old_seqid = be32_to_cpu(data->args.stateid.seqid); + struct pnfs_layout_range range = { + .iomode = IOMODE_ANY, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + + if (nfs4_layout_refresh_old_stateid(&data->args.stateid, + &range, + data->args.inode)) { + struct pnfs_layout_hdr *lo; + + spin_lock(&data->args.inode->i_lock); + lo = NFS_I(data->args.inode)->layout; + if (lo && pnfs_layout_is_valid(lo) && + nfs4_stateid_match_other(&data->args.stateid, + &lo->plh_stateid)) + pnfs_set_layout_stateid(lo, &data->args.stateid, + NULL, false); + spin_unlock(&data->args.inode->i_lock); + + dprintk("%s: refreshed OLD_STATEID inode %llu seq %u->%u\n", + __func__, data->args.inode->i_ino, + old_seqid, + be32_to_cpu(data->args.stateid.seqid)); + + rpc_restart_call_prepare(task); + return; + } + fallthrough; + } + case 0: + break; + default: + if (nfs4_async_handle_error(task, server, NULL, NULL) == -EAGAIN) { + rpc_restart_call_prepare(task); + return; + } + } +} + +static void nfs4_layoutcommit_release(void *calldata) +{ + struct nfs4_layoutcommit_data *data = calldata; + + pnfs_cleanup_layoutcommit(data); + nfs_post_op_update_inode_force_wcc(data->args.inode, + data->res.fattr); + put_cred(data->cred); + nfs_iput_and_deactive(data->inode); + kfree(data); +} + +static const struct rpc_call_ops nfs4_layoutcommit_ops = { + .rpc_call_prepare = nfs4_layoutcommit_prepare, + .rpc_call_done = nfs4_layoutcommit_done, + .rpc_release = nfs4_layoutcommit_release, +}; + +int +nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTCOMMIT], + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->cred, + }; + struct rpc_task_setup task_setup_data = { + .task = &data->task, + .rpc_client = NFS_CLIENT(data->args.inode), + .rpc_message = &msg, + .callback_ops = &nfs4_layoutcommit_ops, + .callback_data = data, + .flags = RPC_TASK_MOVEABLE, + }; + struct rpc_task *task; + int status = 0; + + dprintk("NFS: initiating layoutcommit call. sync %d " + "lbw: %llu inode %llu\n", sync, + data->args.lastbytewritten, + data->args.inode->i_ino); + + if (!sync) { + data->inode = nfs_igrab_and_active(data->args.inode); + if (data->inode == NULL) { + nfs4_layoutcommit_release(data); + return -EAGAIN; + } + task_setup_data.flags = RPC_TASK_ASYNC; + } + nfs4_init_sequence(NFS_SERVER(data->args.inode)->nfs_client, + &data->args.seq_args, &data->res.seq_res, 1, 0); + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + if (sync) + status = task->tk_status; + trace_nfs4_layoutcommit(data->args.inode, &data->args.stateid, status); + dprintk("%s: status %d\n", __func__, status); + rpc_put_task(task); + return status; +} + +/* + * Use the state managment nfs_client cl_rpcclient, which uses krb5i (if + * possible) as per RFC3530bis and RFC5661 Security Considerations sections + */ +static int _nfs41_proc_secinfo_no_name(struct nfs_server *server, + struct nfs_fh *fhandle, + struct nfs4_secinfo_flavors *flavors, + bool use_integrity) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs41_secinfo_no_name_args args = { + .style = SECINFO_STYLE_CURRENT_FH, + }; + struct nfs4_secinfo_res res = { + .flavors = flavors, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SECINFO_NO_NAME], + .rpc_argp = &args, + .rpc_resp = &res, + }; + struct nfs4_call_sync_data data = { + .seq_server = server, + .seq_args = &args.seq_args, + .seq_res = &res.seq_res, + }; + struct rpc_task_setup task_setup = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = clp->cl_mvops->call_sync_ops, + .callback_data = &data, + .flags = RPC_TASK_NO_ROUND_ROBIN, + }; + const struct cred *cred = NULL; + int status; + + if (use_integrity) { + task_setup.rpc_client = clp->cl_rpcclient; + + cred = nfs4_get_clid_cred(clp); + msg.rpc_cred = cred; + } + + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 0); + status = nfs4_call_sync_custom(&task_setup); + dprintk("<-- %s status=%d\n", __func__, status); + + put_cred(cred); + + return status; +} + +static int nfs41_proc_secinfo_no_name(struct nfs_server *server, + struct nfs_fh *fhandle, + struct nfs4_secinfo_flavors *flavors) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + /* first try using integrity protection */ + err = -NFS4ERR_WRONGSEC; + + /* try to use integrity protection with machine cred */ + if (_nfs4_is_integrity_protected(server->nfs_client)) + err = _nfs41_proc_secinfo_no_name(server, fhandle, + flavors, true); + + /* + * if unable to use integrity protection, or SECINFO with + * integrity protection returns NFS4ERR_WRONGSEC (which is + * disallowed by spec, but exists in deployed servers) use + * the current filesystem's rpc_client and the user cred. + */ + if (err == -NFS4ERR_WRONGSEC) + err = _nfs41_proc_secinfo_no_name(server, fhandle, + flavors, false); + + switch (err) { + case 0: + case -NFS4ERR_WRONGSEC: + case -ENOTSUPP: + goto out; + default: + err = nfs4_handle_exception(server, err, &exception); + } + } while (exception.retry); +out: + return err; +} + +static int nfs41_find_root_sec(struct nfs_server *server, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + int err; + struct page *page; + rpc_authflavor_t flavor = RPC_AUTH_MAXFLAVOR; + struct nfs4_secinfo_flavors *flavors; + struct nfs4_secinfo4 *secinfo; + int i; + + page = alloc_page(GFP_KERNEL); + if (!page) { + err = -ENOMEM; + goto out; + } + + flavors = page_address(page); + err = nfs41_proc_secinfo_no_name(server, fhandle, flavors); + + /* + * Fall back on "guess and check" method if + * the server doesn't support SECINFO_NO_NAME + */ + if (err == -NFS4ERR_WRONGSEC || err == -ENOTSUPP) { + err = nfs4_find_root_sec(server, fhandle, fattr); + goto out_freepage; + } + if (err) + goto out_freepage; + + for (i = 0; i < flavors->num_flavors; i++) { + secinfo = &flavors->flavors[i]; + + switch (secinfo->flavor) { + case RPC_AUTH_NULL: + case RPC_AUTH_UNIX: + case RPC_AUTH_GSS: + flavor = rpcauth_get_pseudoflavor(secinfo->flavor, + &secinfo->flavor_info); + break; + default: + flavor = RPC_AUTH_MAXFLAVOR; + break; + } + + if (!nfs_auth_info_match(&server->auth_info, flavor)) + flavor = RPC_AUTH_MAXFLAVOR; + + if (flavor != RPC_AUTH_MAXFLAVOR) { + err = nfs4_lookup_root_sec(server, fhandle, fattr, + flavor); + if (!err) + break; + } + } + + if (flavor == RPC_AUTH_MAXFLAVOR) + err = -EPERM; + +out_freepage: + put_page(page); + if (err == -EACCES) + return -EPERM; +out: + return err; +} + +static int _nfs41_test_stateid(struct nfs_server *server, + const nfs4_stateid *stateid, + const struct cred *cred) +{ + int status; + struct nfs41_test_stateid_args args = { + .stateid = *stateid, + }; + struct nfs41_test_stateid_res res; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_TEST_STATEID], + .rpc_argp = &args, + .rpc_resp = &res, + .rpc_cred = cred, + }; + struct rpc_clnt *rpc_client = server->client; + struct nfs_client *clp = server->nfs_client; + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_STATEID, &rpc_client, &msg); + + dprintk("NFS call test_stateid %p\n", stateid); + nfs4_init_sequence(clp, &args.seq_args, &res.seq_res, 0, 1); + status = nfs4_call_sync_sequence(rpc_client, server, &msg, + &args.seq_args, &res.seq_res); + if (status != NFS_OK) { + dprintk("NFS reply test_stateid: failed, %d\n", status); + return status; + } + dprintk("NFS reply test_stateid: succeeded, %d\n", -res.status); + return -res.status; +} + +static void nfs4_handle_delay_or_session_error(struct nfs_server *server, + int err, struct nfs4_exception *exception) +{ + exception->retry = 0; + switch(err) { + case -NFS4ERR_DELAY: + case -NFS4ERR_RETRY_UNCACHED_REP: + nfs4_handle_exception(server, err, exception); + break; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_DEADSESSION: + nfs4_do_handle_exception(server, err, exception); + } +} + +/** + * nfs41_test_stateid - perform a TEST_STATEID operation + * + * @server: server / transport on which to perform the operation + * @stateid: state ID to test + * @cred: credential + * + * Returns NFS_OK if the server recognizes that "stateid" is valid. + * Otherwise a negative NFS4ERR value is returned if the operation + * failed or the state ID is not currently valid. + */ +static int nfs41_test_stateid(struct nfs_server *server, + const nfs4_stateid *stateid, + const struct cred *cred) +{ + struct nfs4_exception exception = { + .interruptible = true, + }; + int err; + do { + err = _nfs41_test_stateid(server, stateid, cred); + nfs4_handle_delay_or_session_error(server, err, &exception); + } while (exception.retry); + return err; +} + +struct nfs_free_stateid_data { + struct nfs_server *server; + struct nfs41_free_stateid_args args; + struct nfs41_free_stateid_res res; +}; + +static void nfs41_free_stateid_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_free_stateid_data *data = calldata; + nfs4_setup_sequence(data->server->nfs_client, + &data->args.seq_args, + &data->res.seq_res, + task); +} + +static void nfs41_free_stateid_done(struct rpc_task *task, void *calldata) +{ + struct nfs_free_stateid_data *data = calldata; + + nfs41_sequence_done(task, &data->res.seq_res); + + switch (task->tk_status) { + case -NFS4ERR_DELAY: + if (nfs4_async_handle_error(task, data->server, NULL, NULL) == -EAGAIN) + rpc_restart_call_prepare(task); + } +} + +static void nfs41_free_stateid_release(void *calldata) +{ + struct nfs_free_stateid_data *data = calldata; + struct nfs_client *clp = data->server->nfs_client; + + nfs_sb_deactive(data->server->super); + nfs_put_client(clp); + kfree(calldata); +} + +static const struct rpc_call_ops nfs41_free_stateid_ops = { + .rpc_call_prepare = nfs41_free_stateid_prepare, + .rpc_call_done = nfs41_free_stateid_done, + .rpc_release = nfs41_free_stateid_release, +}; + +/** + * nfs41_free_stateid - perform a FREE_STATEID operation + * + * @server: server / transport on which to perform the operation + * @stateid: state ID to release + * @cred: credential + * @privileged: set to true if this call needs to be privileged + * + * Note: this function is always asynchronous. + */ +static int nfs41_free_stateid(struct nfs_server *server, + nfs4_stateid *stateid, + const struct cred *cred, + bool privileged) +{ + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FREE_STATEID], + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup = { + .rpc_client = server->client, + .rpc_message = &msg, + .callback_ops = &nfs41_free_stateid_ops, + .flags = RPC_TASK_ASYNC | RPC_TASK_MOVEABLE, + }; + struct nfs_free_stateid_data *data; + struct rpc_task *task; + struct nfs_client *clp = server->nfs_client; + int ret = -EIO; + + if (!refcount_inc_not_zero(&clp->cl_count)) + return ret; + if (!nfs_sb_active(server->super)) + goto out_put_clp; + + nfs4_state_protect(clp, NFS_SP4_MACH_CRED_STATEID, + &task_setup.rpc_client, &msg); + + dprintk("NFS call free_stateid %p\n", stateid); + data = kmalloc_obj(*data); + if (!data) { + ret = -ENOMEM; + goto out_put_server; + } + data->server = server; + nfs4_stateid_copy(&data->args.stateid, stateid); + + task_setup.callback_data = data; + + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + nfs4_init_sequence(clp, &data->args.seq_args, &data->res.seq_res, 1, + privileged); + task = rpc_run_task(&task_setup); + if (IS_ERR(task)) + return PTR_ERR(task); + rpc_put_task(task); + stateid->type = NFS4_FREED_STATEID_TYPE; + return 0; +out_put_server: + nfs_sb_deactive(server->super); +out_put_clp: + nfs_put_client(clp); + return ret; +} + +static void +nfs41_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp) +{ + const struct cred *cred = lsp->ls_state->owner->so_cred; + + nfs41_free_stateid(server, &lsp->ls_stateid, cred, false); + nfs4_free_lock_state(server, lsp); +} + +static bool nfs41_match_stateid(const nfs4_stateid *s1, + const nfs4_stateid *s2) +{ + trace_nfs41_match_stateid(s1, s2); + + if (s1->type != s2->type) + return false; + + if (memcmp(s1->other, s2->other, sizeof(s1->other)) != 0) + return false; + + if (s1->seqid == s2->seqid) + return true; + + return s1->seqid == 0 || s2->seqid == 0; +} + +bool nfs4_match_stateid(const nfs4_stateid *s1, + const nfs4_stateid *s2) +{ + trace_nfs4_match_stateid(s1, s2); + + return nfs4_stateid_match(s1, s2); +} + + +static const struct nfs4_sequence_slot_ops nfs41_sequence_slot_ops = { + .process = nfs41_sequence_process, + .done = nfs41_sequence_done, + .free_slot = nfs41_sequence_free_slot, +}; + +static const struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT, + .state_flag_bit = NFS_STATE_RECLAIM_REBOOT, + .recover_open = nfs4_open_reclaim, + .recover_lock = nfs4_lock_reclaim, + .establish_clid = nfs41_init_clientid, + .reclaim_complete = nfs41_proc_reclaim_complete, + .detect_trunking = nfs41_discover_server_trunking, +}; + +static const struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = { + .owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE, + .state_flag_bit = NFS_STATE_RECLAIM_NOGRACE, + .recover_open = nfs41_open_expired, + .recover_lock = nfs41_lock_expired, + .establish_clid = nfs41_init_clientid, +}; + +static const struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = { + .sched_state_renewal = nfs41_proc_async_sequence, + .get_state_renewal_cred = nfs4_get_machine_cred, + .renew_lease = nfs4_proc_sequence, +}; + +static const struct nfs4_mig_recovery_ops nfs41_mig_recovery_ops = { + .get_locations = _nfs41_proc_get_locations, + .fsid_present = _nfs41_proc_fsid_present, +}; + +static struct nfs_seqid * +nfs_alloc_no_seqid(struct nfs_seqid_counter *arg1, gfp_t arg2) +{ + return NULL; +} + +static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = { + .minor_version = 1, + .init_caps = NFS_CAP_READDIRPLUS + | NFS_CAP_ATOMIC_OPEN + | NFS_CAP_DIR_DELEG + | NFS_CAP_POSIX_LOCK + | NFS_CAP_STATEID_NFSV41 + | NFS_CAP_ATOMIC_OPEN_V1 + | NFS_CAP_LGOPEN + | NFS_CAP_MOVEABLE, + .init_client = nfs41_init_client, + .shutdown_client = nfs41_shutdown_client, + .match_stateid = nfs41_match_stateid, + .find_root_sec = nfs41_find_root_sec, + .free_lock_state = nfs41_free_lock_state, + .test_and_free_expired = nfs41_test_and_free_expired_stateid, + .alloc_seqid = nfs_alloc_no_seqid, + .session_trunk = nfs4_test_session_trunk, + .call_sync_ops = &nfs41_call_sync_ops, + .sequence_slot_ops = &nfs41_sequence_slot_ops, + .reboot_recovery_ops = &nfs41_reboot_recovery_ops, + .nograce_recovery_ops = &nfs41_nograce_recovery_ops, + .state_renewal_ops = &nfs41_state_renewal_ops, + .mig_recovery_ops = &nfs41_mig_recovery_ops, +}; + +#if defined(CONFIG_NFS_V4_2) +static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = { + .minor_version = 2, + .init_caps = NFS_CAP_READDIRPLUS + | NFS_CAP_ATOMIC_OPEN + | NFS_CAP_DIR_DELEG + | NFS_CAP_POSIX_LOCK + | NFS_CAP_STATEID_NFSV41 + | NFS_CAP_ATOMIC_OPEN_V1 + | NFS_CAP_LGOPEN + | NFS_CAP_ALLOCATE + | NFS_CAP_COPY + | NFS_CAP_OFFLOAD_CANCEL + | NFS_CAP_COPY_NOTIFY + | NFS_CAP_DEALLOCATE + | NFS_CAP_ZERO_RANGE + | NFS_CAP_SEEK + | NFS_CAP_LAYOUTSTATS + | NFS_CAP_CLONE + | NFS_CAP_LAYOUTERROR + | NFS_CAP_READ_PLUS + | NFS_CAP_MOVEABLE + | NFS_CAP_OFFLOAD_STATUS, + .init_client = nfs41_init_client, + .shutdown_client = nfs41_shutdown_client, + .match_stateid = nfs41_match_stateid, + .find_root_sec = nfs41_find_root_sec, + .free_lock_state = nfs41_free_lock_state, + .call_sync_ops = &nfs41_call_sync_ops, + .sequence_slot_ops = &nfs41_sequence_slot_ops, + .test_and_free_expired = nfs41_test_and_free_expired_stateid, + .alloc_seqid = nfs_alloc_no_seqid, + .session_trunk = nfs4_test_session_trunk, + .reboot_recovery_ops = &nfs41_reboot_recovery_ops, + .nograce_recovery_ops = &nfs41_nograce_recovery_ops, + .state_renewal_ops = &nfs41_state_renewal_ops, + .mig_recovery_ops = &nfs41_mig_recovery_ops, +}; +#endif + +const struct nfs4_minor_version_ops *nfs_v4_minor_ops[] = { +#if defined(CONFIG_NFS_V4_0) + [0] = &nfs_v4_0_minor_ops, +#endif /* CONFIG_NFS_V4_0 */ + [1] = &nfs_v4_1_minor_ops, +#if defined(CONFIG_NFS_V4_2) + [2] = &nfs_v4_2_minor_ops, +#endif +}; + +static ssize_t nfs4_listxattr(struct dentry *dentry, char *list, size_t size) +{ + ssize_t error, error2, error3; + ssize_t left = size; + ssize_t left2; + + error = generic_listxattr(dentry, list, left); + if (error < 0) + return error; + if (list) { + list += error; + left -= error; + } + + left2 = left; + error2 = security_inode_listsecurity(d_inode(dentry), &list, &left2); + if (error2 < 0) + return error2; + error2 = left - left2; + if (list) + left -= error2; + + error3 = nfs4_listxattr_nfs4_user(d_inode(dentry), list, left); + if (error3 < 0) + return error3; + + error += error2 + error3; + if (size && error > size) + return -ERANGE; + return error; +} + +static void nfs4_enable_swap(struct inode *inode) +{ + /* The state manager thread must always be running. + * It will notice the client is a swapper, and stay put. + */ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + + nfs4_schedule_state_manager(clp); +} + +static void nfs4_disable_swap(struct inode *inode) +{ + /* The state manager thread will now exit once it is + * woken. + */ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + + set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + wake_up_var(&clp->cl_state); +} + +static const struct inode_operations nfs4_dir_inode_operations = { + .create = nfs_create, + .lookup = nfs_lookup, + .atomic_open = nfs_atomic_open, + .link = nfs_link, + .unlink = nfs_unlink, + .symlink = nfs_symlink, + .mkdir = nfs_mkdir, + .rmdir = nfs_rmdir, + .mknod = nfs_mknod, + .rename = nfs_rename, + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .listxattr = nfs4_listxattr, + .fileattr_get = nfs_fileattr_get, +}; + +static const struct inode_operations nfs4_file_inode_operations = { + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .listxattr = nfs4_listxattr, + .fileattr_get = nfs_fileattr_get, +}; + +static struct nfs_server *nfs4_clone_server(struct nfs_server *source, + struct nfs_fh *fh, struct nfs_fattr *fattr, + rpc_authflavor_t flavor) +{ + struct nfs_server *server; + int error; + + server = nfs_clone_server(source, fh, fattr, flavor); + if (IS_ERR(server)) + return server; + + nfs4_session_limit_rwsize(server); + nfs4_session_limit_xasize(server); + + error = nfs4_delegation_hash_alloc(server); + if (error) { + nfs_free_server(server); + return ERR_PTR(error); + } + + return server; +} + +const struct nfs_rpc_ops nfs_v4_clientops = { + .version = 4, /* protocol version */ + .dentry_ops = &nfs4_dentry_operations, + .dir_inode_ops = &nfs4_dir_inode_operations, + .file_inode_ops = &nfs4_file_inode_operations, + .file_ops = &nfs4_file_operations, + .getroot = nfs4_proc_get_root, + .submount = nfs4_submount, + .try_get_tree = nfs4_try_get_tree, + .getattr = nfs4_proc_getattr, + .setattr = nfs4_proc_setattr, + .lookup = nfs4_proc_lookup, + .lookupp = nfs4_proc_lookupp, + .access = nfs4_proc_access, + .readlink = nfs4_proc_readlink, + .create = nfs4_proc_create, + .remove = nfs4_proc_remove, + .unlink_setup = nfs4_proc_unlink_setup, + .unlink_rpc_prepare = nfs4_proc_unlink_rpc_prepare, + .unlink_done = nfs4_proc_unlink_done, + .rename_setup = nfs4_proc_rename_setup, + .rename_rpc_prepare = nfs4_proc_rename_rpc_prepare, + .rename_done = nfs4_proc_rename_done, + .link = nfs4_proc_link, + .symlink = nfs4_proc_symlink, + .mkdir = nfs4_proc_mkdir, + .rmdir = nfs4_proc_rmdir, + .readdir = nfs4_proc_readdir, + .mknod = nfs4_proc_mknod, + .statfs = nfs4_proc_statfs, + .fsinfo = nfs4_proc_fsinfo, + .pathconf = nfs4_proc_pathconf, + .set_capabilities = nfs4_server_capabilities, + .decode_dirent = nfs4_decode_dirent, + .pgio_rpc_prepare = nfs4_proc_pgio_rpc_prepare, + .read_setup = nfs4_proc_read_setup, + .read_done = nfs4_read_done, + .write_setup = nfs4_proc_write_setup, + .write_done = nfs4_write_done, + .commit_setup = nfs4_proc_commit_setup, + .commit_rpc_prepare = nfs4_proc_commit_rpc_prepare, + .commit_done = nfs4_commit_done, + .lock = nfs4_proc_lock, + .clear_acl_cache = nfs4_zap_acl_attr, + .close_context = nfs4_close_context, + .open_context = nfs4_atomic_open, + .have_delegation = nfs4_have_delegation, + .return_delegation = nfs4_inode_return_delegation, + .alloc_client = nfs4_alloc_client, + .init_client = nfs4_init_client, + .free_client = nfs4_free_client, + .create_server = nfs4_create_server, + .clone_server = nfs4_clone_server, + .discover_trunking = nfs4_discover_trunking, + .enable_swap = nfs4_enable_swap, + .disable_swap = nfs4_disable_swap, +}; + +static const struct xattr_handler nfs4_xattr_nfs4_acl_handler = { + .name = XATTR_NAME_NFSV4_ACL, + .list = nfs4_xattr_list_nfs4_acl, + .get = nfs4_xattr_get_nfs4_acl, + .set = nfs4_xattr_set_nfs4_acl, +}; + +static const struct xattr_handler nfs4_xattr_nfs4_dacl_handler = { + .name = XATTR_NAME_NFSV4_DACL, + .list = nfs4_xattr_list_nfs4_dacl, + .get = nfs4_xattr_get_nfs4_dacl, + .set = nfs4_xattr_set_nfs4_dacl, +}; + +static const struct xattr_handler nfs4_xattr_nfs4_sacl_handler = { + .name = XATTR_NAME_NFSV4_SACL, + .list = nfs4_xattr_list_nfs4_sacl, + .get = nfs4_xattr_get_nfs4_sacl, + .set = nfs4_xattr_set_nfs4_sacl, +}; + +#ifdef CONFIG_NFS_V4_2 +static const struct xattr_handler nfs4_xattr_nfs4_user_handler = { + .prefix = XATTR_USER_PREFIX, + .get = nfs4_xattr_get_nfs4_user, + .set = nfs4_xattr_set_nfs4_user, +}; +#endif + +const struct xattr_handler * const nfs4_xattr_handlers[] = { + &nfs4_xattr_nfs4_acl_handler, + &nfs4_xattr_nfs4_dacl_handler, + &nfs4_xattr_nfs4_sacl_handler, +#ifdef CONFIG_NFS_V4_SECURITY_LABEL + &nfs4_xattr_nfs4_label_handler, +#endif +#ifdef CONFIG_NFS_V4_2 + &nfs4_xattr_nfs4_user_handler, +#endif + NULL +}; diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c new file mode 100644 index 000000000..30065df14 --- /dev/null +++ b/fs/nfs/nfs4renewd.c @@ -0,0 +1,160 @@ +/* + * fs/nfs/nfs4renewd.c + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Kendrick Smith + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Implementation of the NFSv4 "renew daemon", which wakes up periodically to + * send a RENEW, to keep state alive on the server. The daemon is implemented + * as an rpc_task, not a real kernel thread, so it always runs in rpciod's + * context. There is one renewd per nfs_server. + * + */ + +#include +#include +#include +#include + +#include +#include +#include +#include "nfs4_fs.h" +#include "delegation.h" + +#define NFSDBG_FACILITY NFSDBG_STATE + +void +nfs4_renew_state(struct work_struct *work) +{ + const struct nfs4_state_maintenance_ops *ops; + struct nfs_client *clp = + container_of(work, struct nfs_client, cl_renewd.work); + const struct cred *cred; + long lease; + unsigned long last, now; + unsigned renew_flags = 0; + + ops = clp->cl_mvops->state_renewal_ops; + dprintk("%s: start\n", __func__); + + if (test_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state)) + goto out; + + lease = clp->cl_lease_time; + last = clp->cl_last_renewal; + now = jiffies; + /* Are we close to a lease timeout? */ + if (time_after(now, last + lease/3)) + renew_flags |= NFS4_RENEW_TIMEOUT; + if (nfs_delegations_present(clp)) + renew_flags |= NFS4_RENEW_DELEGATION_CB; + + if (renew_flags != 0) { + cred = ops->get_state_renewal_cred(clp); + if (cred == NULL) { + if (!(renew_flags & NFS4_RENEW_DELEGATION_CB)) { + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + goto out; + } + nfs_expire_all_delegations(clp); + } else { + int ret; + + /* Queue an asynchronous RENEW. */ + ret = ops->sched_state_renewal(clp, cred, renew_flags); + put_cred(cred); + switch (ret) { + default: + goto out_exp; + case -EAGAIN: + case -ENOMEM: + break; + } + } + } else { + dprintk("%s: failed to call renewd. Reason: lease not expired \n", + __func__); + } + nfs4_schedule_state_renewal(clp); +out_exp: + nfs_expire_unreferenced_delegations(clp); +out: + dprintk("%s: done\n", __func__); +} + +void +nfs4_schedule_state_renewal(struct nfs_client *clp) +{ + long timeout; + + spin_lock(&clp->cl_lock); + timeout = (2 * clp->cl_lease_time) / 3 + (long)clp->cl_last_renewal + - (long)jiffies; + if (timeout < 5 * HZ) + timeout = 5 * HZ; + dprintk("%s: requeueing work. Lease period = %ld\n", + __func__, (timeout + HZ - 1) / HZ); + mod_delayed_work(system_percpu_wq, &clp->cl_renewd, timeout); + set_bit(NFS_CS_RENEWD, &clp->cl_res_state); + spin_unlock(&clp->cl_lock); +} + +void +nfs4_kill_renewd(struct nfs_client *clp) +{ + cancel_delayed_work_sync(&clp->cl_renewd); +} + +#define MAX_LEASE_PERIOD (60 * 60) /* 1 hour */ + +/** + * nfs4_set_lease_period - Sets the lease period on a nfs_client + * + * @clp: pointer to nfs_client + * @period: new value for lease period (in seconds) + */ +void nfs4_set_lease_period(struct nfs_client *clp, u32 period) +{ + unsigned long lease; + + /* Limit the lease period */ + if (period < MAX_LEASE_PERIOD) + lease = period * HZ; + else + lease = MAX_LEASE_PERIOD * HZ; + + spin_lock(&clp->cl_lock); + clp->cl_lease_time = lease; + spin_unlock(&clp->cl_lock); + + /* Cap maximum reconnect timeout at 1/2 lease period */ + rpc_set_connect_timeout(clp->cl_rpcclient, lease, lease >> 1); +} diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c new file mode 100644 index 000000000..5c128957a --- /dev/null +++ b/fs/nfs/nfs4session.c @@ -0,0 +1,653 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * fs/nfs/nfs4session.c + * + * Copyright (c) 2012 Trond Myklebust + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "internal.h" +#include "nfs4session.h" +#include "callback.h" + +#define NFSDBG_FACILITY NFSDBG_STATE + +static void nfs4_init_slot_table(struct nfs4_slot_table *tbl, const char *queue) +{ + tbl->highest_used_slotid = NFS4_NO_SLOT; + spin_lock_init(&tbl->slot_tbl_lock); + rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, queue); + init_waitqueue_head(&tbl->slot_waitq); + init_completion(&tbl->complete); +} + +/* + * nfs4_shrink_slot_table - free retired slots from the slot table + */ +static void nfs4_shrink_slot_table(struct nfs4_slot_table *tbl, u32 newsize) +{ + struct nfs4_slot **p; + if (newsize >= tbl->max_slots) + return; + + p = &tbl->slots; + while (newsize--) + p = &(*p)->next; + while (*p) { + struct nfs4_slot *slot = *p; + + *p = slot->next; + kfree(slot); + tbl->max_slots--; + } +} + +/** + * nfs4_slot_tbl_drain_complete - wake waiters when drain is complete + * @tbl: controlling slot table + * + */ +void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl) +{ + if (nfs4_slot_tbl_draining(tbl)) + complete(&tbl->complete); +} + +/* + * nfs4_free_slot - free a slot and efficiently update slot table. + * + * freeing a slot is trivially done by clearing its respective bit + * in the bitmap. + * If the freed slotid equals highest_used_slotid we want to update it + * so that the server would be able to size down the slot table if needed, + * otherwise we know that the highest_used_slotid is still in use. + * When updating highest_used_slotid there may be "holes" in the bitmap + * so we need to scan down from highest_used_slotid to 0 looking for the now + * highest slotid in use. + * If none found, highest_used_slotid is set to NFS4_NO_SLOT. + * + * Must be called while holding tbl->slot_tbl_lock + */ +void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot) +{ + u32 slotid = slot->slot_nr; + + /* clear used bit in bitmap */ + __clear_bit(slotid, tbl->used_slots); + + /* update highest_used_slotid when it is freed */ + if (slotid == tbl->highest_used_slotid) { + u32 new_max = find_last_bit(tbl->used_slots, slotid); + if (new_max < slotid) + tbl->highest_used_slotid = new_max; + else { + tbl->highest_used_slotid = NFS4_NO_SLOT; + nfs4_slot_tbl_drain_complete(tbl); + } + } + dprintk("%s: slotid %u highest_used_slotid %u\n", __func__, + slotid, tbl->highest_used_slotid); +} + +static struct nfs4_slot *nfs4_new_slot(struct nfs4_slot_table *tbl, + u32 slotid, u32 seq_init, gfp_t gfp_mask) +{ + struct nfs4_slot *slot; + + slot = kzalloc_obj(*slot, gfp_mask); + if (slot) { + slot->table = tbl; + slot->slot_nr = slotid; + slot->seq_nr = seq_init; + slot->seq_nr_highest_sent = seq_init; + slot->seq_nr_last_acked = seq_init - 1; + } + return slot; +} + +static struct nfs4_slot *nfs4_find_or_create_slot(struct nfs4_slot_table *tbl, + u32 slotid, u32 seq_init, gfp_t gfp_mask) +{ + struct nfs4_slot **p, *slot; + + p = &tbl->slots; + for (;;) { + if (*p == NULL) { + *p = nfs4_new_slot(tbl, tbl->max_slots, + seq_init, gfp_mask); + if (*p == NULL) + break; + tbl->max_slots++; + } + slot = *p; + if (slot->slot_nr == slotid) + return slot; + p = &slot->next; + } + return ERR_PTR(-ENOMEM); +} + +static void nfs4_lock_slot(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot) +{ + u32 slotid = slot->slot_nr; + + __set_bit(slotid, tbl->used_slots); + if (slotid > tbl->highest_used_slotid || + tbl->highest_used_slotid == NFS4_NO_SLOT) + tbl->highest_used_slotid = slotid; + slot->generation = tbl->generation; +} + +/* + * nfs4_try_to_lock_slot - Given a slot try to allocate it + * + * Note: must be called with the slot_tbl_lock held. + */ +bool nfs4_try_to_lock_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot) +{ + if (nfs4_test_locked_slot(tbl, slot->slot_nr)) + return false; + nfs4_lock_slot(tbl, slot); + return true; +} + +/* + * nfs4_lookup_slot - Find a slot but don't allocate it + * + * Note: must be called with the slot_tbl_lock held. + */ +struct nfs4_slot *nfs4_lookup_slot(struct nfs4_slot_table *tbl, u32 slotid) +{ + if (slotid <= tbl->max_slotid) + return nfs4_find_or_create_slot(tbl, slotid, 0, GFP_NOWAIT); + return ERR_PTR(-E2BIG); +} + +static int nfs4_slot_get_seqid(struct nfs4_slot_table *tbl, u32 slotid, + u32 *seq_nr) + __must_hold(&tbl->slot_tbl_lock) +{ + struct nfs4_slot *slot; + int ret; + + slot = nfs4_lookup_slot(tbl, slotid); + ret = PTR_ERR_OR_ZERO(slot); + if (!ret) + *seq_nr = slot->seq_nr; + + return ret; +} + +/* + * nfs4_slot_seqid_in_use - test if a slot sequence id is still in use + * + * Given a slot table, slot id and sequence number, determine if the + * RPC call in question is still in flight. This function is mainly + * intended for use by the callback channel. + */ +static bool nfs4_slot_seqid_in_use(struct nfs4_slot_table *tbl, + u32 slotid, u32 seq_nr) +{ + u32 cur_seq = 0; + bool ret = false; + + spin_lock(&tbl->slot_tbl_lock); + if (nfs4_slot_get_seqid(tbl, slotid, &cur_seq) == 0 && + cur_seq == seq_nr && test_bit(slotid, tbl->used_slots)) + ret = true; + spin_unlock(&tbl->slot_tbl_lock); + return ret; +} + +/* + * nfs4_slot_wait_on_seqid - wait until a slot sequence id is complete + * + * Given a slot table, slot id and sequence number, wait until the + * corresponding RPC call completes. This function is mainly + * intended for use by the callback channel. + */ +int nfs4_slot_wait_on_seqid(struct nfs4_slot_table *tbl, + u32 slotid, u32 seq_nr, + unsigned long timeout) +{ + if (wait_event_timeout(tbl->slot_waitq, + !nfs4_slot_seqid_in_use(tbl, slotid, seq_nr), + timeout) == 0) + return -ETIMEDOUT; + return 0; +} + +/* + * nfs4_alloc_slot - efficiently look for a free slot + * + * nfs4_alloc_slot looks for an unset bit in the used_slots bitmap. + * If found, we mark the slot as used, update the highest_used_slotid, + * and respectively set up the sequence operation args. + * + * Note: must be called with under the slot_tbl_lock. + */ +struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl) +{ + struct nfs4_slot *ret = ERR_PTR(-EBUSY); + u32 slotid; + + dprintk("--> %s used_slots=%04lx highest_used=%u max_slots=%u\n", + __func__, tbl->used_slots[0], tbl->highest_used_slotid, + tbl->max_slotid + 1); + slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slotid + 1); + if (slotid <= tbl->max_slotid) { + ret = nfs4_find_or_create_slot(tbl, slotid, 1, GFP_NOWAIT); + if (!IS_ERR(ret)) + nfs4_lock_slot(tbl, ret); + } + dprintk("<-- %s used_slots=%04lx highest_used=%u slotid=%u\n", + __func__, tbl->used_slots[0], tbl->highest_used_slotid, + !IS_ERR(ret) ? ret->slot_nr : NFS4_NO_SLOT); + return ret; +} + +static int nfs4_grow_slot_table(struct nfs4_slot_table *tbl, + u32 max_reqs, u32 ivalue) +{ + if (max_reqs <= tbl->max_slots) + return 0; + if (!IS_ERR(nfs4_find_or_create_slot(tbl, max_reqs - 1, ivalue, GFP_NOFS))) + return 0; + return -ENOMEM; +} + +static void nfs4_reset_slot_table(struct nfs4_slot_table *tbl, + u32 server_highest_slotid, + u32 ivalue) +{ + struct nfs4_slot **p; + + nfs4_shrink_slot_table(tbl, server_highest_slotid + 1); + p = &tbl->slots; + while (*p) { + (*p)->seq_nr = ivalue; + (*p)->seq_nr_highest_sent = ivalue; + (*p)->seq_nr_last_acked = ivalue - 1; + p = &(*p)->next; + } + tbl->highest_used_slotid = NFS4_NO_SLOT; + tbl->target_highest_slotid = server_highest_slotid; + tbl->server_highest_slotid = server_highest_slotid; + tbl->d_target_highest_slotid = 0; + tbl->d2_target_highest_slotid = 0; + tbl->max_slotid = server_highest_slotid; +} + +/* + * (re)Initialise a slot table + */ +static int nfs4_realloc_slot_table(struct nfs4_slot_table *tbl, + u32 max_reqs, u32 ivalue) +{ + int ret; + + dprintk("--> %s: max_reqs=%u, tbl->max_slots %u\n", __func__, + max_reqs, tbl->max_slots); + + if (max_reqs > NFS4_MAX_SLOT_TABLE) + max_reqs = NFS4_MAX_SLOT_TABLE; + + ret = nfs4_grow_slot_table(tbl, max_reqs, ivalue); + if (ret) + goto out; + + spin_lock(&tbl->slot_tbl_lock); + nfs4_reset_slot_table(tbl, max_reqs - 1, ivalue); + spin_unlock(&tbl->slot_tbl_lock); + + dprintk("%s: tbl=%p slots=%p max_slots=%u\n", __func__, + tbl, tbl->slots, tbl->max_slots); +out: + dprintk("<-- %s: return %d\n", __func__, ret); + return ret; +} + +/* + * nfs4_release_slot_table - release all slot table entries + */ +static void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +{ + nfs4_shrink_slot_table(tbl, 0); +} + +/** + * nfs4_shutdown_slot_table - release resources attached to a slot table + * @tbl: slot table to shut down + * + */ +void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl) +{ + nfs4_release_slot_table(tbl); + rpc_destroy_wait_queue(&tbl->slot_tbl_waitq); +} + +/** + * nfs4_setup_slot_table - prepare a stand-alone slot table for use + * @tbl: slot table to set up + * @max_reqs: maximum number of requests allowed + * @queue: name to give RPC wait queue + * + * Returns zero on success, or a negative errno. + */ +int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs, + const char *queue) +{ + nfs4_init_slot_table(tbl, queue); + return nfs4_realloc_slot_table(tbl, max_reqs, 0); +} + +static bool nfs41_assign_slot(struct rpc_task *task, void *pslot) +{ + struct nfs4_sequence_args *args = task->tk_msg.rpc_argp; + struct nfs4_sequence_res *res = task->tk_msg.rpc_resp; + struct nfs4_slot *slot = pslot; + struct nfs4_slot_table *tbl = slot->table; + + if (nfs4_slot_tbl_draining(tbl) && !args->sa_privileged) + return false; + slot->generation = tbl->generation; + args->sa_slot = slot; + res->sr_timestamp = jiffies; + res->sr_slot = slot; + res->sr_status_flags = 0; + res->sr_status = 1; + return true; +} + +static bool __nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot) +{ + if (rpc_wake_up_first(&tbl->slot_tbl_waitq, nfs41_assign_slot, slot)) + return true; + return false; +} + +bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot) +{ + if (slot->slot_nr > tbl->max_slotid) + return false; + return __nfs41_wake_and_assign_slot(tbl, slot); +} + +static bool nfs41_try_wake_next_slot_table_entry(struct nfs4_slot_table *tbl) +{ + struct nfs4_slot *slot = nfs4_alloc_slot(tbl); + if (!IS_ERR(slot)) { + bool ret = __nfs41_wake_and_assign_slot(tbl, slot); + if (ret) + return ret; + nfs4_free_slot(tbl, slot); + } + return false; +} + +void nfs41_wake_slot_table(struct nfs4_slot_table *tbl) +{ + for (;;) { + if (!nfs41_try_wake_next_slot_table_entry(tbl)) + break; + } +} + +static void nfs41_set_max_slotid_locked(struct nfs4_slot_table *tbl, + u32 target_highest_slotid) +{ + u32 max_slotid; + + max_slotid = min(NFS4_MAX_SLOT_TABLE - 1, target_highest_slotid); + if (max_slotid > tbl->server_highest_slotid) + max_slotid = tbl->server_highest_slotid; + if (max_slotid > tbl->target_highest_slotid) + max_slotid = tbl->target_highest_slotid; + tbl->max_slotid = max_slotid; + nfs41_wake_slot_table(tbl); +} + +/* Update the client's idea of target_highest_slotid */ +static void nfs41_set_target_slotid_locked(struct nfs4_slot_table *tbl, + u32 target_highest_slotid) +{ + if (tbl->target_highest_slotid == target_highest_slotid) + return; + tbl->target_highest_slotid = target_highest_slotid; + tbl->generation++; +} + +void nfs41_set_target_slotid(struct nfs4_slot_table *tbl, + u32 target_highest_slotid) +{ + spin_lock(&tbl->slot_tbl_lock); + nfs41_set_target_slotid_locked(tbl, target_highest_slotid); + tbl->d_target_highest_slotid = 0; + tbl->d2_target_highest_slotid = 0; + nfs41_set_max_slotid_locked(tbl, target_highest_slotid); + spin_unlock(&tbl->slot_tbl_lock); +} + +static void nfs41_set_server_slotid_locked(struct nfs4_slot_table *tbl, + u32 highest_slotid) +{ + if (tbl->server_highest_slotid == highest_slotid) + return; + if (tbl->highest_used_slotid > highest_slotid) + return; + /* Deallocate slots */ + nfs4_shrink_slot_table(tbl, highest_slotid + 1); + tbl->server_highest_slotid = highest_slotid; +} + +static s32 nfs41_derivative_target_slotid(s32 s1, s32 s2) +{ + s1 -= s2; + if (s1 == 0) + return 0; + if (s1 < 0) + return (s1 - 1) >> 1; + return (s1 + 1) >> 1; +} + +static int nfs41_sign_s32(s32 s1) +{ + if (s1 > 0) + return 1; + if (s1 < 0) + return -1; + return 0; +} + +static bool nfs41_same_sign_or_zero_s32(s32 s1, s32 s2) +{ + if (!s1 || !s2) + return true; + return nfs41_sign_s32(s1) == nfs41_sign_s32(s2); +} + +/* Try to eliminate outliers by checking for sharp changes in the + * derivatives and second derivatives + */ +static bool nfs41_is_outlier_target_slotid(struct nfs4_slot_table *tbl, + u32 new_target) +{ + s32 d_target, d2_target; + bool ret = true; + + d_target = nfs41_derivative_target_slotid(new_target, + tbl->target_highest_slotid); + d2_target = nfs41_derivative_target_slotid(d_target, + tbl->d_target_highest_slotid); + /* Is first derivative same sign? */ + if (nfs41_same_sign_or_zero_s32(d_target, tbl->d_target_highest_slotid)) + ret = false; + /* Is second derivative same sign? */ + if (nfs41_same_sign_or_zero_s32(d2_target, tbl->d2_target_highest_slotid)) + ret = false; + tbl->d_target_highest_slotid = d_target; + tbl->d2_target_highest_slotid = d2_target; + return ret; +} + +void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot, + struct nfs4_sequence_res *res) +{ + u32 target_highest_slotid = min(res->sr_target_highest_slotid, + NFS4_MAX_SLOTID); + u32 highest_slotid = min(res->sr_highest_slotid, NFS4_MAX_SLOTID); + + spin_lock(&tbl->slot_tbl_lock); + if (!nfs41_is_outlier_target_slotid(tbl, target_highest_slotid)) + nfs41_set_target_slotid_locked(tbl, target_highest_slotid); + if (tbl->generation == slot->generation) + nfs41_set_server_slotid_locked(tbl, highest_slotid); + nfs41_set_max_slotid_locked(tbl, target_highest_slotid); + spin_unlock(&tbl->slot_tbl_lock); +} + +static void nfs4_release_session_slot_tables(struct nfs4_session *session) +{ + nfs4_release_slot_table(&session->fc_slot_table); + nfs4_release_slot_table(&session->bc_slot_table); +} + +/* + * Initialize or reset the forechannel and backchannel tables + */ +int nfs4_setup_session_slot_tables(struct nfs4_session *ses) +{ + struct nfs4_slot_table *tbl; + int status; + + dprintk("--> %s\n", __func__); + /* Fore channel */ + tbl = &ses->fc_slot_table; + tbl->session = ses; + status = nfs4_realloc_slot_table(tbl, ses->fc_attrs.max_reqs, 1); + if (status || !(ses->flags & SESSION4_BACK_CHAN)) /* -ENOMEM */ + return status; + /* Back channel */ + tbl = &ses->bc_slot_table; + tbl->session = ses; + status = nfs4_realloc_slot_table(tbl, ses->bc_attrs.max_reqs, 0); + if (status && tbl->slots == NULL) + /* Fore and back channel share a connection so get + * both slot tables or neither */ + nfs4_release_session_slot_tables(ses); + return status; +} + +struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp) +{ + struct nfs4_session *session; + + session = kzalloc_obj(struct nfs4_session, GFP_NOFS); + if (!session) + return NULL; + + nfs4_init_slot_table(&session->fc_slot_table, "ForeChannel Slot table"); + nfs4_init_slot_table(&session->bc_slot_table, "BackChannel Slot table"); + session->session_state = 1<clp = clp; + return session; +} + +static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +{ + nfs4_shutdown_slot_table(&session->fc_slot_table); + nfs4_shutdown_slot_table(&session->bc_slot_table); +} + +void nfs4_destroy_session(struct nfs4_session *session) +{ + struct rpc_xprt *xprt; + const struct cred *cred; + + cred = nfs4_get_clid_cred(session->clp); + nfs4_proc_destroy_session(session, cred); + put_cred(cred); + + rcu_read_lock(); + xprt = rcu_dereference(session->clp->cl_rpcclient->cl_xprt); + rcu_read_unlock(); + dprintk("%s Destroy backchannel for xprt %p\n", + __func__, xprt); + xprt_destroy_backchannel(xprt, NFS41_BC_MIN_CALLBACKS); + nfs4_destroy_session_slot_tables(session); + kfree(session); +} + +/* + * With sessions, the client is not marked ready until after a + * successful EXCHANGE_ID and CREATE_SESSION. + * + * Map errors cl_cons_state errors to EPROTONOSUPPORT to indicate + * other versions of NFS can be tried. + */ +static int nfs41_check_session_ready(struct nfs_client *clp) +{ + int ret; + + if (clp->cl_cons_state == NFS_CS_SESSION_INITING) { + ret = nfs4_client_recover_expired_lease(clp); + if (ret) + return ret; + } + if (clp->cl_cons_state < NFS_CS_READY) + return -EPROTONOSUPPORT; + smp_rmb(); + return 0; +} + +int nfs4_init_session(struct nfs_client *clp) +{ + if (!nfs4_has_session(clp)) + return 0; + + clear_bit(NFS4_SESSION_INITING, &clp->cl_session->session_state); + return nfs41_check_session_ready(clp); +} + +int nfs4_init_ds_session(struct nfs_client *clp, unsigned long lease_time) +{ + struct nfs4_session *session = clp->cl_session; + int ret; + + spin_lock(&clp->cl_lock); + if (test_and_clear_bit(NFS4_SESSION_INITING, &session->session_state)) { + /* + * Do not set NFS_CS_CHECK_LEASE_TIME instead set the + * DS lease to be equal to the MDS lease. + */ + clp->cl_lease_time = lease_time; + clp->cl_last_renewal = jiffies; + } + spin_unlock(&clp->cl_lock); + + ret = nfs41_check_session_ready(clp); + if (ret) + return ret; + /* Test for the DS role */ + if (!is_ds_client(clp)) + return -ENODEV; + return 0; +} +EXPORT_SYMBOL_GPL(nfs4_init_ds_session); diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h new file mode 100644 index 000000000..d2569f599 --- /dev/null +++ b/fs/nfs/nfs4session.h @@ -0,0 +1,158 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * fs/nfs/nfs4session.h + * + * Copyright (c) 2012 Trond Myklebust + * + */ +#ifndef __LINUX_FS_NFS_NFS4SESSION_H +#define __LINUX_FS_NFS_NFS4SESSION_H + +/* maximum number of slots to use */ +#define NFS4_DEF_SLOT_TABLE_SIZE (64U) +#define NFS4_DEF_CB_SLOT_TABLE_SIZE (16U) +#define NFS4_MAX_SLOT_TABLE (1024U) +#define NFS4_MAX_SLOTID (NFS4_MAX_SLOT_TABLE - 1U) +#define NFS4_NO_SLOT ((u32)-1) + +#if IS_ENABLED(CONFIG_NFS_V4) + +/* Sessions slot seqid */ +struct nfs4_slot { + struct nfs4_slot_table *table; + struct nfs4_slot *next; + unsigned long generation; + u32 slot_nr; + u32 seq_nr; + u32 seq_nr_last_acked; + u32 seq_nr_highest_sent; + unsigned int privileged : 1, + seq_done : 1; +}; + +/* Sessions */ +enum nfs4_slot_tbl_state { + NFS4_SLOT_TBL_DRAINING, +}; + +#define SLOT_TABLE_SZ DIV_ROUND_UP(NFS4_MAX_SLOT_TABLE, BITS_PER_LONG) +struct nfs4_slot_table { + struct nfs4_session *session; /* Parent session */ + struct nfs4_slot *slots; /* seqid per slot */ + unsigned long used_slots[SLOT_TABLE_SZ]; /* used/unused bitmap */ + spinlock_t slot_tbl_lock; + struct rpc_wait_queue slot_tbl_waitq; /* allocators may wait here */ + wait_queue_head_t slot_waitq; /* Completion wait on slot */ + u32 max_slots; /* # slots in table */ + u32 max_slotid; /* Max allowed slotid value */ + u32 highest_used_slotid; /* sent to server on each SEQ. + * op for dynamic resizing */ + u32 target_highest_slotid; /* Server max_slot target */ + u32 server_highest_slotid; /* Server highest slotid */ + s32 d_target_highest_slotid; /* Derivative */ + s32 d2_target_highest_slotid; /* 2nd derivative */ + unsigned long generation; /* Generation counter for + target_highest_slotid */ + struct completion complete; + unsigned long slot_tbl_state; +}; + +/* + * Session related parameters + */ +struct nfs4_session { + struct nfs4_sessionid sess_id; + u32 flags; + unsigned long session_state; + u32 hash_alg; + u32 ssv_len; + + /* The fore and back channel */ + struct nfs4_channel_attrs fc_attrs; + struct nfs4_slot_table fc_slot_table; + struct nfs4_channel_attrs bc_attrs; + struct nfs4_slot_table bc_slot_table; + struct nfs_client *clp; +}; + +enum nfs4_session_state { + NFS4_SESSION_INITING, + NFS4_SESSION_ESTABLISHED, +}; + +extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, + unsigned int max_reqs, const char *queue); +extern void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl); +extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl); +extern struct nfs4_slot *nfs4_lookup_slot(struct nfs4_slot_table *tbl, u32 slotid); +extern int nfs4_slot_wait_on_seqid(struct nfs4_slot_table *tbl, + u32 slotid, u32 seq_nr, + unsigned long timeout); +extern bool nfs4_try_to_lock_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot); +extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot); +extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); +bool nfs41_wake_and_assign_slot(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot); +void nfs41_wake_slot_table(struct nfs4_slot_table *tbl); + +static inline bool nfs4_slot_tbl_draining(struct nfs4_slot_table *tbl) +{ + return !!test_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); +} + +static inline bool nfs4_test_locked_slot(const struct nfs4_slot_table *tbl, + u32 slotid) +{ + return !!test_bit(slotid, tbl->used_slots); +} + +static inline struct nfs4_session *nfs4_get_session(const struct nfs_client *clp) +{ + return clp->cl_session; +} + +extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl, + u32 target_highest_slotid); +extern void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, + struct nfs4_slot *slot, + struct nfs4_sequence_res *res); + +extern int nfs4_setup_session_slot_tables(struct nfs4_session *ses); + +extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp); +extern void nfs4_destroy_session(struct nfs4_session *session); +extern int nfs4_init_session(struct nfs_client *clp); +extern int nfs4_init_ds_session(struct nfs_client *, unsigned long); + +/* + * Determine if sessions are in use. + */ +static inline int nfs4_has_session(const struct nfs_client *clp) +{ + if (clp->cl_session) + return 1; + return 0; +} + +static inline int nfs4_has_persistent_session(const struct nfs_client *clp) +{ + if (nfs4_has_session(clp)) + return (clp->cl_session->flags & SESSION4_PERSIST); + return 0; +} + +static inline void nfs4_copy_sessionid(struct nfs4_sessionid *dst, + const struct nfs4_sessionid *src) +{ + memcpy(dst->data, src->data, NFS4_MAX_SESSIONID_LEN); +} + +/* + * nfs_session_id_hash - calculate the crc32 hash for the session id + * @session - pointer to session + */ +#define nfs_session_id_hash(sess_id) \ + (~crc32_le(0xFFFFFFFF, &(sess_id)->data[0], sizeof((sess_id)->data))) + +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ +#endif /* __LINUX_FS_NFS_NFS4SESSION_H */ diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c new file mode 100644 index 000000000..305a772e5 --- /dev/null +++ b/fs/nfs/nfs4state.c @@ -0,0 +1,2713 @@ +/* + * fs/nfs/nfs4state.c + * + * Client-side XDR for NFSv4. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Kendrick Smith + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Implementation of the NFSv4 state model. For the time being, + * this is minimal, but will be made much more complex in a + * subsequent patch. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "nfs4_fs.h" +#include "nfs40.h" +#include "callback.h" +#include "delegation.h" +#include "internal.h" +#include "nfs4idmap.h" +#include "nfs4session.h" +#include "pnfs.h" +#include "netns.h" +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_STATE + +#define OPENOWNER_POOL_SIZE 8 + +static void nfs4_state_start_reclaim_reboot(struct nfs_client *clp); + +const nfs4_stateid zero_stateid = { + { .data = { 0 } }, + .type = NFS4_SPECIAL_STATEID_TYPE, +}; +const nfs4_stateid invalid_stateid = { + { + /* Funky initialiser keeps older gcc versions happy */ + .data = { 0xff, 0xff, 0xff, 0xff, 0 }, + }, + .type = NFS4_INVALID_STATEID_TYPE, +}; + +const nfs4_stateid current_stateid = { + { + /* Funky initialiser keeps older gcc versions happy */ + .data = { 0x0, 0x0, 0x0, 0x1, 0 }, + }, + .type = NFS4_SPECIAL_STATEID_TYPE, +}; + +static DEFINE_MUTEX(nfs_clid_init_mutex); + +static int nfs4_setup_state_renewal(struct nfs_client *clp) +{ + int status; + struct nfs_fsinfo fsinfo; + + if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { + nfs4_schedule_state_renewal(clp); + return 0; + } + + status = nfs4_proc_get_lease_time(clp, &fsinfo); + if (status == 0) { + nfs4_set_lease_period(clp, fsinfo.lease_time); + nfs4_schedule_state_renewal(clp); + } + + return status; +} + +int nfs4_init_clientid(struct nfs_client *clp, const struct cred *cred) +{ + struct nfs4_setclientid_res clid = { + .clientid = clp->cl_clientid, + .confirm = clp->cl_confirm, + }; + unsigned short port; + int status; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + + if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state)) + goto do_confirm; + port = nn->nfs_callback_tcpport; + if (clp->cl_addr.ss_family == AF_INET6) + port = nn->nfs_callback_tcpport6; + + status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid); + if (status != 0) + goto out; + clp->cl_clientid = clid.clientid; + clp->cl_confirm = clid.confirm; + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); +do_confirm: + status = nfs4_proc_setclientid_confirm(clp, &clid, cred); + if (status != 0) + goto out; + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + nfs4_setup_state_renewal(clp); +out: + return status; +} + +const struct cred *nfs4_get_machine_cred(struct nfs_client *clp) +{ + return get_cred(rpc_machine_cred()); +} + +static void nfs4_root_machine_cred(struct nfs_client *clp) +{ + + /* Force root creds instead of machine */ + clp->cl_principal = NULL; + clp->cl_rpcclient->cl_principal = NULL; +} + +static const struct cred * +nfs4_get_renew_cred_server_locked(struct nfs_server *server) +{ + const struct cred *cred = NULL; + struct nfs4_state_owner *sp; + struct rb_node *pos; + + for (pos = rb_first(&server->state_owners); + pos != NULL; + pos = rb_next(pos)) { + sp = rb_entry(pos, struct nfs4_state_owner, so_server_node); + if (list_empty(&sp->so_states)) + continue; + cred = get_cred(sp->so_cred); + break; + } + return cred; +} + +/** + * nfs4_get_renew_cred - Acquire credential for a renew operation + * @clp: client state handle + * + * Returns an rpc_cred with reference count bumped, or NULL. + * Caller must hold clp->cl_lock. + */ +const struct cred *nfs4_get_renew_cred(struct nfs_client *clp) +{ + const struct cred *cred = NULL; + struct nfs_server *server; + + /* Use machine credentials if available */ + cred = nfs4_get_machine_cred(clp); + if (cred != NULL) + goto out; + + spin_lock(&clp->cl_lock); + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + cred = nfs4_get_renew_cred_server_locked(server); + if (cred != NULL) + break; + } + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); + +out: + return cred; +} + +static void nfs4_end_drain_slot_table(struct nfs4_slot_table *tbl) +{ + if (test_and_clear_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state)) { + spin_lock(&tbl->slot_tbl_lock); + nfs41_wake_slot_table(tbl); + spin_unlock(&tbl->slot_tbl_lock); + } +} + +static void nfs4_end_drain_session(struct nfs_client *clp) +{ + struct nfs4_session *ses = clp->cl_session; + + if (clp->cl_slot_tbl) { + nfs4_end_drain_slot_table(clp->cl_slot_tbl); + return; + } + + if (ses != NULL) { + nfs4_end_drain_slot_table(&ses->bc_slot_table); + nfs4_end_drain_slot_table(&ses->fc_slot_table); + } +} + +static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl) +{ + set_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state); + spin_lock(&tbl->slot_tbl_lock); + if (tbl->highest_used_slotid != NFS4_NO_SLOT) { + reinit_completion(&tbl->complete); + spin_unlock(&tbl->slot_tbl_lock); + return wait_for_completion_interruptible(&tbl->complete); + } + spin_unlock(&tbl->slot_tbl_lock); + return 0; +} + +static int nfs4_begin_drain_session(struct nfs_client *clp) +{ + struct nfs4_session *ses = clp->cl_session; + int ret; + + if (clp->cl_slot_tbl) + return nfs4_drain_slot_tbl(clp->cl_slot_tbl); + + /* back channel */ + ret = nfs4_drain_slot_tbl(&ses->bc_slot_table); + if (ret) + return ret; + /* fore channel */ + return nfs4_drain_slot_tbl(&ses->fc_slot_table); +} + +static void nfs41_finish_session_reset(struct nfs_client *clp) +{ + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + /* create_session negotiated new slot table */ + clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); + nfs4_setup_state_renewal(clp); +} + +int nfs41_init_clientid(struct nfs_client *clp, const struct cred *cred) +{ + int status; + + if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state)) + goto do_confirm; + status = nfs4_proc_exchange_id(clp, cred); + if (status != 0) + goto out; + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); +do_confirm: + status = nfs4_proc_create_session(clp, cred); + if (status != 0) + goto out; + if (!(clp->cl_exchange_flags & EXCHGID4_FLAG_CONFIRMED_R)) + nfs4_state_start_reclaim_reboot(clp); + nfs41_finish_session_reset(clp); + nfs_mark_client_ready(clp, NFS_CS_READY); +out: + return status; +} + +/** + * nfs41_discover_server_trunking - Detect server IP address trunking (mv1) + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * @cred: credential to use for trunking test + * + * Returns NFS4_OK, a negative errno, or a negative NFS4ERR status. + * If NFS4_OK is returned, an nfs_client pointer is planted in + * "result". + * + * Note: The returned client may not yet be marked ready. + */ +int nfs41_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result, + const struct cred *cred) +{ + int status; + + status = nfs4_proc_exchange_id(clp, cred); + if (status != NFS4_OK) + return status; + + status = nfs41_walk_client_list(clp, result, cred); + if (status < 0) + return status; + if (clp != *result) + return 0; + + /* + * Purge state if the client id was established in a prior + * instance and the client id could not have arrived on the + * server via Transparent State Migration. + */ + if (clp->cl_exchange_flags & EXCHGID4_FLAG_CONFIRMED_R) { + if (!test_bit(NFS_CS_TSM_POSSIBLE, &clp->cl_flags)) + set_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state); + else + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + } + nfs4_schedule_state_manager(clp); + status = nfs_wait_client_init_complete(clp); + if (status < 0) + nfs_put_client(clp); + return status; +} + +/** + * nfs4_get_clid_cred - Acquire credential for a setclientid operation + * @clp: client state handle + * + * Returns a cred with reference count bumped, or NULL. + */ +const struct cred *nfs4_get_clid_cred(struct nfs_client *clp) +{ + const struct cred *cred; + + cred = nfs4_get_machine_cred(clp); + return cred; +} + +static struct nfs4_state_owner * +nfs4_find_state_owner_locked(struct nfs_server *server, const struct cred *cred) +{ + struct rb_node **p = &server->state_owners.rb_node, + *parent = NULL; + struct nfs4_state_owner *sp; + int cmp; + + while (*p != NULL) { + parent = *p; + sp = rb_entry(parent, struct nfs4_state_owner, so_server_node); + cmp = cred_fscmp(cred, sp->so_cred); + + if (cmp < 0) + p = &parent->rb_left; + else if (cmp > 0) + p = &parent->rb_right; + else { + if (!list_empty(&sp->so_lru)) + list_del_init(&sp->so_lru); + atomic_inc(&sp->so_count); + return sp; + } + } + return NULL; +} + +static struct nfs4_state_owner * +nfs4_insert_state_owner_locked(struct nfs4_state_owner *new) +{ + struct nfs_server *server = new->so_server; + struct rb_node **p = &server->state_owners.rb_node, + *parent = NULL; + struct nfs4_state_owner *sp; + int cmp; + + while (*p != NULL) { + parent = *p; + sp = rb_entry(parent, struct nfs4_state_owner, so_server_node); + cmp = cred_fscmp(new->so_cred, sp->so_cred); + + if (cmp < 0) + p = &parent->rb_left; + else if (cmp > 0) + p = &parent->rb_right; + else { + if (!list_empty(&sp->so_lru)) + list_del_init(&sp->so_lru); + atomic_inc(&sp->so_count); + return sp; + } + } + rb_link_node(&new->so_server_node, parent, p); + rb_insert_color(&new->so_server_node, &server->state_owners); + return new; +} + +static void +nfs4_remove_state_owner_locked(struct nfs4_state_owner *sp) +{ + struct nfs_server *server = sp->so_server; + + if (!RB_EMPTY_NODE(&sp->so_server_node)) + rb_erase(&sp->so_server_node, &server->state_owners); +} + +static void +nfs4_init_seqid_counter(struct nfs_seqid_counter *sc) +{ + sc->create_time = ktime_get(); + sc->flags = 0; + sc->counter = 0; + spin_lock_init(&sc->lock); + INIT_LIST_HEAD(&sc->list); + rpc_init_wait_queue(&sc->wait, "Seqid_waitqueue"); +} + +static void +nfs4_destroy_seqid_counter(struct nfs_seqid_counter *sc) +{ + rpc_destroy_wait_queue(&sc->wait); +} + +/* + * nfs4_alloc_state_owner(): this is called on the OPEN or CREATE path to + * create a new state_owner. + * + */ +static struct nfs4_state_owner * +nfs4_alloc_state_owner(struct nfs_server *server, + const struct cred *cred, + gfp_t gfp_flags) +{ + struct nfs4_state_owner *sp; + + sp = kzalloc_obj(*sp, gfp_flags); + if (!sp) + return NULL; + sp->so_seqid.owner_id = atomic64_inc_return(&server->owner_ctr); + sp->so_server = server; + sp->so_cred = get_cred(cred); + spin_lock_init(&sp->so_lock); + INIT_LIST_HEAD(&sp->so_states); + nfs4_init_seqid_counter(&sp->so_seqid); + atomic_set(&sp->so_count, 1); + INIT_LIST_HEAD(&sp->so_lru); + mutex_init(&sp->so_delegreturn_mutex); + return sp; +} + +static void +nfs4_reset_state_owner(struct nfs4_state_owner *sp) +{ + /* This state_owner is no longer usable, but must + * remain in place so that state recovery can find it + * and the opens associated with it. + * It may also be used for new 'open' request to + * return a delegation to the server. + * So update the 'create_time' so that it looks like + * a new state_owner. This will cause the server to + * request an OPEN_CONFIRM to start a new sequence. + */ + sp->so_seqid.create_time = ktime_get(); +} + +static void nfs4_free_state_owner(struct nfs4_state_owner *sp) +{ + nfs4_destroy_seqid_counter(&sp->so_seqid); + put_cred(sp->so_cred); + kfree(sp); +} + +static void nfs4_gc_state_owners(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp, *tmp; + unsigned long time_min, time_max; + LIST_HEAD(doomed); + + spin_lock(&clp->cl_lock); + time_max = jiffies; + time_min = (long)time_max - (long)clp->cl_lease_time; + list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) { + /* NB: LRU is sorted so that oldest is at the head */ + if (time_in_range(sp->so_expires, time_min, time_max)) + break; + list_move(&sp->so_lru, &doomed); + nfs4_remove_state_owner_locked(sp); + } + spin_unlock(&clp->cl_lock); + + list_for_each_entry_safe(sp, tmp, &doomed, so_lru) { + list_del(&sp->so_lru); + nfs4_free_state_owner(sp); + } +} + +/** + * nfs4_get_state_owner - Look up a state owner given a credential + * @server: nfs_server to search + * @cred: RPC credential to match + * @gfp_flags: allocation mode + * + * Returns a pointer to an instantiated nfs4_state_owner struct, or NULL. + */ +struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, + const struct cred *cred, + gfp_t gfp_flags) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp, *new; + + spin_lock(&clp->cl_lock); + sp = nfs4_find_state_owner_locked(server, cred); + spin_unlock(&clp->cl_lock); + if (sp != NULL) + goto out; + new = nfs4_alloc_state_owner(server, cred, gfp_flags); + if (new == NULL) + goto out; + spin_lock(&clp->cl_lock); + sp = nfs4_insert_state_owner_locked(new); + spin_unlock(&clp->cl_lock); + if (sp != new) + nfs4_free_state_owner(new); +out: + nfs4_gc_state_owners(server); + return sp; +} + +/** + * nfs4_put_state_owner - Release a nfs4_state_owner + * @sp: state owner data to release + * + * Note that we keep released state owners on an LRU + * list. + * This caches valid state owners so that they can be + * reused, to avoid the OPEN_CONFIRM on minor version 0. + * It also pins the uniquifier of dropped state owners for + * a while, to ensure that those state owner names are + * never reused. + */ +void nfs4_put_state_owner(struct nfs4_state_owner *sp) +{ + struct nfs_server *server = sp->so_server; + struct nfs_client *clp = server->nfs_client; + + if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock)) + return; + + sp->so_expires = jiffies; + list_add_tail(&sp->so_lru, &server->state_owners_lru); + spin_unlock(&clp->cl_lock); +} + +/** + * nfs4_purge_state_owners - Release all cached state owners + * @server: nfs_server with cached state owners to release + * @head: resulting list of state owners + * + * Called at umount time. Remaining state owners will be on + * the LRU with ref count of zero. + * Note that the state owners are not freed, but are added + * to the list @head, which can later be used as an argument + * to nfs4_free_state_owners. + */ +void nfs4_purge_state_owners(struct nfs_server *server, struct list_head *head) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp, *tmp; + + spin_lock(&clp->cl_lock); + list_for_each_entry_safe(sp, tmp, &server->state_owners_lru, so_lru) { + list_move(&sp->so_lru, head); + nfs4_remove_state_owner_locked(sp); + } + spin_unlock(&clp->cl_lock); +} + +/** + * nfs4_free_state_owners - Release all cached state owners + * @head: resulting list of state owners + * + * Frees a list of state owners that was generated by + * nfs4_purge_state_owners + */ +void nfs4_free_state_owners(struct list_head *head) +{ + struct nfs4_state_owner *sp, *tmp; + + list_for_each_entry_safe(sp, tmp, head, so_lru) { + list_del(&sp->so_lru); + nfs4_free_state_owner(sp); + } +} + +static struct nfs4_state * +nfs4_alloc_open_state(void) +{ + struct nfs4_state *state; + + state = kzalloc_obj(*state, GFP_KERNEL_ACCOUNT); + if (!state) + return NULL; + refcount_set(&state->count, 1); + INIT_LIST_HEAD(&state->lock_states); + spin_lock_init(&state->state_lock); + seqlock_init(&state->seqlock); + init_waitqueue_head(&state->waitq); + return state; +} + +void +nfs4_state_set_mode_locked(struct nfs4_state *state, fmode_t fmode) +{ + if (state->state == fmode) + return; + /* NB! List reordering - see the reclaim code for why. */ + if ((fmode & FMODE_WRITE) != (state->state & FMODE_WRITE)) { + if (fmode & FMODE_WRITE) + list_move(&state->open_states, &state->owner->so_states); + else + list_move_tail(&state->open_states, &state->owner->so_states); + } + state->state = fmode; +} + +static struct nfs4_state * +__nfs4_find_state_byowner(struct inode *inode, struct nfs4_state_owner *owner) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs4_state *state; + + list_for_each_entry_rcu(state, &nfsi->open_states, inode_states) { + if (state->owner != owner) + continue; + if (!nfs4_valid_open_stateid(state)) + continue; + if (refcount_inc_not_zero(&state->count)) + return state; + } + return NULL; +} + +static void +nfs4_free_open_state(struct nfs4_state *state) +{ + kfree_rcu(state, rcu_head); +} + +struct nfs4_state * +nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner) +{ + struct nfs4_state *state, *new; + struct nfs_inode *nfsi = NFS_I(inode); + + rcu_read_lock(); + state = __nfs4_find_state_byowner(inode, owner); + rcu_read_unlock(); + if (state) + goto out; + new = nfs4_alloc_open_state(); + spin_lock(&owner->so_lock); + spin_lock(&inode->i_lock); + state = __nfs4_find_state_byowner(inode, owner); + if (state == NULL && new != NULL) { + state = new; + state->owner = owner; + atomic_inc(&owner->so_count); + ihold(inode); + state->inode = inode; + list_add_rcu(&state->inode_states, &nfsi->open_states); + spin_unlock(&inode->i_lock); + /* Note: The reclaim code dictates that we add stateless + * and read-only stateids to the end of the list */ + list_add_tail(&state->open_states, &owner->so_states); + spin_unlock(&owner->so_lock); + } else { + spin_unlock(&inode->i_lock); + spin_unlock(&owner->so_lock); + if (new) + nfs4_free_open_state(new); + } +out: + return state; +} + +void nfs4_put_open_state(struct nfs4_state *state) +{ + struct inode *inode = state->inode; + struct nfs4_state_owner *owner = state->owner; + + if (!refcount_dec_and_lock(&state->count, &owner->so_lock)) + return; + spin_lock(&inode->i_lock); + list_del_rcu(&state->inode_states); + list_del(&state->open_states); + spin_unlock(&inode->i_lock); + spin_unlock(&owner->so_lock); + nfs4_inode_return_delegation_on_close(inode); + iput(inode); + nfs4_free_open_state(state); + nfs4_put_state_owner(owner); +} + +/* + * Close the current file. + */ +static void __nfs4_close(struct nfs4_state *state, + fmode_t fmode, gfp_t gfp_mask, int wait) +{ + struct nfs4_state_owner *owner = state->owner; + int call_close = 0; + fmode_t newstate; + + atomic_inc(&owner->so_count); + /* Protect against nfs4_find_state() */ + spin_lock(&owner->so_lock); + switch (fmode & (FMODE_READ | FMODE_WRITE)) { + case FMODE_READ: + state->n_rdonly--; + break; + case FMODE_WRITE: + state->n_wronly--; + break; + case FMODE_READ|FMODE_WRITE: + state->n_rdwr--; + } + newstate = FMODE_READ|FMODE_WRITE; + if (state->n_rdwr == 0) { + if (state->n_rdonly == 0) { + newstate &= ~FMODE_READ; + call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags); + call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags); + } + if (state->n_wronly == 0) { + newstate &= ~FMODE_WRITE; + call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags); + call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags); + } + if (newstate == 0) + clear_bit(NFS_DELEGATED_STATE, &state->flags); + } + nfs4_state_set_mode_locked(state, newstate); + spin_unlock(&owner->so_lock); + + if (!call_close) { + nfs4_put_open_state(state); + nfs4_put_state_owner(owner); + } else + nfs4_do_close(state, gfp_mask, wait); +} + +void nfs4_close_state(struct nfs4_state *state, fmode_t fmode) +{ + __nfs4_close(state, fmode, GFP_KERNEL, 0); +} + +void nfs4_close_sync(struct nfs4_state *state, fmode_t fmode) +{ + __nfs4_close(state, fmode, GFP_KERNEL, 1); +} + +/* + * Search the state->lock_states for an existing lock_owner + * that is compatible with either of the given owners. + * If the second is non-zero, then the first refers to a Posix-lock + * owner (current->files) and the second refers to a flock/OFD + * owner (struct file*). In that case, prefer a match for the first + * owner. + * If both sorts of locks are held on the one file we cannot know + * which stateid was intended to be used, so a "correct" choice cannot + * be made. Failing that, a "consistent" choice is preferable. The + * consistent choice we make is to prefer the first owner, that of a + * Posix lock. + */ +static struct nfs4_lock_state * +__nfs4_find_lock_state(struct nfs4_state *state, + fl_owner_t owner, fl_owner_t owner2) +{ + struct nfs4_lock_state *pos, *ret = NULL; + list_for_each_entry(pos, &state->lock_states, ls_locks) { + if (pos->ls_owner == owner) { + ret = pos; + break; + } + if (pos->ls_owner == owner2) + ret = pos; + } + if (ret) + refcount_inc(&ret->ls_count); + return ret; +} + +/* + * Return a compatible lock_state. If no initialized lock_state structure + * exists, return an uninitialized one. + * + */ +static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t owner) +{ + struct nfs4_lock_state *lsp; + struct nfs_server *server = state->owner->so_server; + + lsp = kzalloc_obj(*lsp, GFP_KERNEL_ACCOUNT); + if (lsp == NULL) + return NULL; + nfs4_init_seqid_counter(&lsp->ls_seqid); + refcount_set(&lsp->ls_count, 1); + lsp->ls_state = state; + lsp->ls_owner = owner; + lsp->ls_seqid.owner_id = atomic64_inc_return(&server->owner_ctr); + INIT_LIST_HEAD(&lsp->ls_locks); + return lsp; +} + +void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp) +{ + nfs4_destroy_seqid_counter(&lsp->ls_seqid); + kfree(lsp); +} + +/* + * Return a compatible lock_state. If no initialized lock_state structure + * exists, return an uninitialized one. + * + */ +static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner) +{ + struct nfs4_lock_state *lsp, *new = NULL; + + for(;;) { + spin_lock(&state->state_lock); + lsp = __nfs4_find_lock_state(state, owner, NULL); + if (lsp != NULL) + break; + if (new != NULL) { + list_add(&new->ls_locks, &state->lock_states); + set_bit(LK_STATE_IN_USE, &state->flags); + lsp = new; + new = NULL; + break; + } + spin_unlock(&state->state_lock); + new = nfs4_alloc_lock_state(state, owner); + if (new == NULL) + return NULL; + } + spin_unlock(&state->state_lock); + if (new != NULL) + nfs4_free_lock_state(state->owner->so_server, new); + return lsp; +} + +/* + * Release reference to lock_state, and free it if we see that + * it is no longer in use + */ +void nfs4_put_lock_state(struct nfs4_lock_state *lsp) +{ + struct nfs_server *server; + struct nfs4_state *state; + + if (lsp == NULL) + return; + state = lsp->ls_state; + if (!refcount_dec_and_lock(&lsp->ls_count, &state->state_lock)) + return; + list_del(&lsp->ls_locks); + if (list_empty(&state->lock_states)) + clear_bit(LK_STATE_IN_USE, &state->flags); + spin_unlock(&state->state_lock); + server = state->owner->so_server; + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags)) { + struct nfs_client *clp = server->nfs_client; + + clp->cl_mvops->free_lock_state(server, lsp); + } else + nfs4_free_lock_state(server, lsp); +} + +static void nfs4_fl_copy_lock(struct file_lock *dst, struct file_lock *src) +{ + struct nfs4_lock_state *lsp = src->fl_u.nfs4_fl.owner; + + dst->fl_u.nfs4_fl.owner = lsp; + refcount_inc(&lsp->ls_count); +} + +static void nfs4_fl_release_lock(struct file_lock *fl) +{ + nfs4_put_lock_state(fl->fl_u.nfs4_fl.owner); +} + +static const struct file_lock_operations nfs4_fl_lock_ops = { + .fl_copy_lock = nfs4_fl_copy_lock, + .fl_release_private = nfs4_fl_release_lock, +}; + +int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl) +{ + struct nfs4_lock_state *lsp; + + if (fl->fl_ops != NULL) + return 0; + lsp = nfs4_get_lock_state(state, fl->c.flc_owner); + if (lsp == NULL) + return -ENOMEM; + fl->fl_u.nfs4_fl.owner = lsp; + fl->fl_ops = &nfs4_fl_lock_ops; + return 0; +} + +static int nfs4_copy_lock_stateid(nfs4_stateid *dst, + struct nfs4_state *state, + const struct nfs_lock_context *l_ctx) +{ + struct nfs4_lock_state *lsp; + fl_owner_t owner, fl_flock_owner; + int ret = -ENOENT; + + if (l_ctx == NULL) + goto out; + + if (test_bit(LK_STATE_IN_USE, &state->flags) == 0) + goto out; + + owner = l_ctx->lockowner; + fl_flock_owner = l_ctx->open_context->flock_owner; + + spin_lock(&state->state_lock); + lsp = __nfs4_find_lock_state(state, owner, fl_flock_owner); + if (lsp && test_bit(NFS_LOCK_LOST, &lsp->ls_flags)) + ret = -EIO; + else if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) { + nfs4_stateid_copy(dst, &lsp->ls_stateid); + ret = 0; + } + spin_unlock(&state->state_lock); + nfs4_put_lock_state(lsp); +out: + return ret; +} + +bool nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) +{ + bool ret; + const nfs4_stateid *src; + int seq; + + do { + ret = false; + src = &zero_stateid; + seq = read_seqbegin(&state->seqlock); + if (test_bit(NFS_OPEN_STATE, &state->flags)) { + src = &state->open_stateid; + ret = true; + } + nfs4_stateid_copy(dst, src); + } while (read_seqretry(&state->seqlock, seq)); + return ret; +} + +/* + * Byte-range lock aware utility to initialize the stateid of read/write + * requests. + */ +int nfs4_select_rw_stateid(struct nfs4_state *state, + fmode_t fmode, const struct nfs_lock_context *l_ctx, + nfs4_stateid *dst, const struct cred **cred) +{ + int ret; + + if (!nfs4_valid_open_stateid(state)) + return -EIO; + if (cred != NULL) + *cred = NULL; + ret = nfs4_copy_lock_stateid(dst, state, l_ctx); + if (ret == -EIO) + /* A lost lock - don't even consider delegations */ + goto out; + /* returns true if delegation stateid found and copied */ + if (nfs4_copy_delegation_stateid(state->inode, fmode, dst, cred)) { + ret = 0; + goto out; + } + if (ret != -ENOENT) + /* nfs4_copy_delegation_stateid() didn't over-write + * dst, so it still has the lock stateid which we now + * choose to use. + */ + goto out; + ret = nfs4_copy_open_stateid(dst, state) ? 0 : -EAGAIN; +out: + if (nfs_server_capable(state->inode, NFS_CAP_STATEID_NFSV41)) + dst->seqid = 0; + return ret; +} + +struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask) +{ + struct nfs_seqid *new; + + new = kmalloc_obj(*new, gfp_mask); + if (new == NULL) + return ERR_PTR(-ENOMEM); + new->sequence = counter; + INIT_LIST_HEAD(&new->list); + new->task = NULL; + return new; +} + +void nfs_release_seqid(struct nfs_seqid *seqid) +{ + struct nfs_seqid_counter *sequence; + + if (seqid == NULL || list_empty(&seqid->list)) + return; + sequence = seqid->sequence; + spin_lock(&sequence->lock); + if (list_is_first(&seqid->list, &sequence->list) && + !list_is_singular(&sequence->list)) { + struct nfs_seqid *next = list_next_entry(seqid, list); + rpc_wake_up_queued_task(&sequence->wait, next->task); + } + list_del_init(&seqid->list); + spin_unlock(&sequence->lock); +} + +void nfs_free_seqid(struct nfs_seqid *seqid) +{ + nfs_release_seqid(seqid); + kfree(seqid); +} + +/* + * Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or + * failed with a seqid incrementing error - + * see comments nfs4.h:seqid_mutating_error() + */ +static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) +{ + switch (status) { + case 0: + break; + case -NFS4ERR_BAD_SEQID: + if (seqid->sequence->flags & NFS_SEQID_CONFIRMED) + return; + pr_warn_ratelimited("NFS: v4 server returned a bad" + " sequence-id error on an" + " unconfirmed sequence %p!\n", + seqid->sequence); + return; + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_BADXDR: + case -NFS4ERR_RESOURCE: + case -NFS4ERR_NOFILEHANDLE: + case -NFS4ERR_MOVED: + /* Non-seqid mutating errors */ + return; + } + /* + * Note: no locking needed as we are guaranteed to be first + * on the sequence list + */ + seqid->sequence->counter++; +} + +void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid) +{ + struct nfs4_state_owner *sp; + + if (seqid == NULL) + return; + + sp = container_of(seqid->sequence, struct nfs4_state_owner, so_seqid); + if (status == -NFS4ERR_BAD_SEQID) + nfs4_reset_state_owner(sp); + if (!nfs4_has_session(sp->so_server->nfs_client)) + nfs_increment_seqid(status, seqid); +} + +/* + * Increment the seqid if the LOCK/LOCKU succeeded, or + * failed with a seqid incrementing error - + * see comments nfs4.h:seqid_mutating_error() + */ +void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid) +{ + if (seqid != NULL) + nfs_increment_seqid(status, seqid); +} + +int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task) +{ + struct nfs_seqid_counter *sequence; + int status = 0; + + if (seqid == NULL) + goto out; + sequence = seqid->sequence; + spin_lock(&sequence->lock); + seqid->task = task; + if (list_empty(&seqid->list)) + list_add_tail(&seqid->list, &sequence->list); + if (list_first_entry(&sequence->list, struct nfs_seqid, list) == seqid) + goto unlock; + rpc_sleep_on(&sequence->wait, task, NULL); + status = -EAGAIN; +unlock: + spin_unlock(&sequence->lock); +out: + return status; +} + +static int nfs4_run_state_manager(void *); + +static void nfs4_clear_state_manager_bit(struct nfs_client *clp) +{ + clear_and_wake_up_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); + rpc_wake_up(&clp->cl_rpcwaitq); +} + +/* + * Schedule the nfs_client asynchronous state management routine + */ +void nfs4_schedule_state_manager(struct nfs_client *clp) +{ + struct task_struct *task; + char buf[INET6_ADDRSTRLEN + sizeof("-manager") + 1]; + struct rpc_clnt *clnt = clp->cl_rpcclient; + bool swapon = false; + + if (clp->cl_cons_state < 0) + return; + + set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + + if (atomic_read(&clnt->cl_swapper)) { + swapon = !test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, + &clp->cl_state); + if (!swapon) { + wake_up_var(&clp->cl_state); + return; + } + } + + if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0) + return; + + __module_get(THIS_MODULE); + refcount_inc(&clp->cl_count); + + /* The rcu_read_lock() is not strictly necessary, as the state + * manager is the only thread that ever changes the rpc_xprt + * after it's initialized. At this point, we're single threaded. */ + rcu_read_lock(); + snprintf(buf, sizeof(buf), "%s-manager", + rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + rcu_read_unlock(); + task = kthread_run(nfs4_run_state_manager, clp, "%s", buf); + if (IS_ERR(task)) { + printk(KERN_ERR "%s: kthread_run: %ld\n", + __func__, PTR_ERR(task)); + if (!nfs_client_init_is_complete(clp)) + nfs_mark_client_ready(clp, PTR_ERR(task)); + if (swapon) + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + nfs4_clear_state_manager_bit(clp); + nfs_put_client(clp); + module_put(THIS_MODULE); + } +} + +/* + * Schedule a lease recovery attempt + */ +void nfs4_schedule_lease_recovery(struct nfs_client *clp) +{ + if (!clp) + return; + if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) + set_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state); + dprintk("%s: scheduling lease recovery for server %s\n", __func__, + clp->cl_hostname); + nfs4_schedule_state_manager(clp); +} +EXPORT_SYMBOL_GPL(nfs4_schedule_lease_recovery); + +/** + * nfs4_schedule_migration_recovery - trigger migration recovery + * + * @server: FSID that is migrating + * + * Returns zero if recovery has started, otherwise a negative NFS4ERR + * value is returned. + */ +int nfs4_schedule_migration_recovery(const struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + + if (server->fh_expire_type != NFS4_FH_PERSISTENT) { + pr_err("NFS: volatile file handles not supported (server %s)\n", + clp->cl_hostname); + return -NFS4ERR_IO; + } + + if (test_bit(NFS_MIG_FAILED, &server->mig_status)) + return -NFS4ERR_IO; + + dprintk("%s: scheduling migration recovery for (%llx:%llx) on %s\n", + __func__, + (unsigned long long)server->fsid.major, + (unsigned long long)server->fsid.minor, + clp->cl_hostname); + + set_bit(NFS_MIG_IN_TRANSITION, + &((struct nfs_server *)server)->mig_status); + set_bit(NFS4CLNT_MOVED, &clp->cl_state); + + nfs4_schedule_state_manager(clp); + return 0; +} +EXPORT_SYMBOL_GPL(nfs4_schedule_migration_recovery); + +/** + * nfs4_schedule_lease_moved_recovery - start lease-moved recovery + * + * @clp: server to check for moved leases + * + */ +void nfs4_schedule_lease_moved_recovery(struct nfs_client *clp) +{ + dprintk("%s: scheduling lease-moved recovery for client ID %llx on %s\n", + __func__, clp->cl_clientid, clp->cl_hostname); + + set_bit(NFS4CLNT_LEASE_MOVED, &clp->cl_state); + nfs4_schedule_state_manager(clp); +} +EXPORT_SYMBOL_GPL(nfs4_schedule_lease_moved_recovery); + +int nfs4_wait_clnt_recover(struct nfs_client *clp) +{ + int res; + + might_sleep(); + + refcount_inc(&clp->cl_count); + res = wait_on_bit_action(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING, + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + if (res) + goto out; + if (clp->cl_cons_state < 0) + res = clp->cl_cons_state; +out: + nfs_put_client(clp); + return res; +} + +int nfs4_client_recover_expired_lease(struct nfs_client *clp) +{ + unsigned int loop; + int ret; + + for (loop = NFS4_MAX_LOOP_ON_RECOVER; loop != 0; loop--) { + ret = nfs4_wait_clnt_recover(clp); + if (ret != 0) + break; + if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) && + !test_bit(NFS4CLNT_CHECK_LEASE,&clp->cl_state)) + break; + nfs4_schedule_state_manager(clp); + ret = -EIO; + } + return ret; +} + +static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state) +{ + + if (!nfs4_valid_open_stateid(state)) + return 0; + set_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags); + /* Don't recover state that expired before the reboot */ + if (test_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags)) { + clear_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags); + return 0; + } + set_bit(NFS_OWNER_RECLAIM_REBOOT, &state->owner->so_flags); + set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state); + return 1; +} + +int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state) +{ + if (!nfs4_valid_open_stateid(state)) + return 0; + set_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags); + clear_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags); + set_bit(NFS_OWNER_RECLAIM_NOGRACE, &state->owner->so_flags); + set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state); + return 1; +} + +int nfs4_schedule_stateid_recovery(const struct nfs_server *server, struct nfs4_state *state) +{ + struct nfs_client *clp = server->nfs_client; + + if (!nfs4_state_mark_reclaim_nograce(clp, state)) + return -EBADF; + nfs_inode_find_delegation_state_and_recover(state->inode, + &state->stateid); + dprintk("%s: scheduling stateid recovery for server %s\n", __func__, + clp->cl_hostname); + nfs4_schedule_state_manager(clp); + return clp->cl_cons_state < 0 ? clp->cl_cons_state : 0; +} +EXPORT_SYMBOL_GPL(nfs4_schedule_stateid_recovery); + +static struct nfs4_lock_state * +nfs_state_find_lock_state_by_stateid(struct nfs4_state *state, + const nfs4_stateid *stateid) +{ + struct nfs4_lock_state *pos; + + list_for_each_entry(pos, &state->lock_states, ls_locks) { + if (!test_bit(NFS_LOCK_INITIALIZED, &pos->ls_flags)) + continue; + if (nfs4_stateid_match_or_older(&pos->ls_stateid, stateid)) + return pos; + } + return NULL; +} + +static bool nfs_state_lock_state_matches_stateid(struct nfs4_state *state, + const nfs4_stateid *stateid) +{ + bool found = false; + + if (test_bit(LK_STATE_IN_USE, &state->flags)) { + spin_lock(&state->state_lock); + if (nfs_state_find_lock_state_by_stateid(state, stateid)) + found = true; + spin_unlock(&state->state_lock); + } + return found; +} + +void nfs_inode_find_state_and_recover(struct inode *inode, + const nfs4_stateid *stateid) +{ + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_open_context *ctx; + struct nfs4_state *state; + bool found = false; + + if (!S_ISREG(inode->i_mode)) + goto out; + rcu_read_lock(); + list_for_each_entry_rcu(ctx, &nfsi->open_files, list) { + state = ctx->state; + if (state == NULL) + continue; + if (nfs4_stateid_match_or_older(&state->stateid, stateid) && + nfs4_state_mark_reclaim_nograce(clp, state)) { + found = true; + continue; + } + if (test_bit(NFS_OPEN_STATE, &state->flags) && + nfs4_stateid_match_or_older(&state->open_stateid, stateid) && + nfs4_state_mark_reclaim_nograce(clp, state)) { + found = true; + continue; + } + if (nfs_state_lock_state_matches_stateid(state, stateid) && + nfs4_state_mark_reclaim_nograce(clp, state)) + found = true; + } + rcu_read_unlock(); +out: + nfs_inode_find_delegation_state_and_recover(inode, stateid); + if (found) + nfs4_schedule_state_manager(clp); +} + +static void nfs4_state_mark_open_context_bad(struct nfs4_state *state, int err) +{ + struct inode *inode = state->inode; + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_open_context *ctx; + + if (!S_ISREG(inode->i_mode)) + return; + rcu_read_lock(); + list_for_each_entry_rcu(ctx, &nfsi->open_files, list) { + if (ctx->state != state) + continue; + set_bit(NFS_CONTEXT_BAD, &ctx->flags); + pr_warn("NFSv4: state recovery failed for open file %pd2, " + "error = %d\n", ctx->dentry, err); + } + rcu_read_unlock(); +} + +static void nfs4_state_mark_recovery_failed(struct nfs4_state *state, int error) +{ + set_bit(NFS_STATE_RECOVERY_FAILED, &state->flags); + nfs4_state_mark_open_context_bad(state, error); +} + + +static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_recovery_ops *ops) +{ + struct inode *inode = state->inode; + struct nfs_inode *nfsi = NFS_I(inode); + struct file_lock *fl; + struct nfs4_lock_state *lsp; + int status = 0; + struct file_lock_context *flctx = locks_inode_context(inode); + struct list_head *list; + + if (flctx == NULL) + return 0; + + list = &flctx->flc_posix; + + /* Guard against delegation returns and new lock/unlock calls */ + down_write(&nfsi->rwsem); + spin_lock(&flctx->flc_lock); +restart: + for_each_file_lock(fl, list) { + if (nfs_file_open_context(fl->c.flc_file)->state != state) + continue; + spin_unlock(&flctx->flc_lock); + status = ops->recover_lock(state, fl); + switch (status) { + case 0: + break; + case -ETIMEDOUT: + case -ESTALE: + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_EXPIRED: + case -NFS4ERR_NO_GRACE: + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + goto out; + default: + pr_err("NFS: %s: unhandled error %d\n", + __func__, status); + fallthrough; + case -ENOMEM: + case -NFS4ERR_DENIED: + case -NFS4ERR_RECLAIM_BAD: + case -NFS4ERR_RECLAIM_CONFLICT: + lsp = fl->fl_u.nfs4_fl.owner; + if (lsp) + set_bit(NFS_LOCK_LOST, &lsp->ls_flags); + status = 0; + } + spin_lock(&flctx->flc_lock); + } + if (list == &flctx->flc_posix) { + list = &flctx->flc_flock; + goto restart; + } + spin_unlock(&flctx->flc_lock); +out: + up_write(&nfsi->rwsem); + return status; +} + +#ifdef CONFIG_NFS_V4_2 +static void nfs42_complete_copies(struct nfs4_state_owner *sp, struct nfs4_state *state) +{ + struct nfs4_copy_state *copy; + + if (!test_bit(NFS_CLNT_DST_SSC_COPY_STATE, &state->flags) && + !test_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags)) + return; + + spin_lock(&sp->so_server->nfs_client->cl_lock); + list_for_each_entry(copy, &sp->so_server->ss_copies, copies) { + if ((test_bit(NFS_CLNT_DST_SSC_COPY_STATE, &state->flags) && + !nfs4_stateid_match_other(&state->stateid, + ©->parent_dst_state->stateid))) + continue; + copy->flags = 1; + if (test_and_clear_bit(NFS_CLNT_DST_SSC_COPY_STATE, + &state->flags)) { + clear_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags); + complete(©->completion); + } + } + list_for_each_entry(copy, &sp->so_server->ss_src_copies, src_copies) { + if ((test_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags) && + !nfs4_stateid_match_other(&state->stateid, + ©->parent_src_state->stateid))) + continue; + copy->flags = 1; + if (test_and_clear_bit(NFS_CLNT_DST_SSC_COPY_STATE, + &state->flags)) + complete(©->completion); + } + spin_unlock(&sp->so_server->nfs_client->cl_lock); +} +#else /* !CONFIG_NFS_V4_2 */ +static inline void nfs42_complete_copies(struct nfs4_state_owner *sp, + struct nfs4_state *state) +{ +} +#endif /* CONFIG_NFS_V4_2 */ + +static int __nfs4_reclaim_open_state(struct nfs4_state_owner *sp, struct nfs4_state *state, + const struct nfs4_state_recovery_ops *ops, + int *lost_locks) +{ + struct nfs4_lock_state *lock; + int status; + + status = ops->recover_open(sp, state); + if (status < 0) + return status; + + status = nfs4_reclaim_locks(state, ops); + if (status < 0) + return status; + + if (!test_bit(NFS_DELEGATED_STATE, &state->flags)) { + spin_lock(&state->state_lock); + list_for_each_entry(lock, &state->lock_states, ls_locks) { + trace_nfs4_state_lock_reclaim(state, lock); + if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags) && + !test_bit(NFS_LOCK_UNLOCKING, &lock->ls_flags)) + *lost_locks += 1; + } + spin_unlock(&state->state_lock); + } + + nfs42_complete_copies(sp, state); + clear_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags); + return status; +} + +static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, + const struct nfs4_state_recovery_ops *ops, + int *lost_locks) +{ + struct nfs4_state *state; + unsigned int loop = 0; + int status = 0; +#ifdef CONFIG_NFS_V4_2 + bool found_ssc_copy_state = false; +#endif /* CONFIG_NFS_V4_2 */ + + /* Note: we rely on the sp->so_states list being ordered + * so that we always reclaim open(O_RDWR) and/or open(O_WRITE) + * states first. + * This is needed to ensure that the server won't give us any + * read delegations that we have to return if, say, we are + * recovering after a network partition or a reboot from a + * server that doesn't support a grace period. + */ + spin_lock(&sp->so_lock); +restart: + list_for_each_entry(state, &sp->so_states, open_states) { + if (!test_and_clear_bit(ops->state_flag_bit, &state->flags)) + continue; + if (!nfs4_valid_open_stateid(state)) + continue; + if (state->state == 0) + continue; +#ifdef CONFIG_NFS_V4_2 + if (test_bit(NFS_SRV_SSC_COPY_STATE, &state->flags)) { + nfs4_state_mark_recovery_failed(state, -EIO); + found_ssc_copy_state = true; + continue; + } +#endif /* CONFIG_NFS_V4_2 */ + refcount_inc(&state->count); + spin_unlock(&sp->so_lock); + status = __nfs4_reclaim_open_state(sp, state, ops, lost_locks); + + switch (status) { + default: + if (status >= 0) { + loop = 0; + break; + } + printk(KERN_ERR "NFS: %s: unhandled error %d\n", __func__, status); + fallthrough; + case -ENOENT: + case -ENOMEM: + case -EACCES: + case -EROFS: + case -EIO: + case -ESTALE: + /* Open state on this file cannot be recovered */ + nfs4_state_mark_recovery_failed(state, status); + break; + case -EAGAIN: + ssleep(1); + if (loop++ < 10) { + set_bit(ops->state_flag_bit, &state->flags); + break; + } + fallthrough; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_OLD_STATEID: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_RECLAIM_BAD: + case -NFS4ERR_RECLAIM_CONFLICT: + nfs4_state_mark_reclaim_nograce(sp->so_server->nfs_client, state); + break; + case -NFS4ERR_EXPIRED: + case -NFS4ERR_NO_GRACE: + nfs4_state_mark_reclaim_nograce(sp->so_server->nfs_client, state); + fallthrough; + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -ETIMEDOUT: + goto out_err; + } + nfs4_put_open_state(state); + spin_lock(&sp->so_lock); + goto restart; + } + spin_unlock(&sp->so_lock); +#ifdef CONFIG_NFS_V4_2 + if (found_ssc_copy_state) + return -EIO; +#endif /* CONFIG_NFS_V4_2 */ + return 0; +out_err: + nfs4_put_open_state(state); + spin_lock(&sp->so_lock); + spin_unlock(&sp->so_lock); + return status; +} + +static void nfs4_clear_open_state(struct nfs4_state *state) +{ + struct nfs4_lock_state *lock; + + clear_bit(NFS_DELEGATED_STATE, &state->flags); + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + clear_bit(NFS_O_RDWR_STATE, &state->flags); + spin_lock(&state->state_lock); + list_for_each_entry(lock, &state->lock_states, ls_locks) { + lock->ls_seqid.flags = 0; + clear_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags); + } + spin_unlock(&state->state_lock); +} + +static void nfs4_reset_seqids(struct nfs_server *server, + int (*mark_reclaim)(struct nfs_client *clp, struct nfs4_state *state)) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp; + struct rb_node *pos; + struct nfs4_state *state; + + spin_lock(&clp->cl_lock); + for (pos = rb_first(&server->state_owners); + pos != NULL; + pos = rb_next(pos)) { + sp = rb_entry(pos, struct nfs4_state_owner, so_server_node); + sp->so_seqid.flags = 0; + spin_lock(&sp->so_lock); + list_for_each_entry(state, &sp->so_states, open_states) { + if (mark_reclaim(clp, state)) + nfs4_clear_open_state(state); + } + spin_unlock(&sp->so_lock); + } + spin_unlock(&clp->cl_lock); +} + +static void nfs4_state_mark_reclaim_helper(struct nfs_client *clp, + int (*mark_reclaim)(struct nfs_client *clp, struct nfs4_state *state)) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs4_reset_seqids(server, mark_reclaim); + rcu_read_unlock(); +} + +static void nfs4_state_start_reclaim_reboot(struct nfs_client *clp) +{ + set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state); + /* Mark all delegations for reclaim */ + nfs_delegation_mark_reclaim(clp); + nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_reboot); +} + +static int nfs4_reclaim_complete(struct nfs_client *clp, + const struct nfs4_state_recovery_ops *ops, + const struct cred *cred) +{ + /* Notify the server we're done reclaiming our state */ + if (ops->reclaim_complete) + return ops->reclaim_complete(clp, cred); + return 0; +} + +static void nfs4_clear_reclaim_server(struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_state_owner *sp; + struct rb_node *pos; + struct nfs4_state *state; + + spin_lock(&clp->cl_lock); + for (pos = rb_first(&server->state_owners); + pos != NULL; + pos = rb_next(pos)) { + sp = rb_entry(pos, struct nfs4_state_owner, so_server_node); + spin_lock(&sp->so_lock); + list_for_each_entry(state, &sp->so_states, open_states) { + if (!test_and_clear_bit(NFS_STATE_RECLAIM_REBOOT, + &state->flags)) + continue; + nfs4_state_mark_reclaim_nograce(clp, state); + } + spin_unlock(&sp->so_lock); + } + spin_unlock(&clp->cl_lock); +} + +static int nfs4_state_clear_reclaim_reboot(struct nfs_client *clp) +{ + struct nfs_server *server; + + if (!test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) + return 0; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs4_clear_reclaim_server(server); + rcu_read_unlock(); + + nfs_delegation_reap_unclaimed(clp); + return 1; +} + +static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp) +{ + const struct nfs4_state_recovery_ops *ops; + const struct cred *cred; + int err; + + if (!nfs4_state_clear_reclaim_reboot(clp)) + return; + pnfs_destroy_all_layouts(clp); + ops = clp->cl_mvops->reboot_recovery_ops; + cred = nfs4_get_clid_cred(clp); + err = nfs4_reclaim_complete(clp, ops, cred); + put_cred(cred); + if (err == -NFS4ERR_CONN_NOT_BOUND_TO_SESSION) + set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state); +} + +static void nfs4_state_start_reclaim_nograce(struct nfs_client *clp) +{ + nfs_mark_test_expired_all_delegations(clp); + nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_nograce); +} + +static int nfs4_recovery_handle_error(struct nfs_client *clp, int error) +{ + switch (error) { + case 0: + break; +#if IS_ENABLED(CONFIG_NFS_V4_0) + case -NFS4ERR_CB_PATH_DOWN: + nfs40_handle_cb_pathdown(clp); + break; +#endif /* CONFIG_NFS_V4_0 */ + case -NFS4ERR_NO_GRACE: + nfs4_state_end_reclaim_reboot(clp); + break; + case -NFS4ERR_STALE_CLIENTID: + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + nfs4_state_start_reclaim_reboot(clp); + break; + case -NFS4ERR_EXPIRED: + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + nfs4_state_start_reclaim_nograce(clp); + break; + case -NFS4ERR_BADSESSION: + case -NFS4ERR_BADSLOT: + case -NFS4ERR_BAD_HIGH_SLOT: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_SEQ_FALSE_RETRY: + case -NFS4ERR_SEQ_MISORDERED: + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + /* Zero session reset errors */ + break; + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + set_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); + break; + default: + dprintk("%s: failed to handle error %d for server %s\n", + __func__, error, clp->cl_hostname); + return error; + } + dprintk("%s: handled error %d for server %s\n", __func__, error, + clp->cl_hostname); + return 0; +} + +static int nfs4_do_reclaim(struct nfs_client *clp, const struct nfs4_state_recovery_ops *ops) +{ + struct nfs4_state_owner *sp; + struct nfs_server *server; + struct rb_node *pos; + LIST_HEAD(freeme); + int lost_locks = 0; + int status; + + status = nfs4_begin_drain_session(clp); + if (status < 0) + return status; +restart: + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + nfs4_purge_state_owners(server, &freeme); + spin_lock(&clp->cl_lock); + for (pos = rb_first(&server->state_owners); + pos != NULL; + pos = rb_next(pos)) { + sp = rb_entry(pos, + struct nfs4_state_owner, so_server_node); + if (!test_and_clear_bit(ops->owner_flag_bit, + &sp->so_flags)) + continue; + if (!atomic_inc_not_zero(&sp->so_count)) + continue; + spin_unlock(&clp->cl_lock); + rcu_read_unlock(); + + status = nfs4_reclaim_open_state(sp, ops, &lost_locks); + if (status < 0) { + if (lost_locks) + pr_warn("NFS: %s: lost %d locks\n", + clp->cl_hostname, lost_locks); + set_bit(ops->owner_flag_bit, &sp->so_flags); + nfs4_put_state_owner(sp); + status = nfs4_recovery_handle_error(clp, status); + nfs4_free_state_owners(&freeme); + return (status != 0) ? status : -EAGAIN; + } + + nfs4_put_state_owner(sp); + goto restart; + } + spin_unlock(&clp->cl_lock); + } + rcu_read_unlock(); + nfs4_free_state_owners(&freeme); + nfs_local_probe_async(clp); + if (lost_locks) + pr_warn("NFS: %s: lost %d locks\n", + clp->cl_hostname, lost_locks); + return 0; +} + +static int nfs4_check_lease(struct nfs_client *clp) +{ + const struct cred *cred; + const struct nfs4_state_maintenance_ops *ops = + clp->cl_mvops->state_renewal_ops; + int status; + + /* Is the client already known to have an expired lease? */ + if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) + return 0; + cred = ops->get_state_renewal_cred(clp); + if (cred == NULL) { + cred = nfs4_get_clid_cred(clp); + status = -ENOKEY; + if (cred == NULL) + goto out; + } + status = ops->renew_lease(clp, cred); + put_cred(cred); + if (status == -ETIMEDOUT) { + set_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state); + return 0; + } +out: + return nfs4_recovery_handle_error(clp, status); +} + +/* Set NFS4CLNT_LEASE_EXPIRED and reclaim reboot state for all v4.0 errors + * and for recoverable errors on EXCHANGE_ID for v4.1 + */ +static int nfs4_handle_reclaim_lease_error(struct nfs_client *clp, int status) +{ + switch (status) { + case -NFS4ERR_SEQ_MISORDERED: + if (test_and_set_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) + return -ESERVERFAULT; + /* Lease confirmation error: retry after purging the lease */ + ssleep(1); + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + break; + case -NFS4ERR_STALE_CLIENTID: + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + nfs4_state_start_reclaim_reboot(clp); + break; + case -NFS4ERR_CLID_INUSE: + pr_err("NFS: Server %s reports our clientid is in use\n", + clp->cl_hostname); + nfs_mark_client_ready(clp, -EPERM); + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + return -EPERM; + case -ETIMEDOUT: + if (clp->cl_cons_state == NFS_CS_SESSION_INITING) { + nfs_mark_client_ready(clp, -EIO); + return -EIO; + } + fallthrough; + case -EACCES: + case -NFS4ERR_DELAY: + case -EAGAIN: + ssleep(1); + break; + + case -NFS4ERR_MINOR_VERS_MISMATCH: + if (clp->cl_cons_state == NFS_CS_SESSION_INITING) + nfs_mark_client_ready(clp, -EPROTONOSUPPORT); + dprintk("%s: exit with error %d for server %s\n", + __func__, -EPROTONOSUPPORT, clp->cl_hostname); + return -EPROTONOSUPPORT; + case -ENOSPC: + if (clp->cl_cons_state == NFS_CS_SESSION_INITING) + nfs_mark_client_ready(clp, -EIO); + return -EIO; + case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery + * in nfs4_exchange_id */ + default: + dprintk("%s: exit with error %d for server %s\n", __func__, + status, clp->cl_hostname); + return status; + } + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + dprintk("%s: handled error %d for server %s\n", __func__, status, + clp->cl_hostname); + return 0; +} + +static int nfs4_establish_lease(struct nfs_client *clp) +{ + const struct cred *cred; + const struct nfs4_state_recovery_ops *ops = + clp->cl_mvops->reboot_recovery_ops; + int status; + + status = nfs4_begin_drain_session(clp); + if (status != 0) + return status; + cred = nfs4_get_clid_cred(clp); + if (cred == NULL) + return -ENOENT; + status = ops->establish_clid(clp, cred); + put_cred(cred); + if (status != 0) + return status; + return 0; +} + +/* + * Returns zero or a negative errno. NFS4ERR values are converted + * to local errno values. + */ +static int nfs4_reclaim_lease(struct nfs_client *clp) +{ + int status; + + status = nfs4_establish_lease(clp); + if (status < 0) + return nfs4_handle_reclaim_lease_error(clp, status); + if (test_and_clear_bit(NFS4CLNT_SERVER_SCOPE_MISMATCH, &clp->cl_state)) + nfs4_state_start_reclaim_nograce(clp); + if (!test_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) + set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state); + clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state); + clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + return 0; +} + +static int nfs4_purge_lease(struct nfs_client *clp) +{ + int status; + + status = nfs4_establish_lease(clp); + if (status < 0) + return nfs4_handle_reclaim_lease_error(clp, status); + clear_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state); + set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); + nfs4_state_start_reclaim_nograce(clp); + return 0; +} + +/* + * Try remote migration of one FSID from a source server to a + * destination server. The source server provides a list of + * potential destinations. + * + * Returns zero or a negative NFS4ERR status code. + */ +static int nfs4_try_migration(struct nfs_server *server, const struct cred *cred) +{ + struct nfs_client *clp = server->nfs_client; + struct nfs4_fs_locations *locations = NULL; + struct nfs_fattr *fattr; + struct inode *inode; + struct page *page; + int status, result; + + dprintk("--> %s: FSID %llx:%llx on \"%s\"\n", __func__, + (unsigned long long)server->fsid.major, + (unsigned long long)server->fsid.minor, + clp->cl_hostname); + + page = alloc_page(GFP_KERNEL); + locations = kmalloc_obj(struct nfs4_fs_locations); + fattr = nfs_alloc_fattr(); + if (page == NULL || locations == NULL || fattr == NULL) { + dprintk("<-- %s: no memory\n", __func__); + result = 0; + goto out; + } + + locations->fattr = fattr; + inode = d_inode(server->super->s_root); + result = nfs4_proc_get_locations(server, NFS_FH(inode), locations, + page, cred); + if (result) { + dprintk("<-- %s: failed to retrieve fs_locations: %d\n", + __func__, result); + goto out; + } + + result = -NFS4ERR_NXIO; + if (!locations->nlocations) + goto out; + + if (!(locations->fattr->valid & NFS_ATTR_FATTR_V4_LOCATIONS)) { + dprintk("<-- %s: No fs_locations data, migration skipped\n", + __func__); + goto out; + } + + status = nfs4_begin_drain_session(clp); + if (status != 0) { + result = status; + goto out; + } + + status = nfs4_replace_transport(server, locations); + if (status != 0) { + dprintk("<-- %s: failed to replace transport: %d\n", + __func__, status); + goto out; + } + + result = 0; + dprintk("<-- %s: migration succeeded\n", __func__); + +out: + if (page != NULL) + __free_page(page); + if (locations != NULL) + kfree(locations->fattr); + kfree(locations); + if (result) { + pr_err("NFS: migration recovery failed (server %s)\n", + clp->cl_hostname); + set_bit(NFS_MIG_FAILED, &server->mig_status); + } + return result; +} + +/* + * Returns zero or a negative NFS4ERR status code. + */ +static int nfs4_handle_migration(struct nfs_client *clp) +{ + const struct nfs4_state_maintenance_ops *ops = + clp->cl_mvops->state_renewal_ops; + struct nfs_server *server; + const struct cred *cred; + + dprintk("%s: migration reported on \"%s\"\n", __func__, + clp->cl_hostname); + + cred = ops->get_state_renewal_cred(clp); + if (cred == NULL) + return -NFS4ERR_NOENT; + + clp->cl_mig_gen++; +restart: + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + int status; + + if (server->mig_gen == clp->cl_mig_gen) + continue; + server->mig_gen = clp->cl_mig_gen; + + if (!test_and_clear_bit(NFS_MIG_IN_TRANSITION, + &server->mig_status)) + continue; + + rcu_read_unlock(); + status = nfs4_try_migration(server, cred); + if (status < 0) { + put_cred(cred); + return status; + } + goto restart; + } + rcu_read_unlock(); + put_cred(cred); + return 0; +} + +/* + * Test each nfs_server on the clp's cl_superblocks list to see + * if it's moved to another server. Stop when the server no longer + * returns NFS4ERR_LEASE_MOVED. + */ +static int nfs4_handle_lease_moved(struct nfs_client *clp) +{ + const struct nfs4_state_maintenance_ops *ops = + clp->cl_mvops->state_renewal_ops; + struct nfs_server *server; + const struct cred *cred; + + dprintk("%s: lease moved reported on \"%s\"\n", __func__, + clp->cl_hostname); + + cred = ops->get_state_renewal_cred(clp); + if (cred == NULL) + return -NFS4ERR_NOENT; + + clp->cl_mig_gen++; +restart: + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + struct inode *inode; + int status; + + if (server->mig_gen == clp->cl_mig_gen) + continue; + server->mig_gen = clp->cl_mig_gen; + + rcu_read_unlock(); + + inode = d_inode(server->super->s_root); + status = nfs4_proc_fsid_present(inode, cred); + if (status != -NFS4ERR_MOVED) + goto restart; /* wasn't this one */ + if (nfs4_try_migration(server, cred) == -NFS4ERR_LEASE_MOVED) + goto restart; /* there are more */ + goto out; + } + rcu_read_unlock(); + +out: + put_cred(cred); + return 0; +} + +/** + * nfs4_discover_server_trunking - Detect server IP address trunking + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * + * Returns zero or a negative errno. If zero is returned, + * an nfs_client pointer is planted in "result". + * + * Note: since we are invoked in process context, and + * not from inside the state manager, we cannot use + * nfs4_handle_reclaim_lease_error(). + */ +int nfs4_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result) +{ + const struct nfs4_state_recovery_ops *ops = + clp->cl_mvops->reboot_recovery_ops; + struct rpc_clnt *clnt; + const struct cred *cred; + int i, status; + + dprintk("NFS: %s: testing '%s'\n", __func__, clp->cl_hostname); + + clnt = clp->cl_rpcclient; + i = 0; + + mutex_lock(&nfs_clid_init_mutex); +again: + status = -ENOENT; + cred = nfs4_get_clid_cred(clp); + if (cred == NULL) + goto out_unlock; + + status = ops->detect_trunking(clp, result, cred); + put_cred(cred); + switch (status) { + case 0: + case -EINTR: + case -ERESTARTSYS: + break; + case -ETIMEDOUT: + if (clnt->cl_softrtry) + break; + fallthrough; + case -NFS4ERR_DELAY: + case -EAGAIN: + ssleep(1); + fallthrough; + case -NFS4ERR_STALE_CLIENTID: + dprintk("NFS: %s after status %d, retrying\n", + __func__, status); + goto again; + case -EACCES: + if (i++ == 0) { + nfs4_root_machine_cred(clp); + goto again; + } + if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX) + break; + fallthrough; + case -NFS4ERR_CLID_INUSE: + case -NFS4ERR_WRONGSEC: + /* No point in retrying if we already used RPC_AUTH_UNIX */ + if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX) { + status = -EPERM; + break; + } + clnt = rpc_clone_client_set_auth(clnt, RPC_AUTH_UNIX); + if (IS_ERR(clnt)) { + status = PTR_ERR(clnt); + break; + } + /* Note: this is safe because we haven't yet marked the + * client as ready, so we are the only user of + * clp->cl_rpcclient + */ + clnt = xchg(&clp->cl_rpcclient, clnt); + rpc_shutdown_client(clnt); + clnt = clp->cl_rpcclient; + goto again; + + case -NFS4ERR_MINOR_VERS_MISMATCH: + status = -EPROTONOSUPPORT; + break; + + case -EKEYEXPIRED: + case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery + * in nfs4_exchange_id */ + status = -EKEYEXPIRED; + break; + default: + pr_warn("NFS: %s unhandled error %d. Exiting with error EIO\n", + __func__, status); + status = -EIO; + } + +out_unlock: + mutex_unlock(&nfs_clid_init_mutex); + dprintk("NFS: %s: status = %d\n", __func__, status); + return status; +} + +void nfs4_schedule_session_recovery(struct nfs4_session *session, int err) +{ + struct nfs_client *clp = session->clp; + + switch (err) { + default: + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + break; + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + set_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); + } + nfs4_schedule_state_manager(clp); +} +EXPORT_SYMBOL_GPL(nfs4_schedule_session_recovery); + +void nfs41_notify_server(struct nfs_client *clp) +{ + /* Use CHECK_LEASE to ping the server with a SEQUENCE */ + set_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state); + nfs4_schedule_state_manager(clp); +} + +static void nfs4_reset_all_state(struct nfs_client *clp) +{ + if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) { + set_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state); + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + nfs4_state_start_reclaim_nograce(clp); + dprintk("%s: scheduling reset of all state for server %s!\n", + __func__, clp->cl_hostname); + nfs4_schedule_state_manager(clp); + } +} + +static void nfs41_handle_server_reboot(struct nfs_client *clp) +{ + if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) { + nfs4_state_start_reclaim_reboot(clp); + dprintk("%s: server %s rebooted!\n", __func__, + clp->cl_hostname); + nfs4_schedule_state_manager(clp); + } +} + +static void nfs41_handle_all_state_revoked(struct nfs_client *clp) +{ + nfs4_reset_all_state(clp); + dprintk("%s: state revoked on server %s\n", __func__, clp->cl_hostname); +} + +static void nfs41_handle_some_state_revoked(struct nfs_client *clp) +{ + nfs4_state_start_reclaim_nograce(clp); + nfs4_schedule_state_manager(clp); + + dprintk("%s: state revoked on server %s\n", __func__, clp->cl_hostname); +} + +static void nfs41_handle_recallable_state_revoked(struct nfs_client *clp) +{ + /* FIXME: For now, we destroy all layouts. */ + pnfs_destroy_all_layouts(clp); + nfs_test_expired_all_delegations(clp); + dprintk("%s: Recallable state revoked on server %s!\n", __func__, + clp->cl_hostname); +} + +static void nfs41_handle_backchannel_fault(struct nfs_client *clp) +{ + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + nfs4_schedule_state_manager(clp); + + dprintk("%s: server %s declared a backchannel fault\n", __func__, + clp->cl_hostname); +} + +static void nfs41_handle_cb_path_down(struct nfs_client *clp) +{ + if (test_and_set_bit(NFS4CLNT_BIND_CONN_TO_SESSION, + &clp->cl_state) == 0) + nfs4_schedule_state_manager(clp); +} + +void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags, + bool recovery) +{ + if (!flags) + return; + + dprintk("%s: \"%s\" (client ID %llx) flags=0x%08x\n", + __func__, clp->cl_hostname, clp->cl_clientid, flags); + /* + * If we're called from the state manager thread, then assume we're + * already handling the RECLAIM_NEEDED and/or STATE_REVOKED. + * Those flags are expected to remain set until we're done + * recovering (see RFC5661, section 18.46.3). + */ + if (recovery) + goto out_recovery; + + if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) + nfs41_handle_server_reboot(clp); + if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED)) + nfs41_handle_all_state_revoked(clp); + if (flags & (SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED | + SEQ4_STATUS_ADMIN_STATE_REVOKED)) + nfs41_handle_some_state_revoked(clp); + if (flags & SEQ4_STATUS_LEASE_MOVED) + nfs4_schedule_lease_moved_recovery(clp); + if (flags & SEQ4_STATUS_RECALLABLE_STATE_REVOKED) + nfs41_handle_recallable_state_revoked(clp); +out_recovery: + if (flags & SEQ4_STATUS_BACKCHANNEL_FAULT) + nfs41_handle_backchannel_fault(clp); + else if (flags & (SEQ4_STATUS_CB_PATH_DOWN | + SEQ4_STATUS_CB_PATH_DOWN_SESSION)) + nfs41_handle_cb_path_down(clp); +} + +static int nfs4_reset_session(struct nfs_client *clp) +{ + const struct cred *cred; + int status; + + if (!nfs4_has_session(clp)) + return 0; + status = nfs4_begin_drain_session(clp); + if (status != 0) + return status; + cred = nfs4_get_clid_cred(clp); + status = nfs4_proc_destroy_session(clp->cl_session, cred); + switch (status) { + case 0: + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + break; + case -NFS4ERR_BACK_CHAN_BUSY: + case -NFS4ERR_DELAY: + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + status = 0; + ssleep(1); + goto out; + default: + status = nfs4_recovery_handle_error(clp, status); + goto out; + } + + memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN); + status = nfs4_proc_create_session(clp, cred); + if (status) { + dprintk("%s: session reset failed with status %d for server %s!\n", + __func__, status, clp->cl_hostname); + status = nfs4_handle_reclaim_lease_error(clp, status); + goto out; + } + nfs41_finish_session_reset(clp); + dprintk("%s: session reset was successful for server %s!\n", + __func__, clp->cl_hostname); +out: + put_cred(cred); + return status; +} + +static int nfs4_bind_conn_to_session(struct nfs_client *clp) +{ + const struct cred *cred; + int ret; + + if (!nfs4_has_session(clp)) + return 0; + ret = nfs4_begin_drain_session(clp); + if (ret != 0) + return ret; + cred = nfs4_get_clid_cred(clp); + ret = nfs4_proc_bind_conn_to_session(clp, cred); + put_cred(cred); + clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); + switch (ret) { + case 0: + dprintk("%s: bind_conn_to_session was successful for server %s!\n", + __func__, clp->cl_hostname); + break; + case -NFS4ERR_DELAY: + ssleep(1); + set_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); + break; + default: + return nfs4_recovery_handle_error(clp, ret); + } + return 0; +} + +static void nfs4_layoutreturn_any_run(struct nfs_client *clp) +{ + int iomode = 0; + + if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_READ, &clp->cl_state)) + iomode += IOMODE_READ; + if (test_and_clear_bit(NFS4CLNT_RECALL_ANY_LAYOUT_RW, &clp->cl_state)) + iomode += IOMODE_RW; + /* Note: IOMODE_READ + IOMODE_RW == IOMODE_ANY */ + if (iomode) { + pnfs_layout_return_unused_byclid(clp, iomode); + set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + } +} + +static void nfs4_state_manager(struct nfs_client *clp) +{ + unsigned int memflags; + int status = 0; + const char *section = "", *section_sep = ""; + + /* + * State recovery can deadlock if the direct reclaim code tries + * start NFS writeback. So ensure memory allocations are all + * GFP_NOFS. + */ + memflags = memalloc_nofs_save(); + + /* Ensure exclusive access to NFSv4 state */ + do { + trace_nfs4_state_mgr(clp); + clear_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + if (test_bit(NFS4CLNT_PURGE_STATE, &clp->cl_state)) { + section = "purge state"; + status = nfs4_purge_lease(clp); + if (status < 0) + goto out_error; + continue; + } + + if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) { + section = "lease expired"; + /* We're going to have to re-establish a clientid */ + status = nfs4_reclaim_lease(clp); + if (status < 0) + goto out_error; + continue; + } + + /* Initialize or reset the session */ + if (test_and_clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state)) { + section = "reset session"; + status = nfs4_reset_session(clp); + if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) + continue; + if (status < 0) + goto out_error; + } + + /* Send BIND_CONN_TO_SESSION */ + if (test_and_clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, + &clp->cl_state)) { + section = "bind conn to session"; + status = nfs4_bind_conn_to_session(clp); + if (status < 0) + goto out_error; + continue; + } + + if (test_and_clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state)) { + section = "check lease"; + status = nfs4_check_lease(clp); + if (status < 0) + goto out_error; + continue; + } + + if (test_and_clear_bit(NFS4CLNT_MOVED, &clp->cl_state)) { + section = "migration"; + status = nfs4_handle_migration(clp); + if (status < 0) + goto out_error; + } + + if (test_and_clear_bit(NFS4CLNT_LEASE_MOVED, &clp->cl_state)) { + section = "lease moved"; + status = nfs4_handle_lease_moved(clp); + if (status < 0) + goto out_error; + } + + /* First recover reboot state... */ + if (test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) { + section = "reclaim reboot"; + status = nfs4_do_reclaim(clp, + clp->cl_mvops->reboot_recovery_ops); + if (status == 0) + status = pnfs_layout_handle_reboot(clp); + if (status == -EAGAIN) + continue; + if (status < 0) + goto out_error; + nfs4_state_end_reclaim_reboot(clp); + continue; + } + + /* Detect expired delegations... */ + if (test_and_clear_bit(NFS4CLNT_DELEGATION_EXPIRED, &clp->cl_state)) { + section = "detect expired delegations"; + status = nfs4_begin_drain_session(clp); + if (status < 0) + goto out_error; + nfs_reap_expired_delegations(clp); + continue; + } + + /* Now recover expired state... */ + if (test_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) { + section = "reclaim nograce"; + status = nfs4_do_reclaim(clp, + clp->cl_mvops->nograce_recovery_ops); + if (status == -EAGAIN) + continue; + if (status < 0) + goto out_error; + clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state); + } + + memalloc_nofs_restore(memflags); + nfs4_end_drain_session(clp); + nfs4_clear_state_manager_bit(clp); + + if (test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state) && + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, + &clp->cl_state)) { + memflags = memalloc_nofs_save(); + continue; + } + + if (!test_and_set_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state)) { + if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) { + nfs_client_return_marked_delegations(clp); + set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + } + nfs4_layoutreturn_any_run(clp); + clear_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state); + } + + return; + + } while (refcount_read(&clp->cl_count) > 1 && !signalled()); + goto out_drain; + +out_error: + if (strlen(section)) + section_sep = ": "; + trace_nfs4_state_mgr_failed(clp, section, status); + pr_warn_ratelimited("NFS: state manager%s%s failed on NFSv4 server %s" + " with error %d\n", section_sep, section, + clp->cl_hostname, -status); + switch (status) { + case -ENETDOWN: + case -ENETUNREACH: + nfs_mark_client_ready(clp, -EIO); + break; + case -EINVAL: + nfs_mark_client_ready(clp, status); + break; + default: + ssleep(1); + break; + } +out_drain: + memalloc_nofs_restore(memflags); + nfs4_end_drain_session(clp); + nfs4_clear_state_manager_bit(clp); +} + +static int nfs4_run_state_manager(void *ptr) +{ + struct nfs_client *clp = ptr; + struct rpc_clnt *cl = clp->cl_rpcclient; + + while (cl != cl->cl_parent) + cl = cl->cl_parent; + + allow_signal(SIGKILL); +again: + nfs4_state_manager(clp); + + if (test_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state) && + !test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) { + wait_var_event_interruptible(&clp->cl_state, + test_bit(NFS4CLNT_RUN_MANAGER, + &clp->cl_state)); + if (!atomic_read(&cl->cl_swapper)) + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + if (refcount_read(&clp->cl_count) > 1 && !signalled() && + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) + goto again; + /* Either no longer a swapper, or were signalled */ + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + } + + if (refcount_read(&clp->cl_count) > 1 && !signalled() && + test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state) && + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) + goto again; + + nfs_put_client(clp); + module_put_and_kthread_exit(0); + return 0; +} diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c new file mode 100644 index 000000000..9c5895b55 --- /dev/null +++ b/fs/nfs/nfs4super.c @@ -0,0 +1,317 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012 Bryan Schumaker + */ +#include +#include +#include +#include +#include +#include +#include "delegation.h" +#include "internal.h" +#include "nfs4_fs.h" +#include "nfs4idmap.h" +#include "dns_resolve.h" +#include "pnfs.h" +#include "nfs.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc); +static void nfs4_evict_inode(struct inode *inode); + +static const struct super_operations nfs4_sops = { + .alloc_inode = nfs_alloc_inode, + .free_inode = nfs_free_inode, + .write_inode = nfs4_write_inode, + .drop_inode = nfs_drop_inode, + .statfs = nfs_statfs, + .evict_inode = nfs4_evict_inode, + .umount_begin = nfs_umount_begin, + .show_options = nfs_show_options, + .show_devname = nfs_show_devname, + .show_path = nfs_show_path, + .show_stats = nfs_show_stats, +}; + +struct nfs_subversion nfs_v4 = { + .owner = THIS_MODULE, + .nfs_fs = &nfs4_fs_type, + .rpc_vers = &nfs_version4, + .rpc_ops = &nfs_v4_clientops, + .sops = &nfs4_sops, + .xattr = nfs4_xattr_handlers, +}; + +static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc) +{ + int ret = nfs_write_inode(inode, wbc); + + if (ret == 0) + ret = pnfs_layoutcommit_inode(inode, + wbc->sync_mode == WB_SYNC_ALL); + return ret; +} + +/* + * Clean out any remaining NFSv4 state that might be left over due + * to open() calls that passed nfs_atomic_lookup, but failed to call + * nfs_open(). + */ +static void nfs4_evict_inode(struct inode *inode) +{ + truncate_inode_pages_final(&inode->i_data); + clear_inode(inode); + /* If we are holding a delegation, return and free it */ + nfs_inode_evict_delegation(inode); + /* Note that above delegreturn would trigger pnfs return-on-close */ + pnfs_return_layout(inode); + pnfs_destroy_layout_final(NFS_I(inode)); + /* First call standard NFS clear_inode() code */ + nfs_clear_inode(inode); + nfs4_xattr_cache_zap(inode); +} + +struct nfs_referral_count { + struct list_head list; + const struct task_struct *task; + unsigned int referral_count; +}; + +static LIST_HEAD(nfs_referral_count_list); +static DEFINE_SPINLOCK(nfs_referral_count_list_lock); + +static struct nfs_referral_count *nfs_find_referral_count(void) +{ + struct nfs_referral_count *p; + + list_for_each_entry(p, &nfs_referral_count_list, list) { + if (p->task == current) + return p; + } + return NULL; +} + +#define NFS_MAX_NESTED_REFERRALS 2 + +static int nfs_referral_loop_protect(void) +{ + struct nfs_referral_count *p, *new; + int ret = -ENOMEM; + + new = kmalloc_obj(*new); + if (!new) + goto out; + new->task = current; + new->referral_count = 1; + + ret = 0; + spin_lock(&nfs_referral_count_list_lock); + p = nfs_find_referral_count(); + if (p != NULL) { + if (p->referral_count >= NFS_MAX_NESTED_REFERRALS) + ret = -ELOOP; + else + p->referral_count++; + } else { + list_add(&new->list, &nfs_referral_count_list); + new = NULL; + } + spin_unlock(&nfs_referral_count_list_lock); + kfree(new); +out: + return ret; +} + +static void nfs_referral_loop_unprotect(void) +{ + struct nfs_referral_count *p; + + spin_lock(&nfs_referral_count_list_lock); + p = nfs_find_referral_count(); + p->referral_count--; + if (p->referral_count == 0) + list_del(&p->list); + else + p = NULL; + spin_unlock(&nfs_referral_count_list_lock); + kfree(p); +} + +static int do_nfs4_mount(struct nfs_server *server, + struct fs_context *fc, + const char *hostname, + const char *export_path) +{ + struct nfs_fs_context *root_ctx; + struct nfs_fs_context *ctx; + struct fs_context *root_fc; + struct vfsmount *root_mnt; + struct dentry *dentry; + char *source; + int ret; + + if (IS_ERR(server)) + return PTR_ERR(server); + + root_fc = vfs_dup_fs_context(fc); + if (IS_ERR(root_fc)) { + nfs_free_server(server); + return PTR_ERR(root_fc); + } + kfree(root_fc->source); + root_fc->source = NULL; + + ctx = nfs_fc2context(fc); + root_ctx = nfs_fc2context(root_fc); + root_ctx->internal = true; + root_ctx->server = server; + + if (ctx->fscache_uniq) { + ret = vfs_parse_fs_string(root_fc, "fsc", ctx->fscache_uniq); + if (ret < 0) { + put_fs_context(root_fc); + return ret; + } + } + /* We leave export_path unset as it's not used to find the root. */ + + /* Does hostname needs to be enclosed in brackets? */ + if (strchr(hostname, ':')) + source = kasprintf(GFP_KERNEL, "[%s]:/", hostname); + else + source = kasprintf(GFP_KERNEL, "%s:/", hostname); + + if (!source) { + put_fs_context(root_fc); + return -ENOMEM; + } + ret = vfs_parse_fs_string(root_fc, "source", source); + kfree(source); + if (ret < 0) { + put_fs_context(root_fc); + return ret; + } + root_mnt = fc_mount(root_fc); + put_fs_context(root_fc); + + if (IS_ERR(root_mnt)) + return PTR_ERR(root_mnt); + + ret = nfs_referral_loop_protect(); + if (ret) { + mntput(root_mnt); + return ret; + } + + dentry = mount_subtree(root_mnt, export_path); + nfs_referral_loop_unprotect(); + + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + fc->root = dentry; + return 0; +} + +int nfs4_try_get_tree(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + int err; + + dfprintk(MOUNT, "--> nfs4_try_get_tree()\n"); + + /* We create a mount for the server's root, walk to the requested + * location and then create another mount for that. + */ + err= do_nfs4_mount(nfs4_create_server(fc), + fc, ctx->nfs_server.hostname, + ctx->nfs_server.export_path); + if (err) { + nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path"); + dfprintk(MOUNT, "<-- nfs4_try_get_tree() = %d [error]\n", err); + } else { + dfprintk(MOUNT, "<-- nfs4_try_get_tree() = 0\n"); + } + return err; +} + +/* + * Create an NFS4 server record on referral traversal + */ +int nfs4_get_referral_tree(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + int err; + + dprintk("--> nfs4_referral_mount()\n"); + + /* create a new volume representation */ + err = do_nfs4_mount(nfs4_create_referral_server(fc), + fc, ctx->nfs_server.hostname, + ctx->nfs_server.export_path); + if (err) { + nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path"); + dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = %d [error]\n", err); + } else { + dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = 0\n"); + } + return err; +} + +static int __init init_nfs_v4(void) +{ + int err; + + err = nfs_dns_resolver_init(); + if (err) + goto out; + + err = nfs_idmap_init(); + if (err) + goto out1; + +#ifdef CONFIG_NFS_V4_2 + err = nfs4_xattr_cache_init(); + if (err) + goto out2; +#endif + + err = nfs4_register_sysctl(); + if (err) + goto out2; + +#ifdef CONFIG_NFS_V4_2 + nfs42_ssc_register_ops(); +#endif + register_nfs_version(&nfs_v4); + return 0; +out2: + nfs_idmap_quit(); +out1: + nfs_dns_resolver_destroy(); +out: + return err; +} + +static void __exit exit_nfs_v4(void) +{ + /* Not called in the _init(), conditionally loaded */ + nfs4_pnfs_v3_ds_connect_unload(); + + unregister_nfs_version(&nfs_v4); +#ifdef CONFIG_NFS_V4_2 + nfs4_xattr_cache_exit(); + nfs42_ssc_unregister_ops(); +#endif + nfs4_unregister_sysctl(); + nfs_idmap_quit(); + nfs_dns_resolver_destroy(); +} + +MODULE_DESCRIPTION("NFSv4 client support"); +MODULE_LICENSE("GPL"); + +module_init(init_nfs_v4); +module_exit(exit_nfs_v4); diff --git a/fs/nfs/nfs4sysctl.c b/fs/nfs/nfs4sysctl.c new file mode 100644 index 000000000..d1a92d8f8 --- /dev/null +++ b/fs/nfs/nfs4sysctl.c @@ -0,0 +1,52 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/nfs4sysctl.c + * + * Sysctl interface to NFS v4 parameters + * + * Copyright (c) 2006 Trond Myklebust + */ +#include +#include + +#include "nfs4_fs.h" +#include "nfs4idmap.h" +#include "callback.h" + +static const int nfs_set_port_min; +static const int nfs_set_port_max = 65535; +static struct ctl_table_header *nfs4_callback_sysctl_table; + +static const struct ctl_table nfs4_cb_sysctls[] = { + { + .procname = "nfs_callback_tcpport", + .data = &nfs_callback_set_tcpport, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = (int *)&nfs_set_port_min, + .extra2 = (int *)&nfs_set_port_max, + }, + { + .procname = "idmap_cache_timeout", + .data = &nfs_idmap_cache_timeout, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +}; + +int nfs4_register_sysctl(void) +{ + nfs4_callback_sysctl_table = register_sysctl("fs/nfs", + nfs4_cb_sysctls); + if (nfs4_callback_sysctl_table == NULL) + return -ENOMEM; + return 0; +} + +void nfs4_unregister_sysctl(void) +{ + unregister_sysctl_table(nfs4_callback_sysctl_table); + nfs4_callback_sysctl_table = NULL; +} diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c new file mode 100644 index 000000000..3fdc013f5 --- /dev/null +++ b/fs/nfs/nfs4trace.c @@ -0,0 +1,40 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2013 Trond Myklebust + */ +#include +#include +#include +#include "nfs4_fs.h" +#include "internal.h" +#include "nfs4session.h" +#include "callback.h" +#include "pnfs.h" + +#define CREATE_TRACE_POINTS +#include "nfs4trace.h" + +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds); + +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_read); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_write); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_get_mirror_count); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_ds_connect); + +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_read_error); +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_write_error); +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_commit_error); + +EXPORT_TRACEPOINT_SYMBOL_GPL(bl_ext_tree_prepare_commit); +EXPORT_TRACEPOINT_SYMBOL_GPL(bl_pr_key_reg); +EXPORT_TRACEPOINT_SYMBOL_GPL(bl_pr_key_reg_err); +EXPORT_TRACEPOINT_SYMBOL_GPL(bl_pr_key_unreg); +EXPORT_TRACEPOINT_SYMBOL_GPL(bl_pr_key_unreg_err); + +EXPORT_TRACEPOINT_SYMBOL_GPL(fl_getdevinfo); diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h new file mode 100644 index 000000000..1ed677810 --- /dev/null +++ b/fs/nfs/nfs4trace.h @@ -0,0 +1,2859 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2013 Trond Myklebust + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nfs4 + +#if !defined(_TRACE_NFS4_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NFS4_H + +#include +#include + +#include +#include + +#include "delegation.h" + +#define show_nfs_fattr_flags(valid) \ + __print_flags((unsigned long)valid, "|", \ + { NFS_ATTR_FATTR_TYPE, "TYPE" }, \ + { NFS_ATTR_FATTR_MODE, "MODE" }, \ + { NFS_ATTR_FATTR_NLINK, "NLINK" }, \ + { NFS_ATTR_FATTR_OWNER, "OWNER" }, \ + { NFS_ATTR_FATTR_GROUP, "GROUP" }, \ + { NFS_ATTR_FATTR_RDEV, "RDEV" }, \ + { NFS_ATTR_FATTR_SIZE, "SIZE" }, \ + { NFS_ATTR_FATTR_FSID, "FSID" }, \ + { NFS_ATTR_FATTR_FILEID, "FILEID" }, \ + { NFS_ATTR_FATTR_ATIME, "ATIME" }, \ + { NFS_ATTR_FATTR_MTIME, "MTIME" }, \ + { NFS_ATTR_FATTR_CTIME, "CTIME" }, \ + { NFS_ATTR_FATTR_CHANGE, "CHANGE" }, \ + { NFS_ATTR_FATTR_OWNER_NAME, "OWNER_NAME" }, \ + { NFS_ATTR_FATTR_GROUP_NAME, "GROUP_NAME" }, \ + { NFS_ATTR_FATTR_BTIME, "BTIME" }) + +DECLARE_EVENT_CLASS(nfs4_clientid_event, + TP_PROTO( + const struct nfs_client *clp, + int error + ), + + TP_ARGS(clp, error), + + TP_STRUCT__entry( + __string(dstaddr, clp->cl_hostname) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __assign_str(dstaddr); + ), + + TP_printk( + "error=%ld (%s) dstaddr=%s", + -__entry->error, + show_nfs4_status(__entry->error), + __get_str(dstaddr) + ) +); +#define DEFINE_NFS4_CLIENTID_EVENT(name) \ + DEFINE_EVENT(nfs4_clientid_event, name, \ + TP_PROTO( \ + const struct nfs_client *clp, \ + int error \ + ), \ + TP_ARGS(clp, error)) +DEFINE_NFS4_CLIENTID_EVENT(nfs4_setclientid); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_setclientid_confirm); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_renew); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_renew_async); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_exchange_id); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_create_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_destroy_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_destroy_clientid); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_bind_conn_to_session); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_sequence); +DEFINE_NFS4_CLIENTID_EVENT(nfs4_reclaim_complete); + +TRACE_EVENT(nfs4_trunked_exchange_id, + TP_PROTO( + const struct nfs_client *clp, + const char *addr, + int error + ), + + TP_ARGS(clp, addr, error), + + TP_STRUCT__entry( + __string(main_addr, clp->cl_hostname) + __string(trunk_addr, addr) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __assign_str(main_addr); + __assign_str(trunk_addr); + ), + + TP_printk( + "error=%ld (%s) main_addr=%s trunk_addr=%s", + -__entry->error, + show_nfs4_status(__entry->error), + __get_str(main_addr), + __get_str(trunk_addr) + ) +); + +TRACE_EVENT(nfs4_sequence_done, + TP_PROTO( + const struct nfs4_session *session, + const struct nfs4_sequence_res *res + ), + TP_ARGS(session, res), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, target_highest_slotid) + __field(unsigned long, status_flags) + __field(unsigned long, error) + ), + + TP_fast_assign( + const struct nfs4_slot *sr_slot = res->sr_slot; + __entry->session = nfs_session_id_hash(&session->sess_id); + __entry->slot_nr = sr_slot->slot_nr; + __entry->seq_nr = sr_slot->seq_nr; + __entry->highest_slotid = res->sr_highest_slotid; + __entry->target_highest_slotid = + res->sr_target_highest_slotid; + __entry->status_flags = res->sr_status_flags; + __entry->error = res->sr_status < 0 ? + -res->sr_status : 0; + ), + TP_printk( + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u target_highest_slotid=%u " + "status_flags=0x%lx (%s)", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid, + __entry->target_highest_slotid, + __entry->status_flags, + show_nfs4_seq4_status(__entry->status_flags) + ) +); + +struct cb_sequenceargs; +struct cb_sequenceres; + +TRACE_EVENT(nfs4_cb_sequence, + TP_PROTO( + const struct cb_sequenceargs *args, + const struct cb_sequenceres *res, + __be32 status + ), + TP_ARGS(args, res, status), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, cachethis) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->session = nfs_session_id_hash(&args->csa_sessionid); + __entry->slot_nr = args->csa_slotid; + __entry->seq_nr = args->csa_sequenceid; + __entry->highest_slotid = args->csa_highestslotid; + __entry->cachethis = args->csa_cachethis; + __entry->error = be32_to_cpu(status); + ), + + TP_printk( + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid + ) +); + +TRACE_EVENT(nfs4_cb_seqid_err, + TP_PROTO( + const struct cb_sequenceargs *args, + __be32 status + ), + TP_ARGS(args, status), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, cachethis) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->session = nfs_session_id_hash(&args->csa_sessionid); + __entry->slot_nr = args->csa_slotid; + __entry->seq_nr = args->csa_sequenceid; + __entry->highest_slotid = args->csa_highestslotid; + __entry->cachethis = args->csa_cachethis; + __entry->error = be32_to_cpu(status); + ), + + TP_printk( + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid + ) +); + +TRACE_EVENT(nfs4_cb_offload, + TP_PROTO( + const struct nfs_fh *cb_fh, + const nfs4_stateid *cb_stateid, + uint64_t cb_count, + int cb_error, + int cb_how_stable + ), + + TP_ARGS(cb_fh, cb_stateid, cb_count, cb_error, + cb_how_stable), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(loff_t, cb_count) + __field(int, cb_how) + __field(int, cb_stateid_seq) + __field(u32, cb_stateid_hash) + ), + + TP_fast_assign( + __entry->error = cb_error < 0 ? -cb_error : 0; + __entry->fhandle = nfs_fhandle_hash(cb_fh); + __entry->cb_stateid_seq = + be32_to_cpu(cb_stateid->seqid); + __entry->cb_stateid_hash = + nfs_stateid_hash(cb_stateid); + __entry->cb_count = cb_count; + __entry->cb_how = cb_how_stable; + ), + + TP_printk( + "error=%ld (%s) fhandle=0x%08x cb_stateid=%d:0x%08x " + "cb_count=%llu cb_how=%s", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->fhandle, + __entry->cb_stateid_seq, __entry->cb_stateid_hash, + __entry->cb_count, + show_nfs_stable_how(__entry->cb_how) + ) +); + +TRACE_EVENT(pnfs_ds_connect, + TP_PROTO( + char *ds_remotestr, + int status + ), + + TP_ARGS(ds_remotestr, status), + + TP_STRUCT__entry( + __string(ds_ips, ds_remotestr) + __field(int, status) + ), + + TP_fast_assign( + __assign_str(ds_ips); + __entry->status = status; + ), + + TP_printk( + "ds_ips=%s, status=%d", + __get_str(ds_ips), + __entry->status + ) +); + +TRACE_EVENT(nfs4_setup_sequence, + TP_PROTO( + const struct nfs4_session *session, + const struct nfs4_sequence_args *args + ), + TP_ARGS(session, args), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_used_slotid) + ), + + TP_fast_assign( + const struct nfs4_slot *sa_slot = args->sa_slot; + __entry->session = session ? nfs_session_id_hash(&session->sess_id) : 0; + __entry->slot_nr = sa_slot->slot_nr; + __entry->seq_nr = sa_slot->seq_nr; + __entry->highest_used_slotid = + sa_slot->table->highest_used_slotid; + ), + TP_printk( + "session=0x%08x slot_nr=%u seq_nr=%u " + "highest_used_slotid=%u", + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_used_slotid + ) +); + +TRACE_DEFINE_ENUM(NFS4CLNT_MANAGER_RUNNING); +TRACE_DEFINE_ENUM(NFS4CLNT_CHECK_LEASE); +TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_EXPIRED); +TRACE_DEFINE_ENUM(NFS4CLNT_RECLAIM_REBOOT); +TRACE_DEFINE_ENUM(NFS4CLNT_RECLAIM_NOGRACE); +TRACE_DEFINE_ENUM(NFS4CLNT_DELEGRETURN); +TRACE_DEFINE_ENUM(NFS4CLNT_SESSION_RESET); +TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_CONFIRM); +TRACE_DEFINE_ENUM(NFS4CLNT_SERVER_SCOPE_MISMATCH); +TRACE_DEFINE_ENUM(NFS4CLNT_PURGE_STATE); +TRACE_DEFINE_ENUM(NFS4CLNT_BIND_CONN_TO_SESSION); +TRACE_DEFINE_ENUM(NFS4CLNT_MOVED); +TRACE_DEFINE_ENUM(NFS4CLNT_LEASE_MOVED); +TRACE_DEFINE_ENUM(NFS4CLNT_DELEGATION_EXPIRED); +TRACE_DEFINE_ENUM(NFS4CLNT_RUN_MANAGER); +TRACE_DEFINE_ENUM(NFS4CLNT_MANAGER_AVAILABLE); +TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_RUNNING); +TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_READ); +TRACE_DEFINE_ENUM(NFS4CLNT_RECALL_ANY_LAYOUT_RW); +TRACE_DEFINE_ENUM(NFS4CLNT_DELEGRETURN_DELAYED); + +#define show_nfs4_clp_state(state) \ + __print_flags(state, "|", \ + { BIT(NFS4CLNT_MANAGER_RUNNING), "MANAGER_RUNNING" }, \ + { BIT(NFS4CLNT_CHECK_LEASE), "CHECK_LEASE" }, \ + { BIT(NFS4CLNT_LEASE_EXPIRED), "LEASE_EXPIRED" }, \ + { BIT(NFS4CLNT_RECLAIM_REBOOT), "RECLAIM_REBOOT" }, \ + { BIT(NFS4CLNT_RECLAIM_NOGRACE), "RECLAIM_NOGRACE" }, \ + { BIT(NFS4CLNT_DELEGRETURN), "DELEGRETURN" }, \ + { BIT(NFS4CLNT_SESSION_RESET), "SESSION_RESET" }, \ + { BIT(NFS4CLNT_LEASE_CONFIRM), "LEASE_CONFIRM" }, \ + { BIT(NFS4CLNT_SERVER_SCOPE_MISMATCH), "SERVER_SCOPE_MISMATCH" }, \ + { BIT(NFS4CLNT_PURGE_STATE), "PURGE_STATE" }, \ + { BIT(NFS4CLNT_BIND_CONN_TO_SESSION), "BIND_CONN_TO_SESSION" }, \ + { BIT(NFS4CLNT_MOVED), "MOVED" }, \ + { BIT(NFS4CLNT_LEASE_MOVED), "LEASE_MOVED" }, \ + { BIT(NFS4CLNT_DELEGATION_EXPIRED), "DELEGATION_EXPIRED" }, \ + { BIT(NFS4CLNT_RUN_MANAGER), "RUN_MANAGER" }, \ + { BIT(NFS4CLNT_MANAGER_AVAILABLE), "MANAGER_AVAILABLE" }, \ + { BIT(NFS4CLNT_RECALL_RUNNING), "RECALL_RUNNING" }, \ + { BIT(NFS4CLNT_RECALL_ANY_LAYOUT_READ), "RECALL_ANY_LAYOUT_READ" }, \ + { BIT(NFS4CLNT_RECALL_ANY_LAYOUT_RW), "RECALL_ANY_LAYOUT_RW" }, \ + { BIT(NFS4CLNT_DELEGRETURN_DELAYED), "DELERETURN_DELAYED" }) + +TRACE_EVENT(nfs4_state_mgr, + TP_PROTO( + const struct nfs_client *clp + ), + + TP_ARGS(clp), + + TP_STRUCT__entry( + __field(unsigned long, state) + __string(hostname, clp->cl_hostname) + ), + + TP_fast_assign( + __entry->state = clp->cl_state; + __assign_str(hostname); + ), + + TP_printk( + "hostname=%s clp state=%s", __get_str(hostname), + show_nfs4_clp_state(__entry->state) + ) +) + +TRACE_EVENT(nfs4_state_mgr_failed, + TP_PROTO( + const struct nfs_client *clp, + const char *section, + int status + ), + + TP_ARGS(clp, section, status), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, state) + __string(hostname, clp->cl_hostname) + __string(section, section) + ), + + TP_fast_assign( + __entry->error = status < 0 ? -status : 0; + __entry->state = clp->cl_state; + __assign_str(hostname); + __assign_str(section); + ), + + TP_printk( + "hostname=%s clp state=%s error=%ld (%s) section=%s", + __get_str(hostname), + show_nfs4_clp_state(__entry->state), -__entry->error, + show_nfs4_status(__entry->error), __get_str(section) + + ) +) + +TRACE_EVENT(nfs4_xdr_bad_operation, + TP_PROTO( + const struct xdr_stream *xdr, + u32 op, + u32 expected + ), + + TP_ARGS(xdr, op, expected), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) + __field(u32, op) + __field(u32, expected) + ), + + TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); + __entry->op = op; + __entry->expected = expected; + ), + + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x operation=%u, expected=%u", + __entry->task_id, __entry->client_id, __entry->xid, + __entry->op, __entry->expected + ) +); + +DECLARE_EVENT_CLASS(nfs4_xdr_event, + TP_PROTO( + const struct xdr_stream *xdr, + u32 op, + u32 error + ), + + TP_ARGS(xdr, op, error), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) + __field(u32, op) + __field(unsigned long, error) + ), + + TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); + __entry->op = op; + __entry->error = error; + ), + + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x error=%ld (%s) operation=%u", + __entry->task_id, __entry->client_id, __entry->xid, + -__entry->error, show_nfs4_status(__entry->error), + __entry->op + ) +); +#define DEFINE_NFS4_XDR_EVENT(name) \ + DEFINE_EVENT(nfs4_xdr_event, name, \ + TP_PROTO( \ + const struct xdr_stream *xdr, \ + u32 op, \ + u32 error \ + ), \ + TP_ARGS(xdr, op, error)) +DEFINE_NFS4_XDR_EVENT(nfs4_xdr_status); +DEFINE_NFS4_XDR_EVENT(nfs4_xdr_bad_filehandle); + +DECLARE_EVENT_CLASS(nfs4_cb_error_class, + TP_PROTO( + __be32 xid, + u32 cb_ident + ), + + TP_ARGS(xid, cb_ident), + + TP_STRUCT__entry( + __field(u32, xid) + __field(u32, cbident) + ), + + TP_fast_assign( + __entry->xid = be32_to_cpu(xid); + __entry->cbident = cb_ident; + ), + + TP_printk( + "xid=0x%08x cb_ident=0x%08x", + __entry->xid, __entry->cbident + ) +); + +#define DEFINE_CB_ERROR_EVENT(name) \ + DEFINE_EVENT(nfs4_cb_error_class, nfs_cb_##name, \ + TP_PROTO( \ + __be32 xid, \ + u32 cb_ident \ + ), \ + TP_ARGS(xid, cb_ident)) + +DEFINE_CB_ERROR_EVENT(no_clp); +DEFINE_CB_ERROR_EVENT(badprinc); + +DECLARE_EVENT_CLASS(nfs4_open_event, + TP_PROTO( + const struct nfs_open_context *ctx, + int flags, + int error + ), + + TP_ARGS(ctx, flags, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, flags) + __field(unsigned long, fmode) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, openstateid_seq) + __field(u32, openstateid_hash) + ), + + TP_fast_assign( + const struct nfs4_state *state = ctx->state; + const struct inode *inode = NULL; + + __entry->error = -error; + __entry->flags = flags; + __entry->fmode = (__force unsigned long)ctx->mode; + __entry->dev = ctx->dentry->d_sb->s_dev; + if (!IS_ERR_OR_NULL(state)) { + inode = state->inode; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + __entry->openstateid_seq = + be32_to_cpu(state->open_stateid.seqid); + __entry->openstateid_hash = + nfs_stateid_hash(&state->open_stateid); + } else { + __entry->stateid_seq = 0; + __entry->stateid_hash = 0; + __entry->openstateid_seq = 0; + __entry->openstateid_hash = 0; + } + if (inode != NULL) { + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + } else { + __entry->fileid = 0; + __entry->fhandle = 0; + } + __entry->dir = d_inode(ctx->dentry->d_parent)->i_ino; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) flags=%lu (%s) fmode=%s " + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "name=%02x:%02x:%llu/%s stateid=%d:0x%08x " + "openstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->flags, + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name), + __entry->stateid_seq, __entry->stateid_hash, + __entry->openstateid_seq, __entry->openstateid_hash + ) +); + +#define DEFINE_NFS4_OPEN_EVENT(name) \ + DEFINE_EVENT(nfs4_open_event, name, \ + TP_PROTO( \ + const struct nfs_open_context *ctx, \ + int flags, \ + int error \ + ), \ + TP_ARGS(ctx, flags, error)) +DEFINE_NFS4_OPEN_EVENT(nfs4_open_reclaim); +DEFINE_NFS4_OPEN_EVENT(nfs4_open_expired); +DEFINE_NFS4_OPEN_EVENT(nfs4_open_file); + +TRACE_EVENT(nfs4_cached_open, + TP_PROTO( + const struct nfs4_state *state + ), + TP_ARGS(state), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, fmode) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fmode = (__force unsigned int)state->state; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + ), + + TP_printk( + "fmode=%s fileid=%02x:%02x:%llu " + "fhandle=0x%08x stateid=%d:0x%08x", + __entry->fmode ? show_fs_fmode_flags(__entry->fmode) : + "closed", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +TRACE_EVENT(nfs4_close, + TP_PROTO( + const struct nfs4_state *state, + const struct nfs_closeargs *args, + const struct nfs_closeres *res, + int error + ), + + TP_ARGS(state, args, res, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, fmode) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fmode = (__force unsigned int)state->state; + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(args->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->stateid); + ), + + TP_printk( + "error=%ld (%s) fmode=%s fileid=%02x:%02x:%llu " + "fhandle=0x%08x openstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->fmode ? show_fs_fmode_flags(__entry->fmode) : + "closed", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +DECLARE_EVENT_CLASS(nfs4_lock_event, + TP_PROTO( + const struct file_lock *request, + const struct nfs4_state *state, + int cmd, + int error + ), + + TP_ARGS(request, state, cmd, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, cmd) + __field(unsigned long, type) + __field(loff_t, start) + __field(loff_t, end) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->error = error < 0 ? -error : 0; + __entry->cmd = cmd; + __entry->type = request->c.flc_type; + __entry->start = request->fl_start; + __entry->end = request->fl_end; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + ), + + TP_printk( + "error=%ld (%s) cmd=%s:%s range=%lld:%lld " + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + show_fs_fcntl_cmd(__entry->cmd), + show_fs_fcntl_lock_type(__entry->type), + (long long)__entry->start, + (long long)__entry->end, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +#define DEFINE_NFS4_LOCK_EVENT(name) \ + DEFINE_EVENT(nfs4_lock_event, name, \ + TP_PROTO( \ + const struct file_lock *request, \ + const struct nfs4_state *state, \ + int cmd, \ + int error \ + ), \ + TP_ARGS(request, state, cmd, error)) +DEFINE_NFS4_LOCK_EVENT(nfs4_get_lock); +DEFINE_NFS4_LOCK_EVENT(nfs4_unlock); + +TRACE_EVENT(nfs4_set_lock, + TP_PROTO( + const struct file_lock *request, + const struct nfs4_state *state, + const nfs4_stateid *lockstateid, + int cmd, + int error + ), + + TP_ARGS(request, state, lockstateid, cmd, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, cmd) + __field(unsigned long, type) + __field(loff_t, start) + __field(loff_t, end) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, lockstateid_seq) + __field(u32, lockstateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->error = error < 0 ? -error : 0; + __entry->cmd = cmd; + __entry->type = request->c.flc_type; + __entry->start = request->fl_start; + __entry->end = request->fl_end; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + __entry->lockstateid_seq = + be32_to_cpu(lockstateid->seqid); + __entry->lockstateid_hash = + nfs_stateid_hash(lockstateid); + ), + + TP_printk( + "error=%ld (%s) cmd=%s:%s range=%lld:%lld " + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x lockstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + show_fs_fcntl_cmd(__entry->cmd), + show_fs_fcntl_lock_type(__entry->type), + (long long)__entry->start, + (long long)__entry->end, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __entry->lockstateid_seq, __entry->lockstateid_hash + ) +); + +TRACE_DEFINE_ENUM(LK_STATE_IN_USE); +TRACE_DEFINE_ENUM(NFS_DELEGATED_STATE); +TRACE_DEFINE_ENUM(NFS_OPEN_STATE); +TRACE_DEFINE_ENUM(NFS_O_RDONLY_STATE); +TRACE_DEFINE_ENUM(NFS_O_WRONLY_STATE); +TRACE_DEFINE_ENUM(NFS_O_RDWR_STATE); +TRACE_DEFINE_ENUM(NFS_STATE_RECLAIM_REBOOT); +TRACE_DEFINE_ENUM(NFS_STATE_RECLAIM_NOGRACE); +TRACE_DEFINE_ENUM(NFS_STATE_POSIX_LOCKS); +TRACE_DEFINE_ENUM(NFS_STATE_RECOVERY_FAILED); +TRACE_DEFINE_ENUM(NFS_STATE_MAY_NOTIFY_LOCK); +TRACE_DEFINE_ENUM(NFS_STATE_CHANGE_WAIT); +TRACE_DEFINE_ENUM(NFS_CLNT_DST_SSC_COPY_STATE); +TRACE_DEFINE_ENUM(NFS_CLNT_SRC_SSC_COPY_STATE); +TRACE_DEFINE_ENUM(NFS_SRV_SSC_COPY_STATE); + +#define show_nfs4_state_flags(flags) \ + __print_flags(flags, "|", \ + { LK_STATE_IN_USE, "IN_USE" }, \ + { NFS_DELEGATED_STATE, "DELEGATED" }, \ + { NFS_OPEN_STATE, "OPEN" }, \ + { NFS_O_RDONLY_STATE, "O_RDONLY" }, \ + { NFS_O_WRONLY_STATE, "O_WRONLY" }, \ + { NFS_O_RDWR_STATE, "O_RDWR" }, \ + { NFS_STATE_RECLAIM_REBOOT, "RECLAIM_REBOOT" }, \ + { NFS_STATE_RECLAIM_NOGRACE, "RECLAIM_NOGRACE" }, \ + { NFS_STATE_POSIX_LOCKS, "POSIX_LOCKS" }, \ + { NFS_STATE_RECOVERY_FAILED, "RECOVERY_FAILED" }, \ + { NFS_STATE_MAY_NOTIFY_LOCK, "MAY_NOTIFY_LOCK" }, \ + { NFS_STATE_CHANGE_WAIT, "CHANGE_WAIT" }, \ + { NFS_CLNT_DST_SSC_COPY_STATE, "CLNT_DST_SSC_COPY" }, \ + { NFS_CLNT_SRC_SSC_COPY_STATE, "CLNT_SRC_SSC_COPY" }, \ + { NFS_SRV_SSC_COPY_STATE, "SRV_SSC_COPY" }) + +#define show_nfs4_lock_flags(flags) \ + __print_flags(flags, "|", \ + { BIT(NFS_LOCK_INITIALIZED), "INITIALIZED" }, \ + { BIT(NFS_LOCK_LOST), "LOST" }) + +TRACE_EVENT(nfs4_state_lock_reclaim, + TP_PROTO( + const struct nfs4_state *state, + const struct nfs4_lock_state *lock + ), + + TP_ARGS(state, lock), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned long, state_flags) + __field(unsigned long, lock_flags) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->state_flags = state->flags; + __entry->lock_flags = lock->ls_flags; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x state_flags=%s lock_flags=%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + show_nfs4_state_flags(__entry->state_flags), + show_nfs4_lock_flags(__entry->lock_flags) + ) +) + +DECLARE_EVENT_CLASS(nfs4_set_delegation_event, + TP_PROTO( + const struct inode *inode, + fmode_t fmode + ), + + TP_ARGS(inode, fmode), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, fmode) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fmode = (__force unsigned int)fmode; + ), + + TP_printk( + "fmode=%s fileid=%02x:%02x:%llu fhandle=0x%08x", + show_fs_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); +#define DEFINE_NFS4_SET_DELEGATION_EVENT(name) \ + DEFINE_EVENT(nfs4_set_delegation_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + fmode_t fmode \ + ), \ + TP_ARGS(inode, fmode)) +DEFINE_NFS4_SET_DELEGATION_EVENT(nfs4_set_delegation); +DEFINE_NFS4_SET_DELEGATION_EVENT(nfs4_reclaim_delegation); +DEFINE_NFS4_SET_DELEGATION_EVENT(nfs4_detach_delegation); + +#define show_delegation_flags(flags) \ + __print_flags(flags, "|", \ + { BIT(NFS_DELEGATION_NEED_RECLAIM), "NEED_RECLAIM" }, \ + { BIT(NFS_DELEGATION_RETURN_IF_CLOSED), "RETURN_IF_CLOSED" }, \ + { BIT(NFS_DELEGATION_REFERENCED), "REFERENCED" }, \ + { BIT(NFS_DELEGATION_RETURNING), "RETURNING" }, \ + { BIT(NFS_DELEGATION_REVOKED), "REVOKED" }, \ + { BIT(NFS_DELEGATION_TEST_EXPIRED), "TEST_EXPIRED" }) + +DECLARE_EVENT_CLASS(nfs4_delegation_event, + TP_PROTO( + const struct nfs_delegation *delegation + ), + + TP_ARGS(delegation), + + TP_STRUCT__entry( + __field(u32, fhandle) + __field(unsigned int, fmode) + __field(unsigned long, flags) + ), + + TP_fast_assign( + __entry->fhandle = nfs_fhandle_hash(NFS_FH(delegation->inode)); + __entry->fmode = delegation->type; + __entry->flags = delegation->flags; + ), + + TP_printk( + "fhandle=0x%08x fmode=%s flags=%s", + __entry->fhandle, show_fs_fmode_flags(__entry->fmode), + show_delegation_flags(__entry->flags) + ) +); +#define DEFINE_NFS4_DELEGATION_EVENT(name) \ + DEFINE_EVENT(nfs4_delegation_event, name, \ + TP_PROTO( \ + const struct nfs_delegation *delegation \ + ), \ + TP_ARGS(delegation)) +DEFINE_NFS4_DELEGATION_EVENT(nfs_delegation_need_return); + +TRACE_EVENT(nfs4_delegreturn_exit, + TP_PROTO( + const struct nfs4_delegreturnargs *args, + const struct nfs4_delegreturnres *res, + int error + ), + + TP_ARGS(args, res, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->dev = res->server->s_dev; + __entry->fhandle = nfs_fhandle_hash(args->fhandle); + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(args->stateid->seqid); + __entry->stateid_hash = + nfs_stateid_hash(args->stateid); + ), + + TP_printk( + "error=%ld (%s) dev=%02x:%02x fhandle=0x%08x " + "stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +DECLARE_EVENT_CLASS(nfs4_test_stateid_event, + TP_PROTO( + const struct nfs4_state *state, + const struct nfs4_lock_state *lsp, + int error + ), + + TP_ARGS(state, lsp, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = state->inode; + + __entry->error = error < 0 ? -error : 0; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +#define DEFINE_NFS4_TEST_STATEID_EVENT(name) \ + DEFINE_EVENT(nfs4_test_stateid_event, name, \ + TP_PROTO( \ + const struct nfs4_state *state, \ + const struct nfs4_lock_state *lsp, \ + int error \ + ), \ + TP_ARGS(state, lsp, error)) +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_delegation_stateid); +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_open_stateid); +DEFINE_NFS4_TEST_STATEID_EVENT(nfs4_test_lock_stateid); + +DECLARE_EVENT_CLASS(nfs4_lookup_event, + TP_PROTO( + const struct inode *dir, + const struct qstr *name, + int error + ), + + TP_ARGS(dir, name, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long, error) + __field(u64, dir) + __string(name, name->name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->error = -error; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS4_LOOKUP_EVENT(name) \ + DEFINE_EVENT(nfs4_lookup_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct qstr *name, \ + int error \ + ), \ + TP_ARGS(dir, name, error)) + +DEFINE_NFS4_LOOKUP_EVENT(nfs4_lookup); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_symlink); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_mkdir); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_mknod); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_remove); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_get_fs_locations); +DEFINE_NFS4_LOOKUP_EVENT(nfs4_secinfo); + +TRACE_EVENT(nfs4_lookupp, + TP_PROTO( + const struct inode *inode, + int error + ), + + TP_ARGS(inode, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, ino) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->error = error < 0 ? -error : 0; + ), + + TP_printk( + "error=%ld (%s) inode=%02x:%02x:%llu", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->ino + ) +); + +TRACE_EVENT(nfs4_rename, + TP_PROTO( + const struct inode *olddir, + const struct qstr *oldname, + const struct inode *newdir, + const struct qstr *newname, + int error + ), + + TP_ARGS(olddir, oldname, newdir, newname, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long, error) + __field(u64, olddir) + __string(oldname, oldname->name) + __field(u64, newdir) + __string(newname, newname->name) + ), + + TP_fast_assign( + __entry->dev = olddir->i_sb->s_dev; + __entry->olddir = olddir->i_ino; + __entry->newdir = newdir->i_ino; + __entry->error = error < 0 ? -error : 0; + __assign_str(oldname); + __assign_str(newname); + ), + + TP_printk( + "error=%ld (%s) oldname=%02x:%02x:%llu/%s " + "newname=%02x:%02x:%llu/%s", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->olddir, + __get_str(oldname), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->newdir, + __get_str(newname) + ) +); + +DECLARE_EVENT_CLASS(nfs4_inode_event, + TP_PROTO( + const struct inode *inode, + int error + ), + + TP_ARGS(inode, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->error = error < 0 ? -error : 0; + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +#define DEFINE_NFS4_INODE_EVENT(name) \ + DEFINE_EVENT(nfs4_inode_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + int error \ + ), \ + TP_ARGS(inode, error)) + +DEFINE_NFS4_INODE_EVENT(nfs4_access); +DEFINE_NFS4_INODE_EVENT(nfs4_readlink); +DEFINE_NFS4_INODE_EVENT(nfs4_readdir); +DEFINE_NFS4_INODE_EVENT(nfs4_get_acl); +DEFINE_NFS4_INODE_EVENT(nfs4_set_acl); +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +DEFINE_NFS4_INODE_EVENT(nfs4_get_security_label); +DEFINE_NFS4_INODE_EVENT(nfs4_set_security_label); +#endif /* CONFIG_NFS_V4_SECURITY_LABEL */ + +DECLARE_EVENT_CLASS(nfs4_inode_stateid_event, + TP_PROTO( + const struct inode *inode, + const nfs4_stateid *stateid, + int error + ), + + TP_ARGS(inode, stateid, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(stateid->seqid); + __entry->stateid_hash = + nfs_stateid_hash(stateid); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); + +#define DEFINE_NFS4_INODE_STATEID_EVENT(name) \ + DEFINE_EVENT(nfs4_inode_stateid_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + const nfs4_stateid *stateid, \ + int error \ + ), \ + TP_ARGS(inode, stateid, error)) + +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_setattr); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_delegreturn); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_open_stateid_update); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_open_stateid_update_wait); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_open_stateid_update_skip); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_close_stateid_update_wait); + +DECLARE_EVENT_CLASS(nfs4_getattr_event, + TP_PROTO( + const struct nfs_server *server, + const struct nfs_fh *fhandle, + const struct nfs_fattr *fattr, + int error + ), + + TP_ARGS(server, fhandle, fattr, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned int, valid) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->dev = server->s_dev; + __entry->valid = fattr->valid; + __entry->fhandle = nfs_fhandle_hash(fhandle); + __entry->fileid = (fattr->valid & NFS_ATTR_FATTR_FILEID) ? fattr->fileid : 0; + __entry->error = error < 0 ? -error : 0; + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "valid=%s", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_nfs_fattr_flags(__entry->valid) + ) +); + +#define DEFINE_NFS4_GETATTR_EVENT(name) \ + DEFINE_EVENT(nfs4_getattr_event, name, \ + TP_PROTO( \ + const struct nfs_server *server, \ + const struct nfs_fh *fhandle, \ + const struct nfs_fattr *fattr, \ + int error \ + ), \ + TP_ARGS(server, fhandle, fattr, error)) +DEFINE_NFS4_GETATTR_EVENT(nfs4_getattr); +DEFINE_NFS4_GETATTR_EVENT(nfs4_lookup_root); +DEFINE_NFS4_GETATTR_EVENT(nfs4_fsinfo); + +DECLARE_EVENT_CLASS(nfs4_inode_callback_event, + TP_PROTO( + const struct nfs_client *clp, + const struct nfs_fh *fhandle, + const struct inode *inode, + int error + ), + + TP_ARGS(clp, fhandle, inode, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __string(dstaddr, clp ? clp->cl_hostname : "unknown") + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __entry->fhandle = nfs_fhandle_hash(fhandle); + if (!IS_ERR_OR_NULL(inode)) { + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + } else { + __entry->fileid = 0; + __entry->dev = 0; + } + __assign_str(dstaddr); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "dstaddr=%s", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __get_str(dstaddr) + ) +); + +#define DEFINE_NFS4_INODE_CALLBACK_EVENT(name) \ + DEFINE_EVENT(nfs4_inode_callback_event, name, \ + TP_PROTO( \ + const struct nfs_client *clp, \ + const struct nfs_fh *fhandle, \ + const struct inode *inode, \ + int error \ + ), \ + TP_ARGS(clp, fhandle, inode, error)) +DEFINE_NFS4_INODE_CALLBACK_EVENT(nfs4_cb_getattr); + +DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, + TP_PROTO( + const struct nfs_client *clp, + const struct nfs_fh *fhandle, + const struct inode *inode, + const nfs4_stateid *stateid, + int error + ), + + TP_ARGS(clp, fhandle, inode, stateid, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __string(dstaddr, clp ? clp->cl_hostname : "unknown") + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __entry->fhandle = nfs_fhandle_hash(fhandle); + if (!IS_ERR_OR_NULL(inode)) { + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + } else { + __entry->fileid = 0; + __entry->dev = 0; + } + __assign_str(dstaddr); + __entry->stateid_seq = + be32_to_cpu(stateid->seqid); + __entry->stateid_hash = + nfs_stateid_hash(stateid); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x dstaddr=%s", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __get_str(dstaddr) + ) +); + +#define DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(name) \ + DEFINE_EVENT(nfs4_inode_stateid_callback_event, name, \ + TP_PROTO( \ + const struct nfs_client *clp, \ + const struct nfs_fh *fhandle, \ + const struct inode *inode, \ + const nfs4_stateid *stateid, \ + int error \ + ), \ + TP_ARGS(clp, fhandle, inode, stateid, error)) +DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(nfs4_cb_recall); +DEFINE_NFS4_INODE_STATEID_CALLBACK_EVENT(nfs4_cb_layoutrecall_file); + +#define show_stateid_type(type) \ + __print_symbolic(type, \ + { NFS4_INVALID_STATEID_TYPE, "INVALID" }, \ + { NFS4_SPECIAL_STATEID_TYPE, "SPECIAL" }, \ + { NFS4_OPEN_STATEID_TYPE, "OPEN" }, \ + { NFS4_LOCK_STATEID_TYPE, "LOCK" }, \ + { NFS4_DELEGATION_STATEID_TYPE, "DELEGATION" }, \ + { NFS4_LAYOUT_STATEID_TYPE, "LAYOUT" }, \ + { NFS4_PNFS_DS_STATEID_TYPE, "PNFS_DS" }, \ + { NFS4_REVOKED_STATEID_TYPE, "REVOKED" }, \ + { NFS4_FREED_STATEID_TYPE, "FREED" }) + +DECLARE_EVENT_CLASS(nfs4_match_stateid_event, + TP_PROTO( + const nfs4_stateid *s1, + const nfs4_stateid *s2 + ), + + TP_ARGS(s1, s2), + + TP_STRUCT__entry( + __field(int, s1_seq) + __field(int, s2_seq) + __field(u32, s1_hash) + __field(u32, s2_hash) + __field(int, s1_type) + __field(int, s2_type) + ), + + TP_fast_assign( + __entry->s1_seq = s1->seqid; + __entry->s1_hash = nfs_stateid_hash(s1); + __entry->s1_type = s1->type; + __entry->s2_seq = s2->seqid; + __entry->s2_hash = nfs_stateid_hash(s2); + __entry->s2_type = s2->type; + ), + + TP_printk( + "s1=%s:%x:%u s2=%s:%x:%u", + show_stateid_type(__entry->s1_type), + __entry->s1_hash, __entry->s1_seq, + show_stateid_type(__entry->s2_type), + __entry->s2_hash, __entry->s2_seq + ) +); + +#define DEFINE_NFS4_MATCH_STATEID_EVENT(name) \ + DEFINE_EVENT(nfs4_match_stateid_event, name, \ + TP_PROTO( \ + const nfs4_stateid *s1, \ + const nfs4_stateid *s2 \ + ), \ + TP_ARGS(s1, s2)) +DEFINE_NFS4_MATCH_STATEID_EVENT(nfs41_match_stateid); +DEFINE_NFS4_MATCH_STATEID_EVENT(nfs4_match_stateid); + +DECLARE_EVENT_CLASS(nfs4_idmap_event, + TP_PROTO( + const char *name, + int len, + u32 id, + int error + ), + + TP_ARGS(name, len, id, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, id) + __dynamic_array(char, name, len > 0 ? len + 1 : 1) + ), + + TP_fast_assign( + if (len < 0) + len = 0; + __entry->error = error < 0 ? error : 0; + __entry->id = id; + memcpy(__get_str(name), name, len); + __get_str(name)[len] = 0; + ), + + TP_printk( + "error=%ld (%s) id=%u name=%s", + -__entry->error, show_nfs4_status(__entry->error), + __entry->id, + __get_str(name) + ) +); +#define DEFINE_NFS4_IDMAP_EVENT(name) \ + DEFINE_EVENT(nfs4_idmap_event, name, \ + TP_PROTO( \ + const char *name, \ + int len, \ + u32 id, \ + int error \ + ), \ + TP_ARGS(name, len, id, error)) +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_name_to_uid); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_group_to_gid); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_uid_to_name); +DEFINE_NFS4_IDMAP_EVENT(nfs4_map_gid_to_group); + +#define NFS4_LSEG_LAYOUT_STATEID_HASH(lseg) \ + (lseg ? nfs_stateid_hash(&lseg->pls_layout->plh_stateid) : 0) + +DECLARE_EVENT_CLASS(nfs4_read_event, + TP_PROTO( + const struct nfs_pgio_header *hdr, + int error + ), + + TP_ARGS(hdr, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + const struct nfs4_state *state = + hdr->args.context->state; + const struct pnfs_layout_segment *lseg = hdr->lseg; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + __entry->layoutstateid_seq = lseg ? lseg->pls_seq : 0; + __entry->layoutstateid_hash = + NFS4_LSEG_LAYOUT_STATEID_HASH(lseg); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u stateid=%d:0x%08x " + "layoutstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->arg_count, __entry->res_count, + __entry->stateid_seq, __entry->stateid_hash, + __entry->layoutstateid_seq, __entry->layoutstateid_hash + ) +); +#define DEFINE_NFS4_READ_EVENT(name) \ + DEFINE_EVENT(nfs4_read_event, name, \ + TP_PROTO( \ + const struct nfs_pgio_header *hdr, \ + int error \ + ), \ + TP_ARGS(hdr, error)) +DEFINE_NFS4_READ_EVENT(nfs4_read); +DEFINE_NFS4_READ_EVENT(nfs4_pnfs_read); + +DECLARE_EVENT_CLASS(nfs4_write_event, + TP_PROTO( + const struct nfs_pgio_header *hdr, + int error + ), + + TP_ARGS(hdr, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + const struct nfs4_state *state = + hdr->args.context->state; + const struct pnfs_layout_segment *lseg = hdr->lseg; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + __entry->layoutstateid_seq = lseg ? lseg->pls_seq : 0; + __entry->layoutstateid_hash = + NFS4_LSEG_LAYOUT_STATEID_HASH(lseg); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u stateid=%d:0x%08x " + "layoutstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->arg_count, __entry->res_count, + __entry->stateid_seq, __entry->stateid_hash, + __entry->layoutstateid_seq, __entry->layoutstateid_hash + ) +); + +#define DEFINE_NFS4_WRITE_EVENT(name) \ + DEFINE_EVENT(nfs4_write_event, name, \ + TP_PROTO( \ + const struct nfs_pgio_header *hdr, \ + int error \ + ), \ + TP_ARGS(hdr, error)) +DEFINE_NFS4_WRITE_EVENT(nfs4_write); +DEFINE_NFS4_WRITE_EVENT(nfs4_pnfs_write); + +DECLARE_EVENT_CLASS(nfs4_commit_event, + TP_PROTO( + const struct nfs_commit_data *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(unsigned long, error) + __field(loff_t, offset) + __field(u32, count) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; + const struct pnfs_layout_segment *lseg = data->lseg; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __entry->error = error < 0 ? -error : 0; + __entry->layoutstateid_seq = lseg ? lseg->pls_seq : 0; + __entry->layoutstateid_hash = + NFS4_LSEG_LAYOUT_STATEID_HASH(lseg); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u layoutstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + __entry->count, + __entry->layoutstateid_seq, __entry->layoutstateid_hash + ) +); +#define DEFINE_NFS4_COMMIT_EVENT(name) \ + DEFINE_EVENT(nfs4_commit_event, name, \ + TP_PROTO( \ + const struct nfs_commit_data *data, \ + int error \ + ), \ + TP_ARGS(data, error)) +DEFINE_NFS4_COMMIT_EVENT(nfs4_commit); +DEFINE_NFS4_COMMIT_EVENT(nfs4_pnfs_commit_ds); + +TRACE_EVENT(nfs4_layoutget, + TP_PROTO( + const struct nfs_open_context *ctx, + const struct pnfs_layout_range *args, + const struct pnfs_layout_range *res, + const nfs4_stateid *layout_stateid, + int error + ), + + TP_ARGS(ctx, args, res, layout_stateid, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u32, iomode) + __field(u64, offset) + __field(u64, count) + __field(unsigned long, error) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + ), + + TP_fast_assign( + const struct inode *inode = d_inode(ctx->dentry); + const struct nfs4_state *state = ctx->state; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->iomode = args->iomode; + __entry->offset = args->offset; + __entry->count = args->length; + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(state->stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&state->stateid); + if (!error) { + __entry->layoutstateid_seq = + be32_to_cpu(layout_stateid->seqid); + __entry->layoutstateid_hash = + nfs_stateid_hash(layout_stateid); + } else { + __entry->layoutstateid_seq = 0; + __entry->layoutstateid_hash = 0; + } + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "iomode=%s offset=%llu count=%llu stateid=%d:0x%08x " + "layoutstateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_pnfs_layout_iomode(__entry->iomode), + (unsigned long long)__entry->offset, + (unsigned long long)__entry->count, + __entry->stateid_seq, __entry->stateid_hash, + __entry->layoutstateid_seq, __entry->layoutstateid_hash + ) +); + +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutcommit); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutreturn); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutreturn_on_close); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layouterror); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_layoutstats); + +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_UNKNOWN); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_NO_PNFS); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_RD_ZEROLEN); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_MDSTHRESH); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_NOMEM); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_BULK_RECALL); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_IO_TEST_FAIL); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_FOUND_CACHED); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_RETURN); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_BLOCKED); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_INVALID_OPEN); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_RETRY); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_EXIT); + +#define show_pnfs_update_layout_reason(reason) \ + __print_symbolic(reason, \ + { PNFS_UPDATE_LAYOUT_UNKNOWN, "unknown" }, \ + { PNFS_UPDATE_LAYOUT_NO_PNFS, "no pnfs" }, \ + { PNFS_UPDATE_LAYOUT_RD_ZEROLEN, "read+zerolen" }, \ + { PNFS_UPDATE_LAYOUT_MDSTHRESH, "mdsthresh" }, \ + { PNFS_UPDATE_LAYOUT_NOMEM, "nomem" }, \ + { PNFS_UPDATE_LAYOUT_BULK_RECALL, "bulk recall" }, \ + { PNFS_UPDATE_LAYOUT_IO_TEST_FAIL, "io test fail" }, \ + { PNFS_UPDATE_LAYOUT_FOUND_CACHED, "found cached" }, \ + { PNFS_UPDATE_LAYOUT_RETURN, "layoutreturn" }, \ + { PNFS_UPDATE_LAYOUT_BLOCKED, "layouts blocked" }, \ + { PNFS_UPDATE_LAYOUT_INVALID_OPEN, "invalid open" }, \ + { PNFS_UPDATE_LAYOUT_RETRY, "retrying" }, \ + { PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" }, \ + { PNFS_UPDATE_LAYOUT_EXIT, "exit" }) + +TRACE_EVENT(pnfs_update_layout, + TP_PROTO(struct inode *inode, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg, + enum pnfs_update_layout_reason reason + ), + TP_ARGS(inode, pos, count, iomode, lo, lseg, reason), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u32, fhandle) + __field(loff_t, pos) + __field(u64, count) + __field(enum pnfs_iomode, iomode) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + __field(long, lseg) + __field(enum pnfs_update_layout_reason, reason) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->pos = pos; + __entry->count = count; + __entry->iomode = iomode; + __entry->reason = reason; + if (lo != NULL && pnfs_layout_is_valid(lo)) { + __entry->layoutstateid_seq = + be32_to_cpu(lo->plh_stateid.seqid); + __entry->layoutstateid_hash = + nfs_stateid_hash(&lo->plh_stateid); + } else { + __entry->layoutstateid_seq = 0; + __entry->layoutstateid_hash = 0; + } + __entry->lseg = (long)lseg; + ), + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "iomode=%s pos=%llu count=%llu " + "layoutstateid=%d:0x%08x lseg=0x%lx (%s)", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_pnfs_layout_iomode(__entry->iomode), + (unsigned long long)__entry->pos, + (unsigned long long)__entry->count, + __entry->layoutstateid_seq, __entry->layoutstateid_hash, + __entry->lseg, + show_pnfs_update_layout_reason(__entry->reason) + ) +); + +DECLARE_EVENT_CLASS(pnfs_layout_event, + TP_PROTO(struct inode *inode, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg + ), + TP_ARGS(inode, pos, count, iomode, lo, lseg), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u32, fhandle) + __field(loff_t, pos) + __field(u64, count) + __field(enum pnfs_iomode, iomode) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + __field(long, lseg) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->pos = pos; + __entry->count = count; + __entry->iomode = iomode; + if (lo != NULL && pnfs_layout_is_valid(lo)) { + __entry->layoutstateid_seq = + be32_to_cpu(lo->plh_stateid.seqid); + __entry->layoutstateid_hash = + nfs_stateid_hash(&lo->plh_stateid); + } else { + __entry->layoutstateid_seq = 0; + __entry->layoutstateid_hash = 0; + } + __entry->lseg = (long)lseg; + ), + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "iomode=%s pos=%llu count=%llu " + "layoutstateid=%d:0x%08x lseg=0x%lx", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_pnfs_layout_iomode(__entry->iomode), + (unsigned long long)__entry->pos, + (unsigned long long)__entry->count, + __entry->layoutstateid_seq, __entry->layoutstateid_hash, + __entry->lseg + ) +); + +#define DEFINE_PNFS_LAYOUT_EVENT(name) \ + DEFINE_EVENT(pnfs_layout_event, name, \ + TP_PROTO(struct inode *inode, \ + loff_t pos, \ + u64 count, \ + enum pnfs_iomode iomode, \ + struct pnfs_layout_hdr *lo, \ + struct pnfs_layout_segment *lseg \ + ), \ + TP_ARGS(inode, pos, count, iomode, lo, lseg)) + +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_read); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_write); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_get_mirror_count); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_done); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist); + +DECLARE_EVENT_CLASS(nfs4_deviceid_event, + TP_PROTO( + const struct nfs_client *clp, + const struct nfs4_deviceid *deviceid + ), + + TP_ARGS(clp, deviceid), + + TP_STRUCT__entry( + __string(dstaddr, clp->cl_hostname) + __array(unsigned char, deviceid, NFS4_DEVICEID4_SIZE) + ), + + TP_fast_assign( + __assign_str(dstaddr); + memcpy(__entry->deviceid, deviceid->data, + NFS4_DEVICEID4_SIZE); + ), + + TP_printk( + "deviceid=%s, dstaddr=%s", + __print_hex(__entry->deviceid, NFS4_DEVICEID4_SIZE), + __get_str(dstaddr) + ) +); +#define DEFINE_PNFS_DEVICEID_EVENT(name) \ + DEFINE_EVENT(nfs4_deviceid_event, name, \ + TP_PROTO(const struct nfs_client *clp, \ + const struct nfs4_deviceid *deviceid \ + ), \ + TP_ARGS(clp, deviceid)) +DEFINE_PNFS_DEVICEID_EVENT(nfs4_deviceid_free); + +DECLARE_EVENT_CLASS(nfs4_deviceid_status, + TP_PROTO( + const struct nfs_server *server, + const struct nfs4_deviceid *deviceid, + int status + ), + + TP_ARGS(server, deviceid, status), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, status) + __string(dstaddr, server->nfs_client->cl_hostname) + __array(unsigned char, deviceid, NFS4_DEVICEID4_SIZE) + ), + + TP_fast_assign( + __entry->dev = server->s_dev; + __entry->status = status; + __assign_str(dstaddr); + memcpy(__entry->deviceid, deviceid->data, + NFS4_DEVICEID4_SIZE); + ), + + TP_printk( + "dev=%02x:%02x: deviceid=%s, dstaddr=%s, status=%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __print_hex(__entry->deviceid, NFS4_DEVICEID4_SIZE), + __get_str(dstaddr), + __entry->status + ) +); +#define DEFINE_PNFS_DEVICEID_STATUS(name) \ + DEFINE_EVENT(nfs4_deviceid_status, name, \ + TP_PROTO(const struct nfs_server *server, \ + const struct nfs4_deviceid *deviceid, \ + int status \ + ), \ + TP_ARGS(server, deviceid, status)) +DEFINE_PNFS_DEVICEID_STATUS(nfs4_getdeviceinfo); +DEFINE_PNFS_DEVICEID_STATUS(nfs4_find_deviceid); + +TRACE_EVENT(fl_getdevinfo, + TP_PROTO( + const struct nfs_server *server, + const struct nfs4_deviceid *deviceid, + char *ds_remotestr + ), + TP_ARGS(server, deviceid, ds_remotestr), + + TP_STRUCT__entry( + __string(mds_addr, server->nfs_client->cl_hostname) + __array(unsigned char, deviceid, NFS4_DEVICEID4_SIZE) + __string(ds_ips, ds_remotestr) + ), + + TP_fast_assign( + __assign_str(mds_addr); + __assign_str(ds_ips); + memcpy(__entry->deviceid, deviceid->data, + NFS4_DEVICEID4_SIZE); + ), + TP_printk( + "deviceid=%s, mds_addr=%s, ds_ips=%s", + __print_hex(__entry->deviceid, NFS4_DEVICEID4_SIZE), + __get_str(mds_addr), + __get_str(ds_ips) + ) +); + +DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, + TP_PROTO( + const struct nfs_pgio_header *hdr, + int error + ), + + TP_ARGS(hdr, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, nfs_error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __string(dstaddr, hdr->ds_clp ? + rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown") + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + + __entry->error = -error; + __entry->nfs_error = hdr->res.op_status; + __entry->fhandle = nfs_fhandle_hash(hdr->args.fh); + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; + __entry->stateid_seq = + be32_to_cpu(hdr->args.stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&hdr->args.stateid); + __assign_str(dstaddr); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s " + "nfs_error=%lu (%s)", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->offset, __entry->count, + __entry->stateid_seq, __entry->stateid_hash, + __get_str(dstaddr), __entry->nfs_error, + show_nfs4_status(__entry->nfs_error) + ) +); + +#define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \ + DEFINE_EVENT(nfs4_flexfiles_io_event, name, \ + TP_PROTO( \ + const struct nfs_pgio_header *hdr, \ + int error \ + ), \ + TP_ARGS(hdr, error)) +DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error); +DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error); + +TRACE_EVENT(ff_layout_commit_error, + TP_PROTO( + const struct nfs_commit_data *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, nfs_error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __string(dstaddr, data->ds_clp ? + rpc_peeraddr2str(data->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown") + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + + __entry->error = -error; + __entry->nfs_error = data->res.op_status; + __entry->fhandle = nfs_fhandle_hash(data->args.fh); + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __assign_str(dstaddr); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%llu count=%u dstaddr=%s nfs_error=%lu (%s)", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->offset, __entry->count, + __get_str(dstaddr), __entry->nfs_error, + show_nfs4_status(__entry->nfs_error) + ) +); + +TRACE_EVENT(bl_ext_tree_prepare_commit, + TP_PROTO( + int ret, + size_t count, + u64 lwb, + bool not_all_ranges + ), + + TP_ARGS(ret, count, lwb, not_all_ranges), + + TP_STRUCT__entry( + __field(int, ret) + __field(size_t, count) + __field(u64, lwb) + __field(bool, not_all_ranges) + ), + + TP_fast_assign( + __entry->ret = ret; + __entry->count = count; + __entry->lwb = lwb; + __entry->not_all_ranges = not_all_ranges; + ), + + TP_printk( + "ret=%d, found %zu ranges, lwb=%llu%s", + __entry->ret, + __entry->count, + __entry->lwb, + __entry->not_all_ranges ? ", not all ranges encoded" : + "" + ) +); + +DECLARE_EVENT_CLASS(pnfs_bl_pr_key_class, + TP_PROTO( + const struct block_device *bdev, + u64 key + ), + TP_ARGS(bdev, key), + TP_STRUCT__entry( + __field(u64, key) + __field(dev_t, dev) + __string(device, bdev->bd_disk->disk_name) + ), + TP_fast_assign( + __entry->key = key; + __entry->dev = bdev->bd_dev; + __assign_str(device); + ), + TP_printk("dev=%d,%d (%s) key=0x%016llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __get_str(device), __entry->key + ) +); + +#define DEFINE_NFS4_BLOCK_PRKEY_EVENT(name) \ + DEFINE_EVENT(pnfs_bl_pr_key_class, name, \ + TP_PROTO( \ + const struct block_device *bdev, \ + u64 key \ + ), \ + TP_ARGS(bdev, key)) +DEFINE_NFS4_BLOCK_PRKEY_EVENT(bl_pr_key_reg); +DEFINE_NFS4_BLOCK_PRKEY_EVENT(bl_pr_key_unreg); + +/* + * From uapi/linux/pr.h + */ +TRACE_DEFINE_ENUM(PR_STS_SUCCESS); +TRACE_DEFINE_ENUM(PR_STS_IOERR); +TRACE_DEFINE_ENUM(PR_STS_RESERVATION_CONFLICT); +TRACE_DEFINE_ENUM(PR_STS_RETRY_PATH_FAILURE); +TRACE_DEFINE_ENUM(PR_STS_PATH_FAST_FAILED); +TRACE_DEFINE_ENUM(PR_STS_PATH_FAILED); + +#define show_pr_status(x) \ + __print_symbolic(x, \ + { PR_STS_SUCCESS, "SUCCESS" }, \ + { PR_STS_IOERR, "IOERR" }, \ + { PR_STS_RESERVATION_CONFLICT, "RESERVATION_CONFLICT" }, \ + { PR_STS_RETRY_PATH_FAILURE, "RETRY_PATH_FAILURE" }, \ + { PR_STS_PATH_FAST_FAILED, "PATH_FAST_FAILED" }, \ + { PR_STS_PATH_FAILED, "PATH_FAILED" }) + +DECLARE_EVENT_CLASS(pnfs_bl_pr_key_err_class, + TP_PROTO( + const struct block_device *bdev, + u64 key, + int status + ), + TP_ARGS(bdev, key, status), + TP_STRUCT__entry( + __field(u64, key) + __field(dev_t, dev) + __field(unsigned long, status) + __string(device, bdev->bd_disk->disk_name) + ), + TP_fast_assign( + __entry->key = key; + __entry->dev = bdev->bd_dev; + __entry->status = status; + __assign_str(device); + ), + TP_printk("dev=%d,%d (%s) key=0x%016llx status=%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __get_str(device), __entry->key, + show_pr_status(__entry->status) + ) +); + +#define DEFINE_NFS4_BLOCK_PRKEY_ERR_EVENT(name) \ + DEFINE_EVENT(pnfs_bl_pr_key_err_class, name, \ + TP_PROTO( \ + const struct block_device *bdev, \ + u64 key, \ + int status \ + ), \ + TP_ARGS(bdev, key, status)) +DEFINE_NFS4_BLOCK_PRKEY_ERR_EVENT(bl_pr_key_reg_err); +DEFINE_NFS4_BLOCK_PRKEY_ERR_EVENT(bl_pr_key_unreg_err); + +#ifdef CONFIG_NFS_V4_2 +TRACE_DEFINE_ENUM(NFS4_CONTENT_DATA); +TRACE_DEFINE_ENUM(NFS4_CONTENT_HOLE); + +#define show_llseek_mode(what) \ + __print_symbolic(what, \ + { NFS4_CONTENT_DATA, "DATA" }, \ + { NFS4_CONTENT_HOLE, "HOLE" }) + +TRACE_EVENT(nfs4_llseek, + TP_PROTO( + const struct inode *inode, + const struct nfs42_seek_args *args, + const struct nfs42_seek_res *res, + int error + ), + + TP_ARGS(inode, args, res, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(u64, fileid) + __field(dev_t, dev) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(loff_t, offset_s) + __field(u32, what) + __field(loff_t, offset_r) + __field(u32, eof) + ), + + TP_fast_assign( + const struct nfs_fh *fh = args->sa_fh; + + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset_s = args->sa_offset; + __entry->stateid_seq = + be32_to_cpu(args->sa_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->sa_stateid); + __entry->what = args->sa_what; + if (error) { + __entry->error = -error; + __entry->offset_r = 0; + __entry->eof = 0; + } else { + __entry->error = 0; + __entry->offset_r = res->sr_offset; + __entry->eof = res->sr_eof; + } + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x offset_s=%llu what=%s " + "offset_r=%llu eof=%u", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __entry->offset_s, + show_llseek_mode(__entry->what), + __entry->offset_r, + __entry->eof + ) +); + +DECLARE_EVENT_CLASS(nfs4_sparse_event, + TP_PROTO( + const struct inode *inode, + const struct nfs42_falloc_args *args, + int error + ), + + TP_ARGS(inode, args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(loff_t, offset) + __field(loff_t, len) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __entry->offset = args->falloc_offset; + __entry->len = args->falloc_length; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->stateid_seq = + be32_to_cpu(args->falloc_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->falloc_stateid); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x offset=%llu len=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + (long long)__entry->offset, + (long long)__entry->len + ) +); +#define DEFINE_NFS4_SPARSE_EVENT(name) \ + DEFINE_EVENT(nfs4_sparse_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + const struct nfs42_falloc_args *args, \ + int error \ + ), \ + TP_ARGS(inode, args, error)) +DEFINE_NFS4_SPARSE_EVENT(nfs4_fallocate); +DEFINE_NFS4_SPARSE_EVENT(nfs4_deallocate); + +TRACE_EVENT(nfs4_copy, + TP_PROTO( + const struct inode *src_inode, + const struct inode *dst_inode, + const struct nfs42_copy_args *args, + const struct nfs42_copy_res *res, + const struct nl4_server *nss, + int error + ), + + TP_ARGS(src_inode, dst_inode, args, res, nss, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, src_fhandle) + __field(u32, src_fileid) + __field(u32, dst_fhandle) + __field(u32, dst_fileid) + __field(dev_t, src_dev) + __field(dev_t, dst_dev) + __field(int, src_stateid_seq) + __field(u32, src_stateid_hash) + __field(int, dst_stateid_seq) + __field(u32, dst_stateid_hash) + __field(loff_t, src_offset) + __field(loff_t, dst_offset) + __field(bool, sync) + __field(loff_t, len) + __field(int, res_stateid_seq) + __field(u32, res_stateid_hash) + __field(loff_t, res_count) + __field(bool, res_sync) + __field(bool, res_cons) + __field(bool, intra) + ), + + TP_fast_assign( + __entry->src_fileid = src_inode->i_ino; + __entry->src_dev = src_inode->i_sb->s_dev; + __entry->src_fhandle = nfs_fhandle_hash(args->src_fh); + __entry->src_offset = args->src_pos; + __entry->dst_fileid = dst_inode->i_ino; + __entry->dst_dev = dst_inode->i_sb->s_dev; + __entry->dst_fhandle = nfs_fhandle_hash(args->dst_fh); + __entry->dst_offset = args->dst_pos; + __entry->len = args->count; + __entry->sync = args->sync; + __entry->src_stateid_seq = + be32_to_cpu(args->src_stateid.seqid); + __entry->src_stateid_hash = + nfs_stateid_hash(&args->src_stateid); + __entry->dst_stateid_seq = + be32_to_cpu(args->dst_stateid.seqid); + __entry->dst_stateid_hash = + nfs_stateid_hash(&args->dst_stateid); + __entry->intra = nss ? 0 : 1; + if (error) { + __entry->error = -error; + __entry->res_stateid_seq = 0; + __entry->res_stateid_hash = 0; + __entry->res_count = 0; + __entry->res_sync = 0; + __entry->res_cons = 0; + } else { + __entry->error = 0; + __entry->res_stateid_seq = + be32_to_cpu(res->write_res.stateid.seqid); + __entry->res_stateid_hash = + nfs_stateid_hash(&res->write_res.stateid); + __entry->res_count = res->write_res.count; + __entry->res_sync = res->synchronous; + __entry->res_cons = res->consecutive; + } + ), + + TP_printk( + "error=%ld (%s) intra=%d src_fileid=%02x:%02x:%llu " + "src_fhandle=0x%08x dst_fileid=%02x:%02x:%llu " + "dst_fhandle=0x%08x src_stateid=%d:0x%08x " + "dst_stateid=%d:0x%08x src_offset=%llu dst_offset=%llu " + "len=%llu sync=%d cb_stateid=%d:0x%08x res_sync=%d " + "res_cons=%d res_count=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->intra, + MAJOR(__entry->src_dev), MINOR(__entry->src_dev), + (unsigned long long)__entry->src_fileid, + __entry->src_fhandle, + MAJOR(__entry->dst_dev), MINOR(__entry->dst_dev), + (unsigned long long)__entry->dst_fileid, + __entry->dst_fhandle, + __entry->src_stateid_seq, __entry->src_stateid_hash, + __entry->dst_stateid_seq, __entry->dst_stateid_hash, + __entry->src_offset, + __entry->dst_offset, + __entry->len, + __entry->sync, + __entry->res_stateid_seq, __entry->res_stateid_hash, + __entry->res_sync, + __entry->res_cons, + __entry->res_count + ) +); + +TRACE_EVENT(nfs4_clone, + TP_PROTO( + const struct inode *src_inode, + const struct inode *dst_inode, + const struct nfs42_clone_args *args, + int error + ), + + TP_ARGS(src_inode, dst_inode, args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, src_fhandle) + __field(u32, src_fileid) + __field(u32, dst_fhandle) + __field(u32, dst_fileid) + __field(dev_t, src_dev) + __field(dev_t, dst_dev) + __field(loff_t, src_offset) + __field(loff_t, dst_offset) + __field(int, src_stateid_seq) + __field(u32, src_stateid_hash) + __field(int, dst_stateid_seq) + __field(u32, dst_stateid_hash) + __field(loff_t, len) + ), + + TP_fast_assign( + __entry->src_fileid = src_inode->i_ino; + __entry->src_dev = src_inode->i_sb->s_dev; + __entry->src_fhandle = nfs_fhandle_hash(args->src_fh); + __entry->src_offset = args->src_offset; + __entry->dst_fileid = dst_inode->i_ino; + __entry->dst_dev = dst_inode->i_sb->s_dev; + __entry->dst_fhandle = nfs_fhandle_hash(args->dst_fh); + __entry->dst_offset = args->dst_offset; + __entry->len = args->count; + __entry->error = error < 0 ? -error : 0; + __entry->src_stateid_seq = + be32_to_cpu(args->src_stateid.seqid); + __entry->src_stateid_hash = + nfs_stateid_hash(&args->src_stateid); + __entry->dst_stateid_seq = + be32_to_cpu(args->dst_stateid.seqid); + __entry->dst_stateid_hash = + nfs_stateid_hash(&args->dst_stateid); + ), + + TP_printk( + "error=%ld (%s) src_fileid=%02x:%02x:%llu " + "src_fhandle=0x%08x dst_fileid=%02x:%02x:%llu " + "dst_fhandle=0x%08x src_stateid=%d:0x%08x " + "dst_stateid=%d:0x%08x src_offset=%llu " + "dst_offset=%llu len=%llu", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->src_dev), MINOR(__entry->src_dev), + (unsigned long long)__entry->src_fileid, + __entry->src_fhandle, + MAJOR(__entry->dst_dev), MINOR(__entry->dst_dev), + (unsigned long long)__entry->dst_fileid, + __entry->dst_fhandle, + __entry->src_stateid_seq, __entry->src_stateid_hash, + __entry->dst_stateid_seq, __entry->dst_stateid_hash, + __entry->src_offset, + __entry->dst_offset, + __entry->len + ) +); + +TRACE_EVENT(nfs4_copy_notify, + TP_PROTO( + const struct inode *inode, + const struct nfs42_copy_notify_args *args, + const struct nfs42_copy_notify_res *res, + int error + ), + + TP_ARGS(inode, args, res, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(u64, fileid) + __field(dev_t, dev) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __field(int, res_stateid_seq) + __field(u32, res_stateid_hash) + ), + + TP_fast_assign( + __entry->fileid = inode->i_ino; + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(args->cna_src_fh); + __entry->stateid_seq = + be32_to_cpu(args->cna_src_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->cna_src_stateid); + if (error) { + __entry->error = -error; + __entry->res_stateid_seq = 0; + __entry->res_stateid_hash = 0; + } else { + __entry->error = 0; + __entry->res_stateid_seq = + be32_to_cpu(res->cnr_stateid.seqid); + __entry->res_stateid_hash = + nfs_stateid_hash(&res->cnr_stateid); + } + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "stateid=%d:0x%08x res_stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash, + __entry->res_stateid_seq, __entry->res_stateid_hash + ) +); + +DECLARE_EVENT_CLASS(nfs4_offload_class, + TP_PROTO( + const struct nfs42_offload_status_args *args, + int error + ), + + TP_ARGS(args, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(u32, fhandle) + __field(int, stateid_seq) + __field(u32, stateid_hash) + ), + + TP_fast_assign( + __entry->fhandle = nfs_fhandle_hash(args->osa_src_fh); + __entry->error = error < 0 ? -error : 0; + __entry->stateid_seq = + be32_to_cpu(args->osa_stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&args->osa_stateid); + ), + + TP_printk( + "error=%ld (%s) fhandle=0x%08x stateid=%d:0x%08x", + -__entry->error, + show_nfs4_status(__entry->error), + __entry->fhandle, + __entry->stateid_seq, __entry->stateid_hash + ) +); +#define DEFINE_NFS4_OFFLOAD_EVENT(name) \ + DEFINE_EVENT(nfs4_offload_class, name, \ + TP_PROTO( \ + const struct nfs42_offload_status_args *args, \ + int error \ + ), \ + TP_ARGS(args, error)) +DEFINE_NFS4_OFFLOAD_EVENT(nfs4_offload_cancel); +DEFINE_NFS4_OFFLOAD_EVENT(nfs4_offload_status); + +DECLARE_EVENT_CLASS(nfs4_xattr_event, + TP_PROTO( + const struct inode *inode, + const char *name, + int error + ), + + TP_ARGS(inode, name, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __string(name, name) + ), + + TP_fast_assign( + __entry->error = error < 0 ? -error : 0; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "name=%s", + -__entry->error, show_nfs4_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __get_str(name) + ) +); +#define DEFINE_NFS4_XATTR_EVENT(name) \ + DEFINE_EVENT(nfs4_xattr_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + const char *name, \ + int error \ + ), \ + TP_ARGS(inode, name, error)) +DEFINE_NFS4_XATTR_EVENT(nfs4_getxattr); +DEFINE_NFS4_XATTR_EVENT(nfs4_setxattr); +DEFINE_NFS4_XATTR_EVENT(nfs4_removexattr); + +DEFINE_NFS4_INODE_EVENT(nfs4_listxattr); +#endif /* CONFIG_NFS_V4_2 */ + +#endif /* _TRACE_NFS4_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE nfs4trace +/* This part must be outside protection */ +#include diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c new file mode 100644 index 000000000..c23c2eee1 --- /dev/null +++ b/fs/nfs/nfs4xdr.c @@ -0,0 +1,7786 @@ +/* + * fs/nfs/nfs4xdr.c + * + * Client-side XDR for NFSv4. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Kendrick Smith + * Andy Adamson + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "nfs4trace.h" +#include "internal.h" +#include "nfs4idmap.h" +#include "nfs4session.h" +#include "pnfs.h" +#include "netns.h" + +#define NFSDBG_FACILITY NFSDBG_XDR + +struct compound_hdr; +static void encode_layoutget(struct xdr_stream *xdr, + const struct nfs4_layoutget_args *args, + struct compound_hdr *hdr); +static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs4_layoutget_res *res); + +/* NFSv4 COMPOUND tags are only wanted for debugging purposes */ +#ifdef DEBUG +#define NFS4_MAXTAGLEN 20 +#else +#define NFS4_MAXTAGLEN 0 +#endif + +/* lock,open owner id: + * we currently use size 2 (u64) out of (NFS4_OPAQUE_LIMIT >> 2) + */ +#define pagepad_maxsz (1) +#define open_owner_id_maxsz (2 + 1 + 2 + 2) +#define lock_owner_id_maxsz (2 + 1 + 2) +#define compound_encode_hdr_maxsz (3 + (NFS4_MAXTAGLEN >> 2)) +#define compound_decode_hdr_maxsz (3 + (NFS4_MAXTAGLEN >> 2)) +#define op_encode_hdr_maxsz (1) +#define op_decode_hdr_maxsz (2) +#define encode_stateid_maxsz (XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_stateid_maxsz (XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define encode_verifier_maxsz (XDR_QUADLEN(NFS4_VERIFIER_SIZE)) +#define decode_verifier_maxsz (XDR_QUADLEN(NFS4_VERIFIER_SIZE)) +#define encode_putfh_maxsz (op_encode_hdr_maxsz + 1 + \ + (NFS4_FHSIZE >> 2)) +#define decode_putfh_maxsz (op_decode_hdr_maxsz) +#define encode_putrootfh_maxsz (op_encode_hdr_maxsz) +#define decode_putrootfh_maxsz (op_decode_hdr_maxsz) +#define encode_getfh_maxsz (op_encode_hdr_maxsz) +#define decode_getfh_maxsz (op_decode_hdr_maxsz + 1 + \ + ((3+NFS4_FHSIZE) >> 2)) +#define nfs4_fattr_bitmap_maxsz 4 +#define encode_getattr_maxsz (op_encode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) +#define nfstime4_maxsz (3) +#define nfs4_name_maxsz (1 + ((3 + NFS4_MAXNAMLEN) >> 2)) +#define nfs4_path_maxsz (1 + ((3 + NFS4_MAXPATHLEN) >> 2)) +#define nfs4_owner_maxsz (1 + XDR_QUADLEN(IDMAP_NAMESZ)) +#define nfs4_group_maxsz (1 + XDR_QUADLEN(IDMAP_NAMESZ)) +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +/* PI(4 bytes) + LFS(4 bytes) + 1(for null terminator?) + MAXLABELLEN */ +#define nfs4_label_maxsz (4 + 4 + 1 + XDR_QUADLEN(NFS4_MAXLABELLEN)) +#else +#define nfs4_label_maxsz 0 +#endif +/* We support only one layout type per file system */ +#define decode_mdsthreshold_maxsz (1 + 1 + nfs4_fattr_bitmap_maxsz + 1 + 8) +/* This is based on getfattr, which uses the most attributes: */ +#define nfs4_fattr_value_maxsz (1 + (1 + 2 + 2 + 4 + 2 + 1 + 1 + 2 + 2 + \ + 3*nfstime4_maxsz + \ + nfs4_owner_maxsz + \ + nfs4_group_maxsz + nfs4_label_maxsz + \ + decode_mdsthreshold_maxsz)) +#define nfs4_fattr_maxsz (nfs4_fattr_bitmap_maxsz + \ + nfs4_fattr_value_maxsz) +#define decode_getattr_maxsz (op_decode_hdr_maxsz + nfs4_fattr_maxsz) +#define encode_attrs_maxsz (nfs4_fattr_bitmap_maxsz + \ + 1 + 2 + 1 + \ + nfs4_owner_maxsz + \ + nfs4_group_maxsz + \ + nfs4_label_maxsz + \ + 1 + nfstime4_maxsz + \ + 1 + nfstime4_maxsz) +#define encode_savefh_maxsz (op_encode_hdr_maxsz) +#define decode_savefh_maxsz (op_decode_hdr_maxsz) +#define encode_restorefh_maxsz (op_encode_hdr_maxsz) +#define decode_restorefh_maxsz (op_decode_hdr_maxsz) +#define encode_fsinfo_maxsz (encode_getattr_maxsz) +/* The 5 accounts for the PNFS attributes, and assumes that at most three + * layout types will be returned. + */ +#define decode_fsinfo_maxsz (op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz + 1 + \ + 1 /* lease time */ + \ + 2 /* max filesize */ + \ + 2 /* max read */ + \ + 2 /* max write */ + \ + nfstime4_maxsz /* time delta */ + \ + 5 /* fs layout types */ + \ + 1 /* layout blksize */ + \ + 1 /* clone blksize */ + \ + 1 /* change attr type */ + \ + 1 /* xattr support */) +#define encode_renew_maxsz (op_encode_hdr_maxsz + 3) +#define decode_renew_maxsz (op_decode_hdr_maxsz) +#define encode_setclientid_maxsz \ + (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_VERIFIER_SIZE) + \ + /* client name */ \ + 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 1 /* sc_prog */ + \ + 1 + XDR_QUADLEN(RPCBIND_MAXNETIDLEN) + \ + 1 + XDR_QUADLEN(RPCBIND_MAXUADDRLEN) + \ + 1) /* sc_cb_ident */ +#define decode_setclientid_maxsz \ + (op_decode_hdr_maxsz + \ + 2 /* clientid */ + \ + XDR_QUADLEN(NFS4_VERIFIER_SIZE) + \ + 1 + XDR_QUADLEN(RPCBIND_MAXNETIDLEN) + \ + 1 + XDR_QUADLEN(RPCBIND_MAXUADDRLEN)) +#define encode_setclientid_confirm_maxsz \ + (op_encode_hdr_maxsz + \ + 3 + (NFS4_VERIFIER_SIZE >> 2)) +#define decode_setclientid_confirm_maxsz \ + (op_decode_hdr_maxsz) +#define encode_lookup_maxsz (op_encode_hdr_maxsz + nfs4_name_maxsz) +#define decode_lookup_maxsz (op_decode_hdr_maxsz) +#define encode_lookupp_maxsz (op_encode_hdr_maxsz) +#define decode_lookupp_maxsz (op_decode_hdr_maxsz) +#define encode_share_access_maxsz \ + (2) +#define encode_createmode_maxsz (1 + encode_attrs_maxsz + encode_verifier_maxsz) +#define encode_opentype_maxsz (1 + encode_createmode_maxsz) +#define encode_claim_null_maxsz (1 + nfs4_name_maxsz) +#define encode_open_maxsz (op_encode_hdr_maxsz + \ + 2 + encode_share_access_maxsz + 2 + \ + 1 + open_owner_id_maxsz + \ + encode_opentype_maxsz + \ + encode_claim_null_maxsz) +#define decode_space_limit_maxsz (3) +#define decode_ace_maxsz (3 + nfs4_owner_maxsz) +#define decode_delegation_maxsz (1 + decode_stateid_maxsz + 1 + \ + decode_space_limit_maxsz + \ + decode_ace_maxsz) +#define decode_change_info_maxsz (5) +#define decode_open_maxsz (op_decode_hdr_maxsz + \ + decode_stateid_maxsz + \ + decode_change_info_maxsz + 1 + \ + nfs4_fattr_bitmap_maxsz + \ + decode_delegation_maxsz) +#define encode_open_confirm_maxsz \ + (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 1) +#define decode_open_confirm_maxsz \ + (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_open_downgrade_maxsz \ + (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 1 + \ + encode_share_access_maxsz) +#define decode_open_downgrade_maxsz \ + (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_close_maxsz (op_encode_hdr_maxsz + \ + 1 + encode_stateid_maxsz) +#define decode_close_maxsz (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_setattr_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + encode_attrs_maxsz) +#define decode_setattr_maxsz (op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz) +#define encode_delegattr_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + nfs4_fattr_bitmap_maxsz + \ + 2*nfstime4_maxsz) +#define decode_delegattr_maxsz (decode_setattr_maxsz) +#define encode_read_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define decode_read_maxsz (op_decode_hdr_maxsz + 2 + pagepad_maxsz) +#define encode_readdir_maxsz (op_encode_hdr_maxsz + \ + 2 + encode_verifier_maxsz + 5 + \ + nfs4_label_maxsz) +#define decode_readdir_maxsz (op_decode_hdr_maxsz + \ + decode_verifier_maxsz + pagepad_maxsz) +#define encode_readlink_maxsz (op_encode_hdr_maxsz) +#define decode_readlink_maxsz (op_decode_hdr_maxsz + 1 + pagepad_maxsz) +#define encode_write_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 4) +#define decode_write_maxsz (op_decode_hdr_maxsz + \ + 2 + decode_verifier_maxsz) +#define encode_commit_maxsz (op_encode_hdr_maxsz + 3) +#define decode_commit_maxsz (op_decode_hdr_maxsz + \ + decode_verifier_maxsz) +#define encode_remove_maxsz (op_encode_hdr_maxsz + \ + nfs4_name_maxsz) +#define decode_remove_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz) +#define encode_rename_maxsz (op_encode_hdr_maxsz + \ + 2 * nfs4_name_maxsz) +#define decode_rename_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz + \ + decode_change_info_maxsz) +#define encode_link_maxsz (op_encode_hdr_maxsz + \ + nfs4_name_maxsz) +#define decode_link_maxsz (op_decode_hdr_maxsz + decode_change_info_maxsz) +#define encode_lockowner_maxsz (2 + 1 + lock_owner_id_maxsz) + +#define encode_lock_maxsz (op_encode_hdr_maxsz + \ + 7 + \ + 1 + encode_stateid_maxsz + 1 + \ + encode_lockowner_maxsz) +#define decode_lock_denied_maxsz \ + (2 + 2 + 1 + 2 + 1 + lock_owner_id_maxsz) +#define decode_lock_maxsz (op_decode_hdr_maxsz + \ + decode_lock_denied_maxsz) +#define encode_lockt_maxsz (op_encode_hdr_maxsz + 5 + \ + encode_lockowner_maxsz) +#define decode_lockt_maxsz (op_decode_hdr_maxsz + \ + decode_lock_denied_maxsz) +#define encode_locku_maxsz (op_encode_hdr_maxsz + 3 + \ + encode_stateid_maxsz + \ + 4) +#define decode_locku_maxsz (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_release_lockowner_maxsz \ + (op_encode_hdr_maxsz + \ + encode_lockowner_maxsz) +#define decode_release_lockowner_maxsz \ + (op_decode_hdr_maxsz) +#define encode_access_maxsz (op_encode_hdr_maxsz + 1) +#define decode_access_maxsz (op_decode_hdr_maxsz + 2) +#define encode_symlink_maxsz (op_encode_hdr_maxsz + \ + 1 + nfs4_name_maxsz + \ + 1 + \ + nfs4_fattr_maxsz) +#define decode_symlink_maxsz (op_decode_hdr_maxsz + 8) +#define encode_create_maxsz (op_encode_hdr_maxsz + \ + 1 + 2 + nfs4_name_maxsz + \ + encode_attrs_maxsz) +#define decode_create_maxsz (op_decode_hdr_maxsz + \ + decode_change_info_maxsz + \ + nfs4_fattr_bitmap_maxsz) +#define encode_statfs_maxsz (encode_getattr_maxsz) +#define decode_statfs_maxsz (decode_getattr_maxsz) +#define encode_delegreturn_maxsz (op_encode_hdr_maxsz + 4) +#define decode_delegreturn_maxsz (op_decode_hdr_maxsz) +#define encode_getacl_maxsz (encode_getattr_maxsz) +#define decode_getacl_maxsz (op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz + 1 + pagepad_maxsz) +#define encode_setacl_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define decode_setacl_maxsz (decode_setattr_maxsz) +#define encode_fs_locations_maxsz \ + (encode_getattr_maxsz) +#define decode_fs_locations_maxsz \ + (pagepad_maxsz) +#define encode_secinfo_maxsz (op_encode_hdr_maxsz + nfs4_name_maxsz) +#define decode_secinfo_maxsz (op_decode_hdr_maxsz + 1 + ((NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN)) / 4)) + +#define NFS4_MAX_MACHINE_NAME_LEN (64) +#define IMPL_NAME_LIMIT (sizeof(utsname()->sysname) + sizeof(utsname()->release) + \ + sizeof(utsname()->version) + sizeof(utsname()->machine) + 8) + +#define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \ + encode_verifier_maxsz + \ + 1 /* co_ownerid.len */ + \ + /* eia_clientowner */ \ + 1 + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 1 /* flags */ + \ + 1 /* spa_how */ + \ + /* max is SP4_MACH_CRED (for now) */ + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 1 /* implementation id array of size 1 */ + \ + 1 /* nii_domain */ + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 1 /* nii_name */ + \ + XDR_QUADLEN(IMPL_NAME_LIMIT) + \ + 3 /* nii_date */) +#define decode_exchange_id_maxsz (op_decode_hdr_maxsz + \ + 2 /* eir_clientid */ + \ + 1 /* eir_sequenceid */ + \ + 1 /* eir_flags */ + \ + 1 /* spr_how */ + \ + /* max is SP4_MACH_CRED (for now) */ + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 1 + NFS4_OP_MAP_NUM_WORDS + \ + 2 /* eir_server_owner.so_minor_id */ + \ + /* eir_server_owner.so_major_id<> */ \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \ + /* eir_server_scope<> */ \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \ + 1 /* eir_server_impl_id array length */ + \ + 1 /* nii_domain */ + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 1 /* nii_name */ + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + \ + 3 /* nii_date */) +#define encode_channel_attrs_maxsz (6 + 1 /* ca_rdma_ird.len (0) */) +#define decode_channel_attrs_maxsz (6 + \ + 1 /* ca_rdma_ird.len */ + \ + 1 /* ca_rdma_ird */) +#define encode_create_session_maxsz (op_encode_hdr_maxsz + \ + 2 /* csa_clientid */ + \ + 1 /* csa_sequence */ + \ + 1 /* csa_flags */ + \ + encode_channel_attrs_maxsz + \ + encode_channel_attrs_maxsz + \ + 1 /* csa_cb_program */ + \ + 1 /* csa_sec_parms.len (1) */ + \ + 1 /* cb_secflavor (AUTH_SYS) */ + \ + 1 /* stamp */ + \ + 1 /* machinename.len */ + \ + XDR_QUADLEN(NFS4_MAX_MACHINE_NAME_LEN) + \ + 1 /* uid */ + \ + 1 /* gid */ + \ + 1 /* gids.len (0) */) +#define decode_create_session_maxsz (op_decode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ + 1 /* csr_sequence */ + \ + 1 /* csr_flags */ + \ + decode_channel_attrs_maxsz + \ + decode_channel_attrs_maxsz) +#define encode_bind_conn_to_session_maxsz (op_encode_hdr_maxsz + \ + /* bctsa_sessid */ \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ + 1 /* bctsa_dir */ + \ + 1 /* bctsa_use_conn_in_rdma_mode */) +#define decode_bind_conn_to_session_maxsz (op_decode_hdr_maxsz + \ + /* bctsr_sessid */ \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ + 1 /* bctsr_dir */ + \ + 1 /* bctsr_use_conn_in_rdma_mode */) +#define encode_destroy_session_maxsz (op_encode_hdr_maxsz + 4) +#define decode_destroy_session_maxsz (op_decode_hdr_maxsz) +#define encode_destroy_clientid_maxsz (op_encode_hdr_maxsz + 2) +#define decode_destroy_clientid_maxsz (op_decode_hdr_maxsz) +#define encode_sequence_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4) +#define decode_sequence_maxsz (op_decode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5) +#define encode_reclaim_complete_maxsz (op_encode_hdr_maxsz + 4) +#define decode_reclaim_complete_maxsz (op_decode_hdr_maxsz + 4) +#define encode_get_dir_deleg_maxsz (op_encode_hdr_maxsz + \ + 4 /* gdda_signal_deleg_avail */ + \ + 8 /* gdda_notification_types */ + \ + nfstime4_maxsz /* gdda_child_attr_delay */ + \ + nfstime4_maxsz /* gdda_dir_attr_delay */ + \ + nfs4_fattr_bitmap_maxsz /* gdda_child_attributes */ + \ + nfs4_fattr_bitmap_maxsz /* gdda_dir_attributes */) +#define decode_get_dir_deleg_maxsz (op_decode_hdr_maxsz + \ + 4 /* gddrnf_status */ + \ + encode_verifier_maxsz /* gddr_cookieverf */ + \ + encode_stateid_maxsz /* gddr_stateid */ + \ + 8 /* gddr_notification */ + \ + nfs4_fattr_maxsz /* gddr_child_attributes */ + \ + nfs4_fattr_maxsz /* gddr_dir_attributes */) +#define encode_getdeviceinfo_maxsz (op_encode_hdr_maxsz + \ + XDR_QUADLEN(NFS4_DEVICEID4_SIZE) + \ + 1 /* layout type */ + \ + 1 /* maxcount */ + \ + 1 /* bitmap size */ + \ + 1 /* notification bitmap length */ + \ + 1 /* notification bitmap, word 0 */) +#define decode_getdeviceinfo_maxsz (op_decode_hdr_maxsz + \ + 1 /* layout type */ + \ + 1 /* opaque devaddr4 length */ + \ + /* devaddr4 payload is read into page */ \ + 1 /* notification bitmap length */ + \ + 1 /* notification bitmap, word 0 */ + \ + pagepad_maxsz /* possible XDR padding */) +#define encode_layoutget_maxsz (op_encode_hdr_maxsz + 10 + \ + encode_stateid_maxsz) +#define decode_layoutget_maxsz (op_decode_hdr_maxsz + 8 + \ + decode_stateid_maxsz + \ + XDR_QUADLEN(PNFS_LAYOUT_MAXSIZE) + \ + pagepad_maxsz) +#define encode_layoutcommit_maxsz (op_encode_hdr_maxsz + \ + 2 /* offset */ + \ + 2 /* length */ + \ + 1 /* reclaim */ + \ + encode_stateid_maxsz + \ + 1 /* new offset (true) */ + \ + 2 /* last byte written */ + \ + 1 /* nt_timechanged (false) */ + \ + 1 /* layoutupdate4 layout type */ + \ + 1 /* layoutupdate4 opaqueue len */) + /* the actual content of layoutupdate4 should + be allocated by drivers and spliced in + using xdr_write_pages */ +#define decode_layoutcommit_maxsz (op_decode_hdr_maxsz + 3) +#define encode_layoutreturn_maxsz (8 + op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + 1 + \ + XDR_QUADLEN(NFS4_OPAQUE_LIMIT)) +#define decode_layoutreturn_maxsz (op_decode_hdr_maxsz + \ + 1 + decode_stateid_maxsz) +#define encode_secinfo_no_name_maxsz (op_encode_hdr_maxsz + 1) +#define decode_secinfo_no_name_maxsz decode_secinfo_maxsz +#define encode_test_stateid_maxsz (op_encode_hdr_maxsz + 2 + \ + XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_test_stateid_maxsz (op_decode_hdr_maxsz + 2 + 1) +#define encode_free_stateid_maxsz (op_encode_hdr_maxsz + 1 + \ + XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_free_stateid_maxsz (op_decode_hdr_maxsz) + +#define NFS4_enc_compound_sz (1024) /* XXX: large enough? */ +#define NFS4_dec_compound_sz (1024) /* XXX: large enough? */ +#define NFS4_enc_read_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_read_maxsz) +#define NFS4_dec_read_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_read_maxsz) +#define NFS4_enc_readlink_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_readlink_maxsz) +#define NFS4_dec_readlink_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_readlink_maxsz) +#define NFS4_enc_readdir_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_readdir_maxsz) +#define NFS4_dec_readdir_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_readdir_maxsz) +#define NFS4_enc_write_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_write_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_write_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_write_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_commit_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_commit_maxsz) +#define NFS4_dec_commit_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_commit_maxsz) +#define NFS4_enc_open_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_open_maxsz + \ + encode_access_maxsz + \ + encode_getfh_maxsz + \ + encode_getattr_maxsz + \ + encode_layoutget_maxsz) +#define NFS4_dec_open_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_open_maxsz + \ + decode_access_maxsz + \ + decode_getfh_maxsz + \ + decode_getattr_maxsz + \ + decode_layoutget_maxsz) +#define NFS4_enc_open_confirm_sz \ + (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_open_confirm_maxsz) +#define NFS4_dec_open_confirm_sz \ + (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + decode_open_confirm_maxsz) +#define NFS4_enc_open_noattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_open_maxsz + \ + encode_access_maxsz + \ + encode_getattr_maxsz + \ + encode_layoutget_maxsz) +#define NFS4_dec_open_noattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_open_maxsz + \ + decode_access_maxsz + \ + decode_getattr_maxsz + \ + decode_layoutget_maxsz) +#define NFS4_enc_open_downgrade_sz \ + (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutreturn_maxsz + \ + encode_open_downgrade_maxsz) +#define NFS4_dec_open_downgrade_sz \ + (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutreturn_maxsz + \ + decode_open_downgrade_maxsz) +#define NFS4_enc_close_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutreturn_maxsz + \ + encode_close_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_close_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutreturn_maxsz + \ + decode_close_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_setattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_setattr_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_setattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_setattr_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_fsinfo_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_fsinfo_maxsz) +#define NFS4_dec_fsinfo_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_fsinfo_maxsz) +#define NFS4_enc_renew_sz (compound_encode_hdr_maxsz + \ + encode_renew_maxsz) +#define NFS4_dec_renew_sz (compound_decode_hdr_maxsz + \ + decode_renew_maxsz) +#define NFS4_enc_setclientid_sz (compound_encode_hdr_maxsz + \ + encode_setclientid_maxsz) +#define NFS4_dec_setclientid_sz (compound_decode_hdr_maxsz + \ + decode_setclientid_maxsz) +#define NFS4_enc_setclientid_confirm_sz \ + (compound_encode_hdr_maxsz + \ + encode_setclientid_confirm_maxsz) +#define NFS4_dec_setclientid_confirm_sz \ + (compound_decode_hdr_maxsz + \ + decode_setclientid_confirm_maxsz) +#define NFS4_enc_lock_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_lock_maxsz) +#define NFS4_dec_lock_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_lock_maxsz) +#define NFS4_enc_lockt_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_lockt_maxsz) +#define NFS4_dec_lockt_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_lockt_maxsz) +#define NFS4_enc_locku_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_locku_maxsz) +#define NFS4_dec_locku_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_locku_maxsz) +#define NFS4_enc_release_lockowner_sz \ + (compound_encode_hdr_maxsz + \ + encode_lockowner_maxsz) +#define NFS4_dec_release_lockowner_sz \ + (compound_decode_hdr_maxsz + \ + decode_release_lockowner_maxsz) +#define NFS4_enc_access_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_access_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_access_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_access_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_getattr_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_get_dir_deleg_maxsz + \ + encode_getattr_maxsz + \ + encode_renew_maxsz) +#define NFS4_dec_getattr_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_get_dir_deleg_maxsz + \ + decode_getattr_maxsz + \ + decode_renew_maxsz) +#define NFS4_enc_lookup_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_lookup_maxsz + \ + encode_getattr_maxsz + \ + encode_getfh_maxsz) +#define NFS4_dec_lookup_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_lookup_maxsz + \ + decode_getattr_maxsz + \ + decode_getfh_maxsz) +#define NFS4_enc_lookupp_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_lookupp_maxsz + \ + encode_getattr_maxsz + \ + encode_getfh_maxsz) +#define NFS4_dec_lookupp_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_lookupp_maxsz + \ + decode_getattr_maxsz + \ + decode_getfh_maxsz) +#define NFS4_enc_lookup_root_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putrootfh_maxsz + \ + encode_getattr_maxsz + \ + encode_getfh_maxsz) +#define NFS4_dec_lookup_root_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putrootfh_maxsz + \ + decode_getattr_maxsz + \ + decode_getfh_maxsz) +#define NFS4_enc_remove_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_remove_maxsz) +#define NFS4_dec_remove_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_remove_maxsz) +#define NFS4_enc_rename_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_savefh_maxsz + \ + encode_putfh_maxsz + \ + encode_rename_maxsz) +#define NFS4_dec_rename_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_savefh_maxsz + \ + decode_putfh_maxsz + \ + decode_rename_maxsz) +#define NFS4_enc_link_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_savefh_maxsz + \ + encode_putfh_maxsz + \ + encode_link_maxsz + \ + encode_restorefh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_link_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_savefh_maxsz + \ + decode_putfh_maxsz + \ + decode_link_maxsz + \ + decode_restorefh_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_symlink_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_symlink_maxsz + \ + encode_getattr_maxsz + \ + encode_getfh_maxsz) +#define NFS4_dec_symlink_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_symlink_maxsz + \ + decode_getattr_maxsz + \ + decode_getfh_maxsz) +#define NFS4_enc_create_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_create_maxsz + \ + encode_getfh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_create_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_create_maxsz + \ + decode_getfh_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_pathconf_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_pathconf_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_statfs_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_statfs_maxsz) +#define NFS4_dec_statfs_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_statfs_maxsz) +#define NFS4_enc_server_caps_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_server_caps_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_delegreturn_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutreturn_maxsz + \ + encode_delegattr_maxsz + \ + encode_delegreturn_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutreturn_maxsz + \ + decode_delegattr_maxsz + \ + decode_delegreturn_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_getacl_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_getacl_maxsz) +#define NFS4_dec_getacl_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_getacl_maxsz) +#define NFS4_enc_setacl_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_setacl_maxsz) +#define NFS4_dec_setacl_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_setacl_maxsz) +#define NFS4_enc_fs_locations_sz \ + (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_lookup_maxsz + \ + encode_fs_locations_maxsz + \ + encode_renew_maxsz) +#define NFS4_dec_fs_locations_sz \ + (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_lookup_maxsz + \ + decode_fs_locations_maxsz + \ + decode_renew_maxsz) +#define NFS4_enc_secinfo_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_secinfo_maxsz) +#define NFS4_dec_secinfo_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_secinfo_maxsz) +#define NFS4_enc_fsid_present_sz \ + (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_getfh_maxsz + \ + encode_renew_maxsz) +#define NFS4_dec_fsid_present_sz \ + (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_getfh_maxsz + \ + decode_renew_maxsz) +#define NFS4_enc_bind_conn_to_session_sz \ + (compound_encode_hdr_maxsz + \ + encode_bind_conn_to_session_maxsz) +#define NFS4_dec_bind_conn_to_session_sz \ + (compound_decode_hdr_maxsz + \ + decode_bind_conn_to_session_maxsz) +#define NFS4_enc_exchange_id_sz \ + (compound_encode_hdr_maxsz + \ + encode_exchange_id_maxsz) +#define NFS4_dec_exchange_id_sz \ + (compound_decode_hdr_maxsz + \ + decode_exchange_id_maxsz) +#define NFS4_enc_create_session_sz \ + (compound_encode_hdr_maxsz + \ + encode_create_session_maxsz) +#define NFS4_dec_create_session_sz \ + (compound_decode_hdr_maxsz + \ + decode_create_session_maxsz) +#define NFS4_enc_destroy_session_sz (compound_encode_hdr_maxsz + \ + encode_destroy_session_maxsz) +#define NFS4_dec_destroy_session_sz (compound_decode_hdr_maxsz + \ + decode_destroy_session_maxsz) +#define NFS4_enc_destroy_clientid_sz (compound_encode_hdr_maxsz + \ + encode_destroy_clientid_maxsz) +#define NFS4_dec_destroy_clientid_sz (compound_decode_hdr_maxsz + \ + decode_destroy_clientid_maxsz) +#define NFS4_enc_sequence_sz \ + (compound_decode_hdr_maxsz + \ + encode_sequence_maxsz) +#define NFS4_dec_sequence_sz \ + (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz) +#define NFS4_enc_get_lease_time_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putrootfh_maxsz + \ + encode_fsinfo_maxsz) +#define NFS4_dec_get_lease_time_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putrootfh_maxsz + \ + decode_fsinfo_maxsz) +#define NFS4_enc_reclaim_complete_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_reclaim_complete_maxsz) +#define NFS4_dec_reclaim_complete_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_reclaim_complete_maxsz) +#define NFS4_enc_getdeviceinfo_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz +\ + encode_getdeviceinfo_maxsz) +#define NFS4_dec_getdeviceinfo_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_getdeviceinfo_maxsz) +#define NFS4_enc_layoutget_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutget_maxsz) +#define NFS4_dec_layoutget_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutget_maxsz) +#define NFS4_enc_layoutcommit_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz +\ + encode_putfh_maxsz + \ + encode_layoutcommit_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_layoutcommit_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutcommit_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_layoutreturn_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutreturn_maxsz) +#define NFS4_dec_layoutreturn_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutreturn_maxsz) +#define NFS4_enc_secinfo_no_name_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putrootfh_maxsz +\ + encode_secinfo_no_name_maxsz) +#define NFS4_dec_secinfo_no_name_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putrootfh_maxsz + \ + decode_secinfo_no_name_maxsz) +#define NFS4_enc_test_stateid_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_test_stateid_maxsz) +#define NFS4_dec_test_stateid_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_test_stateid_maxsz) +#define NFS4_enc_free_stateid_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_free_stateid_maxsz) +#define NFS4_dec_free_stateid_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_free_stateid_maxsz) + +const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH + + compound_encode_hdr_maxsz + + encode_sequence_maxsz + + encode_putfh_maxsz + + encode_getattr_maxsz) * + XDR_UNIT); + +const u32 nfs41_maxread_overhead = ((RPC_MAX_HEADER_WITH_AUTH + + compound_decode_hdr_maxsz + + decode_sequence_maxsz + + decode_putfh_maxsz) * + XDR_UNIT); + +const u32 nfs41_maxgetdevinfo_overhead = ((RPC_MAX_REPHEADER_WITH_AUTH + + compound_decode_hdr_maxsz + + decode_sequence_maxsz) * + XDR_UNIT); +EXPORT_SYMBOL_GPL(nfs41_maxgetdevinfo_overhead); + +static const umode_t nfs_type2fmt[] = { + [NF4BAD] = 0, + [NF4REG] = S_IFREG, + [NF4DIR] = S_IFDIR, + [NF4BLK] = S_IFBLK, + [NF4CHR] = S_IFCHR, + [NF4LNK] = S_IFLNK, + [NF4SOCK] = S_IFSOCK, + [NF4FIFO] = S_IFIFO, + [NF4ATTRDIR] = 0, + [NF4NAMEDATTR] = 0, +}; + +struct compound_hdr { + int32_t status; + uint32_t nops; + __be32 * nops_p; + uint32_t taglen; + char * tag; + uint32_t replen; /* expected reply words */ + u32 minorversion; +}; + +static __be32 *reserve_space(struct xdr_stream *xdr, size_t nbytes) +{ + __be32 *p = xdr_reserve_space(xdr, nbytes); + BUG_ON(!p); + return p; +} + +static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) +{ + WARN_ON_ONCE(xdr_stream_encode_opaque(xdr, str, len) < 0); +} + +static void encode_uint32(struct xdr_stream *xdr, u32 n) +{ + WARN_ON_ONCE(xdr_stream_encode_u32(xdr, n) < 0); +} + +static void encode_uint64(struct xdr_stream *xdr, u64 n) +{ + WARN_ON_ONCE(xdr_stream_encode_u64(xdr, n) < 0); +} + +static ssize_t xdr_encode_bitmap4(struct xdr_stream *xdr, + const __u32 *bitmap, size_t len) +{ + ssize_t ret; + + /* Trim empty words */ + while (len > 0 && bitmap[len-1] == 0) + len--; + ret = xdr_stream_encode_uint32_array(xdr, bitmap, len); + if (WARN_ON_ONCE(ret < 0)) + return ret; + return len; +} + +static size_t mask_bitmap4(const __u32 *bitmap, const __u32 *mask, + __u32 *res, size_t len) +{ + size_t i; + __u32 tmp; + + while (len > 0 && (bitmap[len-1] == 0 || mask[len-1] == 0)) + len--; + for (i = len; i-- > 0;) { + tmp = bitmap[i] & mask[i]; + res[i] = tmp; + } + return len; +} + +static void encode_nfs4_seqid(struct xdr_stream *xdr, + const struct nfs_seqid *seqid) +{ + if (seqid != NULL) + encode_uint32(xdr, seqid->sequence->counter); + else + encode_uint32(xdr, 0); +} + +static void encode_compound_hdr(struct xdr_stream *xdr, + struct rpc_rqst *req, + struct compound_hdr *hdr) +{ + __be32 *p; + + /* initialize running count of expected bytes in reply. + * NOTE: the replied tag SHOULD be the same is the one sent, + * but this is not required as a MUST for the server to do so. */ + hdr->replen = 3 + hdr->taglen; + + WARN_ON_ONCE(hdr->taglen > NFS4_MAXTAGLEN); + encode_string(xdr, hdr->taglen, hdr->tag); + p = reserve_space(xdr, 8); + *p++ = cpu_to_be32(hdr->minorversion); + hdr->nops_p = p; + *p = cpu_to_be32(hdr->nops); +} + +static void encode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 op, + uint32_t replen, + struct compound_hdr *hdr) +{ + encode_uint32(xdr, op); + hdr->nops++; + hdr->replen += replen; +} + +static void encode_nops(struct compound_hdr *hdr) +{ + WARN_ON_ONCE(hdr->nops > NFS4_MAX_OPS); + *hdr->nops_p = htonl(hdr->nops); +} + +static void encode_nfs4_stateid(struct xdr_stream *xdr, + const nfs4_stateid *stateid) +{ + encode_opaque_fixed(xdr, stateid->data, NFS4_STATEID_SIZE); +} + +static void encode_nfs4_verifier(struct xdr_stream *xdr, const nfs4_verifier *verf) +{ + encode_opaque_fixed(xdr, verf->data, NFS4_VERIFIER_SIZE); +} + +static __be32 * +xdr_encode_nfstime4(__be32 *p, const struct timespec64 *t) +{ + p = xdr_encode_hyper(p, t->tv_sec); + *p++ = cpu_to_be32(t->tv_nsec); + return p; +} + +static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, + const struct nfs4_label *label, + const umode_t *umask, + const struct nfs_server *server, + const uint32_t attrmask[]) +{ + char owner_name[IDMAP_NAMESZ]; + char owner_group[IDMAP_NAMESZ]; + int owner_namelen = 0; + int owner_grouplen = 0; + __be32 *p; + uint32_t len = 0; + uint32_t bmval[3] = { 0 }; + + /* + * We reserve enough space to write the entire attribute buffer at once. + */ + if ((iap->ia_valid & ATTR_SIZE) && (attrmask[0] & FATTR4_WORD0_SIZE)) { + bmval[0] |= FATTR4_WORD0_SIZE; + len += 8; + } + if (iap->ia_valid & ATTR_MODE) { + if (umask && (attrmask[2] & FATTR4_WORD2_MODE_UMASK)) { + bmval[2] |= FATTR4_WORD2_MODE_UMASK; + len += 8; + } else if (attrmask[1] & FATTR4_WORD1_MODE) { + bmval[1] |= FATTR4_WORD1_MODE; + len += 4; + } + } + if ((iap->ia_valid & ATTR_UID) && (attrmask[1] & FATTR4_WORD1_OWNER)) { + owner_namelen = nfs_map_uid_to_name(server, iap->ia_uid, owner_name, IDMAP_NAMESZ); + if (owner_namelen < 0) { + dprintk("nfs: couldn't resolve uid %d to string\n", + from_kuid(&init_user_ns, iap->ia_uid)); + /* XXX */ + strcpy(owner_name, "nobody"); + owner_namelen = sizeof("nobody") - 1; + /* goto out; */ + } + bmval[1] |= FATTR4_WORD1_OWNER; + len += 4 + (XDR_QUADLEN(owner_namelen) << 2); + } + if ((iap->ia_valid & ATTR_GID) && + (attrmask[1] & FATTR4_WORD1_OWNER_GROUP)) { + owner_grouplen = nfs_map_gid_to_group(server, iap->ia_gid, owner_group, IDMAP_NAMESZ); + if (owner_grouplen < 0) { + dprintk("nfs: couldn't resolve gid %d to string\n", + from_kgid(&init_user_ns, iap->ia_gid)); + strcpy(owner_group, "nobody"); + owner_grouplen = sizeof("nobody") - 1; + /* goto out; */ + } + bmval[1] |= FATTR4_WORD1_OWNER_GROUP; + len += 4 + (XDR_QUADLEN(owner_grouplen) << 2); + } + if (attrmask[1] & FATTR4_WORD1_TIME_ACCESS_SET) { + if (iap->ia_valid & ATTR_ATIME_SET) { + bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; + len += 4 + (nfstime4_maxsz << 2); + } else if (iap->ia_valid & ATTR_ATIME) { + bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; + len += 4; + } + } + if (attrmask[1] & FATTR4_WORD1_TIME_MODIFY_SET) { + if (iap->ia_valid & ATTR_MTIME_SET) { + bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; + len += 4 + (nfstime4_maxsz << 2); + } else if (iap->ia_valid & ATTR_MTIME) { + bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; + len += 4; + } + } + + if (label && (attrmask[2] & FATTR4_WORD2_SECURITY_LABEL)) { + len += 4 + 4 + 4 + (XDR_QUADLEN(label->len) << 2); + bmval[2] |= FATTR4_WORD2_SECURITY_LABEL; + } + + xdr_encode_bitmap4(xdr, bmval, ARRAY_SIZE(bmval)); + xdr_stream_encode_opaque_inline(xdr, (void **)&p, len); + + if (bmval[0] & FATTR4_WORD0_SIZE) + p = xdr_encode_hyper(p, iap->ia_size); + if (bmval[1] & FATTR4_WORD1_MODE) + *p++ = cpu_to_be32(iap->ia_mode & S_IALLUGO); + if (bmval[1] & FATTR4_WORD1_OWNER) + p = xdr_encode_opaque(p, owner_name, owner_namelen); + if (bmval[1] & FATTR4_WORD1_OWNER_GROUP) + p = xdr_encode_opaque(p, owner_group, owner_grouplen); + if (bmval[1] & FATTR4_WORD1_TIME_ACCESS_SET) { + if (iap->ia_valid & ATTR_ATIME_SET) { + *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); + p = xdr_encode_nfstime4(p, &iap->ia_atime); + } else + *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); + } + if (bmval[1] & FATTR4_WORD1_TIME_MODIFY_SET) { + if (iap->ia_valid & ATTR_MTIME_SET) { + *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); + p = xdr_encode_nfstime4(p, &iap->ia_mtime); + } else + *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); + } + if (label && (bmval[2] & FATTR4_WORD2_SECURITY_LABEL)) { + *p++ = cpu_to_be32(label->lfs); + *p++ = cpu_to_be32(label->pi); + *p++ = cpu_to_be32(label->len); + p = xdr_encode_opaque_fixed(p, label->label, label->len); + } + if (bmval[2] & FATTR4_WORD2_MODE_UMASK) { + *p++ = cpu_to_be32(iap->ia_mode & S_IALLUGO); + *p++ = cpu_to_be32(*umask); + } + +/* out: */ +} + +static void encode_access(struct xdr_stream *xdr, u32 access, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_ACCESS, decode_access_maxsz, hdr); + encode_uint32(xdr, access); +} + +static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_CLOSE, decode_close_maxsz, hdr); + encode_nfs4_seqid(xdr, arg->seqid); + encode_nfs4_stateid(xdr, &arg->stateid); +} + +static void encode_commit(struct xdr_stream *xdr, const struct nfs_commitargs *args, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_COMMIT, decode_commit_maxsz, hdr); + p = reserve_space(xdr, 12); + p = xdr_encode_hyper(p, args->offset); + *p = cpu_to_be32(args->count); +} + +static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_CREATE, decode_create_maxsz, hdr); + encode_uint32(xdr, create->ftype); + + switch (create->ftype) { + case NF4LNK: + p = reserve_space(xdr, 4); + *p = cpu_to_be32(create->u.symlink.len); + xdr_write_pages(xdr, create->u.symlink.pages, 0, + create->u.symlink.len); + xdr->buf->flags |= XDRBUF_WRITE; + break; + + case NF4BLK: case NF4CHR: + p = reserve_space(xdr, 8); + *p++ = cpu_to_be32(create->u.device.specdata1); + *p = cpu_to_be32(create->u.device.specdata2); + break; + + default: + break; + } + + encode_string(xdr, create->name->len, create->name->name); + encode_attrs(xdr, create->attrs, create->label, &create->umask, + create->server, create->server->attr_bitmask); +} + +static void encode_getattr(struct xdr_stream *xdr, + const __u32 *bitmap, const __u32 *mask, size_t len, + struct compound_hdr *hdr) +{ + __u32 masked_bitmap[nfs4_fattr_bitmap_maxsz]; + + encode_op_hdr(xdr, OP_GETATTR, decode_getattr_maxsz, hdr); + if (mask) { + if (WARN_ON_ONCE(len > ARRAY_SIZE(masked_bitmap))) + len = ARRAY_SIZE(masked_bitmap); + len = mask_bitmap4(bitmap, mask, masked_bitmap, len); + bitmap = masked_bitmap; + } + xdr_encode_bitmap4(xdr, bitmap, len); +} + +static void encode_getfattr(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) +{ + encode_getattr(xdr, nfs4_fattr_bitmap, bitmask, + ARRAY_SIZE(nfs4_fattr_bitmap), hdr); +} + +static void encode_getfattr_open(struct xdr_stream *xdr, const u32 *bitmask, + const u32 *open_bitmap, + struct compound_hdr *hdr) +{ + encode_getattr(xdr, open_bitmap, bitmask, 3, hdr); +} + +static void encode_fsinfo(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) +{ + encode_getattr(xdr, nfs4_fsinfo_bitmap, bitmask, + ARRAY_SIZE(nfs4_fsinfo_bitmap), hdr); +} + +static void encode_fs_locations(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) +{ + encode_getattr(xdr, nfs4_fs_locations_bitmap, bitmask, + ARRAY_SIZE(nfs4_fs_locations_bitmap), hdr); +} + +static void encode_getfh(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_GETFH, decode_getfh_maxsz, hdr); +} + +static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_LINK, decode_link_maxsz, hdr); + encode_string(xdr, name->len, name->name); +} + +static inline int nfs4_lock_type(struct file_lock *fl, int block) +{ + if (lock_is_read(fl)) + return block ? NFS4_READW_LT : NFS4_READ_LT; + return block ? NFS4_WRITEW_LT : NFS4_WRITE_LT; +} + +static inline uint64_t nfs4_lock_length(struct file_lock *fl) +{ + if (fl->fl_end == OFFSET_MAX) + return ~(uint64_t)0; + return fl->fl_end - fl->fl_start + 1; +} + +static void encode_lockowner(struct xdr_stream *xdr, const struct nfs_lowner *lowner) +{ + __be32 *p; + + p = reserve_space(xdr, 32); + p = xdr_encode_hyper(p, lowner->clientid); + *p++ = cpu_to_be32(20); + p = xdr_encode_opaque_fixed(p, "lock id:", 8); + *p++ = cpu_to_be32(lowner->s_dev); + xdr_encode_hyper(p, lowner->id); +} + +/* + * opcode,type,reclaim,offset,length,new_lock_owner = 32 + * open_seqid,open_stateid,lock_seqid,lock_owner.clientid, lock_owner.id = 40 + */ +static void encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LOCK, decode_lock_maxsz, hdr); + p = reserve_space(xdr, 28); + *p++ = cpu_to_be32(nfs4_lock_type(args->fl, args->block)); + *p++ = cpu_to_be32(args->reclaim); + p = xdr_encode_hyper(p, args->fl->fl_start); + p = xdr_encode_hyper(p, nfs4_lock_length(args->fl)); + *p = cpu_to_be32(args->new_lock_owner); + if (args->new_lock_owner){ + encode_nfs4_seqid(xdr, args->open_seqid); + encode_nfs4_stateid(xdr, &args->open_stateid); + encode_nfs4_seqid(xdr, args->lock_seqid); + encode_lockowner(xdr, &args->lock_owner); + } + else { + encode_nfs4_stateid(xdr, &args->lock_stateid); + encode_nfs4_seqid(xdr, args->lock_seqid); + } +} + +static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LOCKT, decode_lockt_maxsz, hdr); + p = reserve_space(xdr, 20); + *p++ = cpu_to_be32(nfs4_lock_type(args->fl, 0)); + p = xdr_encode_hyper(p, args->fl->fl_start); + p = xdr_encode_hyper(p, nfs4_lock_length(args->fl)); + encode_lockowner(xdr, &args->lock_owner); +} + +static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LOCKU, decode_locku_maxsz, hdr); + encode_uint32(xdr, nfs4_lock_type(args->fl, 0)); + encode_nfs4_seqid(xdr, args->seqid); + encode_nfs4_stateid(xdr, &args->stateid); + p = reserve_space(xdr, 16); + p = xdr_encode_hyper(p, args->fl->fl_start); + xdr_encode_hyper(p, nfs4_lock_length(args->fl)); +} + +#if defined(CONFIG_NFS_V4_0) +static void encode_release_lockowner(struct xdr_stream *xdr, const struct nfs_lowner *lowner, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_RELEASE_LOCKOWNER, decode_release_lockowner_maxsz, hdr); + encode_lockowner(xdr, lowner); +} +#endif /* CONFIG_NFS_V4_0 */ + +static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_LOOKUP, decode_lookup_maxsz, hdr); + encode_string(xdr, name->len, name->name); +} + +static void encode_lookupp(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_LOOKUPP, decode_lookupp_maxsz, hdr); +} + +static void encode_share_access(struct xdr_stream *xdr, u32 share_access) +{ + __be32 *p; + + p = reserve_space(xdr, 8); + *p++ = cpu_to_be32(share_access); + *p = cpu_to_be32(0); /* for linux, share_deny = 0 always */ +} + +static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_openargs *arg) +{ + __be32 *p; + /* + * opcode 4, seqid 4, share_access 4, share_deny 4, clientid 8, ownerlen 4, + * owner 28 + */ + encode_nfs4_seqid(xdr, arg->seqid); + encode_share_access(xdr, arg->share_access); + p = reserve_space(xdr, 40); + p = xdr_encode_hyper(p, arg->clientid); + *p++ = cpu_to_be32(28); + p = xdr_encode_opaque_fixed(p, "open id:", 8); + *p++ = cpu_to_be32(arg->server->s_dev); + p = xdr_encode_hyper(p, arg->id.uniquifier); + xdr_encode_hyper(p, arg->id.create_time); +} + +static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_openargs *arg) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + switch(arg->createmode) { + case NFS4_CREATE_UNCHECKED: + *p = cpu_to_be32(NFS4_CREATE_UNCHECKED); + encode_attrs(xdr, arg->u.attrs, arg->label, &arg->umask, + arg->server, arg->server->attr_bitmask); + break; + case NFS4_CREATE_GUARDED: + *p = cpu_to_be32(NFS4_CREATE_GUARDED); + encode_attrs(xdr, arg->u.attrs, arg->label, &arg->umask, + arg->server, arg->server->attr_bitmask); + break; + case NFS4_CREATE_EXCLUSIVE: + *p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE); + encode_nfs4_verifier(xdr, &arg->u.verifier); + break; + case NFS4_CREATE_EXCLUSIVE4_1: + *p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE4_1); + encode_nfs4_verifier(xdr, &arg->u.verifier); + encode_attrs(xdr, arg->u.attrs, arg->label, &arg->umask, + arg->server, arg->server->exclcreat_bitmask); + } +} + +static void encode_opentype(struct xdr_stream *xdr, const struct nfs_openargs *arg) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + switch (arg->open_flags & O_CREAT) { + case 0: + *p = cpu_to_be32(NFS4_OPEN_NOCREATE); + break; + default: + *p = cpu_to_be32(NFS4_OPEN_CREATE); + encode_createmode(xdr, arg); + } +} + +static inline void encode_delegation_type(struct xdr_stream *xdr, u32 delegation_type) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + switch (delegation_type) { + case NFS4_OPEN_DELEGATE_NONE: + case NFS4_OPEN_DELEGATE_READ: + case NFS4_OPEN_DELEGATE_WRITE: + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + *p = cpu_to_be32(delegation_type); + break; + default: + BUG(); + } +} + +static inline void encode_claim_null(struct xdr_stream *xdr, const struct qstr *name) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + *p = cpu_to_be32(NFS4_OPEN_CLAIM_NULL); + encode_string(xdr, name->len, name->name); +} + +static inline void encode_claim_previous(struct xdr_stream *xdr, u32 type) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + *p = cpu_to_be32(NFS4_OPEN_CLAIM_PREVIOUS); + encode_delegation_type(xdr, type); +} + +static inline void encode_claim_delegate_cur(struct xdr_stream *xdr, const struct qstr *name, const nfs4_stateid *stateid) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + *p = cpu_to_be32(NFS4_OPEN_CLAIM_DELEGATE_CUR); + encode_nfs4_stateid(xdr, stateid); + encode_string(xdr, name->len, name->name); +} + +static inline void encode_claim_fh(struct xdr_stream *xdr) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + *p = cpu_to_be32(NFS4_OPEN_CLAIM_FH); +} + +static inline void encode_claim_delegate_cur_fh(struct xdr_stream *xdr, const nfs4_stateid *stateid) +{ + __be32 *p; + + p = reserve_space(xdr, 4); + *p = cpu_to_be32(NFS4_OPEN_CLAIM_DELEG_CUR_FH); + encode_nfs4_stateid(xdr, stateid); +} + +static void encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_OPEN, decode_open_maxsz, hdr); + encode_openhdr(xdr, arg); + encode_opentype(xdr, arg); + switch (arg->claim) { + case NFS4_OPEN_CLAIM_NULL: + encode_claim_null(xdr, arg->name); + break; + case NFS4_OPEN_CLAIM_PREVIOUS: + encode_claim_previous(xdr, arg->u.delegation_type); + break; + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + encode_claim_delegate_cur(xdr, arg->name, &arg->u.delegation); + break; + case NFS4_OPEN_CLAIM_FH: + encode_claim_fh(xdr); + break; + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: + encode_claim_delegate_cur_fh(xdr, &arg->u.delegation); + break; + default: + BUG(); + } +} + +static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_OPEN_CONFIRM, decode_open_confirm_maxsz, hdr); + encode_nfs4_stateid(xdr, arg->stateid); + encode_nfs4_seqid(xdr, arg->seqid); +} + +static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_OPEN_DOWNGRADE, decode_open_downgrade_maxsz, hdr); + encode_nfs4_stateid(xdr, &arg->stateid); + encode_nfs4_seqid(xdr, arg->seqid); + encode_share_access(xdr, arg->share_access); +} + +static void +encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_PUTFH, decode_putfh_maxsz, hdr); + encode_string(xdr, fh->size, fh->data); +} + +static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_PUTROOTFH, decode_putrootfh_maxsz, hdr); +} + +static void encode_read(struct xdr_stream *xdr, const struct nfs_pgio_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_READ, decode_read_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->stateid); + + p = reserve_space(xdr, 12); + p = xdr_encode_hyper(p, args->offset); + *p = cpu_to_be32(args->count); +} + +static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr) +{ + uint32_t attrs[3] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_RDATTR_ERROR, + FATTR4_WORD1_MOUNTED_ON_FILEID, + }; + uint32_t dircount = readdir->count; + uint32_t maxcount = readdir->count; + __be32 *p, verf[2]; + uint32_t attrlen = 0; + unsigned int i; + + if (readdir->plus) { + attrs[0] |= FATTR4_WORD0_CHANGE + | FATTR4_WORD0_SIZE + | FATTR4_WORD0_FSID + | FATTR4_WORD0_FILEHANDLE + | FATTR4_WORD0_FILEID; + attrs[1] |= FATTR4_WORD1_MODE + | FATTR4_WORD1_NUMLINKS + | FATTR4_WORD1_OWNER + | FATTR4_WORD1_OWNER_GROUP + | FATTR4_WORD1_RAWDEV + | FATTR4_WORD1_SPACE_USED + | FATTR4_WORD1_TIME_ACCESS + | FATTR4_WORD1_TIME_CREATE + | FATTR4_WORD1_TIME_METADATA + | FATTR4_WORD1_TIME_MODIFY; + attrs[2] |= FATTR4_WORD2_SECURITY_LABEL; + } + /* Use mounted_on_fileid only if the server supports it */ + if (!(readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)) + attrs[0] |= FATTR4_WORD0_FILEID; + for (i = 0; i < ARRAY_SIZE(attrs); i++) { + attrs[i] &= readdir->bitmask[i]; + if (attrs[i] != 0) + attrlen = i+1; + } + + encode_op_hdr(xdr, OP_READDIR, decode_readdir_maxsz, hdr); + encode_uint64(xdr, readdir->cookie); + encode_nfs4_verifier(xdr, &readdir->verifier); + p = reserve_space(xdr, 12 + (attrlen << 2)); + *p++ = cpu_to_be32(dircount); + *p++ = cpu_to_be32(maxcount); + *p++ = cpu_to_be32(attrlen); + for (i = 0; i < attrlen; i++) + *p++ = cpu_to_be32(attrs[i]); + memcpy(verf, readdir->verifier.data, sizeof(verf)); + + dprintk("%s: cookie = %llu, verifier = %08x:%08x, bitmap = %08x:%08x:%08x\n", + __func__, + (unsigned long long)readdir->cookie, + verf[0], verf[1], + attrs[0] & readdir->bitmask[0], + attrs[1] & readdir->bitmask[1], + attrs[2] & readdir->bitmask[2]); +} + +static void encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *readlink, struct rpc_rqst *req, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_READLINK, decode_readlink_maxsz, hdr); +} + +static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_REMOVE, decode_remove_maxsz, hdr); + encode_string(xdr, name->len, name->name); +} + +static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_RENAME, decode_rename_maxsz, hdr); + encode_string(xdr, oldname->len, oldname->name); + encode_string(xdr, newname->len, newname->name); +} + +static void encode_renew(struct xdr_stream *xdr, clientid4 clid, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_RENEW, decode_renew_maxsz, hdr); + encode_uint64(xdr, clid); +} + +static void +encode_restorefh(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_RESTOREFH, decode_restorefh_maxsz, hdr); +} + +static void nfs4_acltype_to_bitmap(enum nfs4_acl_type type, __u32 bitmap[2]) +{ + switch (type) { + default: + bitmap[0] = FATTR4_WORD0_ACL; + bitmap[1] = 0; + break; + case NFS4ACL_DACL: + bitmap[0] = 0; + bitmap[1] = FATTR4_WORD1_DACL; + break; + case NFS4ACL_SACL: + bitmap[0] = 0; + bitmap[1] = FATTR4_WORD1_SACL; + } +} + +static void encode_setacl(struct xdr_stream *xdr, + const struct nfs_setaclargs *arg, + struct compound_hdr *hdr) +{ + __u32 bitmap[2]; + + nfs4_acltype_to_bitmap(arg->acl_type, bitmap); + + encode_op_hdr(xdr, OP_SETATTR, decode_setacl_maxsz, hdr); + encode_nfs4_stateid(xdr, &zero_stateid); + xdr_encode_bitmap4(xdr, bitmap, ARRAY_SIZE(bitmap)); + encode_uint32(xdr, arg->acl_len); + xdr_write_pages(xdr, arg->acl_pages, 0, arg->acl_len); +} + +static void +encode_savefh(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SAVEFH, decode_savefh_maxsz, hdr); +} + +static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SETATTR, decode_setattr_maxsz, hdr); + encode_nfs4_stateid(xdr, &arg->stateid); + encode_attrs(xdr, arg->iap, arg->label, NULL, server, + server->attr_bitmask); +} + +static void encode_delegattr(struct xdr_stream *xdr, + const nfs4_stateid *stateid, + const struct nfs4_delegattr *attr, + struct compound_hdr *hdr) +{ + uint32_t bitmap[3] = { 0 }; + uint32_t len = 0; + __be32 *p; + + encode_op_hdr(xdr, OP_SETATTR, encode_delegattr_maxsz, hdr); + encode_nfs4_stateid(xdr, stateid); + if (attr->atime_set) { + bitmap[2] |= FATTR4_WORD2_TIME_DELEG_ACCESS; + len += (nfstime4_maxsz << 2); + } + if (attr->mtime_set) { + bitmap[2] |= FATTR4_WORD2_TIME_DELEG_MODIFY; + len += (nfstime4_maxsz << 2); + } + xdr_encode_bitmap4(xdr, bitmap, ARRAY_SIZE(bitmap)); + xdr_stream_encode_opaque_inline(xdr, (void **)&p, len); + if (bitmap[2] & FATTR4_WORD2_TIME_DELEG_ACCESS) + p = xdr_encode_nfstime4(p, &attr->atime); + if (bitmap[2] & FATTR4_WORD2_TIME_DELEG_MODIFY) + p = xdr_encode_nfstime4(p, &attr->mtime); +} + +static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclientid *setclientid, struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_SETCLIENTID, decode_setclientid_maxsz, hdr); + encode_nfs4_verifier(xdr, setclientid->sc_verifier); + + encode_string(xdr, strlen(setclientid->sc_clnt->cl_owner_id), + setclientid->sc_clnt->cl_owner_id); + p = reserve_space(xdr, 4); + *p = cpu_to_be32(setclientid->sc_prog); + encode_string(xdr, setclientid->sc_netid_len, setclientid->sc_netid); + encode_string(xdr, setclientid->sc_uaddr_len, setclientid->sc_uaddr); + p = reserve_space(xdr, 4); + *p = cpu_to_be32(setclientid->sc_clnt->cl_cb_ident); +} + +static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_setclientid_res *arg, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SETCLIENTID_CONFIRM, + decode_setclientid_confirm_maxsz, hdr); + encode_uint64(xdr, arg->clientid); + encode_nfs4_verifier(xdr, &arg->confirm); +} + +static void encode_write(struct xdr_stream *xdr, const struct nfs_pgio_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_WRITE, decode_write_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->stateid); + + p = reserve_space(xdr, 16); + p = xdr_encode_hyper(p, args->offset); + *p++ = cpu_to_be32(args->stable); + *p = cpu_to_be32(args->count); + + xdr_write_pages(xdr, args->pages, args->pgbase, args->count); +} + +static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_DELEGRETURN, decode_delegreturn_maxsz, hdr); + encode_nfs4_stateid(xdr, stateid); +} + +static void encode_secinfo(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SECINFO, decode_secinfo_maxsz, hdr); + encode_string(xdr, name->len, name->name); +} + +/* NFSv4.1 operations */ +static void encode_bind_conn_to_session(struct xdr_stream *xdr, + const struct nfs41_bind_conn_to_session_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_BIND_CONN_TO_SESSION, + decode_bind_conn_to_session_maxsz, hdr); + encode_opaque_fixed(xdr, args->sessionid.data, NFS4_MAX_SESSIONID_LEN); + p = xdr_reserve_space(xdr, 8); + *p++ = cpu_to_be32(args->dir); + *p = (args->use_conn_in_rdma_mode) ? cpu_to_be32(1) : cpu_to_be32(0); +} + +static void encode_op_map(struct xdr_stream *xdr, const struct nfs4_op_map *op_map) +{ + unsigned int i; + encode_uint32(xdr, NFS4_OP_MAP_NUM_WORDS); + for (i = 0; i < NFS4_OP_MAP_NUM_WORDS; i++) + encode_uint32(xdr, op_map->u.words[i]); +} + +static void encode_exchange_id(struct xdr_stream *xdr, + const struct nfs41_exchange_id_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + char impl_name[IMPL_NAME_LIMIT]; + int len = 0; + + encode_op_hdr(xdr, OP_EXCHANGE_ID, decode_exchange_id_maxsz, hdr); + encode_nfs4_verifier(xdr, &args->verifier); + + encode_string(xdr, strlen(args->client->cl_owner_id), + args->client->cl_owner_id); + + encode_uint32(xdr, args->flags); + encode_uint32(xdr, args->state_protect.how); + + switch (args->state_protect.how) { + case SP4_NONE: + break; + case SP4_MACH_CRED: + encode_op_map(xdr, &args->state_protect.enforce); + encode_op_map(xdr, &args->state_protect.allow); + break; + default: + WARN_ON_ONCE(1); + break; + } + + if (send_implementation_id && + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) > 1 && + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) + <= sizeof(impl_name) + 1) + len = snprintf(impl_name, sizeof(impl_name), "%s %s %s %s", + utsname()->sysname, utsname()->release, + utsname()->version, utsname()->machine); + + if (len > 0) { + encode_uint32(xdr, 1); /* implementation id array length=1 */ + + encode_string(xdr, + sizeof(CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN) - 1, + CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN); + encode_string(xdr, len, impl_name); + /* just send zeros for nii_date - the date is in nii_name */ + p = reserve_space(xdr, 12); + p = xdr_encode_hyper(p, 0); + *p = cpu_to_be32(0); + } else + encode_uint32(xdr, 0); /* implementation id array length=0 */ +} + +static void encode_create_session(struct xdr_stream *xdr, + const struct nfs41_create_session_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + struct nfs_client *clp = args->client; + struct rpc_clnt *clnt = clp->cl_rpcclient; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + u32 max_resp_sz_cached; + + /* + * Assumes OPEN is the biggest non-idempotent compound. + * 2 is the verifier. + */ + max_resp_sz_cached = (NFS4_dec_open_sz + RPC_REPHDRSIZE + 2) + * XDR_UNIT + RPC_MAX_AUTH_SIZE; + + encode_op_hdr(xdr, OP_CREATE_SESSION, decode_create_session_maxsz, hdr); + p = reserve_space(xdr, 16 + 2*28 + 20 + clnt->cl_nodelen + 12); + p = xdr_encode_hyper(p, args->clientid); + *p++ = cpu_to_be32(args->seqid); /*Sequence id */ + *p++ = cpu_to_be32(args->flags); /*flags */ + + /* Fore Channel */ + *p++ = cpu_to_be32(0); /* header padding size */ + *p++ = cpu_to_be32(args->fc_attrs.max_rqst_sz); /* max req size */ + *p++ = cpu_to_be32(args->fc_attrs.max_resp_sz); /* max resp size */ + *p++ = cpu_to_be32(max_resp_sz_cached); /* Max resp sz cached */ + *p++ = cpu_to_be32(args->fc_attrs.max_ops); /* max operations */ + *p++ = cpu_to_be32(args->fc_attrs.max_reqs); /* max requests */ + *p++ = cpu_to_be32(0); /* rdmachannel_attrs */ + + /* Back Channel */ + *p++ = cpu_to_be32(0); /* header padding size */ + *p++ = cpu_to_be32(args->bc_attrs.max_rqst_sz); /* max req size */ + *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz); /* max resp size */ + *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz_cached); /* Max resp sz cached */ + *p++ = cpu_to_be32(args->bc_attrs.max_ops); /* max operations */ + *p++ = cpu_to_be32(args->bc_attrs.max_reqs); /* max requests */ + *p++ = cpu_to_be32(0); /* rdmachannel_attrs */ + + *p++ = cpu_to_be32(args->cb_program); /* cb_program */ + *p++ = cpu_to_be32(1); + *p++ = cpu_to_be32(RPC_AUTH_UNIX); /* auth_sys */ + + /* authsys_parms rfc1831 */ + *p++ = cpu_to_be32(ktime_to_ns(nn->boot_time)); /* stamp */ + p = xdr_encode_array(p, clnt->cl_nodename, clnt->cl_nodelen); + *p++ = cpu_to_be32(0); /* UID */ + *p++ = cpu_to_be32(0); /* GID */ + *p = cpu_to_be32(0); /* No more gids */ +} + +static void encode_destroy_session(struct xdr_stream *xdr, + const struct nfs4_session *session, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_DESTROY_SESSION, decode_destroy_session_maxsz, hdr); + encode_opaque_fixed(xdr, session->sess_id.data, NFS4_MAX_SESSIONID_LEN); +} + +static void encode_destroy_clientid(struct xdr_stream *xdr, + uint64_t clientid, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_DESTROY_CLIENTID, decode_destroy_clientid_maxsz, hdr); + encode_uint64(xdr, clientid); +} + +static void encode_reclaim_complete(struct xdr_stream *xdr, + const struct nfs41_reclaim_complete_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_RECLAIM_COMPLETE, decode_reclaim_complete_maxsz, hdr); + encode_uint32(xdr, args->one_fs); +} + +static void encode_sequence(struct xdr_stream *xdr, + const struct nfs4_sequence_args *args, + struct compound_hdr *hdr) +{ + struct nfs4_session *session; + struct nfs4_slot_table *tp; + struct nfs4_slot *slot = args->sa_slot; + __be32 *p; + + tp = slot->table; + session = tp->session; + if (!session) + return; + + encode_op_hdr(xdr, OP_SEQUENCE, decode_sequence_maxsz, hdr); + + /* + * Sessionid + seqid + slotid + max slotid + cache_this + */ + dprintk("%s: sessionid=%u:%u:%u:%u seqid=%d slotid=%d " + "max_slotid=%d cache_this=%d\n", + __func__, + ((u32 *)session->sess_id.data)[0], + ((u32 *)session->sess_id.data)[1], + ((u32 *)session->sess_id.data)[2], + ((u32 *)session->sess_id.data)[3], + slot->seq_nr, slot->slot_nr, + tp->highest_used_slotid, args->sa_cache_this); + p = reserve_space(xdr, NFS4_MAX_SESSIONID_LEN + 16); + p = xdr_encode_opaque_fixed(p, session->sess_id.data, NFS4_MAX_SESSIONID_LEN); + *p++ = cpu_to_be32(slot->seq_nr); + *p++ = cpu_to_be32(slot->slot_nr); + *p++ = cpu_to_be32(tp->highest_used_slotid); + *p = cpu_to_be32(args->sa_cache_this); +} + +static void +encode_get_dir_delegation(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + struct timespec64 ts = { 0, 0 }; + u32 notifications[1] = { 0 }; + u32 attributes[1] = { 0 }; + __be32 *p; + + encode_op_hdr(xdr, OP_GET_DIR_DELEGATION, decode_get_dir_deleg_maxsz, hdr); + + /* We don't handle CB_RECALLABLE_OBJ_AVAIL yet. */ + xdr_stream_encode_bool(xdr, false); + + xdr_encode_bitmap4(xdr, notifications, ARRAY_SIZE(notifications)); + + /* Request no delay on attribute updates */ + p = reserve_space(xdr, 12 + 12); + p = xdr_encode_nfstime4(p, &ts); + xdr_encode_nfstime4(p, &ts); + + /* Requested child attributes */ + xdr_encode_bitmap4(xdr, attributes, ARRAY_SIZE(attributes)); + + /* Requested dir attributes */ + xdr_encode_bitmap4(xdr, attributes, ARRAY_SIZE(attributes)); +} + +static void +encode_getdeviceinfo(struct xdr_stream *xdr, + const struct nfs4_getdeviceinfo_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_GETDEVICEINFO, decode_getdeviceinfo_maxsz, hdr); + p = reserve_space(xdr, NFS4_DEVICEID4_SIZE + 4 + 4); + p = xdr_encode_opaque_fixed(p, args->pdev->dev_id.data, + NFS4_DEVICEID4_SIZE); + *p++ = cpu_to_be32(args->pdev->layout_type); + *p++ = cpu_to_be32(args->pdev->maxcount); /* gdia_maxcount */ + + p = reserve_space(xdr, 4 + 4); + *p++ = cpu_to_be32(1); /* bitmap length */ + *p++ = cpu_to_be32(args->notify_types); +} + +static void +encode_layoutget(struct xdr_stream *xdr, + const struct nfs4_layoutget_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LAYOUTGET, decode_layoutget_maxsz, hdr); + p = reserve_space(xdr, 36); + *p++ = cpu_to_be32(0); /* Signal layout available */ + *p++ = cpu_to_be32(args->type); + *p++ = cpu_to_be32(args->range.iomode); + p = xdr_encode_hyper(p, args->range.offset); + p = xdr_encode_hyper(p, args->range.length); + p = xdr_encode_hyper(p, args->minlength); + encode_nfs4_stateid(xdr, &args->stateid); + encode_uint32(xdr, args->maxcount); + + dprintk("%s: 1st type:0x%x iomode:%d off:%lu len:%lu mc:%d\n", + __func__, + args->type, + args->range.iomode, + (unsigned long)args->range.offset, + (unsigned long)args->range.length, + args->maxcount); +} + +static int +encode_layoutcommit(struct xdr_stream *xdr, + struct inode *inode, + const struct nfs4_layoutcommit_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + dprintk("%s: lbw: %llu type: %d\n", __func__, args->lastbytewritten, + NFS_SERVER(args->inode)->pnfs_curr_ld->id); + + encode_op_hdr(xdr, OP_LAYOUTCOMMIT, decode_layoutcommit_maxsz, hdr); + p = reserve_space(xdr, 20); + /* Only whole file layouts */ + p = xdr_encode_hyper(p, 0); /* offset */ + p = xdr_encode_hyper(p, args->lastbytewritten + 1); /* length */ + *p = cpu_to_be32(0); /* reclaim */ + encode_nfs4_stateid(xdr, &args->stateid); + if (args->lastbytewritten != U64_MAX) { + p = reserve_space(xdr, 20); + *p++ = cpu_to_be32(1); /* newoffset = TRUE */ + p = xdr_encode_hyper(p, args->lastbytewritten); + } else { + p = reserve_space(xdr, 12); + *p++ = cpu_to_be32(0); /* newoffset = FALSE */ + } + *p++ = cpu_to_be32(0); /* Never send time_modify_changed */ + *p++ = cpu_to_be32(NFS_SERVER(args->inode)->pnfs_curr_ld->id);/* type */ + + encode_uint32(xdr, args->layoutupdate_len); + if (args->layoutupdate_pages) + xdr_write_pages(xdr, args->layoutupdate_pages, 0, + args->layoutupdate_len); + + return 0; +} + +static void +encode_layoutreturn(struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + encode_op_hdr(xdr, OP_LAYOUTRETURN, decode_layoutreturn_maxsz, hdr); + p = reserve_space(xdr, 16); + *p++ = cpu_to_be32(0); /* reclaim. always 0 for now */ + *p++ = cpu_to_be32(args->layout_type); + *p++ = cpu_to_be32(args->range.iomode); + *p = cpu_to_be32(RETURN_FILE); + p = reserve_space(xdr, 16); + p = xdr_encode_hyper(p, args->range.offset); + p = xdr_encode_hyper(p, args->range.length); + spin_lock(&args->inode->i_lock); + encode_nfs4_stateid(xdr, &args->stateid); + spin_unlock(&args->inode->i_lock); + if (args->ld_private->ops && args->ld_private->ops->encode) + args->ld_private->ops->encode(xdr, args, args->ld_private); + else + encode_uint32(xdr, 0); +} + +static int +encode_secinfo_no_name(struct xdr_stream *xdr, + const struct nfs41_secinfo_no_name_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_SECINFO_NO_NAME, decode_secinfo_no_name_maxsz, hdr); + encode_uint32(xdr, args->style); + return 0; +} + +static void encode_test_stateid(struct xdr_stream *xdr, + const struct nfs41_test_stateid_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_TEST_STATEID, decode_test_stateid_maxsz, hdr); + encode_uint32(xdr, 1); + encode_nfs4_stateid(xdr, &args->stateid); +} + +static void encode_free_stateid(struct xdr_stream *xdr, + const struct nfs41_free_stateid_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_FREE_STATEID, decode_free_stateid_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->stateid); +} + +/* + * END OF "GENERIC" ENCODE ROUTINES. + */ + +static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args) +{ + struct nfs4_session *session = args->sa_slot->table->session; + if (session) + return session->clp->cl_mvops->minor_version; + return 0; +} + +/* + * Encode an ACCESS request + */ +static void nfs4_xdr_enc_access(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_accessargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_access(xdr, args->access, &hdr); + if (args->bitmask) + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LOOKUP request + */ +static void nfs4_xdr_enc_lookup(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_lookup_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->dir_fh, &hdr); + encode_lookup(xdr, args->name, &hdr); + encode_getfh(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LOOKUPP request + */ +static void nfs4_xdr_enc_lookupp(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_lookupp_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_lookupp(xdr, &hdr); + encode_getfh(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LOOKUP_ROOT request + */ +static void nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_lookup_root_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putrootfh(xdr, &hdr); + encode_getfh(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode REMOVE request + */ +static void nfs4_xdr_enc_remove(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_removeargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_remove(xdr, &args->name, &hdr); + encode_nops(&hdr); +} + +/* + * Encode RENAME request + */ +static void nfs4_xdr_enc_rename(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_renameargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->old_dir, &hdr); + encode_savefh(xdr, &hdr); + encode_putfh(xdr, args->new_dir, &hdr); + encode_rename(xdr, args->old_name, args->new_name, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LINK request + */ +static void nfs4_xdr_enc_link(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_link_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_savefh(xdr, &hdr); + encode_putfh(xdr, args->dir_fh, &hdr); + encode_link(xdr, args->name, &hdr); + encode_restorefh(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode CREATE request + */ +static void nfs4_xdr_enc_create(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_create_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->dir_fh, &hdr); + encode_create(xdr, args, &hdr); + encode_getfh(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode SYMLINK request + */ +static void nfs4_xdr_enc_symlink(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_create_arg *args = data; + + nfs4_xdr_enc_create(req, xdr, args); +} + +/* + * Encode GETATTR request + */ +static void nfs4_xdr_enc_getattr(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_getattr_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + if (args->get_dir_deleg) + encode_get_dir_delegation(xdr, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode a CLOSE request + */ +static void nfs4_xdr_enc_close(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_closeargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + if (args->lr_args) + encode_layoutreturn(xdr, args->lr_args, &hdr); + if (args->bitmask != NULL) + encode_getfattr(xdr, args->bitmask, &hdr); + encode_close(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode an OPEN request + */ +static void nfs4_xdr_enc_open(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_openargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_open(xdr, args, &hdr); + encode_getfh(xdr, &hdr); + if (args->access) + encode_access(xdr, args->access, &hdr); + encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); + if (args->lg_args) { + encode_layoutget(xdr, args->lg_args, &hdr); + rpc_prepare_reply_pages(req, args->lg_args->layout.pages, 0, + args->lg_args->layout.pglen, + hdr.replen - pagepad_maxsz); + } + encode_nops(&hdr); +} + +/* + * Encode an OPEN_CONFIRM request + */ +static void nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_open_confirmargs *args = data; + struct compound_hdr hdr = { + .nops = 0, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_open_confirm(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode an OPEN request with no attributes. + */ +static void nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_openargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_open(xdr, args, &hdr); + if (args->access) + encode_access(xdr, args->access, &hdr); + encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); + if (args->lg_args) { + encode_layoutget(xdr, args->lg_args, &hdr); + rpc_prepare_reply_pages(req, args->lg_args->layout.pages, 0, + args->lg_args->layout.pglen, + hdr.replen - pagepad_maxsz); + } + encode_nops(&hdr); +} + +/* + * Encode an OPEN_DOWNGRADE request + */ +static void nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_closeargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + if (args->lr_args) + encode_layoutreturn(xdr, args->lr_args, &hdr); + encode_open_downgrade(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode a LOCK request + */ +static void nfs4_xdr_enc_lock(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_lock_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_lock(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode a LOCKT request + */ +static void nfs4_xdr_enc_lockt(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_lockt_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_lockt(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode a LOCKU request + */ +static void nfs4_xdr_enc_locku(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_locku_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_locku(xdr, args, &hdr); + encode_nops(&hdr); +} + +#if defined(CONFIG_NFS_V4_0) +static void nfs4_xdr_enc_release_lockowner(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_release_lockowner_args *args = data; + struct compound_hdr hdr = { + .minorversion = 0, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_release_lockowner(xdr, &args->lock_owner, &hdr); + encode_nops(&hdr); +} +#endif /* CONFIG_NFS_V4_0 */ + +/* + * Encode a READLINK request + */ +static void nfs4_xdr_enc_readlink(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_readlink *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_readlink(xdr, args, req, &hdr); + + rpc_prepare_reply_pages(req, args->pages, args->pgbase, + args->pglen, hdr.replen - pagepad_maxsz); + encode_nops(&hdr); +} + +/* + * Encode a READDIR request + */ +static void nfs4_xdr_enc_readdir(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_readdir_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_readdir(xdr, args, req, &hdr); + + rpc_prepare_reply_pages(req, args->pages, args->pgbase, + args->count, hdr.replen - pagepad_maxsz); + encode_nops(&hdr); +} + +/* + * Encode a READ request + */ +static void nfs4_xdr_enc_read(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_read(xdr, args, &hdr); + + rpc_prepare_reply_pages(req, args->pages, args->pgbase, + args->count, hdr.replen - pagepad_maxsz); + req->rq_rcv_buf.flags |= XDRBUF_READ; + encode_nops(&hdr); +} + +/* + * Encode an SETATTR request + */ +static void nfs4_xdr_enc_setattr(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_setattrargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_setattr(xdr, args, args->server, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode a GETACL request + */ +static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_getaclargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + __u32 nfs4_acl_bitmap[2]; + uint32_t replen; + + nfs4_acltype_to_bitmap(args->acl_type, nfs4_acl_bitmap); + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + replen = hdr.replen + op_decode_hdr_maxsz; + encode_getattr(xdr, nfs4_acl_bitmap, NULL, + ARRAY_SIZE(nfs4_acl_bitmap), &hdr); + + rpc_prepare_reply_pages(req, args->acl_pages, 0, + args->acl_len, replen); + encode_nops(&hdr); +} + +/* + * Encode a WRITE request + */ +static void nfs4_xdr_enc_write(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_write(xdr, args, &hdr); + req->rq_snd_buf.flags |= XDRBUF_WRITE; + if (args->bitmask) + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * a COMMIT request + */ +static void nfs4_xdr_enc_commit(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_commitargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_commit(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * FSINFO request + */ +static void nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_fsinfo_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_fsinfo(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * a PATHCONF request + */ +static void nfs4_xdr_enc_pathconf(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_pathconf_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_getattr(xdr, nfs4_pathconf_bitmap, args->bitmask, + ARRAY_SIZE(nfs4_pathconf_bitmap), &hdr); + encode_nops(&hdr); +} + +/* + * a STATFS request + */ +static void nfs4_xdr_enc_statfs(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_statfs_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_getattr(xdr, nfs4_statfs_bitmap, args->bitmask, + ARRAY_SIZE(nfs4_statfs_bitmap), &hdr); + encode_nops(&hdr); +} + +/* + * GETATTR_BITMAP request + */ +static void nfs4_xdr_enc_server_caps(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_server_caps_arg *args = data; + const u32 *bitmask = args->bitmask; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fhandle, &hdr); + encode_getattr(xdr, bitmask, NULL, 3, &hdr); + encode_nops(&hdr); +} + +/* + * a RENEW request + */ +#if defined(CONFIG_NFS_V4_0) +static void nfs4_xdr_enc_renew(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) + +{ + const struct nfs_client *clp = data; + struct compound_hdr hdr = { + .nops = 0, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_renew(xdr, clp->cl_clientid, &hdr); + encode_nops(&hdr); +} +#endif /* CONFIG_NFS_V4_0 */ + +/* + * a SETCLIENTID request + */ +static void nfs4_xdr_enc_setclientid(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_setclientid *sc = data; + struct compound_hdr hdr = { + .nops = 0, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_setclientid(xdr, sc, &hdr); + encode_nops(&hdr); +} + +/* + * a SETCLIENTID_CONFIRM request + */ +static void nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_setclientid_res *arg = data; + struct compound_hdr hdr = { + .nops = 0, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_setclientid_confirm(xdr, arg, &hdr); + encode_nops(&hdr); +} + +/* + * DELEGRETURN request + */ +static void nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_delegreturnargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fhandle, &hdr); + if (args->lr_args) + encode_layoutreturn(xdr, args->lr_args, &hdr); + if (args->sattr_args) + encode_delegattr(xdr, args->stateid, args->sattr_args, &hdr); + if (args->bitmask) + encode_getfattr(xdr, args->bitmask, &hdr); + encode_delegreturn(xdr, args->stateid, &hdr); + encode_nops(&hdr); +} + +/* + * Encode FS_LOCATIONS request + */ +static void nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_fs_locations_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + uint32_t replen; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + if (args->migration) { + encode_putfh(xdr, args->fh, &hdr); + replen = hdr.replen; + encode_fs_locations(xdr, args->bitmask, &hdr); + if (args->renew) + encode_renew(xdr, args->clientid, &hdr); + } else { + encode_putfh(xdr, args->dir_fh, &hdr); + encode_lookup(xdr, args->name, &hdr); + replen = hdr.replen; + encode_fs_locations(xdr, args->bitmask, &hdr); + } + + rpc_prepare_reply_pages(req, (struct page **)&args->page, 0, + PAGE_SIZE, replen); + encode_nops(&hdr); +} + +/* + * Encode SECINFO request + */ +static void nfs4_xdr_enc_secinfo(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_secinfo_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->dir_fh, &hdr); + encode_secinfo(xdr, args->name, &hdr); + encode_nops(&hdr); +} + +/* + * Encode FSID_PRESENT request + */ +static void nfs4_xdr_enc_fsid_present(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_fsid_present_arg *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_getfh(xdr, &hdr); + if (args->renew) + encode_renew(xdr, args->clientid, &hdr); + encode_nops(&hdr); +} + +/* + * BIND_CONN_TO_SESSION request + */ +static void nfs4_xdr_enc_bind_conn_to_session(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_bind_conn_to_session_args *args = data; + struct compound_hdr hdr = { + .minorversion = args->client->cl_mvops->minor_version, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_bind_conn_to_session(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * EXCHANGE_ID request + */ +static void nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_exchange_id_args *args = data; + struct compound_hdr hdr = { + .minorversion = args->client->cl_mvops->minor_version, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_exchange_id(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * a CREATE_SESSION request + */ +static void nfs4_xdr_enc_create_session(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_create_session_args *args = data; + struct compound_hdr hdr = { + .minorversion = args->client->cl_mvops->minor_version, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_create_session(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * a DESTROY_SESSION request + */ +static void nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_session *session = data; + struct compound_hdr hdr = { + .minorversion = session->clp->cl_mvops->minor_version, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_destroy_session(xdr, session, &hdr); + encode_nops(&hdr); +} + +/* + * a DESTROY_CLIENTID request + */ +static void nfs4_xdr_enc_destroy_clientid(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_client *clp = data; + struct compound_hdr hdr = { + .minorversion = clp->cl_mvops->minor_version, + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_destroy_clientid(xdr, clp->cl_clientid, &hdr); + encode_nops(&hdr); +} + +/* + * a SEQUENCE request + */ +static void nfs4_xdr_enc_sequence(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_sequence_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * a GET_LEASE_TIME request + */ +static void nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_get_lease_time_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->la_seq_args), + }; + const u32 lease_bitmap[3] = { FATTR4_WORD0_LEASE_TIME }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->la_seq_args, &hdr); + encode_putrootfh(xdr, &hdr); + encode_fsinfo(xdr, lease_bitmap, &hdr); + encode_nops(&hdr); +} + +/* + * a RECLAIM_COMPLETE request + */ +static void nfs4_xdr_enc_reclaim_complete(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_reclaim_complete_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args) + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_reclaim_complete(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode GETDEVICEINFO request + */ +static void nfs4_xdr_enc_getdeviceinfo(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_getdeviceinfo_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + uint32_t replen; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + + replen = hdr.replen + op_decode_hdr_maxsz + 2; + + encode_getdeviceinfo(xdr, args, &hdr); + + /* set up reply kvec. device_addr4 opaque data is read into the + * pages */ + rpc_prepare_reply_pages(req, args->pdev->pages, args->pdev->pgbase, + args->pdev->pglen, replen); + encode_nops(&hdr); +} + +/* + * Encode LAYOUTGET request + */ +static void nfs4_xdr_enc_layoutget(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_layoutget_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, NFS_FH(args->inode), &hdr); + encode_layoutget(xdr, args, &hdr); + + rpc_prepare_reply_pages(req, args->layout.pages, 0, + args->layout.pglen, hdr.replen - pagepad_maxsz); + encode_nops(&hdr); +} + +/* + * Encode LAYOUTCOMMIT request + */ +static void nfs4_xdr_enc_layoutcommit(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *priv) +{ + const struct nfs4_layoutcommit_args *args = priv; + struct nfs4_layoutcommit_data *data = + container_of(args, struct nfs4_layoutcommit_data, args); + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, NFS_FH(args->inode), &hdr); + encode_layoutcommit(xdr, data->args.inode, args, &hdr); + encode_getfattr(xdr, args->bitmask, &hdr); + encode_nops(&hdr); +} + +/* + * Encode LAYOUTRETURN request + */ +static void nfs4_xdr_enc_layoutreturn(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs4_layoutreturn_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, NFS_FH(args->inode), &hdr); + encode_layoutreturn(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode SECINFO_NO_NAME request + */ +static void nfs4_xdr_enc_secinfo_no_name(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_secinfo_no_name_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putrootfh(xdr, &hdr); + encode_secinfo_no_name(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode TEST_STATEID request + */ +static void nfs4_xdr_enc_test_stateid(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_test_stateid_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_test_stateid(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Encode FREE_STATEID request + */ +static void nfs4_xdr_enc_free_stateid(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs41_free_stateid_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_free_stateid(xdr, args, &hdr); + encode_nops(&hdr); +} + +static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char **string) +{ + ssize_t ret = xdr_stream_decode_opaque_inline(xdr, (void **)string, + NFS4_OPAQUE_LIMIT); + if (unlikely(ret < 0)) + return -EIO; + *len = ret; + return 0; +} + +static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) +{ + ssize_t ret; + void *ptr; + u32 tmp; + + if (xdr_stream_decode_u32(xdr, &tmp) < 0) + return -EIO; + hdr->status = tmp; + + ret = xdr_stream_decode_opaque_inline(xdr, &ptr, NFS4_OPAQUE_LIMIT); + if (ret < 0) + return -EIO; + hdr->taglen = ret; + hdr->tag = ptr; + + if (xdr_stream_decode_u32(xdr, &tmp) < 0) + return -EIO; + hdr->nops = tmp; + if (unlikely(hdr->nops < 1)) + return nfs4_stat_to_errno(hdr->status); + return 0; +} + +static bool __decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected, + int *nfs_retval) +{ + __be32 *p; + uint32_t opnum; + int32_t nfserr; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + goto out_overflow; + opnum = be32_to_cpup(p++); + if (unlikely(opnum != expected)) + goto out_bad_operation; + if (unlikely(*p != cpu_to_be32(NFS_OK))) + goto out_status; + *nfs_retval = 0; + return true; +out_status: + nfserr = be32_to_cpup(p); + trace_nfs4_xdr_status(xdr, opnum, nfserr); + *nfs_retval = nfs4_stat_to_errno(nfserr); + return true; +out_bad_operation: + trace_nfs4_xdr_bad_operation(xdr, opnum, expected); + *nfs_retval = -EREMOTEIO; + return false; +out_overflow: + *nfs_retval = -EIO; + return false; +} + +static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) +{ + int retval; + + __decode_op_hdr(xdr, expected, &retval); + return retval; +} + +/* Dummy routine */ +static int decode_ace(struct xdr_stream *xdr, void *ace) +{ + __be32 *p; + unsigned int strlen; + char *str; + + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EIO; + return decode_opaque_inline(xdr, &strlen, &str); +} + +static ssize_t +decode_bitmap4(struct xdr_stream *xdr, uint32_t *bitmap, size_t sz) +{ + ssize_t ret; + + ret = xdr_stream_decode_uint32_array(xdr, bitmap, sz); + if (likely(ret >= 0)) + return ret; + if (ret != -EMSGSIZE) + return -EIO; + return sz; +} + +static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +{ + ssize_t ret; + ret = decode_bitmap4(xdr, bitmap, 3); + return ret < 0 ? ret : 0; +} + +static int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, unsigned int *savep) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *attrlen = be32_to_cpup(p); + *savep = xdr_stream_pos(xdr); + return 0; +} + +static int decode_attr_supported(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *bitmask) +{ + if (likely(bitmap[0] & FATTR4_WORD0_SUPPORTED_ATTRS)) { + int ret; + ret = decode_attr_bitmap(xdr, bitmask); + if (unlikely(ret < 0)) + return ret; + bitmap[0] &= ~FATTR4_WORD0_SUPPORTED_ATTRS; + } else + bitmask[0] = bitmask[1] = bitmask[2] = 0; + dprintk("%s: bitmask=%08x:%08x:%08x\n", __func__, + bitmask[0], bitmask[1], bitmask[2]); + return 0; +} + +static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *type) +{ + __be32 *p; + int ret = 0; + + *type = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_TYPE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_TYPE)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *type = be32_to_cpup(p); + if (*type < NF4REG || *type > NF4NAMEDATTR) { + dprintk("%s: bad type %d\n", __func__, *type); + return -EIO; + } + bitmap[0] &= ~FATTR4_WORD0_TYPE; + ret = NFS_ATTR_FATTR_TYPE; + } + dprintk("%s: type=0%o\n", __func__, nfs_type2fmt[*type]); + return ret; +} + +static int decode_attr_fh_expire_type(struct xdr_stream *xdr, + uint32_t *bitmap, uint32_t *type) +{ + __be32 *p; + + *type = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FH_EXPIRE_TYPE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FH_EXPIRE_TYPE)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *type = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_FH_EXPIRE_TYPE; + } + dprintk("%s: expire type=0x%x\n", __func__, *type); + return 0; +} + +static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *change) +{ + __be32 *p; + int ret = 0; + + *change = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_CHANGE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_CHANGE)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, change); + bitmap[0] &= ~FATTR4_WORD0_CHANGE; + ret = NFS_ATTR_FATTR_CHANGE; + } + dprintk("%s: change attribute=%Lu\n", __func__, + (unsigned long long)*change); + return ret; +} + +static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *size) +{ + __be32 *p; + int ret = 0; + + *size = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_SIZE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_SIZE)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, size); + bitmap[0] &= ~FATTR4_WORD0_SIZE; + ret = NFS_ATTR_FATTR_SIZE; + } + dprintk("%s: file size=%Lu\n", __func__, (unsigned long long)*size); + return ret; +} + +static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_LINK_SUPPORT - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_LINK_SUPPORT)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_LINK_SUPPORT; + } + dprintk("%s: link support=%s\n", __func__, str_false_true(*res == 0)); + return 0; +} + +static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_SYMLINK_SUPPORT - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_SYMLINK_SUPPORT)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_SYMLINK_SUPPORT; + } + dprintk("%s: symlink support=%s\n", __func__, str_false_true(*res == 0)); + return 0; +} + +static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fsid *fsid) +{ + __be32 *p; + int ret = 0; + + fsid->major = 0; + fsid->minor = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FSID - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FSID)) { + p = xdr_inline_decode(xdr, 16); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &fsid->major); + xdr_decode_hyper(p, &fsid->minor); + bitmap[0] &= ~FATTR4_WORD0_FSID; + ret = NFS_ATTR_FATTR_FSID; + } + dprintk("%s: fsid=(0x%Lx/0x%Lx)\n", __func__, + (unsigned long long)fsid->major, + (unsigned long long)fsid->minor); + return ret; +} + +static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 60; + if (unlikely(bitmap[0] & (FATTR4_WORD0_LEASE_TIME - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_LEASE_TIME)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_LEASE_TIME; + } + dprintk("%s: lease time=%u\n", __func__, (unsigned int)*res); + return 0; +} + +static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap, int32_t *res) +{ + __be32 *p; + + if (unlikely(bitmap[0] & (FATTR4_WORD0_RDATTR_ERROR - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_RDATTR_ERROR)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + bitmap[0] &= ~FATTR4_WORD0_RDATTR_ERROR; + *res = -be32_to_cpup(p); + } + return 0; +} + +static int decode_attr_exclcreat_supported(struct xdr_stream *xdr, + uint32_t *bitmap, uint32_t *bitmask) +{ + if (likely(bitmap[2] & FATTR4_WORD2_SUPPATTR_EXCLCREAT)) { + int ret; + ret = decode_attr_bitmap(xdr, bitmask); + if (unlikely(ret < 0)) + return ret; + bitmap[2] &= ~FATTR4_WORD2_SUPPATTR_EXCLCREAT; + } else + bitmask[0] = bitmask[1] = bitmask[2] = 0; + dprintk("%s: bitmask=%08x:%08x:%08x\n", __func__, + bitmask[0], bitmask[1], bitmask[2]); + return 0; +} + +static int decode_attr_filehandle(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fh *fh) +{ + __be32 *p; + u32 len; + + if (fh != NULL) + memset(fh, 0, sizeof(*fh)); + + if (unlikely(bitmap[0] & (FATTR4_WORD0_FILEHANDLE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FILEHANDLE)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + if (len > NFS4_FHSIZE || len == 0) { + trace_nfs4_xdr_bad_filehandle(xdr, OP_READDIR, + NFS4ERR_BADHANDLE); + return -EREMOTEIO; + } + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + if (fh != NULL) { + memcpy(fh->data, p, len); + fh->size = len; + } + bitmap[0] &= ~FATTR4_WORD0_FILEHANDLE; + } + return 0; +} + +static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_ACLSUPPORT - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_ACLSUPPORT)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_ACLSUPPORT; + } + dprintk("%s: ACLs supported=%u\n", __func__, (unsigned int)*res); + return 0; +} + +static int decode_attr_case_insensitive(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_CASE_INSENSITIVE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_CASE_INSENSITIVE)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_CASE_INSENSITIVE; + } + dprintk("%s: case_insensitive=%s\n", __func__, str_false_true(*res == 0)); + return 0; +} + +static int decode_attr_case_preserving(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_CASE_PRESERVING - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_CASE_PRESERVING)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_CASE_PRESERVING; + } + dprintk("%s: case_preserving=%s\n", __func__, str_false_true(*res == 0)); + return 0; +} + +static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) +{ + __be32 *p; + int ret = 0; + + *fileid = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FILEID - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FILEID)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, fileid); + bitmap[0] &= ~FATTR4_WORD0_FILEID; + ret = NFS_ATTR_FATTR_FILEID; + } + dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid); + return ret; +} + +static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) +{ + __be32 *p; + int ret = 0; + + *fileid = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_MOUNTED_ON_FILEID - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, fileid); + bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; + ret = NFS_ATTR_FATTR_MOUNTED_ON_FILEID; + } + dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid); + return ret; +} + +static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FILES_AVAIL - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FILES_AVAIL)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[0] &= ~FATTR4_WORD0_FILES_AVAIL; + } + dprintk("%s: files avail=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_files_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FILES_FREE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FILES_FREE)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[0] &= ~FATTR4_WORD0_FILES_FREE; + } + dprintk("%s: files free=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_FILES_TOTAL - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_FILES_TOTAL)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[0] &= ~FATTR4_WORD0_FILES_TOTAL; + } + dprintk("%s: files total=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_pathname(struct xdr_stream *xdr, struct nfs4_pathname *path) +{ + u32 n; + __be32 *p; + int status = 0; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + n = be32_to_cpup(p); + if (n == 0) + goto root_path; + dprintk("pathname4: "); + if (n > NFS4_PATHNAME_MAXCOMPONENTS) { + dprintk("cannot parse %d components in path\n", n); + goto out_eio; + } + for (path->ncomponents = 0; path->ncomponents < n; path->ncomponents++) { + struct nfs4_string *component = &path->components[path->ncomponents]; + status = decode_opaque_inline(xdr, &component->len, &component->data); + if (unlikely(status != 0)) + goto out_eio; + ifdebug (XDR) + pr_cont("%s%.*s ", + (path->ncomponents != n ? "/ " : ""), + component->len, component->data); + } +out: + return status; +root_path: +/* a root pathname is sent as a zero component4 */ + path->ncomponents = 1; + path->components[0].len=0; + path->components[0].data=NULL; + dprintk("pathname4: /\n"); + goto out; +out_eio: + dprintk(" status %d", status); + status = -EIO; + goto out; +} + +static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_fs_locations *res) +{ + int n; + __be32 *p; + int status = -EIO; + + if (unlikely(bitmap[0] & (FATTR4_WORD0_FS_LOCATIONS -1U))) + goto out; + status = 0; + if (unlikely(!(bitmap[0] & FATTR4_WORD0_FS_LOCATIONS))) + goto out; + bitmap[0] &= ~FATTR4_WORD0_FS_LOCATIONS; + status = -EIO; + /* Ignore borken servers that return unrequested attrs */ + if (unlikely(res == NULL)) + goto out; + dprintk("%s: fsroot:\n", __func__); + status = decode_pathname(xdr, &res->fs_path); + if (unlikely(status != 0)) + goto out; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out_eio; + n = be32_to_cpup(p); + for (res->nlocations = 0; res->nlocations < n; res->nlocations++) { + u32 m; + struct nfs4_fs_location *loc; + + if (res->nlocations == NFS4_FS_LOCATIONS_MAXENTRIES) + break; + loc = &res->locations[res->nlocations]; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out_eio; + m = be32_to_cpup(p); + + dprintk("%s: servers:\n", __func__); + for (loc->nservers = 0; loc->nservers < m; loc->nservers++) { + struct nfs4_string *server; + + if (loc->nservers == NFS4_FS_LOCATION_MAXSERVERS) { + unsigned int i; + dprintk("%s: using first %u of %u servers " + "returned for location %u\n", + __func__, + NFS4_FS_LOCATION_MAXSERVERS, + m, res->nlocations); + for (i = loc->nservers; i < m; i++) { + unsigned int len; + char *data; + status = decode_opaque_inline(xdr, &len, &data); + if (unlikely(status != 0)) + goto out_eio; + } + break; + } + server = &loc->servers[loc->nservers]; + status = decode_opaque_inline(xdr, &server->len, &server->data); + if (unlikely(status != 0)) + goto out_eio; + dprintk("%s ", server->data); + } + status = decode_pathname(xdr, &loc->rootpath); + if (unlikely(status != 0)) + goto out_eio; + } + if (res->nlocations != 0) + status = NFS_ATTR_FATTR_V4_LOCATIONS; +out: + dprintk("%s: fs_locations done, error = %d\n", __func__, status); + return status; +out_eio: + status = -EIO; + goto out; +} + +static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[0] & (FATTR4_WORD0_MAXFILESIZE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_MAXFILESIZE)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[0] &= ~FATTR4_WORD0_MAXFILESIZE; + } + dprintk("%s: maxfilesize=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_maxlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxlink) +{ + __be32 *p; + int status = 0; + + *maxlink = 1; + if (unlikely(bitmap[0] & (FATTR4_WORD0_MAXLINK - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_MAXLINK)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *maxlink = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_MAXLINK; + } + dprintk("%s: maxlink=%u\n", __func__, *maxlink); + return status; +} + +static int decode_attr_maxname(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxname) +{ + __be32 *p; + int status = 0; + + *maxname = 1024; + if (unlikely(bitmap[0] & (FATTR4_WORD0_MAXNAME - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_MAXNAME)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *maxname = be32_to_cpup(p); + bitmap[0] &= ~FATTR4_WORD0_MAXNAME; + } + dprintk("%s: maxname=%u\n", __func__, *maxname); + return status; +} + +static int decode_attr_maxread(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + int status = 0; + + *res = 1024; + if (unlikely(bitmap[0] & (FATTR4_WORD0_MAXREAD - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_MAXREAD)) { + uint64_t maxread; + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, &maxread); + if (maxread > 0x7FFFFFFF) + maxread = 0x7FFFFFFF; + *res = (uint32_t)maxread; + bitmap[0] &= ~FATTR4_WORD0_MAXREAD; + } + dprintk("%s: maxread=%lu\n", __func__, (unsigned long)*res); + return status; +} + +static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) +{ + __be32 *p; + int status = 0; + + *res = 1024; + if (unlikely(bitmap[0] & (FATTR4_WORD0_MAXWRITE - 1U))) + return -EIO; + if (likely(bitmap[0] & FATTR4_WORD0_MAXWRITE)) { + uint64_t maxwrite; + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, &maxwrite); + if (maxwrite > 0x7FFFFFFF) + maxwrite = 0x7FFFFFFF; + *res = (uint32_t)maxwrite; + bitmap[0] &= ~FATTR4_WORD0_MAXWRITE; + } + dprintk("%s: maxwrite=%lu\n", __func__, (unsigned long)*res); + return status; +} + +static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, umode_t *mode) +{ + uint32_t tmp; + __be32 *p; + int ret = 0; + + *mode = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_MODE - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_MODE)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + tmp = be32_to_cpup(p); + *mode = tmp & ~S_IFMT; + bitmap[1] &= ~FATTR4_WORD1_MODE; + ret = NFS_ATTR_FATTR_MODE; + } + dprintk("%s: file mode=0%o\n", __func__, (unsigned int)*mode); + return ret; +} + +static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *nlink) +{ + __be32 *p; + int ret = 0; + + *nlink = 1; + if (unlikely(bitmap[1] & (FATTR4_WORD1_NUMLINKS - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_NUMLINKS)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *nlink = be32_to_cpup(p); + bitmap[1] &= ~FATTR4_WORD1_NUMLINKS; + ret = NFS_ATTR_FATTR_NLINK; + } + dprintk("%s: nlink=%u\n", __func__, (unsigned int)*nlink); + return ret; +} + +static ssize_t decode_nfs4_string(struct xdr_stream *xdr, + struct nfs4_string *name, gfp_t gfp_flags) +{ + ssize_t ret; + + ret = xdr_stream_decode_string_dup(xdr, &name->data, + XDR_MAX_NETOBJ, gfp_flags); + name->len = 0; + if (ret > 0) + name->len = ret; + return ret; +} + +static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, + const struct nfs_server *server, kuid_t *uid, + struct nfs4_string *owner_name) +{ + ssize_t len; + char *p; + + *uid = make_kuid(&init_user_ns, -2); + if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER - 1U))) + return -EIO; + if (!(bitmap[1] & FATTR4_WORD1_OWNER)) + return 0; + bitmap[1] &= ~FATTR4_WORD1_OWNER; + + if (owner_name != NULL) { + len = decode_nfs4_string(xdr, owner_name, GFP_NOIO); + if (len <= 0) + goto out; + dprintk("%s: name=%s\n", __func__, owner_name->data); + return NFS_ATTR_FATTR_OWNER_NAME; + } else { + len = xdr_stream_decode_opaque_inline(xdr, (void **)&p, + XDR_MAX_NETOBJ); + if (len <= 0 || nfs_map_name_to_uid(server, p, len, uid) != 0) + goto out; + dprintk("%s: uid=%d\n", __func__, (int)from_kuid(&init_user_ns, *uid)); + return NFS_ATTR_FATTR_OWNER; + } +out: + if (len == -EBADMSG) + return -EIO; + return 0; +} + +static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, + const struct nfs_server *server, kgid_t *gid, + struct nfs4_string *group_name) +{ + ssize_t len; + char *p; + + *gid = make_kgid(&init_user_ns, -2); + if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER_GROUP - 1U))) + return -EIO; + if (!(bitmap[1] & FATTR4_WORD1_OWNER_GROUP)) + return 0; + bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP; + + if (group_name != NULL) { + len = decode_nfs4_string(xdr, group_name, GFP_NOIO); + if (len <= 0) + goto out; + dprintk("%s: name=%s\n", __func__, group_name->data); + return NFS_ATTR_FATTR_GROUP_NAME; + } else { + len = xdr_stream_decode_opaque_inline(xdr, (void **)&p, + XDR_MAX_NETOBJ); + if (len <= 0 || nfs_map_group_to_gid(server, p, len, gid) != 0) + goto out; + dprintk("%s: gid=%d\n", __func__, (int)from_kgid(&init_user_ns, *gid)); + return NFS_ATTR_FATTR_GROUP; + } +out: + if (len == -EBADMSG) + return -EIO; + return 0; +} + +static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rdev) +{ + uint32_t major = 0, minor = 0; + __be32 *p; + int ret = 0; + + *rdev = MKDEV(0,0); + if (unlikely(bitmap[1] & (FATTR4_WORD1_RAWDEV - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_RAWDEV)) { + dev_t tmp; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + major = be32_to_cpup(p++); + minor = be32_to_cpup(p); + tmp = MKDEV(major, minor); + if (MAJOR(tmp) == major && MINOR(tmp) == minor) + *rdev = tmp; + bitmap[1] &= ~ FATTR4_WORD1_RAWDEV; + ret = NFS_ATTR_FATTR_RDEV; + } + dprintk("%s: rdev=(0x%x:0x%x)\n", __func__, major, minor); + return ret; +} + +static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_AVAIL - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_SPACE_AVAIL)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[1] &= ~FATTR4_WORD1_SPACE_AVAIL; + } + dprintk("%s: space avail=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_space_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_FREE - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_SPACE_FREE)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[1] &= ~FATTR4_WORD1_SPACE_FREE; + } + dprintk("%s: space free=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) +{ + __be32 *p; + int status = 0; + + *res = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_TOTAL - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_SPACE_TOTAL)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + bitmap[1] &= ~FATTR4_WORD1_SPACE_TOTAL; + } + dprintk("%s: space total=%Lu\n", __func__, (unsigned long long)*res); + return status; +} + +static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *used) +{ + __be32 *p; + int ret = 0; + + *used = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_USED - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_SPACE_USED)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, used); + bitmap[1] &= ~FATTR4_WORD1_SPACE_USED; + ret = NFS_ATTR_FATTR_SPACE_USED; + } + dprintk("%s: space used=%Lu\n", __func__, + (unsigned long long)*used); + return ret; +} + +static __be32 * +xdr_decode_nfstime4(__be32 *p, struct timespec64 *t) +{ + __u64 sec; + + p = xdr_decode_hyper(p, &sec); + t-> tv_sec = sec; + t->tv_nsec = be32_to_cpup(p++); + return p; +} + +static int decode_attr_time(struct xdr_stream *xdr, struct timespec64 *time) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, nfstime4_maxsz << 2); + if (unlikely(!p)) + return -EIO; + xdr_decode_nfstime4(p, time); + return 0; +} + +static int decode_attr_time_access(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time) +{ + int status = 0; + + time->tv_sec = 0; + time->tv_nsec = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_TIME_ACCESS - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_TIME_ACCESS)) { + status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_ATIME; + bitmap[1] &= ~FATTR4_WORD1_TIME_ACCESS; + } + dprintk("%s: atime=%lld\n", __func__, time->tv_sec); + return status; +} + +static int decode_attr_time_create(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time) +{ + int status = 0; + + time->tv_sec = 0; + time->tv_nsec = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_TIME_CREATE - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_TIME_CREATE)) { + status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_BTIME; + bitmap[1] &= ~FATTR4_WORD1_TIME_CREATE; + } + dprintk("%s: btime=%lld\n", __func__, time->tv_sec); + return status; +} + +static int decode_attr_time_metadata(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time) +{ + int status = 0; + + time->tv_sec = 0; + time->tv_nsec = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_TIME_METADATA - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_TIME_METADATA)) { + status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_CTIME; + bitmap[1] &= ~FATTR4_WORD1_TIME_METADATA; + } + dprintk("%s: ctime=%lld\n", __func__, time->tv_sec); + return status; +} + +static int decode_attr_time_delta(struct xdr_stream *xdr, uint32_t *bitmap, + struct timespec64 *time) +{ + int status = 0; + + time->tv_sec = 0; + time->tv_nsec = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_TIME_DELTA - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_TIME_DELTA)) { + status = decode_attr_time(xdr, time); + bitmap[1] &= ~FATTR4_WORD1_TIME_DELTA; + } + dprintk("%s: time_delta=%lld %ld\n", __func__, time->tv_sec, + time->tv_nsec); + return status; +} + +static int decode_attr_security_label(struct xdr_stream *xdr, uint32_t *bitmap, + struct nfs4_label *label) +{ + uint32_t pi = 0; + uint32_t lfs = 0; + __u32 len; + __be32 *p; + int status = 0; + + if (unlikely(bitmap[2] & (FATTR4_WORD2_SECURITY_LABEL - 1U))) + return -EIO; + if (likely(bitmap[2] & FATTR4_WORD2_SECURITY_LABEL)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + lfs = be32_to_cpup(p++); + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + pi = be32_to_cpup(p++); + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p++); + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + bitmap[2] &= ~FATTR4_WORD2_SECURITY_LABEL; + if (len < NFS4_MAXLABELLEN) { + if (label && label->len) { + if (label->len < len) + return -ERANGE; + memcpy(label->label, p, len); + label->len = len; + label->pi = pi; + label->lfs = lfs; + status = NFS_ATTR_FATTR_V4_SECURITY_LABEL; + } + } else + printk(KERN_WARNING "%s: label too long (%u)!\n", + __func__, len); + if (label && label->label) + dprintk("%s: label=%.*s, len=%d, PI=%d, LFS=%d\n", + __func__, label->len, (char *)label->label, + label->len, label->pi, label->lfs); + } + return status; +} + +static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, struct timespec64 *time) +{ + int status = 0; + + time->tv_sec = 0; + time->tv_nsec = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_TIME_MODIFY - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_TIME_MODIFY)) { + status = decode_attr_time(xdr, time); + if (status == 0) + status = NFS_ATTR_FATTR_MTIME; + bitmap[1] &= ~FATTR4_WORD1_TIME_MODIFY; + } + dprintk("%s: mtime=%lld\n", __func__, time->tv_sec); + return status; +} + +static int decode_attr_xattrsupport(struct xdr_stream *xdr, uint32_t *bitmap, + uint32_t *res) +{ + __be32 *p; + + *res = 0; + if (unlikely(bitmap[2] & (FATTR4_WORD2_XATTR_SUPPORT - 1U))) + return -EIO; + if (likely(bitmap[2] & FATTR4_WORD2_XATTR_SUPPORT)) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[2] &= ~FATTR4_WORD2_XATTR_SUPPORT; + } + dprintk("%s: XATTR support=%s\n", __func__, str_false_true(*res == 0)); + return 0; +} + +static int decode_attr_open_arguments(struct xdr_stream *xdr, uint32_t *bitmap, + struct nfs4_open_caps *res) +{ + memset(res, 0, sizeof(*res)); + if (unlikely(bitmap[2] & (FATTR4_WORD2_OPEN_ARGUMENTS - 1U))) + return -EIO; + if (likely(bitmap[2] & FATTR4_WORD2_OPEN_ARGUMENTS)) { + if (decode_bitmap4(xdr, res->oa_share_access, ARRAY_SIZE(res->oa_share_access)) < 0) + return -EIO; + if (decode_bitmap4(xdr, res->oa_share_deny, ARRAY_SIZE(res->oa_share_deny)) < 0) + return -EIO; + if (decode_bitmap4(xdr, res->oa_share_access_want, ARRAY_SIZE(res->oa_share_access_want)) < 0) + return -EIO; + if (decode_bitmap4(xdr, res->oa_open_claim, ARRAY_SIZE(res->oa_open_claim)) < 0) + return -EIO; + if (decode_bitmap4(xdr, res->oa_createmode, ARRAY_SIZE(res->oa_createmode)) < 0) + return -EIO; + bitmap[2] &= ~FATTR4_WORD2_OPEN_ARGUMENTS; + } + return 0; +} + +static int verify_attr_len(struct xdr_stream *xdr, unsigned int savep, uint32_t attrlen) +{ + unsigned int attrwords = XDR_QUADLEN(attrlen); + unsigned int nwords = (xdr_stream_pos(xdr) - savep) >> 2; + + if (unlikely(attrwords != nwords)) { + dprintk("%s: server returned incorrect attribute length: " + "%u %c %u\n", + __func__, + attrwords << 2, + (attrwords < nwords) ? '<' : '>', + nwords << 2); + return -EIO; + } + return 0; +} + +static int decode_change_info(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 20); + if (unlikely(!p)) + return -EIO; + cinfo->atomic = be32_to_cpup(p++); + p = xdr_decode_hyper(p, &cinfo->before); + xdr_decode_hyper(p, &cinfo->after); + return 0; +} + +static int decode_access(struct xdr_stream *xdr, u32 *supported, u32 *access) +{ + __be32 *p; + uint32_t supp, acc; + int status; + + status = decode_op_hdr(xdr, OP_ACCESS); + if (status) + return status; + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + supp = be32_to_cpup(p++); + acc = be32_to_cpup(p); + *supported = supp; + *access = acc; + return 0; +} + +static int decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + return decode_opaque_fixed(xdr, stateid, NFS4_STATEID_SIZE); +} + +static int decode_open_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_OPEN_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static int decode_lock_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_LOCK_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static int decode_delegation_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_DELEGATION_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static int decode_invalid_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + nfs4_stateid dummy; + + nfs4_stateid_copy(stateid, &invalid_stateid); + return decode_stateid(xdr, &dummy); +} + +static int decode_close(struct xdr_stream *xdr, struct nfs_closeres *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_CLOSE); + if (status != -EIO) + nfs_increment_open_seqid(status, res->seqid); + if (!status) + status = decode_invalid_stateid(xdr, &res->stateid); + return status; +} + +static int decode_verifier(struct xdr_stream *xdr, void *verifier) +{ + return decode_opaque_fixed(xdr, verifier, NFS4_VERIFIER_SIZE); +} + +static int decode_write_verifier(struct xdr_stream *xdr, struct nfs_write_verifier *verifier) +{ + return decode_opaque_fixed(xdr, verifier->data, NFS4_VERIFIER_SIZE); +} + +static int decode_commit(struct xdr_stream *xdr, struct nfs_commitres *res) +{ + struct nfs_writeverf *verf = res->verf; + int status; + + status = decode_op_hdr(xdr, OP_COMMIT); + if (!status) + status = decode_write_verifier(xdr, &verf->verifier); + if (!status) + verf->committed = NFS_FILE_SYNC; + return status; +} + +static int decode_create(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) +{ + __be32 *p; + uint32_t bmlen; + int status; + + status = decode_op_hdr(xdr, OP_CREATE); + if (status) + return status; + if ((status = decode_change_info(xdr, cinfo))) + return status; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + bmlen = be32_to_cpup(p); + p = xdr_inline_decode(xdr, bmlen << 2); + if (likely(p)) + return 0; + return -EIO; +} + +static int decode_server_caps(struct xdr_stream *xdr, struct nfs4_server_caps_res *res) +{ + unsigned int savep; + uint32_t attrlen, bitmap[3] = {0}; + int status; + + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto xdr_error; + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto xdr_error; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto xdr_error; + if ((status = decode_attr_supported(xdr, bitmap, res->attr_bitmask)) != 0) + goto xdr_error; + if ((status = decode_attr_fh_expire_type(xdr, bitmap, + &res->fh_expire_type)) != 0) + goto xdr_error; + if ((status = decode_attr_link_support(xdr, bitmap, &res->has_links)) != 0) + goto xdr_error; + if ((status = decode_attr_symlink_support(xdr, bitmap, &res->has_symlinks)) != 0) + goto xdr_error; + if ((status = decode_attr_aclsupport(xdr, bitmap, &res->acl_bitmask)) != 0) + goto xdr_error; + if ((status = decode_attr_case_insensitive(xdr, bitmap, &res->case_insensitive)) != 0) + goto xdr_error; + if ((status = decode_attr_case_preserving(xdr, bitmap, &res->case_preserving)) != 0) + goto xdr_error; + if ((status = decode_attr_exclcreat_supported(xdr, bitmap, + res->exclcreat_bitmask)) != 0) + goto xdr_error; + if ((status = decode_attr_open_arguments(xdr, bitmap, &res->open_caps)) != 0) + goto xdr_error; + status = verify_attr_len(xdr, savep, attrlen); +xdr_error: + dprintk("%s: xdr returned %d!\n", __func__, -status); + return status; +} + +static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat) +{ + unsigned int savep; + uint32_t attrlen, bitmap[3] = {0}; + int status; + + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto xdr_error; + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto xdr_error; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto xdr_error; + + if ((status = decode_attr_files_avail(xdr, bitmap, &fsstat->afiles)) != 0) + goto xdr_error; + if ((status = decode_attr_files_free(xdr, bitmap, &fsstat->ffiles)) != 0) + goto xdr_error; + if ((status = decode_attr_files_total(xdr, bitmap, &fsstat->tfiles)) != 0) + goto xdr_error; + + status = -EIO; + if (unlikely(bitmap[0])) + goto xdr_error; + + if ((status = decode_attr_space_avail(xdr, bitmap, &fsstat->abytes)) != 0) + goto xdr_error; + if ((status = decode_attr_space_free(xdr, bitmap, &fsstat->fbytes)) != 0) + goto xdr_error; + if ((status = decode_attr_space_total(xdr, bitmap, &fsstat->tbytes)) != 0) + goto xdr_error; + + status = verify_attr_len(xdr, savep, attrlen); +xdr_error: + dprintk("%s: xdr returned %d!\n", __func__, -status); + return status; +} + +static int decode_pathconf(struct xdr_stream *xdr, struct nfs_pathconf *pathconf) +{ + unsigned int savep; + uint32_t attrlen, bitmap[3] = {0}; + int status; + + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto xdr_error; + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto xdr_error; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto xdr_error; + + if ((status = decode_attr_maxlink(xdr, bitmap, &pathconf->max_link)) != 0) + goto xdr_error; + if ((status = decode_attr_maxname(xdr, bitmap, &pathconf->max_namelen)) != 0) + goto xdr_error; + + status = verify_attr_len(xdr, savep, attrlen); +xdr_error: + dprintk("%s: xdr returned %d!\n", __func__, -status); + return status; +} + +static int decode_threshold_hint(struct xdr_stream *xdr, + uint32_t *bitmap, + uint64_t *res, + uint32_t hint_bit) +{ + __be32 *p; + + *res = 0; + if (likely(bitmap[0] & hint_bit)) { + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, res); + } + return 0; +} + +static int decode_first_threshold_item4(struct xdr_stream *xdr, + struct nfs4_threshold *res) +{ + __be32 *p; + unsigned int savep; + uint32_t bitmap[3] = {0,}, attrlen; + int status; + + /* layout type */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->l_type = be32_to_cpup(p); + + /* thi_hintset bitmap */ + status = decode_attr_bitmap(xdr, bitmap); + if (status < 0) + goto xdr_error; + + /* thi_hintlist length */ + status = decode_attr_length(xdr, &attrlen, &savep); + if (status < 0) + goto xdr_error; + /* thi_hintlist */ + status = decode_threshold_hint(xdr, bitmap, &res->rd_sz, THRESHOLD_RD); + if (status < 0) + goto xdr_error; + status = decode_threshold_hint(xdr, bitmap, &res->wr_sz, THRESHOLD_WR); + if (status < 0) + goto xdr_error; + status = decode_threshold_hint(xdr, bitmap, &res->rd_io_sz, + THRESHOLD_RD_IO); + if (status < 0) + goto xdr_error; + status = decode_threshold_hint(xdr, bitmap, &res->wr_io_sz, + THRESHOLD_WR_IO); + if (status < 0) + goto xdr_error; + + status = verify_attr_len(xdr, savep, attrlen); + res->bm = bitmap[0]; + + dprintk("%s bm=0x%x rd_sz=%llu wr_sz=%llu rd_io=%llu wr_io=%llu\n", + __func__, res->bm, res->rd_sz, res->wr_sz, res->rd_io_sz, + res->wr_io_sz); +xdr_error: + dprintk("%s ret=%d!\n", __func__, status); + return status; +} + +/* + * Thresholds on pNFS direct I/O vrs MDS I/O + */ +static int decode_attr_mdsthreshold(struct xdr_stream *xdr, + uint32_t *bitmap, + struct nfs4_threshold *res) +{ + __be32 *p; + int status = 0; + uint32_t num; + + if (unlikely(bitmap[2] & (FATTR4_WORD2_MDSTHRESHOLD - 1U))) + return -EIO; + if (bitmap[2] & FATTR4_WORD2_MDSTHRESHOLD) { + /* Did the server return an unrequested attribute? */ + if (unlikely(res == NULL)) + return -EREMOTEIO; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + num = be32_to_cpup(p); + if (num == 0) + return 0; + if (num > 1) + printk(KERN_INFO "%s: Warning: Multiple pNFS layout " + "drivers per filesystem not supported\n", + __func__); + + status = decode_first_threshold_item4(xdr, res); + bitmap[2] &= ~FATTR4_WORD2_MDSTHRESHOLD; + } + return status; +} + +static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, + struct nfs_fattr *fattr, struct nfs_fh *fh, + struct nfs4_fs_locations *fs_loc, const struct nfs_server *server) +{ + int status; + umode_t fmode = 0; + uint32_t type; + int32_t err; + + status = decode_attr_type(xdr, bitmap, &type); + if (status < 0) + goto xdr_error; + fattr->mode = 0; + if (status != 0) { + fattr->mode |= nfs_type2fmt[type]; + fattr->valid |= status; + } + + status = decode_attr_change(xdr, bitmap, &fattr->change_attr); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_size(xdr, bitmap, &fattr->size); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_fsid(xdr, bitmap, &fattr->fsid); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + err = 0; + status = decode_attr_error(xdr, bitmap, &err); + if (status < 0) + goto xdr_error; + + status = decode_attr_filehandle(xdr, bitmap, fh); + if (status < 0) + goto xdr_error; + + status = decode_attr_fileid(xdr, bitmap, &fattr->fileid); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_fs_locations(xdr, bitmap, fs_loc); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = -EIO; + if (unlikely(bitmap[0])) + goto xdr_error; + + status = decode_attr_mode(xdr, bitmap, &fmode); + if (status < 0) + goto xdr_error; + if (status != 0) { + fattr->mode |= fmode; + fattr->valid |= status; + } + + status = decode_attr_nlink(xdr, bitmap, &fattr->nlink); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_owner(xdr, bitmap, server, &fattr->uid, fattr->owner_name); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_group(xdr, bitmap, server, &fattr->gid, fattr->group_name); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_rdev(xdr, bitmap, &fattr->rdev); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_space_used(xdr, bitmap, &fattr->du.nfs3.used); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_time_access(xdr, bitmap, &fattr->atime); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_time_create(xdr, bitmap, &fattr->btime); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_time_metadata(xdr, bitmap, &fattr->ctime); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = decode_attr_mounted_on_fileid(xdr, bitmap, &fattr->mounted_on_fileid); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + + status = -EIO; + if (unlikely(bitmap[1])) + goto xdr_error; + + status = decode_attr_mdsthreshold(xdr, bitmap, fattr->mdsthreshold); + if (status < 0) + goto xdr_error; + + status = decode_attr_security_label(xdr, bitmap, fattr->label); + if (status < 0) + goto xdr_error; + fattr->valid |= status; + +xdr_error: + dprintk("%s: xdr returned %d\n", __func__, -status); + return status; +} + +static int decode_getfattr_generic(struct xdr_stream *xdr, struct nfs_fattr *fattr, + struct nfs_fh *fh, struct nfs4_fs_locations *fs_loc, + const struct nfs_server *server) +{ + unsigned int savep; + uint32_t attrlen, + bitmap[3] = {0}; + int status; + + status = decode_op_hdr(xdr, OP_GETATTR); + if (status < 0) + goto xdr_error; + + status = decode_attr_bitmap(xdr, bitmap); + if (status < 0) + goto xdr_error; + + status = decode_attr_length(xdr, &attrlen, &savep); + if (status < 0) + goto xdr_error; + + status = decode_getfattr_attrs(xdr, bitmap, fattr, fh, fs_loc, server); + if (status < 0) + goto xdr_error; + + status = verify_attr_len(xdr, savep, attrlen); +xdr_error: + dprintk("%s: xdr returned %d\n", __func__, -status); + return status; +} + +static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, + const struct nfs_server *server) +{ + return decode_getfattr_generic(xdr, fattr, NULL, NULL, server); +} + +/* + * Decode potentially multiple layout types. + */ +static int decode_pnfs_layout_types(struct xdr_stream *xdr, + struct nfs_fsinfo *fsinfo) +{ + __be32 *p; + uint32_t i; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + fsinfo->nlayouttypes = be32_to_cpup(p); + + /* pNFS is not supported by the underlying file system */ + if (fsinfo->nlayouttypes == 0) + return 0; + + /* Decode and set first layout type, move xdr->p past unused types */ + p = xdr_inline_decode(xdr, fsinfo->nlayouttypes * 4); + if (unlikely(!p)) + return -EIO; + + /* If we get too many, then just cap it at the max */ + if (fsinfo->nlayouttypes > NFS_MAX_LAYOUT_TYPES) { + printk(KERN_INFO "NFS: %s: Warning: Too many (%u) pNFS layout types\n", + __func__, fsinfo->nlayouttypes); + fsinfo->nlayouttypes = NFS_MAX_LAYOUT_TYPES; + } + + for(i = 0; i < fsinfo->nlayouttypes; ++i) + fsinfo->layouttype[i] = be32_to_cpup(p++); + return 0; +} + +/* + * The type of file system exported. + * Note we must ensure that layouttype is set in any non-error case. + */ +static int decode_attr_pnfstype(struct xdr_stream *xdr, uint32_t *bitmap, + struct nfs_fsinfo *fsinfo) +{ + int status = 0; + + dprintk("%s: bitmap is %x\n", __func__, bitmap[1]); + if (unlikely(bitmap[1] & (FATTR4_WORD1_FS_LAYOUT_TYPES - 1U))) + return -EIO; + if (bitmap[1] & FATTR4_WORD1_FS_LAYOUT_TYPES) { + status = decode_pnfs_layout_types(xdr, fsinfo); + bitmap[1] &= ~FATTR4_WORD1_FS_LAYOUT_TYPES; + } + return status; +} + +/* + * The preferred block size for layout directed io + */ +static int decode_attr_layout_blksize(struct xdr_stream *xdr, uint32_t *bitmap, + uint32_t *res) +{ + __be32 *p; + + dprintk("%s: bitmap is %x\n", __func__, bitmap[2]); + *res = 0; + if (bitmap[2] & FATTR4_WORD2_LAYOUT_BLKSIZE) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[2] &= ~FATTR4_WORD2_LAYOUT_BLKSIZE; + } + return 0; +} + +/* + * The granularity of a CLONE operation. + */ +static int decode_attr_clone_blksize(struct xdr_stream *xdr, uint32_t *bitmap, + uint32_t *res) +{ + __be32 *p; + + dprintk("%s: bitmap is %x\n", __func__, bitmap[2]); + *res = 0; + if (bitmap[2] & FATTR4_WORD2_CLONE_BLKSIZE) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + *res = be32_to_cpup(p); + bitmap[2] &= ~FATTR4_WORD2_CLONE_BLKSIZE; + } + return 0; +} + +static int decode_attr_change_attr_type(struct xdr_stream *xdr, + uint32_t *bitmap, + enum nfs4_change_attr_type *res) +{ + u32 tmp = NFS4_CHANGE_TYPE_IS_UNDEFINED; + + dprintk("%s: bitmap is %x\n", __func__, bitmap[2]); + if (bitmap[2] & FATTR4_WORD2_CHANGE_ATTR_TYPE) { + if (xdr_stream_decode_u32(xdr, &tmp)) + return -EIO; + bitmap[2] &= ~FATTR4_WORD2_CHANGE_ATTR_TYPE; + } + + switch(tmp) { + case NFS4_CHANGE_TYPE_IS_MONOTONIC_INCR: + case NFS4_CHANGE_TYPE_IS_VERSION_COUNTER: + case NFS4_CHANGE_TYPE_IS_VERSION_COUNTER_NOPNFS: + case NFS4_CHANGE_TYPE_IS_TIME_METADATA: + *res = tmp; + break; + default: + *res = NFS4_CHANGE_TYPE_IS_UNDEFINED; + } + return 0; +} + +static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo) +{ + unsigned int savep; + uint32_t attrlen, bitmap[3]; + int status; + + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto xdr_error; + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto xdr_error; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto xdr_error; + + fsinfo->rtmult = fsinfo->wtmult = 512; /* ??? */ + + if ((status = decode_attr_lease_time(xdr, bitmap, &fsinfo->lease_time)) != 0) + goto xdr_error; + if ((status = decode_attr_maxfilesize(xdr, bitmap, &fsinfo->maxfilesize)) != 0) + goto xdr_error; + if ((status = decode_attr_maxread(xdr, bitmap, &fsinfo->rtmax)) != 0) + goto xdr_error; + fsinfo->rtpref = fsinfo->dtpref = fsinfo->rtmax; + if ((status = decode_attr_maxwrite(xdr, bitmap, &fsinfo->wtmax)) != 0) + goto xdr_error; + fsinfo->wtpref = fsinfo->wtmax; + + status = -EIO; + if (unlikely(bitmap[0])) + goto xdr_error; + + status = decode_attr_time_delta(xdr, bitmap, &fsinfo->time_delta); + if (status != 0) + goto xdr_error; + status = decode_attr_pnfstype(xdr, bitmap, fsinfo); + if (status != 0) + goto xdr_error; + + status = -EIO; + if (unlikely(bitmap[1])) + goto xdr_error; + + status = decode_attr_layout_blksize(xdr, bitmap, &fsinfo->blksize); + if (status) + goto xdr_error; + status = decode_attr_clone_blksize(xdr, bitmap, &fsinfo->clone_blksize); + if (status) + goto xdr_error; + + status = decode_attr_change_attr_type(xdr, bitmap, + &fsinfo->change_attr_type); + if (status) + goto xdr_error; + + status = decode_attr_xattrsupport(xdr, bitmap, + &fsinfo->xattr_support); + if (status) + goto xdr_error; + + status = verify_attr_len(xdr, savep, attrlen); +xdr_error: + dprintk("%s: xdr returned %d!\n", __func__, -status); + return status; +} + +static int decode_getfh(struct xdr_stream *xdr, struct nfs_fh *fh) +{ + __be32 *p; + uint32_t len; + int status; + + /* Zero handle first to allow comparisons */ + memset(fh, 0, sizeof(*fh)); + + status = decode_op_hdr(xdr, OP_GETFH); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + if (len > NFS4_FHSIZE || len == 0) { + trace_nfs4_xdr_bad_filehandle(xdr, OP_GETFH, NFS4ERR_BADHANDLE); + return -EREMOTEIO; + } + fh->size = len; + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + memcpy(fh->data, p, len); + return 0; +} + +static int decode_link(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) +{ + int status; + + status = decode_op_hdr(xdr, OP_LINK); + if (status) + return status; + return decode_change_info(xdr, cinfo); +} + +/* + * We create the owner, so we know a proper owner.id length is 4. + */ +static int decode_lock_denied(struct xdr_stream *xdr, struct file_lock *fl) +{ + uint64_t offset, length, clientid; + __be32 *p; + uint32_t namelen, type; + + p = xdr_inline_decode(xdr, 32); /* read 32 bytes */ + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &offset); /* read 2 8-byte long words */ + p = xdr_decode_hyper(p, &length); + type = be32_to_cpup(p++); /* 4 byte read */ + if (fl != NULL) { /* manipulate file lock */ + fl->fl_start = (loff_t)offset; + fl->fl_end = fl->fl_start + (loff_t)length - 1; + if (length == ~(uint64_t)0) + fl->fl_end = OFFSET_MAX; + fl->c.flc_type = F_WRLCK; + if (type & 1) + fl->c.flc_type = F_RDLCK; + fl->c.flc_pid = 0; + } + p = xdr_decode_hyper(p, &clientid); /* read 8 bytes */ + namelen = be32_to_cpup(p); /* read 4 bytes */ /* have read all 32 bytes now */ + p = xdr_inline_decode(xdr, namelen); /* variable size field */ + if (likely(!p)) + return -EIO; + return -NFS4ERR_DENIED; +} + +static int decode_lock(struct xdr_stream *xdr, struct nfs_lock_res *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_LOCK); + if (status == -EIO) + goto out; + if (status == 0) { + status = decode_lock_stateid(xdr, &res->stateid); + if (unlikely(status)) + goto out; + } else if (status == -NFS4ERR_DENIED) + status = decode_lock_denied(xdr, NULL); + if (res->open_seqid != NULL) + nfs_increment_open_seqid(status, res->open_seqid); + nfs_increment_lock_seqid(status, res->lock_seqid); +out: + return status; +} + +static int decode_lockt(struct xdr_stream *xdr, struct nfs_lockt_res *res) +{ + int status; + status = decode_op_hdr(xdr, OP_LOCKT); + if (status == -NFS4ERR_DENIED) + return decode_lock_denied(xdr, res->denied); + return status; +} + +static int decode_locku(struct xdr_stream *xdr, struct nfs_locku_res *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_LOCKU); + if (status != -EIO) + nfs_increment_lock_seqid(status, res->seqid); + if (status == 0) + status = decode_lock_stateid(xdr, &res->stateid); + return status; +} + +#if defined(CONFIG_NFS_V4_0) +static int decode_release_lockowner(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_RELEASE_LOCKOWNER); +} +#endif /* CONFIG_NFS_V4_0 */ + +static int decode_lookup(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_LOOKUP); +} + +static int decode_lookupp(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_LOOKUPP); +} + +/* This is too sick! */ +static int decode_space_limit(struct xdr_stream *xdr, + unsigned long *pagemod_limit) +{ + __be32 *p; + uint32_t limit_type, nblocks, blocksize; + u64 maxsize = 0; + + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EIO; + limit_type = be32_to_cpup(p++); + switch (limit_type) { + case NFS4_LIMIT_SIZE: + xdr_decode_hyper(p, &maxsize); + break; + case NFS4_LIMIT_BLOCKS: + nblocks = be32_to_cpup(p++); + blocksize = be32_to_cpup(p); + maxsize = (uint64_t)nblocks * (uint64_t)blocksize; + } + maxsize >>= PAGE_SHIFT; + *pagemod_limit = min_t(u64, maxsize, ULONG_MAX); + return 0; +} + +static int decode_rw_delegation(struct xdr_stream *xdr, + struct nfs4_open_delegation *res) +{ + __be32 *p; + int status; + + status = decode_delegation_stateid(xdr, &res->stateid); + if (unlikely(status)) + return status; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->do_recall = be32_to_cpup(p); + + switch (res->open_delegation_type) { + case NFS4_OPEN_DELEGATE_READ: + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + res->type = FMODE_READ; + break; + case NFS4_OPEN_DELEGATE_WRITE: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + res->type = FMODE_WRITE|FMODE_READ; + if (decode_space_limit(xdr, &res->pagemod_limit) < 0) + return -EIO; + } + return decode_ace(xdr, NULL); +} + +static int decode_no_delegation(struct xdr_stream *xdr, + struct nfs4_open_delegation *res) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->why_no_delegation = be32_to_cpup(p); + switch (res->why_no_delegation) { + case WND4_CONTENTION: + case WND4_RESOURCE: + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->will_notify = be32_to_cpup(p); + } + return 0; +} + +static int decode_delegation(struct xdr_stream *xdr, + struct nfs4_open_delegation *res) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->open_delegation_type = be32_to_cpup(p); + switch (res->open_delegation_type) { + case NFS4_OPEN_DELEGATE_NONE: + return 0; + case NFS4_OPEN_DELEGATE_READ: + case NFS4_OPEN_DELEGATE_WRITE: + case NFS4_OPEN_DELEGATE_READ_ATTRS_DELEG: + case NFS4_OPEN_DELEGATE_WRITE_ATTRS_DELEG: + return decode_rw_delegation(xdr, res); + case NFS4_OPEN_DELEGATE_NONE_EXT: + return decode_no_delegation(xdr, res); + } + return -EIO; +} + +static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) +{ + __be32 *p; + uint32_t savewords, bmlen, i; + int status; + + if (!__decode_op_hdr(xdr, OP_OPEN, &status)) + return status; + nfs_increment_open_seqid(status, res->seqid); + if (status) + return status; + status = decode_open_stateid(xdr, &res->stateid); + if (unlikely(status)) + return status; + + decode_change_info(xdr, &res->cinfo); + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + res->rflags = be32_to_cpup(p++); + bmlen = be32_to_cpup(p); + if (bmlen > 10) + goto xdr_error; + + p = xdr_inline_decode(xdr, bmlen << 2); + if (unlikely(!p)) + return -EIO; + savewords = min_t(uint32_t, bmlen, NFS4_BITMAP_SIZE); + for (i = 0; i < savewords; ++i) + res->attrset[i] = be32_to_cpup(p++); + for (; i < NFS4_BITMAP_SIZE; i++) + res->attrset[i] = 0; + + return decode_delegation(xdr, &res->delegation); +xdr_error: + dprintk("%s: Bitmap too large! Length = %u\n", __func__, bmlen); + return -EIO; +} + +static int decode_open_confirm(struct xdr_stream *xdr, struct nfs_open_confirmres *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_OPEN_CONFIRM); + if (status != -EIO) + nfs_increment_open_seqid(status, res->seqid); + if (!status) + status = decode_open_stateid(xdr, &res->stateid); + return status; +} + +static int decode_open_downgrade(struct xdr_stream *xdr, struct nfs_closeres *res) +{ + int status; + + status = decode_op_hdr(xdr, OP_OPEN_DOWNGRADE); + if (status != -EIO) + nfs_increment_open_seqid(status, res->seqid); + if (!status) + status = decode_open_stateid(xdr, &res->stateid); + return status; +} + +static int decode_putfh(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_PUTFH); +} + +static int decode_putrootfh(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_PUTROOTFH); +} + +static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs_pgio_res *res) +{ + __be32 *p; + uint32_t count, eof, recvd; + int status; + + status = decode_op_hdr(xdr, OP_READ); + if (status) + return status; + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + eof = be32_to_cpup(p++); + count = be32_to_cpup(p); + recvd = xdr_read_pages(xdr, count); + if (count > recvd) { + dprintk("NFS: server cheating in read reply: " + "count %u > recvd %u\n", count, recvd); + count = recvd; + eof = 0; + } + res->eof = eof; + res->count = count; + return 0; +} + +static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs4_readdir_res *readdir) +{ + int status; + __be32 verf[2]; + + status = decode_op_hdr(xdr, OP_READDIR); + if (!status) + status = decode_verifier(xdr, readdir->verifier.data); + if (unlikely(status)) + return status; + memcpy(verf, readdir->verifier.data, sizeof(verf)); + dprintk("%s: verifier = %08x:%08x\n", + __func__, verf[0], verf[1]); + return xdr_read_pages(xdr, xdr->buf->page_len); +} + +static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req) +{ + struct xdr_buf *rcvbuf = &req->rq_rcv_buf; + u32 len, recvd; + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_READLINK); + if (status) + return status; + + /* Convert length of symlink */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + if (len >= rcvbuf->page_len || len <= 0) { + dprintk("nfs: server returned giant symlink!\n"); + return -ENAMETOOLONG; + } + recvd = xdr_read_pages(xdr, len); + if (recvd < len) { + dprintk("NFS: server cheating in readlink reply: " + "count %u > recvd %u\n", len, recvd); + return -EIO; + } + /* + * The XDR encode routine has set things up so that + * the link text will be copied directly into the + * buffer. We just have to do overflow-checking, + * and null-terminate the text (the VFS expects + * null-termination). + */ + xdr_terminate_string(rcvbuf, len); + return 0; +} + +static int decode_remove(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) +{ + int status; + + status = decode_op_hdr(xdr, OP_REMOVE); + if (status) + goto out; + status = decode_change_info(xdr, cinfo); +out: + return status; +} + +static int decode_rename(struct xdr_stream *xdr, struct nfs4_change_info *old_cinfo, + struct nfs4_change_info *new_cinfo) +{ + int status; + + status = decode_op_hdr(xdr, OP_RENAME); + if (status) + goto out; + if ((status = decode_change_info(xdr, old_cinfo))) + goto out; + status = decode_change_info(xdr, new_cinfo); +out: + return status; +} + +static int decode_renew(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_RENEW); +} + +static int +decode_restorefh(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_RESTOREFH); +} + +static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs_getaclres *res, enum nfs4_acl_type type) +{ + unsigned int savep; + uint32_t attrlen, + bitmap[3] = {0}; + int status; + + res->acl_len = 0; + if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) + goto out; + + xdr_enter_page(xdr, xdr->buf->page_len); + + if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) + goto out; + if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) + goto out; + + switch (type) { + default: + if (unlikely(bitmap[0] & (FATTR4_WORD0_ACL - 1U))) + return -EIO; + if (!(bitmap[0] & FATTR4_WORD0_ACL)) + return -EOPNOTSUPP; + break; + case NFS4ACL_DACL: + if (unlikely(bitmap[0] || bitmap[1] & (FATTR4_WORD1_DACL - 1U))) + return -EIO; + if (!(bitmap[1] & FATTR4_WORD1_DACL)) + return -EOPNOTSUPP; + break; + case NFS4ACL_SACL: + if (unlikely(bitmap[0] || bitmap[1] & (FATTR4_WORD1_SACL - 1U))) + return -EIO; + if (!(bitmap[1] & FATTR4_WORD1_SACL)) + return -EOPNOTSUPP; + } + + /* The bitmap (xdr len + bitmaps) and the attr xdr len words + * are stored with the acl data to handle the problem of + * variable length bitmaps.*/ + res->acl_data_offset = xdr_page_pos(xdr); + res->acl_len = attrlen; + + /* Check for receive buffer overflow */ + if (res->acl_len > xdr_stream_remaining(xdr) || + res->acl_len + res->acl_data_offset > xdr->buf->page_len) { + res->acl_flags |= NFS4_ACL_TRUNC; + dprintk("NFS: acl reply: attrlen %u > page_len %zu\n", + attrlen, xdr_stream_remaining(xdr)); + } +out: + return status; +} + +static int +decode_savefh(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_SAVEFH); +} + +static int decode_setattr(struct xdr_stream *xdr) +{ + int status; + + status = decode_op_hdr(xdr, OP_SETATTR); + if (status) + return status; + if (decode_bitmap4(xdr, NULL, 0) >= 0) + return 0; + return -EIO; +} + +static int decode_delegattr(struct xdr_stream *xdr) +{ + return decode_setattr(xdr); +} + +static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_setclientid_res *res) +{ + __be32 *p; + uint32_t opnum; + int32_t nfserr; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + opnum = be32_to_cpup(p++); + if (opnum != OP_SETCLIENTID) { + dprintk("nfs: decode_setclientid: Server returned operation" + " %d\n", opnum); + return -EIO; + } + nfserr = be32_to_cpup(p); + if (nfserr == NFS_OK) { + p = xdr_inline_decode(xdr, 8 + NFS4_VERIFIER_SIZE); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &res->clientid); + memcpy(res->confirm.data, p, NFS4_VERIFIER_SIZE); + } else if (nfserr == NFSERR_CLID_INUSE) { + uint32_t len; + + /* skip netid string */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + + /* skip uaddr string */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return -EIO; + return -NFSERR_CLID_INUSE; + } else + return nfs4_stat_to_errno(nfserr); + + return 0; +} + +static int decode_setclientid_confirm(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_SETCLIENTID_CONFIRM); +} + +static int decode_write(struct xdr_stream *xdr, struct nfs_pgio_res *res) +{ + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_WRITE); + if (status) + return status; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + res->count = be32_to_cpup(p++); + res->verf->committed = be32_to_cpup(p++); + return decode_write_verifier(xdr, &res->verf->verifier); +} + +static int decode_delegreturn(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_DELEGRETURN); +} + +static int decode_secinfo_gss(struct xdr_stream *xdr, + struct nfs4_secinfo4 *flavor) +{ + u32 oid_len; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + oid_len = be32_to_cpup(p); + if (oid_len > GSS_OID_MAX_LEN) + return -EINVAL; + + p = xdr_inline_decode(xdr, oid_len); + if (unlikely(!p)) + return -EIO; + memcpy(flavor->flavor_info.oid.data, p, oid_len); + flavor->flavor_info.oid.len = oid_len; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + flavor->flavor_info.qop = be32_to_cpup(p++); + flavor->flavor_info.service = be32_to_cpup(p); + + return 0; +} + +static int decode_secinfo_common(struct xdr_stream *xdr, struct nfs4_secinfo_res *res) +{ + struct nfs4_secinfo4 *sec_flavor; + unsigned int i, num_flavors; + int status; + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + res->flavors->num_flavors = 0; + num_flavors = be32_to_cpup(p); + + for (i = 0; i < num_flavors; i++) { + sec_flavor = &res->flavors->flavors[i]; + if ((char *)&sec_flavor[1] - (char *)res->flavors > PAGE_SIZE) + break; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + sec_flavor->flavor = be32_to_cpup(p); + + if (sec_flavor->flavor == RPC_AUTH_GSS) { + status = decode_secinfo_gss(xdr, sec_flavor); + if (status) + goto out; + } + res->flavors->num_flavors++; + } + + status = 0; +out: + return status; +} + +static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res) +{ + int status = decode_op_hdr(xdr, OP_SECINFO); + if (status) + return status; + return decode_secinfo_common(xdr, res); +} + +static int decode_secinfo_no_name(struct xdr_stream *xdr, struct nfs4_secinfo_res *res) +{ + int status = decode_op_hdr(xdr, OP_SECINFO_NO_NAME); + if (status) + return status; + return decode_secinfo_common(xdr, res); +} + +static int decode_op_map(struct xdr_stream *xdr, struct nfs4_op_map *op_map) +{ + if (xdr_stream_decode_uint32_array(xdr, op_map->u.words, + ARRAY_SIZE(op_map->u.words)) < 0) + return -EIO; + return 0; +} + +static int decode_exchange_id(struct xdr_stream *xdr, + struct nfs41_exchange_id_res *res) +{ + __be32 *p; + uint32_t dummy; + char *dummy_str; + int status; + uint32_t impl_id_count; + + status = decode_op_hdr(xdr, OP_EXCHANGE_ID); + if (status) + return status; + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + xdr_decode_hyper(p, &res->clientid); + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EIO; + res->seqid = be32_to_cpup(p++); + res->flags = be32_to_cpup(p++); + + res->state_protect.how = be32_to_cpup(p); + switch (res->state_protect.how) { + case SP4_NONE: + break; + case SP4_MACH_CRED: + status = decode_op_map(xdr, &res->state_protect.enforce); + if (status) + return status; + status = decode_op_map(xdr, &res->state_protect.allow); + if (status) + return status; + break; + default: + WARN_ON_ONCE(1); + return -EIO; + } + + /* server_owner4.so_minor_id */ + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &res->server_owner->minor_id); + + /* server_owner4.so_major_id */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + memcpy(res->server_owner->major_id, dummy_str, dummy); + res->server_owner->major_id_sz = dummy; + + /* server_scope4 */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + memcpy(res->server_scope->server_scope, dummy_str, dummy); + res->server_scope->server_scope_sz = dummy; + + /* Implementation Id */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + impl_id_count = be32_to_cpup(p++); + + if (impl_id_count) { + /* nii_domain */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + memcpy(res->impl_id->domain, dummy_str, dummy); + + /* nii_name */ + status = decode_opaque_inline(xdr, &dummy, &dummy_str); + if (unlikely(status)) + return status; + memcpy(res->impl_id->name, dummy_str, dummy); + + /* nii_date */ + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EIO; + p = xdr_decode_hyper(p, &res->impl_id->date.seconds); + res->impl_id->date.nseconds = be32_to_cpup(p); + + /* if there's more than one entry, ignore the rest */ + } + return 0; +} + +static int decode_chan_attrs(struct xdr_stream *xdr, + struct nfs4_channel_attrs *attrs) +{ + __be32 *p; + u32 nr_attrs, val; + + p = xdr_inline_decode(xdr, 28); + if (unlikely(!p)) + return -EIO; + val = be32_to_cpup(p++); /* headerpadsz */ + if (val) + return -EINVAL; /* no support for header padding yet */ + attrs->max_rqst_sz = be32_to_cpup(p++); + attrs->max_resp_sz = be32_to_cpup(p++); + attrs->max_resp_sz_cached = be32_to_cpup(p++); + attrs->max_ops = be32_to_cpup(p++); + attrs->max_reqs = be32_to_cpup(p++); + nr_attrs = be32_to_cpup(p); + if (unlikely(nr_attrs > 1)) { + printk(KERN_WARNING "NFS: %s: Invalid rdma channel attrs " + "count %u\n", __func__, nr_attrs); + return -EINVAL; + } + if (nr_attrs == 1) { + p = xdr_inline_decode(xdr, 4); /* skip rdma_attrs */ + if (unlikely(!p)) + return -EIO; + } + return 0; +} + +static int decode_sessionid(struct xdr_stream *xdr, struct nfs4_sessionid *sid) +{ + return decode_opaque_fixed(xdr, sid->data, NFS4_MAX_SESSIONID_LEN); +} + +static int decode_bind_conn_to_session(struct xdr_stream *xdr, + struct nfs41_bind_conn_to_session_res *res) +{ + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_BIND_CONN_TO_SESSION); + if (!status) + status = decode_sessionid(xdr, &res->sessionid); + if (unlikely(status)) + return status; + + /* dir flags, rdma mode bool */ + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + + res->dir = be32_to_cpup(p++); + if (res->dir == 0 || res->dir > NFS4_CDFS4_BOTH) + return -EIO; + if (be32_to_cpup(p) == 0) + res->use_conn_in_rdma_mode = false; + else + res->use_conn_in_rdma_mode = true; + + return 0; +} + +static int decode_create_session(struct xdr_stream *xdr, + struct nfs41_create_session_res *res) +{ + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_CREATE_SESSION); + if (!status) + status = decode_sessionid(xdr, &res->sessionid); + if (unlikely(status)) + return status; + + /* seqid, flags */ + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + res->seqid = be32_to_cpup(p++); + res->flags = be32_to_cpup(p); + + /* Channel attributes */ + status = decode_chan_attrs(xdr, &res->fc_attrs); + if (!status) + status = decode_chan_attrs(xdr, &res->bc_attrs); + return status; +} + +static int decode_destroy_session(struct xdr_stream *xdr, void *dummy) +{ + return decode_op_hdr(xdr, OP_DESTROY_SESSION); +} + +static int decode_destroy_clientid(struct xdr_stream *xdr, void *dummy) +{ + return decode_op_hdr(xdr, OP_DESTROY_CLIENTID); +} + +static int decode_reclaim_complete(struct xdr_stream *xdr, void *dummy) +{ + return decode_op_hdr(xdr, OP_RECLAIM_COMPLETE); +} + +static int decode_sequence(struct xdr_stream *xdr, + struct nfs4_sequence_res *res, + struct rpc_rqst *rqstp) +{ + struct nfs4_session *session; + struct nfs4_sessionid id; + u32 dummy; + int status; + __be32 *p; + + if (res->sr_slot == NULL) + return 0; + if (!res->sr_slot->table->session) + return 0; + + status = decode_op_hdr(xdr, OP_SEQUENCE); + if (!status) + status = decode_sessionid(xdr, &id); + if (unlikely(status)) + goto out_err; + + /* + * If the server returns different values for sessionID, slotID or + * sequence number, the server is looney tunes. + */ + status = -EREMOTEIO; + session = res->sr_slot->table->session; + + if (memcmp(id.data, session->sess_id.data, + NFS4_MAX_SESSIONID_LEN)) { + dprintk("%s Invalid session id\n", __func__); + goto out_err; + } + + p = xdr_inline_decode(xdr, 20); + if (unlikely(!p)) + goto out_overflow; + + /* seqid */ + dummy = be32_to_cpup(p++); + if (dummy != res->sr_slot->seq_nr) { + dprintk("%s Invalid sequence number\n", __func__); + goto out_err; + } + /* slot id */ + dummy = be32_to_cpup(p++); + if (dummy != res->sr_slot->slot_nr) { + dprintk("%s Invalid slot id\n", __func__); + goto out_err; + } + /* highest slot id */ + res->sr_highest_slotid = be32_to_cpup(p++); + /* target highest slot id */ + res->sr_target_highest_slotid = be32_to_cpup(p++); + /* result flags */ + res->sr_status_flags = be32_to_cpup(p); + status = 0; +out_err: + res->sr_status = status; + return status; +out_overflow: + status = -EIO; + goto out_err; +} + +static int decode_layout_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +{ + stateid->type = NFS4_LAYOUT_STATEID_TYPE; + return decode_stateid(xdr, stateid); +} + +static int decode_get_dir_delegation(struct xdr_stream *xdr, + struct nfs4_getattr_res *res) +{ + struct nfs4_gdd_res *gdd_res = res->gdd_res; + nfs4_verifier cookieverf; + u32 bitmap[1]; + int status; + + status = decode_op_hdr(xdr, OP_GET_DIR_DELEGATION); + if (status) + return status; + + if (xdr_stream_decode_u32(xdr, &gdd_res->status)) + return -EIO; + + if (gdd_res->status == GDD4_UNAVAIL) + return xdr_inline_decode(xdr, 4) ? 0 : -EIO; + + status = decode_verifier(xdr, &cookieverf); + if (status) + return status; + + status = decode_delegation_stateid(xdr, &gdd_res->deleg); + if (status) + return status; + + /* Decode supported notification types. */ + status = decode_bitmap4(xdr, bitmap, ARRAY_SIZE(bitmap)); + if (status < 0) + return status; + + /* Decode supported child attributes. */ + status = decode_bitmap4(xdr, bitmap, ARRAY_SIZE(bitmap)); + if (status < 0) + return status; + + /* Decode supported attributes. */ + status = decode_bitmap4(xdr, bitmap, ARRAY_SIZE(bitmap)); + if (status < 0) + return status; + return 0; +} + +static int decode_getdeviceinfo(struct xdr_stream *xdr, + struct nfs4_getdeviceinfo_res *res) +{ + struct pnfs_device *pdev = res->pdev; + __be32 *p; + uint32_t len, type; + int status; + + status = decode_op_hdr(xdr, OP_GETDEVICEINFO); + if (status) { + if (status == -ETOOSMALL) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + pdev->mincount = be32_to_cpup(p); + dprintk("%s: Min count too small. mincnt = %u\n", + __func__, pdev->mincount); + } + return status; + } + + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + type = be32_to_cpup(p++); + if (type != pdev->layout_type) { + dprintk("%s: layout mismatch req: %u pdev: %u\n", + __func__, pdev->layout_type, type); + return -EINVAL; + } + /* + * Get the length of the opaque device_addr4. xdr_read_pages places + * the opaque device_addr4 in the xdr_buf->pages (pnfs_device->pages) + * and places the remaining xdr data in xdr_buf->tail + */ + pdev->mincount = be32_to_cpup(p); + if (xdr_read_pages(xdr, pdev->mincount) != pdev->mincount) + return -EIO; + + /* Parse notification bitmap, verifying that it is zero. */ + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + len = be32_to_cpup(p); + if (len) { + uint32_t i; + + p = xdr_inline_decode(xdr, 4 * len); + if (unlikely(!p)) + return -EIO; + + res->notification = be32_to_cpup(p++); + for (i = 1; i < len; i++) { + if (be32_to_cpup(p++)) { + dprintk("%s: unsupported notification\n", + __func__); + return -EIO; + } + } + } + return 0; +} + +static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs4_layoutget_res *res) +{ + __be32 *p; + int status; + u32 layout_count; + u32 recvd; + + status = decode_op_hdr(xdr, OP_LAYOUTGET); + if (status) + goto out; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out_overflow; + res->return_on_close = be32_to_cpup(p); + decode_layout_stateid(xdr, &res->stateid); + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out_overflow; + layout_count = be32_to_cpup(p); + if (!layout_count) { + dprintk("%s: server responded with empty layout array\n", + __func__); + status = -EINVAL; + goto out; + } + + p = xdr_inline_decode(xdr, 28); + if (unlikely(!p)) + goto out_overflow; + p = xdr_decode_hyper(p, &res->range.offset); + p = xdr_decode_hyper(p, &res->range.length); + res->range.iomode = be32_to_cpup(p++); + res->type = be32_to_cpup(p++); + res->layoutp->len = be32_to_cpup(p); + + dprintk("%s roff:%lu rlen:%lu riomode:%d, lo_type:0x%x, lo.len:%d\n", + __func__, + (unsigned long)res->range.offset, + (unsigned long)res->range.length, + res->range.iomode, + res->type, + res->layoutp->len); + + recvd = xdr_read_pages(xdr, res->layoutp->len); + if (res->layoutp->len > recvd) { + dprintk("NFS: server cheating in layoutget reply: " + "layout len %u > recvd %u\n", + res->layoutp->len, recvd); + status = -EINVAL; + goto out; + } + + if (layout_count > 1) { + /* We only handle a length one array at the moment. Any + * further entries are just ignored. Note that this means + * the client may see a response that is less than the + * minimum it requested. + */ + dprintk("%s: server responded with %d layouts, dropping tail\n", + __func__, layout_count); + } + +out: + res->status = status; + return status; +out_overflow: + status = -EIO; + goto out; +} + +static int decode_layoutreturn(struct xdr_stream *xdr, + struct nfs4_layoutreturn_res *res) +{ + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_LAYOUTRETURN); + if (status) + return status; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->lrs_present = be32_to_cpup(p); + if (res->lrs_present) + status = decode_layout_stateid(xdr, &res->stateid); + else + nfs4_stateid_copy(&res->stateid, &invalid_stateid); + return status; +} + +static int decode_layoutcommit(struct xdr_stream *xdr, + struct rpc_rqst *req, + struct nfs4_layoutcommit_res *res) +{ + __be32 *p; + __u32 sizechanged; + int status; + + status = decode_op_hdr(xdr, OP_LAYOUTCOMMIT); + res->status = status; + if (status) + return status; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + sizechanged = be32_to_cpup(p); + + if (sizechanged) { + /* throw away new size */ + p = xdr_inline_decode(xdr, 8); + if (unlikely(!p)) + return -EIO; + } + return 0; +} + +static int decode_test_stateid(struct xdr_stream *xdr, + struct nfs41_test_stateid_res *res) +{ + __be32 *p; + int status; + int num_res; + + status = decode_op_hdr(xdr, OP_TEST_STATEID); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + num_res = be32_to_cpup(p++); + if (num_res != 1) + return -EIO; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + res->status = be32_to_cpup(p++); + + return status; +} + +static int decode_free_stateid(struct xdr_stream *xdr, + struct nfs41_free_stateid_res *res) +{ + res->status = decode_op_hdr(xdr, OP_FREE_STATEID); + return res->status; +} + +/* + * END OF "GENERIC" DECODE ROUTINES. + */ + +/* + * Decode OPEN_DOWNGRADE response + */ +static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_closeres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + if (res->lr_res) { + status = decode_layoutreturn(xdr, res->lr_res); + res->lr_ret = status; + if (status) + goto out; + } + status = decode_open_downgrade(xdr, res); +out: + return status; +} + +/* + * Decode ACCESS response + */ +static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_accessres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status != 0) + goto out; + status = decode_access(xdr, &res->supported, &res->access); + if (status != 0) + goto out; + if (res->fattr) + decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode LOOKUP response + */ +static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_lookup_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_lookup(xdr); + if (status) + goto out; + status = decode_getfh(xdr, res->fh); + if (status) + goto out; + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode LOOKUPP response + */ +static int nfs4_xdr_dec_lookupp(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_lookupp_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_lookupp(xdr); + if (status) + goto out; + status = decode_getfh(xdr, res->fh); + if (status) + goto out; + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode LOOKUP_ROOT response + */ +static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_lookup_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putrootfh(xdr); + if (status) + goto out; + status = decode_getfh(xdr, res->fh); + if (status == 0) + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode REMOVE response + */ +static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_removeres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_remove(xdr, &res->cinfo); +out: + return status; +} + +/* + * Decode RENAME response + */ +static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_renameres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_savefh(xdr); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_rename(xdr, &res->old_cinfo, &res->new_cinfo); +out: + return status; +} + +/* + * Decode LINK response + */ +static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_link_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_savefh(xdr); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_link(xdr, &res->cinfo); + if (status) + goto out; + /* + * Note order: OP_LINK leaves the directory as the current + * filehandle. + */ + status = decode_restorefh(xdr); + if (status) + goto out; + decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode CREATE response + */ +static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_create_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_create(xdr, &res->dir_cinfo); + if (status) + goto out; + status = decode_getfh(xdr, res->fh); + if (status) + goto out; + decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode SYMLINK response + */ +static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *res) +{ + return nfs4_xdr_dec_create(rqstp, xdr, res); +} + +/* + * Decode GETATTR response + */ +static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_getattr_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + if (res->gdd_res) { + status = decode_get_dir_delegation(xdr, res); + if (status) + goto out; + } + status = decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Encode an SETACL request + */ +static void nfs4_xdr_enc_setacl(struct rpc_rqst *req, struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_setaclargs *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_setacl(xdr, args, &hdr); + encode_nops(&hdr); +} + +/* + * Decode SETACL response + */ +static int +nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_setaclres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_setattr(xdr); +out: + return status; +} + +/* + * Decode GETACL response + */ +static int +nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_getaclres *res = data; + struct compound_hdr hdr; + int status; + + if (res->acl_scratch != NULL) + xdr_set_scratch_folio(xdr, res->acl_scratch); + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_getacl(xdr, rqstp, res, res->acl_type); + +out: + return status; +} + +/* + * Decode CLOSE response + */ +static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_closeres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + if (res->lr_res) { + status = decode_layoutreturn(xdr, res->lr_res); + res->lr_ret = status; + if (status) + goto out; + } + if (res->fattr != NULL) { + status = decode_getfattr(xdr, res->fattr, res->server); + if (status != 0) + goto out; + } + status = decode_close(xdr, res); +out: + return status; +} + +/* + * Decode OPEN response + */ +static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_openres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_open(xdr, res); + if (status) + goto out; + status = decode_getfh(xdr, &res->fh); + if (status) + goto out; + if (res->access_request) + decode_access(xdr, &res->access_supported, &res->access_result); + decode_getfattr(xdr, res->f_attr, res->server); + if (res->lg_res) + decode_layoutget(xdr, rqstp, res->lg_res); +out: + return status; +} + +/* + * Decode OPEN_CONFIRM response + */ +static int nfs4_xdr_dec_open_confirm(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_open_confirmres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_open_confirm(xdr, res); +out: + return status; +} + +/* + * Decode OPEN response + */ +static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_openres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_open(xdr, res); + if (status) + goto out; + if (res->access_request) + decode_access(xdr, &res->access_supported, &res->access_result); + decode_getfattr(xdr, res->f_attr, res->server); + if (res->lg_res) + decode_layoutget(xdr, rqstp, res->lg_res); +out: + return status; +} + +/* + * Decode SETATTR response + */ +static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_setattrres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_setattr(xdr); + if (status) + goto out; + decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode LOCK response + */ +static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_lock_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_lock(xdr, res); +out: + return status; +} + +/* + * Decode LOCKT response + */ +static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_lockt_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_lockt(xdr, res); +out: + return status; +} + +/* + * Decode LOCKU response + */ +static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_locku_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_locku(xdr, res); +out: + return status; +} + +#if defined(CONFIG_NFS_V4_0) +static int nfs4_xdr_dec_release_lockowner(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, void *dummy) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_release_lockowner(xdr); + return status; +} +#endif /* CONFIG_NFS_V4_0 */ + +/* + * Decode READLINK response + */ +static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_readlink_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_readlink(xdr, rqstp); +out: + return status; +} + +/* + * Decode READDIR response + */ +static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_readdir_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_readdir(xdr, rqstp, res); +out: + return status; +} + +/* + * Decode Read response + */ +static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + res->op_status = hdr.status; + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_read(xdr, rqstp, res); + if (!status) + status = res->count; +out: + return status; +} + +/* + * Decode WRITE response + */ +static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + res->op_status = hdr.status; + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_write(xdr, res); + if (status) + goto out; + if (res->fattr) + decode_getfattr(xdr, res->fattr, res->server); + if (!status) + status = res->count; +out: + return status; +} + +/* + * Decode COMMIT response + */ +static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *data) +{ + struct nfs_commitres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + res->op_status = hdr.status; + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_commit(xdr, res); +out: + return status; +} + +/* + * Decode FSINFO response + */ +static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_fsinfo_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, &res->seq_res, req); + if (!status) + status = decode_putfh(xdr); + if (!status) + status = decode_fsinfo(xdr, res->fsinfo); + return status; +} + +/* + * Decode PATHCONF response + */ +static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_pathconf_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, &res->seq_res, req); + if (!status) + status = decode_putfh(xdr); + if (!status) + status = decode_pathconf(xdr, res->pathconf); + return status; +} + +/* + * Decode STATFS response + */ +static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, struct xdr_stream *xdr, + void *data) +{ + struct nfs4_statfs_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, &res->seq_res, req); + if (!status) + status = decode_putfh(xdr); + if (!status) + status = decode_statfs(xdr, res->fsstat); + return status; +} + +/* + * Decode GETATTR_BITMAP response + */ +static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_server_caps_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, req); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_server_caps(xdr, res); +out: + return status; +} + +/* + * Decode RENEW response + */ +#if defined(CONFIG_NFS_V4_0) +static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, struct xdr_stream *xdr, + void *__unused) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_renew(xdr); + return status; +} +#endif /* CONFIG_NFS_V4_0 */ + +/* + * Decode SETCLIENTID response + */ +static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_setclientid_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_setclientid(xdr, res); + return status; +} + +/* + * Decode SETCLIENTID_CONFIRM response + */ +static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_setclientid_confirm(xdr); + return status; +} + +/* + * Decode DELEGRETURN response + */ +static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_delegreturnres *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status != 0) + goto out; + if (res->lr_res) { + status = decode_layoutreturn(xdr, res->lr_res); + res->lr_ret = status; + if (status) + goto out; + } + if (res->sattr_res) { + status = decode_delegattr(xdr); + res->sattr_ret = status; + if (status) + goto out; + } + if (res->fattr) { + status = decode_getfattr(xdr, res->fattr, res->server); + if (status != 0) + goto out; + } + status = decode_delegreturn(xdr); +out: + return status; +} + +/* + * Decode FS_LOCATIONS response + */ +static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_fs_locations_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, req); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + if (res->migration) { + xdr_enter_page(xdr, PAGE_SIZE); + status = decode_getfattr_generic(xdr, + res->fs_locations->fattr, + NULL, res->fs_locations, + res->fs_locations->server); + if (status) + goto out; + if (res->renew) + status = decode_renew(xdr); + } else { + status = decode_lookup(xdr); + if (status) + goto out; + xdr_enter_page(xdr, PAGE_SIZE); + status = decode_getfattr_generic(xdr, + res->fs_locations->fattr, + NULL, res->fs_locations, + res->fs_locations->server); + } +out: + return status; +} + +/* + * Decode SECINFO response + */ +static int nfs4_xdr_dec_secinfo(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_secinfo_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_secinfo(xdr, res); +out: + return status; +} + +/* + * Decode FSID_PRESENT response + */ +static int nfs4_xdr_dec_fsid_present(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_fsid_present_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_getfh(xdr, res->fh); + if (status) + goto out; + if (res->renew) + status = decode_renew(xdr); +out: + return status; +} + +/* + * Decode BIND_CONN_TO_SESSION response + */ +static int nfs4_xdr_dec_bind_conn_to_session(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_bind_conn_to_session(xdr, res); + return status; +} + +/* + * Decode EXCHANGE_ID response + */ +static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_exchange_id(xdr, res); + return status; +} + +/* + * Decode CREATE_SESSION response + */ +static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_create_session(xdr, res); + return status; +} + +/* + * Decode DESTROY_SESSION response + */ +static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_destroy_session(xdr, res); + return status; +} + +/* + * Decode DESTROY_CLIENTID response + */ +static int nfs4_xdr_dec_destroy_clientid(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_destroy_clientid(xdr, res); + return status; +} + +/* + * Decode SEQUENCE response + */ +static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, res, rqstp); + return status; +} + +/* + * Decode GET_LEASE_TIME response + */ +static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_get_lease_time_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, &res->lr_seq_res, rqstp); + if (!status) + status = decode_putrootfh(xdr); + if (!status) + status = decode_fsinfo(xdr, res->lr_fsinfo); + return status; +} + +/* + * Decode RECLAIM_COMPLETE response + */ +static int nfs4_xdr_dec_reclaim_complete(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs41_reclaim_complete_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (!status) + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (!status) + status = decode_reclaim_complete(xdr, NULL); + return status; +} + +/* + * Decode GETDEVINFO response + */ +static int nfs4_xdr_dec_getdeviceinfo(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_getdeviceinfo_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status != 0) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status != 0) + goto out; + status = decode_getdeviceinfo(xdr, res); +out: + return status; +} + +/* + * Decode LAYOUTGET response + */ +static int nfs4_xdr_dec_layoutget(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_layoutget_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_layoutget(xdr, rqstp, res); +out: + return status; +} + +/* + * Decode LAYOUTRETURN response + */ +static int nfs4_xdr_dec_layoutreturn(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_layoutreturn_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_layoutreturn(xdr, res); +out: + return status; +} + +/* + * Decode LAYOUTCOMMIT response + */ +static int nfs4_xdr_dec_layoutcommit(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_layoutcommit_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_layoutcommit(xdr, rqstp, res); + if (status) + goto out; + decode_getfattr(xdr, res->fattr, res->server); +out: + return status; +} + +/* + * Decode SECINFO_NO_NAME response + */ +static int nfs4_xdr_dec_secinfo_no_name(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs4_secinfo_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putrootfh(xdr); + if (status) + goto out; + status = decode_secinfo_no_name(xdr, res); +out: + return status; +} + +/* + * Decode TEST_STATEID response + */ +static int nfs4_xdr_dec_test_stateid(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs41_test_stateid_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_test_stateid(xdr, res); +out: + return status; +} + +/* + * Decode FREE_STATEID response + */ +static int nfs4_xdr_dec_free_stateid(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs41_free_stateid_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_free_stateid(xdr, res); +out: + return status; +} + +/** + * nfs4_decode_dirent - Decode a single NFSv4 directory entry stored in + * the local page cache. + * @xdr: XDR stream where entry resides + * @entry: buffer to fill in with entry data + * @plus: boolean indicating whether this should be a readdirplus entry + * + * Returns zero if successful, otherwise a negative errno value is + * returned. + * + * This function is not invoked during READDIR reply decoding, but + * rather whenever an application invokes the getdents(2) system call + * on a directory already in our cache. + */ +int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, + bool plus) +{ + unsigned int savep; + uint32_t bitmap[3] = {0}; + uint32_t len; + uint64_t new_cookie; + __be32 *p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p == xdr_zero) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EAGAIN; + if (*p == xdr_zero) + return -EAGAIN; + entry->eof = 1; + return -EBADCOOKIE; + } + + p = xdr_inline_decode(xdr, 12); + if (unlikely(!p)) + return -EAGAIN; + p = xdr_decode_hyper(p, &new_cookie); + entry->len = be32_to_cpup(p); + + p = xdr_inline_decode(xdr, entry->len); + if (unlikely(!p)) + return -EAGAIN; + entry->name = (const char *) p; + + /* + * In case the server doesn't return an inode number, + * we fake one here. (We don't use inode number 0, + * since glibc seems to choke on it...) + */ + entry->ino = 1; + entry->fattr->valid = 0; + + if (decode_attr_bitmap(xdr, bitmap) < 0) + return -EAGAIN; + + if (decode_attr_length(xdr, &len, &savep) < 0) + return -EAGAIN; + + if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh, + NULL, entry->server) < 0) + return -EAGAIN; + if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) + entry->ino = entry->fattr->mounted_on_fileid; + else if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID) + entry->ino = entry->fattr->fileid; + + entry->d_type = DT_UNKNOWN; + if (entry->fattr->valid & NFS_ATTR_FATTR_TYPE) + entry->d_type = nfs_umode_to_dtype(entry->fattr->mode); + + entry->cookie = new_cookie; + + return 0; +} + +#ifdef CONFIG_NFS_V4_2 +#include "nfs42xdr.c" +#endif /* CONFIG_NFS_V4_2 */ + +#define PROC(proc, argtype, restype) \ +[NFSPROC4_CLNT_##proc] = { \ + .p_proc = NFSPROC4_COMPOUND, \ + .p_encode = nfs4_xdr_##argtype, \ + .p_decode = nfs4_xdr_##restype, \ + .p_arglen = NFS4_##argtype##_sz, \ + .p_replen = NFS4_##restype##_sz, \ + .p_statidx = NFSPROC4_CLNT_##proc, \ + .p_name = #proc, \ +} + +#define STUB(proc) \ +[NFSPROC4_CLNT_##proc] = { \ + .p_name = #proc, \ +} + +#if defined(CONFIG_NFS_V4_0) +#define PROC40(proc, argtype, restype) \ + PROC(proc, argtype, restype) +#else +#define PROC40(proc, argtype, restype) \ + STUB(proc) +#endif /* CONFIG_NFS_V4_0 */ + +#define PROC41(proc, argtype, restype) \ + PROC(proc, argtype, restype) + +#if defined(CONFIG_NFS_V4_2) +#define PROC42(proc, argtype, restype) \ + PROC(proc, argtype, restype) +#else +#define PROC42(proc, argtype, restype) \ + STUB(proc) +#endif + +const struct rpc_procinfo nfs4_procedures[] = { + PROC(READ, enc_read, dec_read), + PROC(WRITE, enc_write, dec_write), + PROC(COMMIT, enc_commit, dec_commit), + PROC(OPEN, enc_open, dec_open), + PROC(OPEN_CONFIRM, enc_open_confirm, dec_open_confirm), + PROC(OPEN_NOATTR, enc_open_noattr, dec_open_noattr), + PROC(OPEN_DOWNGRADE, enc_open_downgrade, dec_open_downgrade), + PROC(CLOSE, enc_close, dec_close), + PROC(SETATTR, enc_setattr, dec_setattr), + PROC(FSINFO, enc_fsinfo, dec_fsinfo), + PROC40(RENEW, enc_renew, dec_renew), + PROC(SETCLIENTID, enc_setclientid, dec_setclientid), + PROC(SETCLIENTID_CONFIRM, enc_setclientid_confirm, dec_setclientid_confirm), + PROC(LOCK, enc_lock, dec_lock), + PROC(LOCKT, enc_lockt, dec_lockt), + PROC(LOCKU, enc_locku, dec_locku), + PROC(ACCESS, enc_access, dec_access), + PROC(GETATTR, enc_getattr, dec_getattr), + PROC(LOOKUP, enc_lookup, dec_lookup), + PROC(LOOKUP_ROOT, enc_lookup_root, dec_lookup_root), + PROC(REMOVE, enc_remove, dec_remove), + PROC(RENAME, enc_rename, dec_rename), + PROC(LINK, enc_link, dec_link), + PROC(SYMLINK, enc_symlink, dec_symlink), + PROC(CREATE, enc_create, dec_create), + PROC(PATHCONF, enc_pathconf, dec_pathconf), + PROC(STATFS, enc_statfs, dec_statfs), + PROC(READLINK, enc_readlink, dec_readlink), + PROC(READDIR, enc_readdir, dec_readdir), + PROC(SERVER_CAPS, enc_server_caps, dec_server_caps), + PROC(DELEGRETURN, enc_delegreturn, dec_delegreturn), + PROC(GETACL, enc_getacl, dec_getacl), + PROC(SETACL, enc_setacl, dec_setacl), + PROC(FS_LOCATIONS, enc_fs_locations, dec_fs_locations), + PROC40(RELEASE_LOCKOWNER, enc_release_lockowner, dec_release_lockowner), + PROC(SECINFO, enc_secinfo, dec_secinfo), + PROC(FSID_PRESENT, enc_fsid_present, dec_fsid_present), + PROC41(EXCHANGE_ID, enc_exchange_id, dec_exchange_id), + PROC41(CREATE_SESSION, enc_create_session, dec_create_session), + PROC41(DESTROY_SESSION, enc_destroy_session, dec_destroy_session), + PROC41(SEQUENCE, enc_sequence, dec_sequence), + PROC(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time), + PROC41(RECLAIM_COMPLETE,enc_reclaim_complete, dec_reclaim_complete), + PROC41(GETDEVICEINFO, enc_getdeviceinfo, dec_getdeviceinfo), + PROC41(LAYOUTGET, enc_layoutget, dec_layoutget), + PROC41(LAYOUTCOMMIT, enc_layoutcommit, dec_layoutcommit), + PROC41(LAYOUTRETURN, enc_layoutreturn, dec_layoutreturn), + PROC41(SECINFO_NO_NAME, enc_secinfo_no_name, dec_secinfo_no_name), + PROC41(TEST_STATEID, enc_test_stateid, dec_test_stateid), + PROC41(FREE_STATEID, enc_free_stateid, dec_free_stateid), + STUB(GETDEVICELIST), + PROC41(BIND_CONN_TO_SESSION, + enc_bind_conn_to_session, dec_bind_conn_to_session), + PROC41(DESTROY_CLIENTID,enc_destroy_clientid, dec_destroy_clientid), + PROC42(SEEK, enc_seek, dec_seek), + PROC42(ALLOCATE, enc_allocate, dec_allocate), + PROC42(DEALLOCATE, enc_deallocate, dec_deallocate), + PROC42(LAYOUTSTATS, enc_layoutstats, dec_layoutstats), + PROC42(CLONE, enc_clone, dec_clone), + PROC42(COPY, enc_copy, dec_copy), + PROC42(OFFLOAD_CANCEL, enc_offload_cancel, dec_offload_cancel), + PROC42(OFFLOAD_STATUS, enc_offload_status, dec_offload_status), + PROC42(COPY_NOTIFY, enc_copy_notify, dec_copy_notify), + PROC(LOOKUPP, enc_lookupp, dec_lookupp), + PROC42(LAYOUTERROR, enc_layouterror, dec_layouterror), + PROC42(GETXATTR, enc_getxattr, dec_getxattr), + PROC42(SETXATTR, enc_setxattr, dec_setxattr), + PROC42(LISTXATTRS, enc_listxattrs, dec_listxattrs), + PROC42(REMOVEXATTR, enc_removexattr, dec_removexattr), + PROC42(READ_PLUS, enc_read_plus, dec_read_plus), + PROC42(ZERO_RANGE, enc_zero_range, dec_zero_range), +}; + +static unsigned int nfs_version4_counts[ARRAY_SIZE(nfs4_procedures)]; +const struct rpc_version nfs_version4 = { + .number = 4, + .nrprocs = ARRAY_SIZE(nfs4_procedures), + .procs = nfs4_procedures, + .counts = nfs_version4_counts, +}; diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c new file mode 100644 index 000000000..432612d22 --- /dev/null +++ b/fs/nfs/nfsroot.c @@ -0,0 +1,316 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 1995, 1996 Gero Kuhlmann + * + * Allow an NFS filesystem to be mounted as root. The way this works is: + * (1) Use the IP autoconfig mechanism to set local IP addresses and routes. + * (2) Construct the device string and the options string using DHCP + * option 17 and/or kernel command line options. + * (3) When mount_root() sets up the root file system, pass these strings + * to the NFS client's regular mount interface via sys_mount(). + * + * + * Changes: + * + * Alan Cox : Removed get_address name clash with FPU. + * Alan Cox : Reformatted a bit. + * Gero Kuhlmann : Code cleanup + * Michael Rausch : Fixed recognition of an incoming RARP answer. + * Martin Mares : (2.0) Auto-configuration via BOOTP supported. + * Martin Mares : Manual selection of interface & BOOTP/RARP. + * Martin Mares : Using network routes instead of host routes, + * allowing the default configuration to be used + * for normal operation of the host. + * Martin Mares : Randomized timer with exponential backoff + * installed to minimize network congestion. + * Martin Mares : Code cleanup. + * Martin Mares : (2.1) BOOTP and RARP made configuration options. + * Martin Mares : Server hostname generation fixed. + * Gerd Knorr : Fixed wired inode handling + * Martin Mares : (2.2) "0.0.0.0" addresses from command line ignored. + * Martin Mares : RARP replies not tested for server address. + * Gero Kuhlmann : (2.3) Some bug fixes and code cleanup again (please + * send me your new patches _before_ bothering + * Linus so that I don' always have to cleanup + * _afterwards_ - thanks) + * Gero Kuhlmann : Last changes of Martin Mares undone. + * Gero Kuhlmann : RARP replies are tested for specified server + * again. However, it's now possible to have + * different RARP and NFS servers. + * Gero Kuhlmann : "0.0.0.0" addresses from command line are + * now mapped to INADDR_NONE. + * Gero Kuhlmann : Fixed a bug which prevented BOOTP path name + * from being used (thanks to Leo Spiekman) + * Andy Walker : Allow to specify the NFS server in nfs_root + * without giving a path name + * Swen Thümmler : Allow to specify the NFS options in nfs_root + * without giving a path name. Fix BOOTP request + * for domainname (domainname is NIS domain, not + * DNS domain!). Skip dummy devices for BOOTP. + * Jacek Zapala : Fixed a bug which prevented server-ip address + * from nfsroot parameter from being used. + * Olaf Kirch : Adapted to new NFS code. + * Jakub Jelinek : Free used code segment. + * Marko Kohtala : Fixed some bugs. + * Martin Mares : Debug message cleanup + * Martin Mares : Changed to use the new generic IP layer autoconfig + * code. BOOTP and RARP moved there. + * Martin Mares : Default path now contains host name instead of + * host IP address (but host name defaults to IP + * address anyway). + * Martin Mares : Use root_server_addr appropriately during setup. + * Martin Mares : Rewrote parameter parsing, now hopefully giving + * correct overriding. + * Trond Myklebust : Add in preliminary support for NFSv3 and TCP. + * Fix bug in root_nfs_addr(). nfs_data.namlen + * is NOT for the length of the hostname. + * Hua Qin : Support for mounting root file system via + * NFS over TCP. + * Fabian Frederick: Option parser rebuilt (using parser lib) + * Chuck Lever : Use super.c's text-based mount option parsing + * Chuck Lever : Add "nfsrootdebug". + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_ROOT + +/* Default path we try to mount. "%s" gets replaced by our IP address */ +#define NFS_ROOT "/tftpboot/%s" + +/* Default NFSROOT mount options. */ +#if defined(CONFIG_NFS_V2) +#define NFS_DEF_OPTIONS "vers=2,tcp,rsize=4096,wsize=4096" +#elif defined(CONFIG_NFS_V3) +#define NFS_DEF_OPTIONS "vers=3,tcp,rsize=4096,wsize=4096" +#else +#define NFS_DEF_OPTIONS "vers=4,tcp,rsize=4096,wsize=4096" +#endif + +/* Parameters passed from the kernel command line */ +static char nfs_root_parms[NFS_MAXPATHLEN + 1] __initdata = ""; + +/* Text-based mount options passed to super.c */ +static char nfs_root_options[256] __initdata = NFS_DEF_OPTIONS; + +/* Address of NFS server */ +static __be32 servaddr __initdata = htonl(INADDR_NONE); + +/* Name of directory to mount */ +static char nfs_export_path[NFS_MAXPATHLEN + 1] __initdata = ""; + +/* server:export path string passed to super.c */ +static char nfs_root_device[NFS_MAXPATHLEN + 1] __initdata = ""; + +#ifdef NFS_DEBUG +/* + * When the "nfsrootdebug" kernel command line option is specified, + * enable debugging messages for NFSROOT. + */ +static int __init nfs_root_debug(char *__unused) +{ + nfs_debug |= NFSDBG_ROOT | NFSDBG_MOUNT; + return 1; +} + +__setup("nfsrootdebug", nfs_root_debug); +#endif + +/* + * Parse NFS server and directory information passed on the kernel + * command line. + * + * nfsroot=[:][,] + * + * If there is a "%s" token in the string, it is replaced + * by the ASCII-representation of the client's IP address. + */ +static int __init nfs_root_setup(char *line) +{ + ROOT_DEV = Root_NFS; + + if (line[0] == '/' || line[0] == ',' || (line[0] >= '0' && line[0] <= '9')) { + strscpy(nfs_root_parms, line, sizeof(nfs_root_parms)); + } else { + size_t n = strlen(line) + sizeof(NFS_ROOT) - 1; + if (n >= sizeof(nfs_root_parms)) + line[sizeof(nfs_root_parms) - sizeof(NFS_ROOT) - 2] = '\0'; + sprintf(nfs_root_parms, NFS_ROOT, line); + } + + /* + * Extract the IP address of the NFS server containing our + * root file system, if one was specified. + * + * Note: root_nfs_parse_addr() removes the server-ip from + * nfs_root_parms, if it exists. + */ + root_server_addr = root_nfs_parse_addr(nfs_root_parms); + + return 1; +} + +__setup("nfsroot=", nfs_root_setup); + +static int __init root_nfs_copy(char *dest, const char *src, + const size_t destlen) +{ + if (strscpy(dest, src, destlen) == -E2BIG) + return -1; + return 0; +} + +static int __init root_nfs_cat(char *dest, const char *src, + const size_t destlen) +{ + size_t len = strlen(dest); + + if (len && dest[len - 1] != ',') + if (strlcat(dest, ",", destlen) >= destlen) + return -1; + + if (strlcat(dest, src, destlen) >= destlen) + return -1; + return 0; +} + +/* + * Parse out root export path and mount options from + * passed-in string @incoming. + * + * Copy the export path into @exppath. + */ +static int __init root_nfs_parse_options(char *incoming, char *exppath, + const size_t exppathlen) +{ + char *p; + + /* + * Set the NFS remote path + */ + p = strsep(&incoming, ","); + if (*p != '\0' && strcmp(p, "default") != 0) + if (root_nfs_copy(exppath, p, exppathlen)) + return -1; + + /* + * @incoming now points to the rest of the string; if it + * contains something, append it to our root options buffer + */ + if (incoming != NULL && *incoming != '\0') + if (root_nfs_cat(nfs_root_options, incoming, + sizeof(nfs_root_options))) + return -1; + return 0; +} + +/* + * Decode the export directory path name and NFS options from + * the kernel command line. This has to be done late in order to + * use a dynamically acquired client IP address for the remote + * root directory path. + * + * Returns zero if successful; otherwise -1 is returned. + */ +static int __init root_nfs_data(char *cmdline) +{ + char mand_options[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1]; + int len, retval = -1; + char *tmp = NULL; + const size_t tmplen = sizeof(nfs_export_path); + + tmp = kzalloc(tmplen, GFP_KERNEL); + if (tmp == NULL) + goto out_nomem; + strcpy(tmp, NFS_ROOT); + + if (root_server_path[0] != '\0') { + dprintk("Root-NFS: DHCPv4 option 17: %s\n", + root_server_path); + if (root_nfs_parse_options(root_server_path, tmp, tmplen)) + goto out_optionstoolong; + } + + if (cmdline[0] != '\0') { + dprintk("Root-NFS: nfsroot=%s\n", cmdline); + if (root_nfs_parse_options(cmdline, tmp, tmplen)) + goto out_optionstoolong; + } + + /* + * Append mandatory options for nfsroot so they override + * what has come before + */ + snprintf(mand_options, sizeof(mand_options), "nolock,addr=%pI4", + &servaddr); + if (root_nfs_cat(nfs_root_options, mand_options, + sizeof(nfs_root_options))) + goto out_optionstoolong; + + /* + * Set up nfs_root_device. For NFS mounts, this looks like + * + * server:/path + * + * At this point, utsname()->nodename contains our local + * IP address or hostname, set by ipconfig. If "%s" exists + * in tmp, substitute the nodename, then shovel the whole + * mess into nfs_root_device. + */ + len = snprintf(nfs_export_path, sizeof(nfs_export_path), + tmp, utsname()->nodename); + if (len >= (int)sizeof(nfs_export_path)) + goto out_devnametoolong; + len = snprintf(nfs_root_device, sizeof(nfs_root_device), + "%pI4:%s", &servaddr, nfs_export_path); + if (len >= (int)sizeof(nfs_root_device)) + goto out_devnametoolong; + + retval = 0; + +out: + kfree(tmp); + return retval; +out_nomem: + printk(KERN_ERR "Root-NFS: could not allocate memory\n"); + goto out; +out_optionstoolong: + printk(KERN_ERR "Root-NFS: mount options string too long\n"); + goto out; +out_devnametoolong: + printk(KERN_ERR "Root-NFS: root device name too long.\n"); + goto out; +} + +/** + * nfs_root_data - Return prepared 'data' for NFSROOT mount + * @root_device: OUT: address of string containing NFSROOT device + * @root_data: OUT: address of string containing NFSROOT mount options + * + * Returns zero and sets @root_device and @root_data if successful, + * otherwise -1 is returned. + */ +int __init nfs_root_data(char **root_device, char **root_data) +{ + servaddr = root_server_addr; + if (servaddr == htonl(INADDR_NONE)) { + printk(KERN_ERR "Root-NFS: no NFS server address\n"); + return -1; + } + + if (root_nfs_data(nfs_root_parms) < 0) + return -1; + + *root_device = nfs_root_device; + *root_data = nfs_root_options; + return 0; +} diff --git a/fs/nfs/nfstrace.c b/fs/nfs/nfstrace.c new file mode 100644 index 000000000..5d1bfccbb --- /dev/null +++ b/fs/nfs/nfstrace.c @@ -0,0 +1,15 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2013 Trond Myklebust + */ +#include +#include +#include "internal.h" + +#define CREATE_TRACE_POINTS +#include "nfstrace.h" + +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs_fsync_enter); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs_fsync_exit); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs_xdr_status); +EXPORT_TRACEPOINT_SYMBOL_GPL(nfs_xdr_bad_filehandle); diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h new file mode 100644 index 000000000..4ada21f4e --- /dev/null +++ b/fs/nfs/nfstrace.h @@ -0,0 +1,1986 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2013 Trond Myklebust + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nfs + +#if !defined(_TRACE_NFS_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NFS_H + +#include +#include + +#include +#include +#include + +#define nfs_show_cache_validity(v) \ + __print_flags(v, "|", \ + { NFS_INO_INVALID_DATA, "INVALID_DATA" }, \ + { NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \ + { NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \ + { NFS_INO_INVALID_ACL, "INVALID_ACL" }, \ + { NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \ + { NFS_INO_INVALID_LABEL, "INVALID_LABEL" }, \ + { NFS_INO_INVALID_CHANGE, "INVALID_CHANGE" }, \ + { NFS_INO_INVALID_CTIME, "INVALID_CTIME" }, \ + { NFS_INO_INVALID_MTIME, "INVALID_MTIME" }, \ + { NFS_INO_INVALID_SIZE, "INVALID_SIZE" }, \ + { NFS_INO_INVALID_OTHER, "INVALID_OTHER" }, \ + { NFS_INO_DATA_INVAL_DEFER, "DATA_INVAL_DEFER" }, \ + { NFS_INO_INVALID_BLOCKS, "INVALID_BLOCKS" }, \ + { NFS_INO_INVALID_XATTR, "INVALID_XATTR" }, \ + { NFS_INO_INVALID_NLINK, "INVALID_NLINK" }, \ + { NFS_INO_INVALID_MODE, "INVALID_MODE" }, \ + { NFS_INO_INVALID_BTIME, "INVALID_BTIME" }) + +#define nfs_show_nfsi_flags(v) \ + __print_flags(v, "|", \ + { BIT(NFS_INO_STALE), "STALE" }, \ + { BIT(NFS_INO_ACL_LRU_SET), "ACL_LRU_SET" }, \ + { BIT(NFS_INO_INVALIDATING), "INVALIDATING" }, \ + { BIT(NFS_INO_LAYOUTCOMMIT), "NEED_LAYOUTCOMMIT" }, \ + { BIT(NFS_INO_LAYOUTCOMMITTING), "LAYOUTCOMMIT" }, \ + { BIT(NFS_INO_LAYOUTSTATS), "LAYOUTSTATS" }, \ + { BIT(NFS_INO_ODIRECT), "ODIRECT" }) + +#define nfs_show_wb_flags(v) \ + __print_flags(v, "|", \ + { BIT(PG_BUSY), "BUSY" }, \ + { BIT(PG_MAPPED), "MAPPED" }, \ + { BIT(PG_FOLIO), "FOLIO" }, \ + { BIT(PG_CLEAN), "CLEAN" }, \ + { BIT(PG_COMMIT_TO_DS), "COMMIT_TO_DS" }, \ + { BIT(PG_INODE_REF), "INODE_REF" }, \ + { BIT(PG_HEADLOCK), "HEADLOCK" }, \ + { BIT(PG_TEARDOWN), "TEARDOWN" }, \ + { BIT(PG_UNLOCKPAGE), "UNLOCKPAGE" }, \ + { BIT(PG_UPTODATE), "UPTODATE" }, \ + { BIT(PG_WB_END), "WB_END" }, \ + { BIT(PG_REMOVE), "REMOVE" }, \ + { BIT(PG_CONTENDED1), "CONTENDED1" }, \ + { BIT(PG_CONTENDED2), "CONTENDED2" }) + +DECLARE_EVENT_CLASS(nfs_inode_event, + TP_PROTO( + const struct inode *inode + ), + + TP_ARGS(inode), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(unsigned long, cache_validity) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->cache_validity = nfsi->cache_validity; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu cache_validity=0x%lx (%s)", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (unsigned long long)__entry->version, + __entry->cache_validity, + nfs_show_cache_validity(__entry->cache_validity) + ) +); + +DECLARE_EVENT_CLASS(nfs_inode_event_done, + TP_PROTO( + const struct inode *inode, + int error + ), + + TP_ARGS(inode, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(unsigned char, type) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, size) + __field(unsigned long, nfsi_flags) + __field(unsigned long, cache_validity) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->error = error < 0 ? -error : 0; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->type = nfs_umode_to_dtype(inode->i_mode); + __entry->version = inode_peek_iversion_raw(inode); + __entry->size = i_size_read(inode); + __entry->nfsi_flags = nfsi->flags; + __entry->cache_validity = nfsi->cache_validity; + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "type=%u (%s) version=%llu size=%lld " + "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s)", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->type, + show_fs_dirent_type(__entry->type), + (unsigned long long)__entry->version, + (long long)__entry->size, + __entry->cache_validity, + nfs_show_cache_validity(__entry->cache_validity), + __entry->nfsi_flags, + nfs_show_nfsi_flags(__entry->nfsi_flags) + ) +); + +#define DEFINE_NFS_INODE_EVENT(name) \ + DEFINE_EVENT(nfs_inode_event, name, \ + TP_PROTO( \ + const struct inode *inode \ + ), \ + TP_ARGS(inode)) +#define DEFINE_NFS_INODE_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_inode_event_done, name, \ + TP_PROTO( \ + const struct inode *inode, \ + int error \ + ), \ + TP_ARGS(inode, error)) +DEFINE_NFS_INODE_EVENT(nfs_set_inode_stale); +DEFINE_NFS_INODE_EVENT(nfs_refresh_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_refresh_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_revalidate_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_revalidate_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_invalidate_mapping_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_invalidate_mapping_exit); +DEFINE_NFS_INODE_EVENT(nfs_getattr_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_getattr_exit); +DEFINE_NFS_INODE_EVENT(nfs_setattr_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_setattr_exit); +DEFINE_NFS_INODE_EVENT(nfs_writeback_inode_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit); +DEFINE_NFS_INODE_EVENT(nfs_fsync_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit); +DEFINE_NFS_INODE_EVENT(nfs_access_enter); +DEFINE_NFS_INODE_EVENT_DONE(nfs_set_cache_invalid); +DEFINE_NFS_INODE_EVENT(nfs_readdir_force_readdirplus); +DEFINE_NFS_INODE_EVENT_DONE(nfs_readdir_cache_fill_done); +DEFINE_NFS_INODE_EVENT_DONE(nfs_readdir_uncached_done); + +TRACE_EVENT(nfs_access_exit, + TP_PROTO( + const struct inode *inode, + unsigned int mask, + unsigned int permitted, + int error + ), + + TP_ARGS(inode, mask, permitted, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(unsigned char, type) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, size) + __field(unsigned long, nfsi_flags) + __field(unsigned long, cache_validity) + __field(unsigned int, mask) + __field(unsigned int, permitted) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->error = error < 0 ? -error : 0; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->type = nfs_umode_to_dtype(inode->i_mode); + __entry->version = inode_peek_iversion_raw(inode); + __entry->size = i_size_read(inode); + __entry->nfsi_flags = nfsi->flags; + __entry->cache_validity = nfsi->cache_validity; + __entry->mask = mask; + __entry->permitted = permitted; + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "type=%u (%s) version=%llu size=%lld " + "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s) " + "mask=0x%x permitted=0x%x", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->type, + show_fs_dirent_type(__entry->type), + (unsigned long long)__entry->version, + (long long)__entry->size, + __entry->cache_validity, + nfs_show_cache_validity(__entry->cache_validity), + __entry->nfsi_flags, + nfs_show_nfsi_flags(__entry->nfsi_flags), + __entry->mask, __entry->permitted + ) +); + +DECLARE_EVENT_CLASS(nfs_update_size_class, + TP_PROTO( + const struct inode *inode, + loff_t new_size + ), + + TP_ARGS(inode, new_size), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, cur_size) + __field(loff_t, new_size) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fileid = inode->i_ino; + __entry->version = inode_peek_iversion_raw(inode); + __entry->cur_size = i_size_read(inode); + __entry->new_size = new_size; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu cursize=%lld newsize=%lld", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->cur_size, __entry->new_size + ) +); + +#define DEFINE_NFS_UPDATE_SIZE_EVENT(name) \ + DEFINE_EVENT(nfs_update_size_class, nfs_size_##name, \ + TP_PROTO( \ + const struct inode *inode, \ + loff_t new_size \ + ), \ + TP_ARGS(inode, new_size)) + +DEFINE_NFS_UPDATE_SIZE_EVENT(truncate); +DEFINE_NFS_UPDATE_SIZE_EVENT(truncate_folio); +DEFINE_NFS_UPDATE_SIZE_EVENT(wcc); +DEFINE_NFS_UPDATE_SIZE_EVENT(update); +DEFINE_NFS_UPDATE_SIZE_EVENT(grow); + +DECLARE_EVENT_CLASS(nfs_inode_range_event, + TP_PROTO( + const struct inode *inode, + loff_t range_start, + loff_t range_end + ), + + TP_ARGS(inode, range_start, range_end), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, range_start) + __field(loff_t, range_end) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fileid = inode->i_ino; + __entry->version = inode_peek_iversion_raw(inode); + __entry->range_start = range_start; + __entry->range_end = range_end; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu " + "range=[%lld, %lld]", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->range_start, __entry->range_end + ) +); + +#define DEFINE_NFS_INODE_RANGE_EVENT(name) \ + DEFINE_EVENT(nfs_inode_range_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + loff_t range_start, \ + loff_t range_end \ + ), \ + TP_ARGS(inode, range_start, range_end)) + +DEFINE_NFS_INODE_RANGE_EVENT(nfs_readdir_invalidate_cache_range); + +DECLARE_EVENT_CLASS(nfs_readdir_event, + TP_PROTO( + const struct file *file, + const __be32 *verifier, + u64 cookie, + pgoff_t page_index, + unsigned int dtsize + ), + + TP_ARGS(file, verifier, cookie, page_index, dtsize), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __array(char, verifier, NFS4_VERIFIER_SIZE) + __field(u64, cookie) + __field(pgoff_t, index) + __field(unsigned int, dtsize) + ), + + TP_fast_assign( + const struct inode *dir = file_inode(file); + const struct nfs_inode *nfsi = NFS_I(dir); + + __entry->dev = dir->i_sb->s_dev; + __entry->fileid = dir->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(dir); + if (cookie != 0) + memcpy(__entry->verifier, verifier, + NFS4_VERIFIER_SIZE); + else + memset(__entry->verifier, 0, + NFS4_VERIFIER_SIZE); + __entry->cookie = cookie; + __entry->index = page_index; + __entry->dtsize = dtsize; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu " + "cookie=%s:0x%llx cache_index=%lu dtsize=%u", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, __entry->fhandle, + __entry->version, show_nfs4_verifier(__entry->verifier), + (unsigned long long)__entry->cookie, __entry->index, + __entry->dtsize + ) +); + +#define DEFINE_NFS_READDIR_EVENT(name) \ + DEFINE_EVENT(nfs_readdir_event, name, \ + TP_PROTO( \ + const struct file *file, \ + const __be32 *verifier, \ + u64 cookie, \ + pgoff_t page_index, \ + unsigned int dtsize \ + ), \ + TP_ARGS(file, verifier, cookie, page_index, dtsize)) + +DEFINE_NFS_READDIR_EVENT(nfs_readdir_cache_fill); +DEFINE_NFS_READDIR_EVENT(nfs_readdir_uncached); + +DECLARE_EVENT_CLASS(nfs_lookup_event, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags + ), + + TP_ARGS(dir, dentry, flags), + + TP_STRUCT__entry( + __field(unsigned long, flags) + __field(dev_t, dev) + __field(u64, dir) + __field(u64, fileid) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->flags = flags; + __entry->fileid = d_is_negative(dentry) ? 0 : d_inode(dentry)->i_ino; + __assign_str(name); + ), + + TP_printk( + "flags=0x%lx (%s) name=%02x:%02x:%llu/%s fileid=%llu", + __entry->flags, + show_fs_lookup_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name), + __entry->fileid + ) +); + +#define DEFINE_NFS_LOOKUP_EVENT(name) \ + DEFINE_EVENT(nfs_lookup_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + unsigned int flags \ + ), \ + TP_ARGS(dir, dentry, flags)) + +DECLARE_EVENT_CLASS(nfs_lookup_event_done, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags, + int error + ), + + TP_ARGS(dir, dentry, flags, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, flags) + __field(dev_t, dev) + __field(u64, dir) + __field(u64, fileid) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->error = error < 0 ? -error : 0; + __entry->flags = flags; + __entry->fileid = d_is_negative(dentry) ? 0 : d_inode(dentry)->i_ino; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s fileid=%llu", + -__entry->error, show_nfs_status(__entry->error), + __entry->flags, + show_fs_lookup_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name), + __entry->fileid + ) +); + +#define DEFINE_NFS_LOOKUP_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_lookup_event_done, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + unsigned int flags, \ + int error \ + ), \ + TP_ARGS(dir, dentry, flags, error)) + +DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_enter); +DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit); +DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter); +DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit); +DEFINE_NFS_LOOKUP_EVENT(nfs_readdir_lookup); +DEFINE_NFS_LOOKUP_EVENT(nfs_readdir_lookup_revalidate_failed); +DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_readdir_lookup_revalidate); + +TRACE_EVENT(nfs_atomic_open_enter, + TP_PROTO( + const struct inode *dir, + const struct nfs_open_context *ctx, + unsigned int flags + ), + + TP_ARGS(dir, ctx, flags), + + TP_STRUCT__entry( + __field(unsigned long, flags) + __field(unsigned long, fmode) + __field(dev_t, dev) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->flags = flags; + __entry->fmode = (__force unsigned long)ctx->mode; + __assign_str(name); + ), + + TP_printk( + "flags=0x%lx (%s) fmode=%s name=%02x:%02x:%llu/%s", + __entry->flags, + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_atomic_open_exit, + TP_PROTO( + const struct inode *dir, + const struct nfs_open_context *ctx, + unsigned int flags, + int error + ), + + TP_ARGS(dir, ctx, flags, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, flags) + __field(unsigned long, fmode) + __field(dev_t, dev) + __field(u64, dir) + __string(name, ctx->dentry->d_name.name) + ), + + TP_fast_assign( + __entry->error = -error; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->flags = flags; + __entry->fmode = (__force unsigned long)ctx->mode; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) flags=0x%lx (%s) fmode=%s " + "name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + __entry->flags, + show_fs_fcntl_open_flags(__entry->flags), + show_fs_fmode_flags(__entry->fmode), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_create_enter, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags + ), + + TP_ARGS(dir, dentry, flags), + + TP_STRUCT__entry( + __field(unsigned long, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->flags = flags; + __assign_str(name); + ), + + TP_printk( + "flags=0x%lx (%s) name=%02x:%02x:%llu/%s", + __entry->flags, + show_fs_fcntl_open_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_create_exit, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + unsigned int flags, + int error + ), + + TP_ARGS(dir, dentry, flags, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(unsigned long, flags) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->error = -error; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->flags = flags; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + __entry->flags, + show_fs_fcntl_open_flags(__entry->flags), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +DECLARE_EVENT_CLASS(nfs_directory_event, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry + ), + + TP_ARGS(dir, dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __assign_str(name); + ), + + TP_printk( + "name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_DIRECTORY_EVENT(name) \ + DEFINE_EVENT(nfs_directory_event, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry \ + ), \ + TP_ARGS(dir, dentry)) + +DECLARE_EVENT_CLASS(nfs_directory_event_done, + TP_PROTO( + const struct inode *dir, + const struct dentry *dentry, + int error + ), + + TP_ARGS(dir, dentry, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->error = error < 0 ? -error : 0; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +#define DEFINE_NFS_DIRECTORY_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_directory_event_done, name, \ + TP_PROTO( \ + const struct inode *dir, \ + const struct dentry *dentry, \ + int error \ + ), \ + TP_ARGS(dir, dentry, error)) + +DEFINE_NFS_DIRECTORY_EVENT(nfs_mknod_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mknod_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_mkdir_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_mkdir_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_rmdir_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_rmdir_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_remove_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_remove_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_unlink_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_unlink_exit); +DEFINE_NFS_DIRECTORY_EVENT(nfs_symlink_enter); +DEFINE_NFS_DIRECTORY_EVENT_DONE(nfs_symlink_exit); + +TRACE_EVENT(nfs_link_enter, + TP_PROTO( + const struct inode *inode, + const struct inode *dir, + const struct dentry *dentry + ), + + TP_ARGS(inode, dir, dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->dir = dir->i_ino; + __assign_str(name); + ), + + TP_printk( + "fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fileid, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +TRACE_EVENT(nfs_link_exit, + TP_PROTO( + const struct inode *inode, + const struct inode *dir, + const struct dentry *dentry, + int error + ), + + TP_ARGS(inode, dir, dentry, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u64, fileid) + __field(u64, dir) + __string(name, dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->dir = dir->i_ino; + __entry->error = error < 0 ? -error : 0; + __assign_str(name); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->fileid, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +DECLARE_EVENT_CLASS(nfs_rename_event, + TP_PROTO( + const struct inode *old_dir, + const struct dentry *old_dentry, + const struct inode *new_dir, + const struct dentry *new_dentry + ), + + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, old_dir) + __field(u64, new_dir) + __string(old_name, old_dentry->d_name.name) + __string(new_name, new_dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = old_dir->i_sb->s_dev; + __entry->old_dir = old_dir->i_ino; + __entry->new_dir = new_dir->i_ino; + __assign_str(old_name); + __assign_str(new_name); + ), + + TP_printk( + "old_name=%02x:%02x:%llu/%s new_name=%02x:%02x:%llu/%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->old_dir, + __get_str(old_name), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->new_dir, + __get_str(new_name) + ) +); +#define DEFINE_NFS_RENAME_EVENT(name) \ + DEFINE_EVENT(nfs_rename_event, name, \ + TP_PROTO( \ + const struct inode *old_dir, \ + const struct dentry *old_dentry, \ + const struct inode *new_dir, \ + const struct dentry *new_dentry \ + ), \ + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry)) + +DECLARE_EVENT_CLASS(nfs_rename_event_done, + TP_PROTO( + const struct inode *old_dir, + const struct dentry *old_dentry, + const struct inode *new_dir, + const struct dentry *new_dentry, + int error + ), + + TP_ARGS(old_dir, old_dentry, new_dir, new_dentry, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long, error) + __field(u64, old_dir) + __string(old_name, old_dentry->d_name.name) + __field(u64, new_dir) + __string(new_name, new_dentry->d_name.name) + ), + + TP_fast_assign( + __entry->dev = old_dir->i_sb->s_dev; + __entry->error = -error; + __entry->old_dir = old_dir->i_ino; + __entry->new_dir = new_dir->i_ino; + __assign_str(old_name); + __assign_str(new_name); + ), + + TP_printk( + "error=%ld (%s) old_name=%02x:%02x:%llu/%s " + "new_name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->old_dir, + __get_str(old_name), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->new_dir, + __get_str(new_name) + ) +); +#define DEFINE_NFS_RENAME_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_rename_event_done, name, \ + TP_PROTO( \ + const struct inode *old_dir, \ + const struct dentry *old_dentry, \ + const struct inode *new_dir, \ + const struct dentry *new_dentry, \ + int error \ + ), \ + TP_ARGS(old_dir, old_dentry, new_dir, \ + new_dentry, error)) + +DEFINE_NFS_RENAME_EVENT(nfs_rename_enter); +DEFINE_NFS_RENAME_EVENT_DONE(nfs_rename_exit); + +DEFINE_NFS_RENAME_EVENT_DONE(nfs_async_rename_done); + +TRACE_EVENT(nfs_sillyrename_unlink, + TP_PROTO( + const struct nfs_unlinkdata *data, + int error + ), + + TP_ARGS(data, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long, error) + __field(u64, dir) + __dynamic_array(char, name, data->args.name.len + 1) + ), + + TP_fast_assign( + struct inode *dir = d_inode(data->dentry->d_parent); + size_t len = data->args.name.len; + __entry->dev = dir->i_sb->s_dev; + __entry->dir = dir->i_ino; + __entry->error = -error; + memcpy(__get_str(name), + data->args.name.name, len); + __get_str(name)[len] = 0; + ), + + TP_printk( + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, show_nfs_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->dir, + __get_str(name) + ) +); + +DECLARE_EVENT_CLASS(nfs_folio_event, + TP_PROTO( + const struct inode *inode, + loff_t offset, + size_t count + ), + + TP_ARGS(inode, offset, count), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + __field(size_t, count) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = offset; + __entry->count = count; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu " + "offset=%lld count=%zu", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset, __entry->count + ) +); + +#define DEFINE_NFS_FOLIO_EVENT(name) \ + DEFINE_EVENT(nfs_folio_event, name, \ + TP_PROTO( \ + const struct inode *inode, \ + loff_t offset, \ + size_t count \ + ), \ + TP_ARGS(inode, offset, count)) + +DECLARE_EVENT_CLASS(nfs_folio_event_done, + TP_PROTO( + const struct inode *inode, + loff_t offset, + size_t count, + int ret + ), + + TP_ARGS(inode, offset, count, ret), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(int, ret) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + __field(size_t, count) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = offset; + __entry->count = count; + __entry->ret = ret; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu " + "offset=%lld count=%zu ret=%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset, __entry->count, __entry->ret + ) +); + +#define DEFINE_NFS_FOLIO_EVENT_DONE(name) \ + DEFINE_EVENT(nfs_folio_event_done, name, \ + TP_PROTO( \ + const struct inode *inode, \ + loff_t offset, \ + size_t count, \ + int ret \ + ), \ + TP_ARGS(inode, offset, count, ret)) + +DEFINE_NFS_FOLIO_EVENT(nfs_aop_readpage); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_aop_readpage_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_writeback_folio_reclaim); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_writeback_folio_reclaim_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_writeback_folio); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_writeback_folio_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_invalidate_folio); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_launder_folio_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_try_to_update_request); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_try_to_update_request_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_update_folio); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_update_folio_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_write_begin); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_write_begin_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_write_end); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_write_end_done); + +DEFINE_NFS_FOLIO_EVENT(nfs_writepages); +DEFINE_NFS_FOLIO_EVENT_DONE(nfs_writepages_done); + +DECLARE_EVENT_CLASS(nfs_kiocb_event, + TP_PROTO( + const struct kiocb *iocb, + const struct iov_iter *iter + ), + + TP_ARGS(iocb, iter), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + __field(size_t, count) + __field(int, flags) + ), + + TP_fast_assign( + const struct inode *inode = file_inode(iocb->ki_filp); + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = iocb->ki_pos; + __entry->count = iov_iter_count(iter); + __entry->flags = iocb->ki_flags; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu offset=%lld count=%zu ki_flags=%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset, __entry->count, + __print_flags(__entry->flags, "|", TRACE_IOCB_STRINGS) + ) +); + +#define DEFINE_NFS_KIOCB_EVENT(name) \ + DEFINE_EVENT(nfs_kiocb_event, name, \ + TP_PROTO( \ + const struct kiocb *iocb, \ + const struct iov_iter *iter \ + ), \ + TP_ARGS(iocb, iter)) + +DEFINE_NFS_KIOCB_EVENT(nfs_file_read); +DEFINE_NFS_KIOCB_EVENT(nfs_file_write); + +TRACE_EVENT(nfs_aop_readahead, + TP_PROTO( + const struct inode *inode, + loff_t pos, + unsigned int nr_pages + ), + + TP_ARGS(inode, pos, nr_pages), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + __field(unsigned int, nr_pages) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->offset = pos; + __entry->nr_pages = nr_pages; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu offset=%lld nr_pages=%u", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->offset, __entry->nr_pages + ) +); + +TRACE_EVENT(nfs_aop_readahead_done, + TP_PROTO( + const struct inode *inode, + unsigned int nr_pages, + int ret + ), + + TP_ARGS(inode, nr_pages, ret), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(int, ret) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, offset) + __field(unsigned int, nr_pages) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->version = inode_peek_iversion_raw(inode); + __entry->nr_pages = nr_pages; + __entry->ret = ret; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu nr_pages=%u ret=%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->version, + __entry->nr_pages, __entry->ret + ) +); + +TRACE_EVENT(nfs_initiate_read, + TP_PROTO( + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(hdr), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->count + ) +); + +TRACE_EVENT(nfs_readpage_done, + TP_PROTO( + const struct rpc_task *task, + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(task, hdr), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(bool, eof) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->error = task->tk_status; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->eof = hdr->res.eof; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u%s", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->arg_count, + __entry->res_count, __entry->eof ? " eof" : "" + ) +); + +TRACE_EVENT(nfs_readpage_short, + TP_PROTO( + const struct rpc_task *task, + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(task, hdr), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(bool, eof) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->error = task->tk_status; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->eof = hdr->res.eof; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u%s", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->arg_count, + __entry->res_count, __entry->eof ? " eof" : "" + ) +); + + +TRACE_EVENT(nfs_pgio_error, + TP_PROTO( + const struct nfs_pgio_header *hdr, + int error, + loff_t pos + ), + + TP_ARGS(hdr, error, pos), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(loff_t, pos) + __field(int, error) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->error = error; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk("error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u pos=%llu", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, __entry->fhandle, + (long long)__entry->offset, __entry->arg_count, __entry->res_count, + __entry->pos + ) +); + +TRACE_EVENT(nfs_initiate_write, + TP_PROTO( + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(hdr), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __field(unsigned long, stable) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; + __entry->stable = hdr->args.stable; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u stable=%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->count, + show_nfs_stable_how(__entry->stable) + ) +); + +TRACE_EVENT(nfs_writeback_done, + TP_PROTO( + const struct rpc_task *task, + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(task, hdr), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(int, error) + __field(unsigned long, stable) + __array(char, verifier, NFS4_VERIFIER_SIZE) + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + const struct nfs_writeverf *verf = hdr->res.verf; + + __entry->error = task->tk_status; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->stable = verf->committed; + memcpy(__entry->verifier, + &verf->verifier, + NFS4_VERIFIER_SIZE); + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u res=%u stable=%s " + "verifier=%s", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->arg_count, + __entry->res_count, + show_nfs_stable_how(__entry->stable), + show_nfs4_verifier(__entry->verifier) + ) +); + +DECLARE_EVENT_CLASS(nfs_page_class, + TP_PROTO( + const struct nfs_page *req + ), + + TP_ARGS(req), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(const struct nfs_page *, req) + __field(loff_t, offset) + __field(unsigned int, count) + __field(unsigned long, flags) + ), + + TP_fast_assign( + const struct inode *inode = folio_inode(req->wb_folio); + const struct nfs_inode *nfsi = NFS_I(inode); + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->req = req; + __entry->offset = req_offset(req); + __entry->count = req->wb_bytes; + __entry->flags = req->wb_flags; + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x req=%p offset=%lld count=%u flags=%s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, __entry->fhandle, + __entry->req, __entry->offset, __entry->count, + nfs_show_wb_flags(__entry->flags) + ) +); + +#define DEFINE_NFS_PAGE_EVENT(name) \ + DEFINE_EVENT(nfs_page_class, name, \ + TP_PROTO( \ + const struct nfs_page *req \ + ), \ + TP_ARGS(req)) + +DEFINE_NFS_PAGE_EVENT(nfs_writepage_setup); +DEFINE_NFS_PAGE_EVENT(nfs_do_writepage); + +DECLARE_EVENT_CLASS(nfs_page_error_class, + TP_PROTO( + const struct inode *inode, + const struct nfs_page *req, + int error + ), + + TP_ARGS(inode, req, error), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(unsigned int, count) + __field(int, error) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->offset = req_offset(req); + __entry->count = req->wb_bytes; + __entry->error = error; + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->offset, + __entry->count + ) +); + +#define DEFINE_NFS_PAGEERR_EVENT(name) \ + DEFINE_EVENT(nfs_page_error_class, name, \ + TP_PROTO( \ + const struct inode *inode, \ + const struct nfs_page *req, \ + int error \ + ), \ + TP_ARGS(inode, req, error)) + +DEFINE_NFS_PAGEERR_EVENT(nfs_write_error); +DEFINE_NFS_PAGEERR_EVENT(nfs_comp_error); +DEFINE_NFS_PAGEERR_EVENT(nfs_commit_error); + +TRACE_EVENT(nfs_initiate_commit, + TP_PROTO( + const struct nfs_commit_data *data + ), + + TP_ARGS(data), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; + + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%u", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, __entry->count + ) +); + +TRACE_EVENT(nfs_commit_done, + TP_PROTO( + const struct rpc_task *task, + const struct nfs_commit_data *data + ), + + TP_ARGS(task, data), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(int, error) + __field(unsigned long, stable) + __array(char, verifier, NFS4_VERIFIER_SIZE) + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; + const struct nfs_writeverf *verf = data->res.verf; + + __entry->error = task->tk_status; + __entry->offset = data->args.offset; + __entry->stable = verf->committed; + memcpy(__entry->verifier, + &verf->verifier, + NFS4_VERIFIER_SIZE); + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld stable=%s verifier=%s", __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + (long long)__entry->offset, + show_nfs_stable_how(__entry->stable), + show_nfs4_verifier(__entry->verifier) + ) +); + +#define nfs_show_direct_req_flags(v) \ + __print_flags(v, "|", \ + { NFS_ODIRECT_DO_COMMIT, "DO_COMMIT" }, \ + { NFS_ODIRECT_RESCHED_WRITES, "RESCHED_WRITES" }, \ + { NFS_ODIRECT_SHOULD_DIRTY, "SHOULD DIRTY" }, \ + { NFS_ODIRECT_DONE, "DONE" } ) + +DECLARE_EVENT_CLASS(nfs_direct_req_class, + TP_PROTO( + const struct nfs_direct_req *dreq + ), + + TP_ARGS(dreq), + + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u32, fhandle) + __field(loff_t, offset) + __field(ssize_t, count) + __field(ssize_t, error) + __field(int, flags) + ), + + TP_fast_assign( + const struct inode *inode = dreq->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = &nfsi->fh; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset = dreq->io_start; + __entry->count = dreq->count; + __entry->error = dreq->error; + __entry->flags = dreq->flags; + ), + + TP_printk( + "error=%zd fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%zd flags=%s", + __entry->error, MAJOR(__entry->dev), + MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->offset, + __entry->count, + nfs_show_direct_req_flags(__entry->flags) + ) +); + +#define DEFINE_NFS_DIRECT_REQ_EVENT(name) \ + DEFINE_EVENT(nfs_direct_req_class, name, \ + TP_PROTO( \ + const struct nfs_direct_req *dreq \ + ), \ + TP_ARGS(dreq)) + +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_commit_complete); +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_resched_write); +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_write_complete); +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_write_completion); +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_write_schedule_iovec); +DEFINE_NFS_DIRECT_REQ_EVENT(nfs_direct_write_reschedule_io); + +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + +DECLARE_EVENT_CLASS(nfs_local_dio_class, + TP_PROTO( + const struct inode *inode, + loff_t offset, + ssize_t count, + const struct nfs_local_dio *local_dio + ), + TP_ARGS(inode, offset, count, local_dio), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u32, fhandle) + __field(loff_t, offset) + __field(ssize_t, count) + __field(u32, mem_align) + __field(u32, offset_align) + __field(loff_t, start) + __field(ssize_t, start_len) + __field(loff_t, middle) + __field(ssize_t, middle_len) + __field(loff_t, end) + __field(ssize_t, end_len) + ), + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = &nfsi->fh; + + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = inode->i_ino; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->offset = offset; + __entry->count = count; + __entry->mem_align = local_dio->mem_align; + __entry->offset_align = local_dio->offset_align; + __entry->start = offset; + __entry->start_len = local_dio->start_len; + __entry->middle = local_dio->middle_offset; + __entry->middle_len = local_dio->middle_len; + __entry->end = local_dio->end_offset; + __entry->end_len = local_dio->end_len; + ), + TP_printk("fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%lld count=%zd " + "mem_align=%u offset_align=%u " + "start=%llu+%zd middle=%llu+%zd end=%llu+%zd", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, __entry->offset, __entry->count, + __entry->mem_align, __entry->offset_align, + __entry->start, __entry->start_len, + __entry->middle, __entry->middle_len, + __entry->end, __entry->end_len) +) + +#define DEFINE_NFS_LOCAL_DIO_EVENT(name) \ +DEFINE_EVENT(nfs_local_dio_class, nfs_local_dio_##name, \ + TP_PROTO(const struct inode *inode, \ + loff_t offset, \ + ssize_t count, \ + const struct nfs_local_dio *local_dio),\ + TP_ARGS(inode, offset, count, local_dio)) + +DEFINE_NFS_LOCAL_DIO_EVENT(read); +DEFINE_NFS_LOCAL_DIO_EVENT(write); +DEFINE_NFS_LOCAL_DIO_EVENT(misaligned); + +#endif /* CONFIG_NFS_LOCALIO */ + +TRACE_EVENT(nfs_fh_to_dentry, + TP_PROTO( + const struct super_block *sb, + const struct nfs_fh *fh, + u64 fileid, + int error + ), + + TP_ARGS(sb, fh, fileid, error), + + TP_STRUCT__entry( + __field(int, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + ), + + TP_fast_assign( + __entry->error = error; + __entry->dev = sb->s_dev; + __entry->fileid = fileid; + __entry->fhandle = nfs_fhandle_hash(fh); + ), + + TP_printk( + "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x ", + __entry->error, + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle + ) +); + +TRACE_EVENT(nfs_mount_assign, + TP_PROTO( + const char *option, + const char *value + ), + + TP_ARGS(option, value), + + TP_STRUCT__entry( + __string(option, option) + __string(value, value) + ), + + TP_fast_assign( + __assign_str(option); + __assign_str(value); + ), + + TP_printk("option %s=%s", + __get_str(option), __get_str(value) + ) +); + +TRACE_EVENT(nfs_mount_option, + TP_PROTO( + const struct fs_parameter *param + ), + + TP_ARGS(param), + + TP_STRUCT__entry( + __string(option, param->key) + ), + + TP_fast_assign( + __assign_str(option); + ), + + TP_printk("option %s", __get_str(option)) +); + +TRACE_EVENT(nfs_mount_path, + TP_PROTO( + const char *path + ), + + TP_ARGS(path), + + TP_STRUCT__entry( + __string(path, path) + ), + + TP_fast_assign( + __assign_str(path); + ), + + TP_printk("path='%s'", __get_str(path)) +); + +TRACE_EVENT(nfs_local_open_fh, + TP_PROTO( + const struct nfs_fh *fh, + fmode_t fmode, + int error + ), + + TP_ARGS(fh, fmode, error), + + TP_STRUCT__entry( + __field(int, error) + __field(u32, fhandle) + __field(unsigned int, fmode) + ), + + TP_fast_assign( + __entry->error = error; + __entry->fhandle = nfs_fhandle_hash(fh); + __entry->fmode = (__force unsigned int)fmode; + ), + + TP_printk( + "fhandle=0x%08x mode=%s result=%d", + __entry->fhandle, + show_fs_fmode_flags(__entry->fmode), + __entry->error + ) +); + +DECLARE_EVENT_CLASS(nfs_xdr_event, + TP_PROTO( + const struct xdr_stream *xdr, + int error + ), + + TP_ARGS(xdr, error), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) + __field(int, version) + __field(unsigned long, error) + __string(program, + xdr->rqst->rq_task->tk_client->cl_program->name) + __string(procedure, + xdr->rqst->rq_task->tk_msg.rpc_proc->p_name) + ), + + TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); + __entry->version = task->tk_client->cl_vers; + __entry->error = error; + __assign_str(program); + __assign_str(procedure); + ), + + TP_printk(SUNRPC_TRACE_TASK_SPECIFIER + " xid=0x%08x %sv%d %s error=%ld (%s)", + __entry->task_id, __entry->client_id, __entry->xid, + __get_str(program), __entry->version, + __get_str(procedure), -__entry->error, + show_nfs_status(__entry->error) + ) +); +#define DEFINE_NFS_XDR_EVENT(name) \ + DEFINE_EVENT(nfs_xdr_event, name, \ + TP_PROTO( \ + const struct xdr_stream *xdr, \ + int error \ + ), \ + TP_ARGS(xdr, error)) +DEFINE_NFS_XDR_EVENT(nfs_xdr_status); +DEFINE_NFS_XDR_EVENT(nfs_xdr_bad_filehandle); + +#endif /* _TRACE_NFS_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE nfstrace +/* This part must be outside protection */ +#include diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c new file mode 100644 index 000000000..7dd478ffc --- /dev/null +++ b/fs/nfs/pagelist.c @@ -0,0 +1,1496 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/pagelist.c + * + * A set of helper functions for managing NFS read and write requests. + * The main purpose of these routines is to provide support for the + * coalescing of several requests into a single RPC call. + * + * Copyright 2000, 2001 (c) Trond Myklebust + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "pnfs.h" +#include "nfstrace.h" +#include "fscache.h" + +#define NFSDBG_FACILITY NFSDBG_PAGECACHE + +static struct kmem_cache *nfs_page_cachep; +static const struct rpc_call_ops nfs_pgio_common_ops; + +struct nfs_page_iter_page { + const struct nfs_page *req; + size_t count; +}; + +static void nfs_page_iter_page_init(struct nfs_page_iter_page *i, + const struct nfs_page *req) +{ + i->req = req; + i->count = 0; +} + +static void nfs_page_iter_page_advance(struct nfs_page_iter_page *i, size_t sz) +{ + const struct nfs_page *req = i->req; + size_t tmp = i->count + sz; + + i->count = (tmp < req->wb_bytes) ? tmp : req->wb_bytes; +} + +static struct page *nfs_page_iter_page_get(struct nfs_page_iter_page *i) +{ + const struct nfs_page *req = i->req; + struct page *page; + + if (i->count != req->wb_bytes) { + size_t base = i->count + req->wb_pgbase; + size_t len = PAGE_SIZE - offset_in_page(base); + + page = nfs_page_to_page(req, base); + nfs_page_iter_page_advance(i, len); + return page; + } + return NULL; +} + +static struct nfs_pgio_mirror * +nfs_pgio_get_mirror(struct nfs_pageio_descriptor *desc, u32 idx) +{ + if (desc->pg_ops->pg_get_mirror) + return desc->pg_ops->pg_get_mirror(desc, idx); + return &desc->pg_mirrors[0]; +} + +struct nfs_pgio_mirror * +nfs_pgio_current_mirror(struct nfs_pageio_descriptor *desc) +{ + return nfs_pgio_get_mirror(desc, desc->pg_mirror_idx); +} +EXPORT_SYMBOL_GPL(nfs_pgio_current_mirror); + +static u32 +nfs_pgio_set_current_mirror(struct nfs_pageio_descriptor *desc, u32 idx) +{ + if (desc->pg_ops->pg_set_mirror) + return desc->pg_ops->pg_set_mirror(desc, idx); + return desc->pg_mirror_idx; +} + +void nfs_pgheader_init(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr, + void (*release)(struct nfs_pgio_header *hdr)) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + + hdr->req = nfs_list_entry(mirror->pg_list.next); + hdr->inode = desc->pg_inode; + hdr->cred = nfs_req_openctx(hdr->req)->cred; + hdr->io_start = req_offset(hdr->req); + hdr->good_bytes = mirror->pg_count; + hdr->io_completion = desc->pg_io_completion; + hdr->dreq = desc->pg_dreq; + nfs_netfs_set_pgio_header(hdr, desc); + hdr->release = release; + hdr->completion_ops = desc->pg_completion_ops; + if (hdr->completion_ops->init_hdr) + hdr->completion_ops->init_hdr(hdr); + + hdr->pgio_mirror_idx = desc->pg_mirror_idx; +} +EXPORT_SYMBOL_GPL(nfs_pgheader_init); + +void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos) +{ + unsigned int new = pos - hdr->io_start; + + trace_nfs_pgio_error(hdr, error, pos); + if (hdr->good_bytes > new) { + hdr->good_bytes = new; + clear_bit(NFS_IOHDR_EOF, &hdr->flags); + if (!test_and_set_bit(NFS_IOHDR_ERROR, &hdr->flags)) + hdr->error = error; + } +} + +static inline struct nfs_page *nfs_page_alloc(void) +{ + struct nfs_page *p = + kmem_cache_zalloc(nfs_page_cachep, nfs_io_gfp_mask()); + if (p) + INIT_LIST_HEAD(&p->wb_list); + return p; +} + +static inline void +nfs_page_free(struct nfs_page *p) +{ + kmem_cache_free(nfs_page_cachep, p); +} + +/** + * nfs_iocounter_wait - wait for i/o to complete + * @l_ctx: nfs_lock_context with io_counter to use + * + * returns -ERESTARTSYS if interrupted by a fatal signal. + * Otherwise returns 0 once the io_count hits 0. + */ +int +nfs_iocounter_wait(struct nfs_lock_context *l_ctx) +{ + return wait_var_event_killable(&l_ctx->io_count, + !atomic_read(&l_ctx->io_count)); +} + +/** + * nfs_async_iocounter_wait - wait on a rpc_waitqueue for I/O + * to complete + * @task: the rpc_task that should wait + * @l_ctx: nfs_lock_context with io_counter to check + * + * Returns true if there is outstanding I/O to wait on and the + * task has been put to sleep. + */ +bool +nfs_async_iocounter_wait(struct rpc_task *task, struct nfs_lock_context *l_ctx) +{ + struct inode *inode = d_inode(l_ctx->open_context->dentry); + bool ret = false; + + if (atomic_read(&l_ctx->io_count) > 0) { + rpc_sleep_on(&NFS_SERVER(inode)->uoc_rpcwaitq, task, NULL); + ret = true; + } + + if (atomic_read(&l_ctx->io_count) == 0) { + rpc_wake_up_queued_task(&NFS_SERVER(inode)->uoc_rpcwaitq, task); + ret = false; + } + + return ret; +} +EXPORT_SYMBOL_GPL(nfs_async_iocounter_wait); + +/* + * nfs_page_set_headlock - set the request PG_HEADLOCK + * @req: request that is to be locked + * + * this lock must be held when modifying req->wb_head + * + * return 0 on success, < 0 on error + */ +int +nfs_page_set_headlock(struct nfs_page *req) +{ + if (!test_and_set_bit(PG_HEADLOCK, &req->wb_flags)) + return 0; + + set_bit(PG_CONTENDED1, &req->wb_flags); + smp_mb__after_atomic(); + return wait_on_bit_lock(&req->wb_flags, PG_HEADLOCK, + TASK_UNINTERRUPTIBLE); +} + +/* + * nfs_page_clear_headlock - clear the request PG_HEADLOCK + * @req: request that is to be locked + */ +void +nfs_page_clear_headlock(struct nfs_page *req) +{ + clear_bit_unlock(PG_HEADLOCK, &req->wb_flags); + smp_mb__after_atomic(); + if (!test_bit(PG_CONTENDED1, &req->wb_flags)) + return; + wake_up_bit(&req->wb_flags, PG_HEADLOCK); +} + +/* + * nfs_page_group_lock - lock the head of the page group + * @req: request in group that is to be locked + * + * this lock must be held when traversing or modifying the page + * group list + * + * return 0 on success, < 0 on error + */ +int +nfs_page_group_lock(struct nfs_page *req) +{ + int ret; + + ret = nfs_page_set_headlock(req); + if (ret || req->wb_head == req) + return ret; + return nfs_page_set_headlock(req->wb_head); +} + +/* + * nfs_page_group_unlock - unlock the head of the page group + * @req: request in group that is to be unlocked + */ +void +nfs_page_group_unlock(struct nfs_page *req) +{ + if (req != req->wb_head) + nfs_page_clear_headlock(req->wb_head); + nfs_page_clear_headlock(req); +} + +/** + * nfs_page_group_sync_on_bit_locked - Test if all requests have @bit set + * @req: request in page group + * @bit: PG_* bit that is used to sync page group + * + * must be called with page group lock held + */ +bool nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) +{ + struct nfs_page *head = req->wb_head; + struct nfs_page *tmp; + + WARN_ON_ONCE(!test_bit(PG_HEADLOCK, &head->wb_flags)); + WARN_ON_ONCE(test_and_set_bit(bit, &req->wb_flags)); + + tmp = req->wb_this_page; + while (tmp != req) { + if (!test_bit(bit, &tmp->wb_flags)) + return false; + tmp = tmp->wb_this_page; + } + + /* true! reset all bits */ + tmp = req; + do { + clear_bit(bit, &tmp->wb_flags); + tmp = tmp->wb_this_page; + } while (tmp != req); + + return true; +} + +/* + * nfs_page_group_sync_on_bit - set bit on current request, but only + * return true if the bit is set for all requests in page group + * @req - request in page group + * @bit - PG_* bit that is used to sync page group + */ +bool nfs_page_group_sync_on_bit(struct nfs_page *req, unsigned int bit) +{ + bool ret; + + nfs_page_group_lock(req); + ret = nfs_page_group_sync_on_bit_locked(req, bit); + nfs_page_group_unlock(req); + + return ret; +} + +/* + * nfs_page_group_init - Initialize the page group linkage for @req + * @req - a new nfs request + * @prev - the previous request in page group, or NULL if @req is the first + * or only request in the group (the head). + */ +static inline void +nfs_page_group_init(struct nfs_page *req, struct nfs_page *prev) +{ + struct inode *inode; + WARN_ON_ONCE(prev == req); + + if (!prev) { + /* a head request */ + req->wb_head = req; + req->wb_this_page = req; + } else { + /* a subrequest */ + WARN_ON_ONCE(prev->wb_this_page != prev->wb_head); + WARN_ON_ONCE(!test_bit(PG_HEADLOCK, &prev->wb_head->wb_flags)); + req->wb_head = prev->wb_head; + req->wb_this_page = prev->wb_this_page; + prev->wb_this_page = req; + + /* All subrequests take a ref on the head request until + * nfs_page_group_destroy is called */ + kref_get(&req->wb_head->wb_kref); + + /* grab extra ref and bump the request count if head request + * has extra ref from the write/commit path to handle handoff + * between write and commit lists. */ + if (test_bit(PG_INODE_REF, &prev->wb_head->wb_flags)) { + inode = nfs_page_to_inode(req); + set_bit(PG_INODE_REF, &req->wb_flags); + kref_get(&req->wb_kref); + atomic_long_inc(&NFS_I(inode)->nrequests); + } + } +} + +/* + * nfs_page_group_destroy - sync the destruction of page groups + * @req - request that no longer needs the page group + * + * releases the page group reference from each member once all + * members have called this function. + */ +static void +nfs_page_group_destroy(struct kref *kref) +{ + struct nfs_page *req = container_of(kref, struct nfs_page, wb_kref); + struct nfs_page *head = req->wb_head; + struct nfs_page *tmp, *next; + + if (!nfs_page_group_sync_on_bit(req, PG_TEARDOWN)) + goto out; + + tmp = req; + do { + next = tmp->wb_this_page; + /* unlink and free */ + tmp->wb_this_page = tmp; + tmp->wb_head = tmp; + nfs_free_request(tmp); + tmp = next; + } while (tmp != req); +out: + /* subrequests must release the ref on the head request */ + if (head != req) + nfs_release_request(head); +} + +static struct nfs_page *nfs_page_create(struct nfs_lock_context *l_ctx, + unsigned int pgbase, pgoff_t index, + unsigned int offset, unsigned int count) +{ + struct nfs_page *req; + struct nfs_open_context *ctx = l_ctx->open_context; + + if (test_bit(NFS_CONTEXT_BAD, &ctx->flags)) + return ERR_PTR(-EBADF); + /* try to allocate the request struct */ + req = nfs_page_alloc(); + if (req == NULL) + return ERR_PTR(-ENOMEM); + + req->wb_lock_context = l_ctx; + refcount_inc(&l_ctx->count); + atomic_inc(&l_ctx->io_count); + + /* Initialize the request struct. Initially, we assume a + * long write-back delay. This will be adjusted in + * update_nfs_request below if the region is not locked. */ + req->wb_pgbase = pgbase; + req->wb_index = index; + req->wb_offset = offset; + req->wb_bytes = count; + kref_init(&req->wb_kref); + req->wb_nio = 0; + return req; +} + +static void nfs_page_assign_folio(struct nfs_page *req, struct folio *folio) +{ + if (folio != NULL) { + req->wb_folio = folio; + folio_get(folio); + set_bit(PG_FOLIO, &req->wb_flags); + } +} + +static void nfs_page_assign_page(struct nfs_page *req, struct page *page) +{ + if (page != NULL) { + req->wb_page = page; + get_page(page); + } +} + +/** + * nfs_page_create_from_page - Create an NFS read/write request. + * @ctx: open context to use + * @page: page to write + * @pgbase: starting offset within the page for the write + * @offset: file offset for the write + * @count: number of bytes to read/write + * + * The page must be locked by the caller. This makes sure we never + * create two different requests for the same page. + * User should ensure it is safe to sleep in this function. + */ +struct nfs_page *nfs_page_create_from_page(struct nfs_open_context *ctx, + struct page *page, + unsigned int pgbase, loff_t offset, + unsigned int count) +{ + struct nfs_lock_context *l_ctx = nfs_get_lock_context(ctx); + struct nfs_page *ret; + + if (IS_ERR(l_ctx)) + return ERR_CAST(l_ctx); + ret = nfs_page_create(l_ctx, pgbase, offset >> PAGE_SHIFT, + offset_in_page(offset), count); + if (!IS_ERR(ret)) { + nfs_page_assign_page(ret, page); + nfs_page_group_init(ret, NULL); + } + nfs_put_lock_context(l_ctx); + return ret; +} + +/** + * nfs_page_create_from_folio - Create an NFS read/write request. + * @ctx: open context to use + * @folio: folio to write + * @offset: starting offset within the folio for the write + * @count: number of bytes to read/write + * + * The page must be locked by the caller. This makes sure we never + * create two different requests for the same page. + * User should ensure it is safe to sleep in this function. + */ +struct nfs_page *nfs_page_create_from_folio(struct nfs_open_context *ctx, + struct folio *folio, + unsigned int offset, + unsigned int count) +{ + struct nfs_lock_context *l_ctx = nfs_get_lock_context(ctx); + struct nfs_page *ret; + + if (IS_ERR(l_ctx)) + return ERR_CAST(l_ctx); + ret = nfs_page_create(l_ctx, offset, folio->index, offset, count); + if (!IS_ERR(ret)) { + nfs_page_assign_folio(ret, folio); + nfs_page_group_init(ret, NULL); + } + nfs_put_lock_context(l_ctx); + return ret; +} + +static struct nfs_page * +nfs_create_subreq(struct nfs_page *req, + unsigned int pgbase, + unsigned int offset, + unsigned int count) +{ + struct nfs_page *last; + struct nfs_page *ret; + struct folio *folio = nfs_page_to_folio(req); + struct page *page = nfs_page_to_page(req, pgbase); + + ret = nfs_page_create(req->wb_lock_context, pgbase, req->wb_index, + offset, count); + if (!IS_ERR(ret)) { + if (folio) + nfs_page_assign_folio(ret, folio); + else + nfs_page_assign_page(ret, page); + /* find the last request */ + for (last = req->wb_head; + last->wb_this_page != req->wb_head; + last = last->wb_this_page) + ; + + nfs_lock_request(ret); + nfs_page_group_init(ret, last); + ret->wb_nio = req->wb_nio; + } + return ret; +} + +/** + * nfs_unlock_request - Unlock request and wake up sleepers. + * @req: pointer to request + */ +void nfs_unlock_request(struct nfs_page *req) +{ + clear_bit_unlock(PG_BUSY, &req->wb_flags); + smp_mb__after_atomic(); + if (!test_bit(PG_CONTENDED2, &req->wb_flags)) + return; + wake_up_bit(&req->wb_flags, PG_BUSY); +} + +/** + * nfs_unlock_and_release_request - Unlock request and release the nfs_page + * @req: pointer to request + */ +void nfs_unlock_and_release_request(struct nfs_page *req) +{ + nfs_unlock_request(req); + nfs_release_request(req); +} + +/* + * nfs_clear_request - Free up all resources allocated to the request + * @req: + * + * Release page and open context resources associated with a read/write + * request after it has completed. + */ +static void nfs_clear_request(struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + struct page *page = req->wb_page; + struct nfs_lock_context *l_ctx = req->wb_lock_context; + struct nfs_open_context *ctx; + + if (folio != NULL) { + folio_put(folio); + req->wb_folio = NULL; + clear_bit(PG_FOLIO, &req->wb_flags); + } else if (page != NULL) { + put_page(page); + req->wb_page = NULL; + } + if (l_ctx != NULL) { + if (atomic_dec_and_test(&l_ctx->io_count)) { + wake_up_var(&l_ctx->io_count); + ctx = l_ctx->open_context; + if (test_bit(NFS_CONTEXT_UNLOCK, &ctx->flags)) + rpc_wake_up(&NFS_SERVER(d_inode(ctx->dentry))->uoc_rpcwaitq); + } + nfs_put_lock_context(l_ctx); + req->wb_lock_context = NULL; + } +} + +/** + * nfs_free_request - Release the count on an NFS read/write request + * @req: request to release + * + * Note: Should never be called with the spinlock held! + */ +void nfs_free_request(struct nfs_page *req) +{ + WARN_ON_ONCE(req->wb_this_page != req); + + /* extra debug: make sure no sync bits are still set */ + WARN_ON_ONCE(test_bit(PG_TEARDOWN, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_UNLOCKPAGE, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_UPTODATE, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_WB_END, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_REMOVE, &req->wb_flags)); + + /* Release struct file and open context */ + nfs_clear_request(req); + nfs_page_free(req); +} + +void nfs_release_request(struct nfs_page *req) +{ + kref_put(&req->wb_kref, nfs_page_group_destroy); +} +EXPORT_SYMBOL_GPL(nfs_release_request); + +/* + * nfs_generic_pg_test - determine if requests can be coalesced + * @desc: pointer to descriptor + * @prev: previous request in desc, or NULL + * @req: this request + * + * Returns zero if @req cannot be coalesced into @desc, otherwise it returns + * the size of the request. + */ +size_t nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, + struct nfs_page *prev, struct nfs_page *req) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + + if (mirror->pg_count > mirror->pg_bsize) { + /* should never happen */ + WARN_ON_ONCE(1); + return 0; + } + + /* + * Limit the request size so that we can still allocate a page array + * for it without upsetting the slab allocator. + */ + if (((mirror->pg_count + req->wb_bytes) >> PAGE_SHIFT) * + sizeof(struct page *) > PAGE_SIZE) + return 0; + + return min(mirror->pg_bsize - mirror->pg_count, (size_t)req->wb_bytes); +} +EXPORT_SYMBOL_GPL(nfs_generic_pg_test); + +struct nfs_pgio_header *nfs_pgio_header_alloc(const struct nfs_rw_ops *ops) +{ + struct nfs_pgio_header *hdr = ops->rw_alloc_header(); + + if (hdr) { + INIT_LIST_HEAD(&hdr->pages); + hdr->rw_ops = ops; + } + return hdr; +} +EXPORT_SYMBOL_GPL(nfs_pgio_header_alloc); + +/** + * nfs_pgio_data_destroy - make @hdr suitable for reuse + * + * Frees memory and releases refs from nfs_generic_pgio, so that it may + * be called again. + * + * @hdr: A header that has had nfs_generic_pgio called + */ +static void nfs_pgio_data_destroy(struct nfs_pgio_header *hdr) +{ + if (hdr->args.context) + put_nfs_open_context(hdr->args.context); + if (hdr->page_array.pagevec != hdr->page_array.page_array) + kfree(hdr->page_array.pagevec); +} + +/* + * nfs_pgio_header_free - Free a read or write header + * @hdr: The header to free + */ +void nfs_pgio_header_free(struct nfs_pgio_header *hdr) +{ + nfs_pgio_data_destroy(hdr); + hdr->rw_ops->rw_free_header(hdr); +} +EXPORT_SYMBOL_GPL(nfs_pgio_header_free); + +/** + * nfs_pgio_rpcsetup - Set up arguments for a pageio call + * @hdr: The pageio hdr + * @pgbase: base + * @count: Number of bytes to read + * @how: How to commit data (writes only) + * @cinfo: Commit information for the call (writes only) + */ +static void nfs_pgio_rpcsetup(struct nfs_pgio_header *hdr, unsigned int pgbase, + unsigned int count, int how, + struct nfs_commit_info *cinfo) +{ + struct nfs_page *req = hdr->req; + + /* Set up the RPC argument and reply structs + * NB: take care not to mess about with hdr->commit et al. */ + + hdr->args.fh = NFS_FH(hdr->inode); + hdr->args.offset = req_offset(req); + /* pnfs_set_layoutcommit needs this */ + hdr->mds_offset = hdr->args.offset; + hdr->args.pgbase = pgbase; + hdr->args.pages = hdr->page_array.pagevec; + hdr->args.count = count; + hdr->args.context = get_nfs_open_context(nfs_req_openctx(req)); + hdr->args.lock_context = req->wb_lock_context; + hdr->args.stable = NFS_UNSTABLE; + switch (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) { + case 0: + break; + case FLUSH_COND_STABLE: + if (nfs_reqs_to_commit(cinfo)) + break; + fallthrough; + default: + hdr->args.stable = NFS_FILE_SYNC; + } + + hdr->res.fattr = &hdr->fattr; + hdr->res.count = 0; + hdr->res.eof = 0; + hdr->res.verf = &hdr->verf; + nfs_fattr_init(&hdr->fattr); +} + +/** + * nfs_pgio_prepare - Prepare pageio hdr to go over the wire + * @task: The current task + * @calldata: pageio header to prepare + */ +static void nfs_pgio_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_pgio_header *hdr = calldata; + int err; + err = NFS_PROTO(hdr->inode)->pgio_rpc_prepare(task, hdr); + if (err) + rpc_exit(task, err); +} + +int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_header *hdr, + const struct cred *cred, const struct nfs_rpc_ops *rpc_ops, + const struct rpc_call_ops *call_ops, int how, int flags, + struct nfsd_file *localio) +{ + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &hdr->args, + .rpc_resp = &hdr->res, + .rpc_cred = cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clnt, + .task = &hdr->task, + .rpc_message = &msg, + .callback_ops = call_ops, + .callback_data = hdr, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | flags, + }; + + if (nfs_server_capable(hdr->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + hdr->rw_ops->rw_initiate(hdr, &msg, rpc_ops, &task_setup_data, how); + + dprintk("NFS: initiated pgio call " + "(req %s/%llu, %u bytes @ offset %llu)\n", + hdr->inode->i_sb->s_id, + (unsigned long long)hdr->inode->i_ino, + hdr->args.count, + (unsigned long long)hdr->args.offset); + + if (localio) + return nfs_local_doio(NFS_SERVER(hdr->inode)->nfs_client, + localio, hdr, call_ops); + + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + rpc_put_task(task); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_initiate_pgio); + +/** + * nfs_pgio_error - Clean up from a pageio error + * @hdr: pageio header + */ +static void nfs_pgio_error(struct nfs_pgio_header *hdr) +{ + set_bit(NFS_IOHDR_REDO, &hdr->flags); + hdr->completion_ops->completion(hdr); +} + +/** + * nfs_pgio_release - Release pageio data + * @calldata: The pageio header to release + */ +static void nfs_pgio_release(void *calldata) +{ + struct nfs_pgio_header *hdr = calldata; + hdr->completion_ops->completion(hdr); +} + +static void nfs_pageio_mirror_init(struct nfs_pgio_mirror *mirror, + unsigned int bsize) +{ + INIT_LIST_HEAD(&mirror->pg_list); + mirror->pg_bytes_written = 0; + mirror->pg_count = 0; + mirror->pg_bsize = bsize; + mirror->pg_base = 0; + mirror->pg_recoalesce = 0; +} + +/** + * nfs_pageio_init - initialise a page io descriptor + * @desc: pointer to descriptor + * @inode: pointer to inode + * @pg_ops: pointer to pageio operations + * @compl_ops: pointer to pageio completion operations + * @rw_ops: pointer to nfs read/write operations + * @bsize: io block size + * @io_flags: extra parameters for the io function + */ +void nfs_pageio_init(struct nfs_pageio_descriptor *desc, + struct inode *inode, + const struct nfs_pageio_ops *pg_ops, + const struct nfs_pgio_completion_ops *compl_ops, + const struct nfs_rw_ops *rw_ops, + size_t bsize, + int io_flags) +{ + desc->pg_moreio = 0; + desc->pg_inode = inode; + desc->pg_ops = pg_ops; + desc->pg_completion_ops = compl_ops; + desc->pg_rw_ops = rw_ops; + desc->pg_ioflags = io_flags; + desc->pg_error = 0; + desc->pg_lseg = NULL; + desc->pg_io_completion = NULL; + desc->pg_dreq = NULL; + nfs_netfs_reset_pageio_descriptor(desc); + desc->pg_bsize = bsize; + + desc->pg_mirror_count = 1; + desc->pg_mirror_idx = 0; + + desc->pg_mirrors_dynamic = NULL; + desc->pg_mirrors = desc->pg_mirrors_static; + nfs_pageio_mirror_init(&desc->pg_mirrors[0], bsize); + desc->pg_maxretrans = 0; +} + +/** + * nfs_pgio_result - Basic pageio error handling + * @task: The task that ran + * @calldata: Pageio header to check + */ +static void nfs_pgio_result(struct rpc_task *task, void *calldata) +{ + struct nfs_pgio_header *hdr = calldata; + struct inode *inode = hdr->inode; + + if (hdr->rw_ops->rw_done(task, hdr, inode) != 0) + return; + if (task->tk_status < 0) + nfs_set_pgio_error(hdr, task->tk_status, hdr->args.offset); + else + hdr->rw_ops->rw_result(task, hdr); +} + +/* + * Create an RPC task for the given read or write request and kick it. + * The page must have been locked by the caller. + * + * It may happen that the page we're passed is not marked dirty. + * This is the case if nfs_updatepage detects a conflicting request + * that has been written but not committed. + */ +int nfs_generic_pgio(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + struct nfs_page *req; + struct page **pages, + *last_page; + struct list_head *head = &mirror->pg_list; + struct nfs_commit_info cinfo; + struct nfs_page_array *pg_array = &hdr->page_array; + unsigned int pagecount, pageused; + unsigned int pg_base = offset_in_page(mirror->pg_base); + gfp_t gfp_flags = nfs_io_gfp_mask(); + + pagecount = nfs_page_array_len(pg_base, mirror->pg_count); + pg_array->npages = pagecount; + + if (pagecount <= ARRAY_SIZE(pg_array->page_array)) + pg_array->pagevec = pg_array->page_array; + else { + pg_array->pagevec = kzalloc_objs(struct page *, pagecount, + gfp_flags); + if (!pg_array->pagevec) { + pg_array->npages = 0; + nfs_pgio_error(hdr); + desc->pg_error = -ENOMEM; + return desc->pg_error; + } + } + + nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq); + pages = hdr->page_array.pagevec; + last_page = NULL; + pageused = 0; + while (!list_empty(head)) { + struct nfs_page_iter_page i; + struct page *page; + + req = nfs_list_entry(head->next); + nfs_list_move_request(req, &hdr->pages); + + if (req->wb_pgbase == 0) + last_page = NULL; + + nfs_page_iter_page_init(&i, req); + while ((page = nfs_page_iter_page_get(&i)) != NULL) { + if (last_page != page) { + pageused++; + if (pageused > pagecount) + goto full; + *pages++ = last_page = page; + } + } + } +full: + if (WARN_ON_ONCE(pageused != pagecount)) { + nfs_pgio_error(hdr); + desc->pg_error = -EINVAL; + return desc->pg_error; + } + + if ((desc->pg_ioflags & FLUSH_COND_STABLE) && + (desc->pg_moreio || nfs_reqs_to_commit(&cinfo))) + desc->pg_ioflags &= ~FLUSH_COND_STABLE; + + /* Set up the argument struct */ + nfs_pgio_rpcsetup(hdr, pg_base, mirror->pg_count, desc->pg_ioflags, + &cinfo); + desc->pg_rpc_callops = &nfs_pgio_common_ops; + return 0; +} +EXPORT_SYMBOL_GPL(nfs_generic_pgio); + +static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc) +{ + struct nfs_pgio_header *hdr; + int ret; + unsigned short task_flags = 0; + + hdr = nfs_pgio_header_alloc(desc->pg_rw_ops); + if (!hdr) { + desc->pg_error = -ENOMEM; + return desc->pg_error; + } + nfs_pgheader_init(desc, hdr, nfs_pgio_header_free); + ret = nfs_generic_pgio(desc, hdr); + if (ret == 0) { + struct nfs_client *clp = NFS_SERVER(hdr->inode)->nfs_client; + + struct nfsd_file *localio = + nfs_local_open_fh(clp, hdr->cred, hdr->args.fh, + &hdr->args.context->nfl, + hdr->args.context->mode); + + if (NFS_SERVER(hdr->inode)->nfs_client->cl_minorversion) + task_flags = RPC_TASK_MOVEABLE; + ret = nfs_initiate_pgio(NFS_CLIENT(hdr->inode), + hdr, + hdr->cred, + NFS_PROTO(hdr->inode), + desc->pg_rpc_callops, + desc->pg_ioflags, + RPC_TASK_CRED_NOREF | task_flags, + localio); + } + return ret; +} + +static struct nfs_pgio_mirror * +nfs_pageio_alloc_mirrors(struct nfs_pageio_descriptor *desc, + unsigned int mirror_count) +{ + struct nfs_pgio_mirror *ret; + unsigned int i; + + kfree(desc->pg_mirrors_dynamic); + desc->pg_mirrors_dynamic = NULL; + if (mirror_count == 1) + return desc->pg_mirrors_static; + ret = kmalloc_objs(*ret, mirror_count, nfs_io_gfp_mask()); + if (ret != NULL) { + for (i = 0; i < mirror_count; i++) + nfs_pageio_mirror_init(&ret[i], desc->pg_bsize); + desc->pg_mirrors_dynamic = ret; + } + return ret; +} + +/* + * nfs_pageio_setup_mirroring - determine if mirroring is to be used + * by calling the pg_get_mirror_count op + */ +static void nfs_pageio_setup_mirroring(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + unsigned int mirror_count = 1; + + if (pgio->pg_ops->pg_get_mirror_count) + mirror_count = pgio->pg_ops->pg_get_mirror_count(pgio, req); + if (mirror_count == pgio->pg_mirror_count || pgio->pg_error < 0) + return; + + if (!mirror_count || mirror_count > NFS_PAGEIO_DESCRIPTOR_MIRROR_MAX) { + pgio->pg_error = -EINVAL; + return; + } + + pgio->pg_mirrors = nfs_pageio_alloc_mirrors(pgio, mirror_count); + if (pgio->pg_mirrors == NULL) { + pgio->pg_error = -ENOMEM; + pgio->pg_mirrors = pgio->pg_mirrors_static; + mirror_count = 1; + } + pgio->pg_mirror_count = mirror_count; +} + +static void nfs_pageio_cleanup_mirroring(struct nfs_pageio_descriptor *pgio) +{ + pgio->pg_mirror_count = 1; + pgio->pg_mirror_idx = 0; + pgio->pg_mirrors = pgio->pg_mirrors_static; + kfree(pgio->pg_mirrors_dynamic); + pgio->pg_mirrors_dynamic = NULL; +} + +static bool nfs_match_lock_context(const struct nfs_lock_context *l1, + const struct nfs_lock_context *l2) +{ + return l1->lockowner == l2->lockowner; +} + +static bool nfs_page_is_contiguous(const struct nfs_page *prev, + const struct nfs_page *req) +{ + size_t prev_end = prev->wb_pgbase + prev->wb_bytes; + + if (req_offset(req) != req_offset(prev) + prev->wb_bytes) + return false; + if (req->wb_pgbase == 0) + return prev_end == nfs_page_max_length(prev); + if (req->wb_pgbase == prev_end) { + struct folio *folio = nfs_page_to_folio(req); + if (folio) + return folio == nfs_page_to_folio(prev); + return req->wb_page == prev->wb_page; + } + return false; +} + +/** + * nfs_coalesce_size - test two requests for compatibility + * @prev: pointer to nfs_page + * @req: pointer to nfs_page + * @pgio: pointer to nfs_pagio_descriptor + * + * The nfs_page structures 'prev' and 'req' are compared to ensure that the + * page data area they describe is contiguous, and that their RPC + * credentials, NFSv4 open state, and lockowners are the same. + * + * Returns size of the request that can be coalesced + */ +static unsigned int nfs_coalesce_size(struct nfs_page *prev, + struct nfs_page *req, + struct nfs_pageio_descriptor *pgio) +{ + struct file_lock_context *flctx; + + if (prev) { + if (!nfs_match_open_context(nfs_req_openctx(req), nfs_req_openctx(prev))) + return 0; + flctx = locks_inode_context(d_inode(nfs_req_openctx(req)->dentry)); + if (flctx != NULL && + !(list_empty_careful(&flctx->flc_posix) && + list_empty_careful(&flctx->flc_flock)) && + !nfs_match_lock_context(req->wb_lock_context, + prev->wb_lock_context)) + return 0; + if (!nfs_page_is_contiguous(prev, req)) + return 0; + } + return pgio->pg_ops->pg_test(pgio, prev, req); +} + +/** + * nfs_pageio_do_add_request - Attempt to coalesce a request into a page list. + * @desc: destination io descriptor + * @req: request + * + * If the request 'req' was successfully coalesced into the existing list + * of pages 'desc', it returns the size of req. + */ +static unsigned int +nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc, + struct nfs_page *req) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + struct nfs_page *prev = NULL; + unsigned int size; + + if (list_empty(&mirror->pg_list)) { + if (desc->pg_ops->pg_init) + desc->pg_ops->pg_init(desc, req); + if (desc->pg_error < 0) + return 0; + mirror->pg_base = req->wb_pgbase; + mirror->pg_count = 0; + mirror->pg_recoalesce = 0; + } else + prev = nfs_list_entry(mirror->pg_list.prev); + + if (desc->pg_maxretrans && req->wb_nio > desc->pg_maxretrans) { + if (NFS_SERVER(desc->pg_inode)->flags & NFS_MOUNT_SOFTERR) + desc->pg_error = -ETIMEDOUT; + else + desc->pg_error = -EIO; + return 0; + } + + size = nfs_coalesce_size(prev, req, desc); + if (size < req->wb_bytes) + return size; + nfs_list_move_request(req, &mirror->pg_list); + mirror->pg_count += req->wb_bytes; + return req->wb_bytes; +} + +/* + * Helper for nfs_pageio_add_request and nfs_pageio_complete + */ +static void nfs_pageio_doio(struct nfs_pageio_descriptor *desc) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + if (!list_empty(&mirror->pg_list)) { + int error = desc->pg_ops->pg_doio(desc); + if (error < 0) + desc->pg_error = error; + if (list_empty(&mirror->pg_list)) + mirror->pg_bytes_written += mirror->pg_count; + } +} + +static void +nfs_pageio_cleanup_request(struct nfs_pageio_descriptor *desc, + struct nfs_page *req) +{ + LIST_HEAD(head); + + nfs_list_move_request(req, &head); + desc->pg_completion_ops->error_cleanup(&head, desc->pg_error); +} + +/** + * __nfs_pageio_add_request - Attempt to coalesce a request into a page list. + * @desc: destination io descriptor + * @req: request + * + * This may split a request into subrequests which are all part of the + * same page group. If so, it will submit @req as the last one, to ensure + * the pointer to @req is still valid in case of failure. + * + * Returns true if the request 'req' was successfully coalesced into the + * existing list of pages 'desc'. + */ +static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, + struct nfs_page *req) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + struct nfs_page *subreq; + unsigned int size, subreq_size; + + nfs_page_group_lock(req); + + if (test_bit(NFS_CONTEXT_WRITE_SYNC, + &req->wb_lock_context->open_context->flags)) + desc->pg_ioflags |= FLUSH_STABLE; + subreq = req; + subreq_size = subreq->wb_bytes; + for(;;) { + size = nfs_pageio_do_add_request(desc, subreq); + if (size == subreq_size) { + /* We successfully submitted a request */ + if (subreq == req) + break; + req->wb_pgbase += size; + req->wb_bytes -= size; + req->wb_offset += size; + subreq_size = req->wb_bytes; + subreq = req; + continue; + } + if (WARN_ON_ONCE(subreq != req)) { + nfs_page_group_unlock(req); + nfs_pageio_cleanup_request(desc, subreq); + subreq = req; + subreq_size = req->wb_bytes; + nfs_page_group_lock(req); + } + if (!size) { + /* Can't coalesce any more, so do I/O */ + nfs_page_group_unlock(req); + desc->pg_moreio = 1; + nfs_pageio_doio(desc); + if (desc->pg_error < 0 || mirror->pg_recoalesce) + return 0; + /* retry add_request for this subreq */ + nfs_page_group_lock(req); + continue; + } + subreq = nfs_create_subreq(req, req->wb_pgbase, + req->wb_offset, size); + if (IS_ERR(subreq)) + goto err_ptr; + subreq_size = size; + } + + nfs_page_group_unlock(req); + return 1; +err_ptr: + desc->pg_error = PTR_ERR(subreq); + nfs_page_group_unlock(req); + return 0; +} + +static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + LIST_HEAD(head); + + do { + list_splice_init(&mirror->pg_list, &head); + mirror->pg_recoalesce = 0; + + while (!list_empty(&head)) { + struct nfs_page *req; + + req = list_first_entry(&head, struct nfs_page, wb_list); + if (__nfs_pageio_add_request(desc, req)) + continue; + if (desc->pg_error < 0) { + list_splice_tail(&head, &mirror->pg_list); + mirror->pg_recoalesce = 1; + return 0; + } + break; + } + } while (mirror->pg_recoalesce); + return 1; +} + +static int nfs_pageio_add_request_mirror(struct nfs_pageio_descriptor *desc, + struct nfs_page *req) +{ + int ret; + + do { + ret = __nfs_pageio_add_request(desc, req); + if (ret) + break; + if (desc->pg_error < 0) + break; + ret = nfs_do_recoalesce(desc); + } while (ret); + + return ret; +} + +static void nfs_pageio_error_cleanup(struct nfs_pageio_descriptor *desc) +{ + u32 midx; + struct nfs_pgio_mirror *mirror; + + if (!desc->pg_error) + return; + + for (midx = 0; midx < desc->pg_mirror_count; midx++) { + mirror = nfs_pgio_get_mirror(desc, midx); + desc->pg_completion_ops->error_cleanup(&mirror->pg_list, + desc->pg_error); + } +} + +int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, + struct nfs_page *req) +{ + u32 midx; + unsigned int pgbase, offset, bytes; + struct nfs_page *dupreq; + + pgbase = req->wb_pgbase; + offset = req->wb_offset; + bytes = req->wb_bytes; + + nfs_pageio_setup_mirroring(desc, req); + if (desc->pg_error < 0) + goto out_failed; + + /* Create the mirror instances first, and fire them off */ + for (midx = 1; midx < desc->pg_mirror_count; midx++) { + nfs_page_group_lock(req); + + dupreq = nfs_create_subreq(req, + pgbase, offset, bytes); + + nfs_page_group_unlock(req); + if (IS_ERR(dupreq)) { + desc->pg_error = PTR_ERR(dupreq); + goto out_failed; + } + + nfs_pgio_set_current_mirror(desc, midx); + if (!nfs_pageio_add_request_mirror(desc, dupreq)) + goto out_cleanup_subreq; + } + + nfs_pgio_set_current_mirror(desc, 0); + if (!nfs_pageio_add_request_mirror(desc, req)) + goto out_failed; + + return 1; + +out_cleanup_subreq: + nfs_pageio_cleanup_request(desc, dupreq); +out_failed: + nfs_pageio_error_cleanup(desc); + return 0; +} + +/* + * nfs_pageio_complete_mirror - Complete I/O on the current mirror of an + * nfs_pageio_descriptor + * @desc: pointer to io descriptor + * @mirror_idx: pointer to mirror index + */ +static void nfs_pageio_complete_mirror(struct nfs_pageio_descriptor *desc, + u32 mirror_idx) +{ + struct nfs_pgio_mirror *mirror; + u32 restore_idx; + + restore_idx = nfs_pgio_set_current_mirror(desc, mirror_idx); + mirror = nfs_pgio_current_mirror(desc); + + for (;;) { + nfs_pageio_doio(desc); + if (desc->pg_error < 0 || !mirror->pg_recoalesce) + break; + if (!nfs_do_recoalesce(desc)) + break; + } + nfs_pgio_set_current_mirror(desc, restore_idx); +} + +/* + * nfs_pageio_resend - Transfer requests to new descriptor and resend + * @hdr - the pgio header to move request from + * @desc - the pageio descriptor to add requests to + * + * Try to move each request (nfs_page) from @hdr to @desc then attempt + * to send them. + * + * Returns 0 on success and < 0 on error. + */ +int nfs_pageio_resend(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + LIST_HEAD(pages); + + desc->pg_io_completion = hdr->io_completion; + desc->pg_dreq = hdr->dreq; + nfs_netfs_set_pageio_descriptor(desc, hdr); + list_splice_init(&hdr->pages, &pages); + while (!list_empty(&pages)) { + struct nfs_page *req = nfs_list_entry(pages.next); + + if (!nfs_pageio_add_request(desc, req)) + break; + } + nfs_pageio_complete(desc); + if (!list_empty(&pages)) { + int err = desc->pg_error < 0 ? desc->pg_error : -EIO; + hdr->completion_ops->error_cleanup(&pages, err); + nfs_set_pgio_error(hdr, err, hdr->io_start); + return err; + } + return 0; +} +EXPORT_SYMBOL_GPL(nfs_pageio_resend); + +/** + * nfs_pageio_complete - Complete I/O then cleanup an nfs_pageio_descriptor + * @desc: pointer to io descriptor + */ +void nfs_pageio_complete(struct nfs_pageio_descriptor *desc) +{ + u32 midx; + + for (midx = 0; midx < desc->pg_mirror_count; midx++) + nfs_pageio_complete_mirror(desc, midx); + + if (desc->pg_error < 0) + nfs_pageio_error_cleanup(desc); + if (desc->pg_ops->pg_cleanup) + desc->pg_ops->pg_cleanup(desc); + nfs_pageio_cleanup_mirroring(desc); +} + +/** + * nfs_pageio_cond_complete - Conditional I/O completion + * @desc: pointer to io descriptor + * @index: page index + * + * It is important to ensure that processes don't try to take locks + * on non-contiguous ranges of pages as that might deadlock. This + * function should be called before attempting to wait on a locked + * nfs_page. It will complete the I/O if the page index 'index' + * is not contiguous with the existing list of pages in 'desc'. + */ +void nfs_pageio_cond_complete(struct nfs_pageio_descriptor *desc, pgoff_t index) +{ + struct nfs_pgio_mirror *mirror; + struct nfs_page *prev; + struct folio *folio; + u32 midx; + + for (midx = 0; midx < desc->pg_mirror_count; midx++) { + mirror = nfs_pgio_get_mirror(desc, midx); + if (!list_empty(&mirror->pg_list)) { + prev = nfs_list_entry(mirror->pg_list.prev); + folio = nfs_page_to_folio(prev); + if (folio) { + if (index == folio_next_index(folio)) + continue; + } else if (index == prev->wb_index + 1) + continue; + /* + * We will submit more requests after these. Indicate + * this to the underlying layers. + */ + desc->pg_moreio = 1; + nfs_pageio_complete(desc); + break; + } + } +} + +/* + * nfs_pageio_stop_mirroring - stop using mirroring (set mirror count to 1) + */ +void nfs_pageio_stop_mirroring(struct nfs_pageio_descriptor *pgio) +{ + nfs_pageio_complete(pgio); +} + +int __init nfs_init_nfspagecache(void) +{ + nfs_page_cachep = kmem_cache_create("nfs_page", + sizeof(struct nfs_page), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (nfs_page_cachep == NULL) + return -ENOMEM; + + return 0; +} + +void nfs_destroy_nfspagecache(void) +{ + kmem_cache_destroy(nfs_page_cachep); +} + +static const struct rpc_call_ops nfs_pgio_common_ops = { + .rpc_call_prepare = nfs_pgio_prepare, + .rpc_call_done = nfs_pgio_result, + .rpc_release = nfs_pgio_release, +}; + +const struct nfs_pageio_ops nfs_pgio_rw_ops = { + .pg_test = nfs_generic_pg_test, + .pg_doio = nfs_generic_pg_pgios, +}; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c new file mode 100644 index 000000000..7715e2bd5 --- /dev/null +++ b/fs/nfs/pnfs.c @@ -0,0 +1,3535 @@ +/* + * pNFS functions to call and manage layout drivers. + * + * Copyright (c) 2002 [year of first publication] + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#include +#include +#include +#include +#include "internal.h" +#include "pnfs.h" +#include "iostat.h" +#include "nfs4trace.h" +#include "delegation.h" +#include "nfs42.h" +#include "nfs4_fs.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS +#define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ) + +/* Locking: + * + * pnfs_spinlock: + * protects pnfs_modules_tbl. + */ +static DEFINE_SPINLOCK(pnfs_spinlock); + +/* + * pnfs_modules_tbl holds all pnfs modules + */ +static LIST_HEAD(pnfs_modules_tbl); + +static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo); +static void pnfs_free_returned_lsegs(struct pnfs_layout_hdr *lo, + struct list_head *free_me, + const struct pnfs_layout_range *range, + u32 seq); +static bool pnfs_lseg_dec_and_remove_zero(struct pnfs_layout_segment *lseg, + struct list_head *tmp_list); +static int pnfs_layout_return_on_reboot(struct pnfs_layout_hdr *lo); + +/* Return the registered pnfs layout driver module matching given id */ +static struct pnfs_layoutdriver_type * +find_pnfs_driver_locked(u32 id) +{ + struct pnfs_layoutdriver_type *local; + + list_for_each_entry(local, &pnfs_modules_tbl, pnfs_tblid) + if (local->id == id) + goto out; + local = NULL; +out: + dprintk("%s: Searching for id %u, found %p\n", __func__, id, local); + return local; +} + +static struct pnfs_layoutdriver_type * +find_pnfs_driver(u32 id) +{ + struct pnfs_layoutdriver_type *local; + + spin_lock(&pnfs_spinlock); + local = find_pnfs_driver_locked(id); + if (local != NULL && !try_module_get(local->owner)) { + dprintk("%s: Could not grab reference on module\n", __func__); + local = NULL; + } + spin_unlock(&pnfs_spinlock); + return local; +} + +const struct pnfs_layoutdriver_type *pnfs_find_layoutdriver(u32 id) +{ + return find_pnfs_driver(id); +} + +void pnfs_put_layoutdriver(const struct pnfs_layoutdriver_type *ld) +{ + if (ld) + module_put(ld->owner); +} + +void +unset_pnfs_layoutdriver(struct nfs_server *nfss) +{ + if (nfss->pnfs_curr_ld) { + if (nfss->pnfs_curr_ld->clear_layoutdriver) + nfss->pnfs_curr_ld->clear_layoutdriver(nfss); + /* Decrement the MDS count. Purge the deviceid cache if zero */ + if (atomic_dec_and_test(&nfss->nfs_client->cl_mds_count)) + nfs4_deviceid_purge_client(nfss->nfs_client); + module_put(nfss->pnfs_curr_ld->owner); + } + nfss->pnfs_curr_ld = NULL; +} + +/* + * When the server sends a list of layout types, we choose one in the order + * given in the list below. + * + * FIXME: should this list be configurable in some fashion? module param? + * mount option? something else? + */ +static const u32 ld_prefs[] = { + LAYOUT_SCSI, + LAYOUT_BLOCK_VOLUME, + LAYOUT_OSD2_OBJECTS, + LAYOUT_FLEX_FILES, + LAYOUT_NFSV4_1_FILES, + 0 +}; + +static int +ld_cmp(const void *e1, const void *e2) +{ + u32 ld1 = *((u32 *)e1); + u32 ld2 = *((u32 *)e2); + int i; + + for (i = 0; ld_prefs[i] != 0; i++) { + if (ld1 == ld_prefs[i]) + return -1; + + if (ld2 == ld_prefs[i]) + return 1; + } + return 0; +} + +/* + * Try to set the server's pnfs module to the pnfs layout type specified by id. + * Currently only one pNFS layout driver per filesystem is supported. + * + * @ids array of layout types supported by MDS. + */ +void +set_pnfs_layoutdriver(struct nfs_server *server, const struct nfs_fh *mntfh, + struct nfs_fsinfo *fsinfo) +{ + struct pnfs_layoutdriver_type *ld_type = NULL; + u32 id; + int i; + + if (fsinfo->nlayouttypes == 0) + goto out_no_driver; + if (!(server->nfs_client->cl_exchange_flags & + (EXCHGID4_FLAG_USE_NON_PNFS | EXCHGID4_FLAG_USE_PNFS_MDS))) { + printk(KERN_ERR "NFS: %s: cl_exchange_flags 0x%x\n", + __func__, server->nfs_client->cl_exchange_flags); + goto out_no_driver; + } + + sort(fsinfo->layouttype, fsinfo->nlayouttypes, + sizeof(*fsinfo->layouttype), ld_cmp, NULL); + + for (i = 0; i < fsinfo->nlayouttypes; i++) { + id = fsinfo->layouttype[i]; + ld_type = find_pnfs_driver(id); + if (!ld_type) { + request_module("%s-%u", LAYOUT_NFSV4_1_MODULE_PREFIX, + id); + ld_type = find_pnfs_driver(id); + } + if (ld_type) + break; + } + + if (!ld_type) { + dprintk("%s: No pNFS module found!\n", __func__); + goto out_no_driver; + } + + server->pnfs_curr_ld = ld_type; + if (ld_type->set_layoutdriver + && ld_type->set_layoutdriver(server, mntfh)) { + printk(KERN_ERR "NFS: %s: Error initializing pNFS layout " + "driver %u.\n", __func__, id); + module_put(ld_type->owner); + goto out_no_driver; + } + /* Bump the MDS count */ + atomic_inc(&server->nfs_client->cl_mds_count); + + dprintk("%s: pNFS module for %u set\n", __func__, id); + return; + +out_no_driver: + dprintk("%s: Using NFSv4 I/O\n", __func__); + server->pnfs_curr_ld = NULL; +} + +int +pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *ld_type) +{ + int status = -EINVAL; + struct pnfs_layoutdriver_type *tmp; + + if (ld_type->id == 0) { + printk(KERN_ERR "NFS: %s id 0 is reserved\n", __func__); + return status; + } + if (!ld_type->alloc_lseg || !ld_type->free_lseg) { + printk(KERN_ERR "NFS: %s Layout driver must provide " + "alloc_lseg and free_lseg.\n", __func__); + return status; + } + + spin_lock(&pnfs_spinlock); + tmp = find_pnfs_driver_locked(ld_type->id); + if (!tmp) { + list_add(&ld_type->pnfs_tblid, &pnfs_modules_tbl); + status = 0; + dprintk("%s Registering id:%u name:%s\n", __func__, ld_type->id, + ld_type->name); + } else { + printk(KERN_ERR "NFS: %s Module with id %d already loaded!\n", + __func__, ld_type->id); + } + spin_unlock(&pnfs_spinlock); + + return status; +} +EXPORT_SYMBOL_GPL(pnfs_register_layoutdriver); + +void +pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *ld_type) +{ + dprintk("%s Deregistering id:%u\n", __func__, ld_type->id); + spin_lock(&pnfs_spinlock); + list_del(&ld_type->pnfs_tblid); + spin_unlock(&pnfs_spinlock); +} +EXPORT_SYMBOL_GPL(pnfs_unregister_layoutdriver); + +/* + * pNFS client layout cache + */ + +/* Need to hold i_lock if caller does not already hold reference */ +void +pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo) +{ + refcount_inc(&lo->plh_refcount); +} + +static struct pnfs_layout_hdr * +pnfs_alloc_layout_hdr(struct inode *ino, gfp_t gfp_flags) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld; + return ld->alloc_layout_hdr(ino, gfp_flags); +} + +static void +pnfs_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct nfs_server *server = NFS_SERVER(lo->plh_inode); + struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld; + + if (test_and_clear_bit(NFS_LAYOUT_HASHED, &lo->plh_flags)) { + struct nfs_client *clp = server->nfs_client; + + spin_lock(&clp->cl_lock); + list_del_rcu(&lo->plh_layouts); + spin_unlock(&clp->cl_lock); + } + put_cred(lo->plh_lc_cred); + return ld->free_layout_hdr(lo); +} + +static void +pnfs_detach_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct nfs_inode *nfsi = NFS_I(lo->plh_inode); + dprintk("%s: freeing layout cache %p\n", __func__, lo); + nfsi->layout = NULL; + /* Reset MDS Threshold I/O counters */ + nfsi->write_io = 0; + nfsi->read_io = 0; +} + +void +pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct inode *inode; + + if (!lo) + return; + inode = lo->plh_inode; + pnfs_layoutreturn_before_put_layout_hdr(lo); + + if (refcount_dec_and_lock(&lo->plh_refcount, &inode->i_lock)) { + if (!list_empty(&lo->plh_segs)) + WARN_ONCE(1, "NFS: BUG unfreed layout segments.\n"); + pnfs_detach_layout_hdr(lo); + /* Notify pnfs_destroy_layout_final() that we're done */ + if (inode_state_read(inode) & (I_FREEING | I_CLEAR)) + wake_up_var_locked(lo, &inode->i_lock); + spin_unlock(&inode->i_lock); + pnfs_free_layout_hdr(lo); + } +} + +static struct inode * +pnfs_grab_inode_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct inode *inode = igrab(lo->plh_inode); + if (inode) + return inode; + set_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags); + return NULL; +} + +/* + * Compare 2 layout stateid sequence ids, to see which is newer, + * taking into account wraparound issues. + */ +static bool pnfs_seqid_is_newer(u32 s1, u32 s2) +{ + return (s32)(s1 - s2) > 0; +} + +static void pnfs_barrier_update(struct pnfs_layout_hdr *lo, u32 newseq) +{ + if (pnfs_seqid_is_newer(newseq, lo->plh_barrier) || !lo->plh_barrier) + lo->plh_barrier = newseq; +} + +static void +pnfs_set_plh_return_info(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode, + u32 seq) +{ + if (lo->plh_return_iomode != 0 && lo->plh_return_iomode != iomode) + iomode = IOMODE_ANY; + lo->plh_return_iomode = iomode; + set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); + /* + * We must set lo->plh_return_seq to avoid livelocks with + * pnfs_layout_need_return() + */ + if (seq == 0) + seq = be32_to_cpu(lo->plh_stateid.seqid); + if (!lo->plh_return_seq || pnfs_seqid_is_newer(seq, lo->plh_return_seq)) + lo->plh_return_seq = seq; + pnfs_barrier_update(lo, seq); +} + +static void +pnfs_clear_layoutreturn_info(struct pnfs_layout_hdr *lo) +{ + struct pnfs_layout_segment *lseg; + lo->plh_return_iomode = 0; + lo->plh_return_seq = 0; + clear_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); + list_for_each_entry(lseg, &lo->plh_segs, pls_list) { + if (!test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) + continue; + pnfs_set_plh_return_info(lo, lseg->pls_range.iomode, 0); + } +} + +static void pnfs_clear_layoutreturn_waitbit(struct pnfs_layout_hdr *lo) +{ + clear_bit_unlock(NFS_LAYOUT_RETURN, &lo->plh_flags); + clear_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags); + smp_mb__after_atomic(); + wake_up_bit(&lo->plh_flags, NFS_LAYOUT_RETURN); + rpc_wake_up(&NFS_SERVER(lo->plh_inode)->roc_rpcwaitq); +} + +static void +pnfs_clear_lseg_state(struct pnfs_layout_segment *lseg, + struct list_head *free_me) +{ + clear_bit(NFS_LSEG_ROC, &lseg->pls_flags); + clear_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags); + if (test_and_clear_bit(NFS_LSEG_VALID, &lseg->pls_flags)) + pnfs_lseg_dec_and_remove_zero(lseg, free_me); + if (test_and_clear_bit(NFS_LSEG_LAYOUTCOMMIT, &lseg->pls_flags)) + pnfs_lseg_dec_and_remove_zero(lseg, free_me); +} + +/* + * Update the seqid of a layout stateid after receiving + * NFS4ERR_OLD_STATEID + */ +bool nfs4_layout_refresh_old_stateid(nfs4_stateid *dst, + struct pnfs_layout_range *dst_range, + struct inode *inode) +{ + struct pnfs_layout_hdr *lo; + struct pnfs_layout_range range = { + .iomode = IOMODE_ANY, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + bool ret = false; + LIST_HEAD(head); + int err; + + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (lo && pnfs_layout_is_valid(lo) && + nfs4_stateid_match_other(dst, &lo->plh_stateid)) { + /* Is our call using the most recent seqid? If so, bump it */ + if (!nfs4_stateid_is_newer(&lo->plh_stateid, dst)) { + nfs4_stateid_seqid_inc(dst); + ret = true; + goto out; + } + /* Try to update the seqid to the most recent */ + err = pnfs_mark_matching_lsegs_return(lo, &head, &range, 0); + if (err != -EBUSY) { + dst->seqid = lo->plh_stateid.seqid; + *dst_range = range; + ret = true; + } + } +out: + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&head); + return ret; +} + +/* + * Mark a pnfs_layout_hdr and all associated layout segments as invalid + * + * In order to continue using the pnfs_layout_hdr, a full recovery + * is required. + * Note that caller must hold inode->i_lock. + */ +int +pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo, + struct list_head *lseg_list) +{ + struct pnfs_layout_range range = { + .iomode = IOMODE_ANY, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + struct pnfs_layout_segment *lseg, *next; + + if (test_and_set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) + return !list_empty(&lo->plh_segs); + clear_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(lo->plh_inode)->flags); + list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) + pnfs_clear_lseg_state(lseg, lseg_list); + pnfs_clear_layoutreturn_info(lo); + pnfs_free_returned_lsegs(lo, lseg_list, &range, 0); + set_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags); + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags) && + !test_and_set_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) + pnfs_clear_layoutreturn_waitbit(lo); + return !list_empty(&lo->plh_segs); +} + +static int pnfs_mark_layout_stateid_return(struct pnfs_layout_hdr *lo, + struct list_head *lseg_list, + enum pnfs_iomode iomode, u32 seq) +{ + struct pnfs_layout_range range = { + .iomode = iomode, + .length = NFS4_MAX_UINT64, + }; + + return pnfs_mark_matching_lsegs_return(lo, lseg_list, &range, seq); +} + +static int +pnfs_iomode_to_fail_bit(u32 iomode) +{ + return iomode == IOMODE_RW ? + NFS_LAYOUT_RW_FAILED : NFS_LAYOUT_RO_FAILED; +} + +static void +pnfs_layout_set_fail_bit(struct pnfs_layout_hdr *lo, int fail_bit) +{ + lo->plh_retry_timestamp = jiffies; + if (!test_and_set_bit(fail_bit, &lo->plh_flags)) + refcount_inc(&lo->plh_refcount); +} + +static void +pnfs_layout_clear_fail_bit(struct pnfs_layout_hdr *lo, int fail_bit) +{ + if (test_and_clear_bit(fail_bit, &lo->plh_flags)) + refcount_dec(&lo->plh_refcount); +} + +static void +pnfs_layout_io_set_failed(struct pnfs_layout_hdr *lo, u32 iomode) +{ + struct inode *inode = lo->plh_inode; + struct pnfs_layout_range range = { + .iomode = iomode, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + LIST_HEAD(head); + + spin_lock(&inode->i_lock); + pnfs_layout_set_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + pnfs_mark_matching_lsegs_return(lo, &head, &range, 0); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&head); + dprintk("%s Setting layout IOMODE_%s fail bit\n", __func__, + iomode == IOMODE_RW ? "RW" : "READ"); +} + +static bool +pnfs_layout_io_test_failed(struct pnfs_layout_hdr *lo, u32 iomode) +{ + unsigned long start, end; + int fail_bit = pnfs_iomode_to_fail_bit(iomode); + + if (test_bit(fail_bit, &lo->plh_flags) == 0) + return false; + end = jiffies; + start = end - PNFS_LAYOUTGET_RETRY_TIMEOUT; + if (!time_in_range(lo->plh_retry_timestamp, start, end)) { + /* It is time to retry the failed layoutgets */ + pnfs_layout_clear_fail_bit(lo, fail_bit); + return false; + } + return true; +} + +static void +pnfs_init_lseg(struct pnfs_layout_hdr *lo, struct pnfs_layout_segment *lseg, + const struct pnfs_layout_range *range, + const nfs4_stateid *stateid) +{ + INIT_LIST_HEAD(&lseg->pls_list); + INIT_LIST_HEAD(&lseg->pls_lc_list); + INIT_LIST_HEAD(&lseg->pls_commits); + refcount_set(&lseg->pls_refcount, 1); + set_bit(NFS_LSEG_VALID, &lseg->pls_flags); + lseg->pls_layout = lo; + lseg->pls_range = *range; + lseg->pls_seq = be32_to_cpu(stateid->seqid); +} + +static void pnfs_free_lseg(struct pnfs_layout_segment *lseg) +{ + if (lseg != NULL) { + struct inode *inode = lseg->pls_layout->plh_inode; + NFS_SERVER(inode)->pnfs_curr_ld->free_lseg(lseg); + } +} + +static void +pnfs_layout_remove_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg) +{ + WARN_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); + list_del_init(&lseg->pls_list); + /* Matched by pnfs_get_layout_hdr in pnfs_layout_insert_lseg */ + refcount_dec(&lo->plh_refcount); + if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) + return; + if (list_empty(&lo->plh_segs) && + !test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags) && + !test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) { + if (atomic_read(&lo->plh_outstanding) == 0) + set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); + clear_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); + } +} + +static bool +pnfs_cache_lseg_for_layoutreturn(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg) +{ + if (test_and_clear_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags) && + pnfs_layout_is_valid(lo)) { + pnfs_set_plh_return_info(lo, lseg->pls_range.iomode, 0); + list_move_tail(&lseg->pls_list, &lo->plh_return_segs); + return true; + } + return false; +} + +void +pnfs_put_lseg(struct pnfs_layout_segment *lseg) +{ + struct pnfs_layout_hdr *lo; + struct inode *inode; + + if (!lseg) + return; + + dprintk("%s: lseg %p ref %d valid %d\n", __func__, lseg, + refcount_read(&lseg->pls_refcount), + test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); + + lo = lseg->pls_layout; + inode = lo->plh_inode; + + if (refcount_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) { + pnfs_get_layout_hdr(lo); + pnfs_layout_remove_lseg(lo, lseg); + if (pnfs_cache_lseg_for_layoutreturn(lo, lseg)) + lseg = NULL; + spin_unlock(&inode->i_lock); + pnfs_free_lseg(lseg); + pnfs_put_layout_hdr(lo); + } +} +EXPORT_SYMBOL_GPL(pnfs_put_lseg); + +/* + * is l2 fully contained in l1? + * start1 end1 + * [----------------------------------) + * start2 end2 + * [----------------) + */ +static bool +pnfs_lseg_range_contained(const struct pnfs_layout_range *l1, + const struct pnfs_layout_range *l2) +{ + u64 start1 = l1->offset; + u64 end1 = pnfs_end_offset(start1, l1->length); + u64 start2 = l2->offset; + u64 end2 = pnfs_end_offset(start2, l2->length); + + return (start1 <= start2) && (end1 >= end2); +} + +static bool pnfs_lseg_dec_and_remove_zero(struct pnfs_layout_segment *lseg, + struct list_head *tmp_list) +{ + if (!refcount_dec_and_test(&lseg->pls_refcount)) + return false; + pnfs_layout_remove_lseg(lseg->pls_layout, lseg); + list_add(&lseg->pls_list, tmp_list); + return true; +} + +/* Returns 1 if lseg is removed from list, 0 otherwise */ +static int mark_lseg_invalid(struct pnfs_layout_segment *lseg, + struct list_head *tmp_list) +{ + int rv = 0; + + if (test_and_clear_bit(NFS_LSEG_VALID, &lseg->pls_flags)) { + /* Remove the reference keeping the lseg in the + * list. It will now be removed when all + * outstanding io is finished. + */ + dprintk("%s: lseg %p ref %d\n", __func__, lseg, + refcount_read(&lseg->pls_refcount)); + if (pnfs_lseg_dec_and_remove_zero(lseg, tmp_list)) + rv = 1; + } + return rv; +} + +static bool +pnfs_should_free_range(const struct pnfs_layout_range *lseg_range, + const struct pnfs_layout_range *recall_range) +{ + return (recall_range->iomode == IOMODE_ANY || + lseg_range->iomode == recall_range->iomode) && + pnfs_lseg_range_intersecting(lseg_range, recall_range); +} + +static bool +pnfs_match_lseg_recall(const struct pnfs_layout_segment *lseg, + const struct pnfs_layout_range *recall_range, + u32 seq) +{ + if (seq != 0 && pnfs_seqid_is_newer(lseg->pls_seq, seq)) + return false; + if (recall_range == NULL) + return true; + return pnfs_should_free_range(&lseg->pls_range, recall_range); +} + +/** + * pnfs_mark_matching_lsegs_invalid - tear down lsegs or mark them for later + * @lo: layout header containing the lsegs + * @tmp_list: list head where doomed lsegs should go + * @recall_range: optional recall range argument to match (may be NULL) + * @seq: only invalidate lsegs obtained prior to this sequence (may be 0) + * + * Walk the list of lsegs in the layout header, and tear down any that should + * be destroyed. If "recall_range" is specified then the segment must match + * that range. If "seq" is non-zero, then only match segments that were handed + * out at or before that sequence. + * + * Returns number of matching invalid lsegs remaining in list after scanning + * it and purging them. + */ +int +pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, + struct list_head *tmp_list, + const struct pnfs_layout_range *recall_range, + u32 seq) +{ + struct pnfs_layout_segment *lseg, *next; + struct nfs_server *server = NFS_SERVER(lo->plh_inode); + int remaining = 0; + + dprintk("%s:Begin lo %p\n", __func__, lo); + + if (list_empty(&lo->plh_segs)) + return 0; + list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) + if (pnfs_match_lseg_recall(lseg, recall_range, seq)) { + dprintk("%s: freeing lseg %p iomode %d seq %u " + "offset %llu length %llu\n", __func__, + lseg, lseg->pls_range.iomode, lseg->pls_seq, + lseg->pls_range.offset, lseg->pls_range.length); + if (mark_lseg_invalid(lseg, tmp_list)) + continue; + remaining++; + pnfs_lseg_cancel_io(server, lseg); + } + dprintk("%s:Return %i\n", __func__, remaining); + return remaining; +} + +static void pnfs_reset_return_info(struct pnfs_layout_hdr *lo) +{ + struct pnfs_layout_segment *lseg; + + list_for_each_entry(lseg, &lo->plh_return_segs, pls_list) + pnfs_set_plh_return_info(lo, lseg->pls_range.iomode, 0); +} + +static void +pnfs_free_returned_lsegs(struct pnfs_layout_hdr *lo, + struct list_head *free_me, + const struct pnfs_layout_range *range, + u32 seq) +{ + struct pnfs_layout_segment *lseg, *next; + + list_for_each_entry_safe(lseg, next, &lo->plh_return_segs, pls_list) { + if (pnfs_match_lseg_recall(lseg, range, seq)) + list_move_tail(&lseg->pls_list, free_me); + } +} + +/* note free_me must contain lsegs from a single layout_hdr */ +void +pnfs_free_lseg_list(struct list_head *free_me) +{ + struct pnfs_layout_segment *lseg, *tmp; + + if (list_empty(free_me)) + return; + + list_for_each_entry_safe(lseg, tmp, free_me, pls_list) { + list_del(&lseg->pls_list); + pnfs_free_lseg(lseg); + } +} + +static struct pnfs_layout_hdr *__pnfs_destroy_layout(struct nfs_inode *nfsi) +{ + struct pnfs_layout_hdr *lo; + LIST_HEAD(tmp_list); + + spin_lock(&nfsi->vfs_inode.i_lock); + lo = nfsi->layout; + if (lo) { + pnfs_get_layout_hdr(lo); + pnfs_mark_layout_stateid_invalid(lo, &tmp_list); + pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RO_FAILED); + pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RW_FAILED); + spin_unlock(&nfsi->vfs_inode.i_lock); + pnfs_free_lseg_list(&tmp_list); + nfs_commit_inode(&nfsi->vfs_inode, 0); + pnfs_put_layout_hdr(lo); + } else + spin_unlock(&nfsi->vfs_inode.i_lock); + return lo; +} + +void pnfs_destroy_layout(struct nfs_inode *nfsi) +{ + __pnfs_destroy_layout(nfsi); +} +EXPORT_SYMBOL_GPL(pnfs_destroy_layout); + +void pnfs_destroy_layout_final(struct nfs_inode *nfsi) +{ + struct pnfs_layout_hdr *lo = __pnfs_destroy_layout(nfsi); + struct inode *inode = &nfsi->vfs_inode; + + if (lo) { + spin_lock(&inode->i_lock); + wait_var_event_spinlock(lo, nfsi->layout != lo, + &inode->i_lock); + spin_unlock(&inode->i_lock); + } +} + +static bool +pnfs_layout_add_bulk_destroy_list(struct inode *inode, + struct list_head *layout_list) +{ + struct pnfs_layout_hdr *lo; + bool ret = false; + + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (lo != NULL && list_empty(&lo->plh_bulk_destroy)) { + pnfs_get_layout_hdr(lo); + list_add(&lo->plh_bulk_destroy, layout_list); + ret = true; + } + spin_unlock(&inode->i_lock); + return ret; +} + +/* Caller must hold rcu_read_lock and clp->cl_lock */ +static int +pnfs_layout_bulk_destroy_byserver_locked(struct nfs_client *clp, + struct nfs_server *server, + struct list_head *layout_list) + __must_hold(&clp->cl_lock) + __must_hold(RCU) +{ + struct pnfs_layout_hdr *lo, *next; + struct inode *inode; + + list_for_each_entry_safe(lo, next, &server->layouts, plh_layouts) { + if (test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags) || + test_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags) || + !list_empty(&lo->plh_bulk_destroy)) + continue; + /* If the sb is being destroyed, just bail */ + if (!nfs_sb_active(server->super)) + break; + inode = pnfs_grab_inode_layout_hdr(lo); + if (inode != NULL) { + if (pnfs_layout_add_bulk_destroy_list(inode, + layout_list)) + continue; + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); + iput(inode); + } else { + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); + } + nfs_sb_deactive(server->super); + spin_lock(&clp->cl_lock); + rcu_read_lock(); + return -EAGAIN; + } + return 0; +} + +static int +pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list, + enum pnfs_layout_destroy_mode mode) +{ + struct pnfs_layout_hdr *lo; + struct inode *inode; + LIST_HEAD(lseg_list); + int ret = 0; + + while (!list_empty(layout_list)) { + lo = list_entry(layout_list->next, struct pnfs_layout_hdr, + plh_bulk_destroy); + dprintk("%s freeing layout for inode %llu\n", __func__, + lo->plh_inode->i_ino); + inode = lo->plh_inode; + + pnfs_layoutcommit_inode(inode, false); + + spin_lock(&inode->i_lock); + list_del_init(&lo->plh_bulk_destroy); + if (mode == PNFS_LAYOUT_FILE_BULK_RETURN) { + pnfs_mark_layout_stateid_return(lo, &lseg_list, + IOMODE_ANY, 0); + } else if (pnfs_mark_layout_stateid_invalid(lo, &lseg_list)) { + if (mode == PNFS_LAYOUT_BULK_RETURN) + set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); + ret = -EAGAIN; + } + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&lseg_list); + /* Free all lsegs that are attached to commit buckets */ + nfs_commit_inode(inode, 0); + pnfs_put_layout_hdr(lo); + nfs_iput_and_deactive(inode); + } + return ret; +} + +int pnfs_layout_destroy_byfsid(struct nfs_client *clp, struct nfs_fsid *fsid, + enum pnfs_layout_destroy_mode mode) +{ + struct nfs_server *server; + LIST_HEAD(layout_list); + + spin_lock(&clp->cl_lock); + rcu_read_lock(); +restart: + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (memcmp(&server->fsid, fsid, sizeof(*fsid)) != 0) + continue; + if (pnfs_layout_bulk_destroy_byserver_locked(clp, + server, + &layout_list) != 0) + goto restart; + } + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); + + return pnfs_layout_free_bulk_destroy_list(&layout_list, mode); +} + +static void pnfs_layout_build_destroy_list_byclient(struct nfs_client *clp, + struct list_head *list) +{ + struct nfs_server *server; + + spin_lock(&clp->cl_lock); + rcu_read_lock(); +restart: + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (pnfs_layout_bulk_destroy_byserver_locked(clp, server, + list) != 0) + goto restart; + } + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); +} + +static int pnfs_layout_do_destroy_byclid(struct nfs_client *clp, + struct list_head *list, + enum pnfs_layout_destroy_mode mode) +{ + pnfs_layout_build_destroy_list_byclient(clp, list); + return pnfs_layout_free_bulk_destroy_list(list, mode); +} + +int pnfs_layout_destroy_byclid(struct nfs_client *clp, + enum pnfs_layout_destroy_mode mode) +{ + LIST_HEAD(layout_list); + + return pnfs_layout_do_destroy_byclid(clp, &layout_list, mode); +} + +/* + * Called by the state manager to remove all layouts established under an + * expired lease. + */ +void +pnfs_destroy_all_layouts(struct nfs_client *clp) +{ + nfs4_deviceid_mark_client_invalid(clp); + nfs4_deviceid_purge_client(clp); + + pnfs_layout_destroy_byclid(clp, PNFS_LAYOUT_INVALIDATE); +} + +static void pnfs_layout_build_recover_list_byclient(struct nfs_client *clp, + struct list_head *list) +{ + struct nfs_server *server; + + spin_lock(&clp->cl_lock); + rcu_read_lock(); +restart: + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (!(server->caps & NFS_CAP_REBOOT_LAYOUTRETURN)) + continue; + if (pnfs_layout_bulk_destroy_byserver_locked(clp, server, + list) != 0) + goto restart; + } + rcu_read_unlock(); + spin_unlock(&clp->cl_lock); +} + +static int pnfs_layout_bulk_list_reboot(struct list_head *list) +{ + struct pnfs_layout_hdr *lo; + struct nfs_server *server; + int ret; + + list_for_each_entry(lo, list, plh_bulk_destroy) { + server = NFS_SERVER(lo->plh_inode); + ret = pnfs_layout_return_on_reboot(lo); + switch (ret) { + case 0: + continue; + case -NFS4ERR_BAD_STATEID: + server->caps &= ~NFS_CAP_REBOOT_LAYOUTRETURN; + break; + case -NFS4ERR_NO_GRACE: + break; + default: + goto err; + } + break; + } + return 0; +err: + return ret; +} + +int pnfs_layout_handle_reboot(struct nfs_client *clp) +{ + LIST_HEAD(list); + int ret = 0, ret2; + + pnfs_layout_build_recover_list_byclient(clp, &list); + if (!list_empty(&list)) + ret = pnfs_layout_bulk_list_reboot(&list); + ret2 = pnfs_layout_do_destroy_byclid(clp, &list, + PNFS_LAYOUT_INVALIDATE); + if (!ret) + ret = ret2; + return (ret == 0) ? 0 : -EAGAIN; +} + +static void +pnfs_set_layout_cred(struct pnfs_layout_hdr *lo, const struct cred *cred) +{ + const struct cred *old; + + if (cred && cred_fscmp(lo->plh_lc_cred, cred) != 0) { + old = xchg(&lo->plh_lc_cred, get_cred(cred)); + put_cred(old); + } +} + +/* update lo->plh_stateid with new if is more recent */ +void +pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new, + const struct cred *cred, bool update_barrier) +{ + u32 oldseq = be32_to_cpu(lo->plh_stateid.seqid); + u32 newseq = be32_to_cpu(new->seqid); + + if (!pnfs_layout_is_valid(lo)) { + pnfs_set_layout_cred(lo, cred); + nfs4_stateid_copy(&lo->plh_stateid, new); + lo->plh_barrier = newseq; + pnfs_clear_layoutreturn_info(lo); + clear_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags); + return; + } + + if (pnfs_seqid_is_newer(newseq, oldseq)) + nfs4_stateid_copy(&lo->plh_stateid, new); + + if (update_barrier) { + pnfs_barrier_update(lo, newseq); + return; + } + /* + * Because of wraparound, we want to keep the barrier + * "close" to the current seqids. We really only want to + * get here from a layoutget call. + */ + if (atomic_read(&lo->plh_outstanding) == 1) + pnfs_barrier_update(lo, be32_to_cpu(lo->plh_stateid.seqid)); +} + +static bool +pnfs_layout_stateid_blocked(const struct pnfs_layout_hdr *lo, + const nfs4_stateid *stateid) +{ + u32 seqid = be32_to_cpu(stateid->seqid); + + return lo->plh_barrier && pnfs_seqid_is_newer(lo->plh_barrier, seqid); +} + +/* lget is set to 1 if called from inside send_layoutget call chain */ +static bool +pnfs_layoutgets_blocked(const struct pnfs_layout_hdr *lo) +{ + return lo->plh_block_lgets || + test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); +} + +static struct nfs_server * +pnfs_find_server(struct inode *inode, struct nfs_open_context *ctx) +{ + struct nfs_server *server; + + if (inode) { + server = NFS_SERVER(inode); + } else { + struct dentry *parent_dir = dget_parent(ctx->dentry); + server = NFS_SERVER(parent_dir->d_inode); + dput(parent_dir); + } + return server; +} + +static void nfs4_free_pages(struct page **pages, size_t size) +{ + int i; + + if (!pages) + return; + + for (i = 0; i < size; i++) { + if (!pages[i]) + break; + __free_page(pages[i]); + } + kfree(pages); +} + +static struct page **nfs4_alloc_pages(size_t size, gfp_t gfp_flags) +{ + struct page **pages; + int i; + + pages = kmalloc_objs(struct page *, size, gfp_flags); + if (!pages) { + dprintk("%s: can't alloc array of %zu pages\n", __func__, size); + return NULL; + } + + for (i = 0; i < size; i++) { + pages[i] = alloc_page(gfp_flags); + if (!pages[i]) { + dprintk("%s: failed to allocate page\n", __func__); + nfs4_free_pages(pages, i); + return NULL; + } + } + + return pages; +} + +static struct nfs4_layoutget * +pnfs_alloc_init_layoutget_args(struct inode *ino, + struct nfs_open_context *ctx, + const nfs4_stateid *stateid, + const struct pnfs_layout_range *range, + gfp_t gfp_flags) +{ + struct nfs_server *server = pnfs_find_server(ino, ctx); + size_t max_reply_sz = server->pnfs_curr_ld->max_layoutget_response; + size_t max_pages = max_response_pages(server); + struct nfs4_layoutget *lgp; + + dprintk("--> %s\n", __func__); + + lgp = kzalloc_obj(*lgp, gfp_flags); + if (lgp == NULL) + return NULL; + + if (max_reply_sz) { + size_t npages = (max_reply_sz + PAGE_SIZE - 1) >> PAGE_SHIFT; + if (npages < max_pages) + max_pages = npages; + } + + lgp->args.layout.pages = nfs4_alloc_pages(max_pages, gfp_flags); + if (!lgp->args.layout.pages) { + kfree(lgp); + return NULL; + } + lgp->args.layout.pglen = max_pages * PAGE_SIZE; + lgp->res.layoutp = &lgp->args.layout; + + /* Don't confuse uninitialised result and success */ + lgp->res.status = -NFS4ERR_DELAY; + + lgp->args.minlength = PAGE_SIZE; + if (lgp->args.minlength > range->length) + lgp->args.minlength = range->length; + if (ino) { + loff_t i_size = i_size_read(ino); + + if (range->iomode == IOMODE_READ) { + if (range->offset >= i_size) + lgp->args.minlength = 0; + else if (i_size - range->offset < lgp->args.minlength) + lgp->args.minlength = i_size - range->offset; + } + } + lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE; + pnfs_copy_range(&lgp->args.range, range); + lgp->args.type = server->pnfs_curr_ld->id; + lgp->args.inode = ino; + lgp->args.ctx = get_nfs_open_context(ctx); + nfs4_stateid_copy(&lgp->args.stateid, stateid); + lgp->gfp_flags = gfp_flags; + lgp->cred = ctx->cred; + return lgp; +} + +void pnfs_layoutget_free(struct nfs4_layoutget *lgp) +{ + size_t max_pages = lgp->args.layout.pglen / PAGE_SIZE; + + nfs4_free_pages(lgp->args.layout.pages, max_pages); + pnfs_put_layout_hdr(lgp->lo); + put_nfs_open_context(lgp->args.ctx); + kfree(lgp); +} + +static void pnfs_clear_layoutcommit(struct inode *inode, + struct list_head *head) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct pnfs_layout_segment *lseg, *tmp; + + if (!test_and_clear_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) + return; + list_for_each_entry_safe(lseg, tmp, &nfsi->layout->plh_segs, pls_list) { + if (!test_and_clear_bit(NFS_LSEG_LAYOUTCOMMIT, &lseg->pls_flags)) + continue; + pnfs_lseg_dec_and_remove_zero(lseg, head); + } +} + +static void +pnfs_layoutreturn_retry_later_locked(struct pnfs_layout_hdr *lo, + const nfs4_stateid *arg_stateid, + const struct pnfs_layout_range *range, + struct list_head *freeme) +{ + if (pnfs_layout_is_valid(lo) && + nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid)) + pnfs_reset_return_info(lo); + else + pnfs_mark_layout_stateid_invalid(lo, freeme); + pnfs_clear_layoutreturn_waitbit(lo); +} + +void pnfs_layoutreturn_retry_later(struct pnfs_layout_hdr *lo, + const nfs4_stateid *arg_stateid, + const struct pnfs_layout_range *range) +{ + struct inode *inode = lo->plh_inode; + LIST_HEAD(freeme); + + spin_lock(&inode->i_lock); + pnfs_layoutreturn_retry_later_locked(lo, arg_stateid, range, &freeme); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&freeme); +} + +void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo, + const nfs4_stateid *arg_stateid, + const struct pnfs_layout_range *range, + const nfs4_stateid *stateid) +{ + struct inode *inode = lo->plh_inode; + LIST_HEAD(freeme); + + spin_lock(&inode->i_lock); + if (!nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid)) + goto out_unlock; + if (stateid && pnfs_layout_is_valid(lo)) { + u32 seq = be32_to_cpu(arg_stateid->seqid); + + pnfs_mark_matching_lsegs_invalid(lo, &freeme, range, seq); + pnfs_free_returned_lsegs(lo, &freeme, range, seq); + pnfs_set_layout_stateid(lo, stateid, NULL, true); + pnfs_reset_return_info(lo); + } else + pnfs_mark_layout_stateid_invalid(lo, &freeme); +out_unlock: + pnfs_clear_layoutreturn_waitbit(lo); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&freeme); + +} + +static bool +pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo, + nfs4_stateid *stateid, + const struct cred **cred, + enum pnfs_iomode *iomode) +{ + /* Serialise LAYOUTGET/LAYOUTRETURN */ + if (atomic_read(&lo->plh_outstanding) != 0 && lo->plh_return_seq == 0) + return false; + if (test_and_set_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) + return false; + set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags); + pnfs_get_layout_hdr(lo); + nfs4_stateid_copy(stateid, &lo->plh_stateid); + *cred = get_cred(lo->plh_lc_cred); + if (test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) { + if (lo->plh_return_seq != 0) + stateid->seqid = cpu_to_be32(lo->plh_return_seq); + if (iomode != NULL) + *iomode = lo->plh_return_iomode; + pnfs_clear_layoutreturn_info(lo); + } else if (iomode != NULL) + *iomode = IOMODE_ANY; + pnfs_barrier_update(lo, be32_to_cpu(stateid->seqid)); + return true; +} + +static void +pnfs_init_layoutreturn_args(struct nfs4_layoutreturn_args *args, + struct pnfs_layout_hdr *lo, + const nfs4_stateid *stateid, + enum pnfs_iomode iomode) +{ + struct inode *inode = lo->plh_inode; + + args->layout_type = NFS_SERVER(inode)->pnfs_curr_ld->id; + args->inode = inode; + args->range.iomode = iomode; + args->range.offset = 0; + args->range.length = NFS4_MAX_UINT64; + args->layout = lo; + nfs4_stateid_copy(&args->stateid, stateid); +} + +static int +pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, + const nfs4_stateid *stateid, + const struct cred **pcred, + enum pnfs_iomode iomode, + unsigned int flags) +{ + struct inode *ino = lo->plh_inode; + struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld; + struct nfs4_layoutreturn *lrp; + const struct cred *cred = *pcred; + int status = 0; + + *pcred = NULL; + lrp = kzalloc_obj(*lrp, nfs_io_gfp_mask()); + if (unlikely(lrp == NULL)) { + status = -ENOMEM; + spin_lock(&ino->i_lock); + pnfs_clear_layoutreturn_waitbit(lo); + spin_unlock(&ino->i_lock); + put_cred(cred); + pnfs_put_layout_hdr(lo); + goto out; + } + + pnfs_init_layoutreturn_args(&lrp->args, lo, stateid, iomode); + lrp->args.ld_private = &lrp->ld_private; + lrp->clp = NFS_SERVER(ino)->nfs_client; + lrp->cred = cred; + if (ld->prepare_layoutreturn) + ld->prepare_layoutreturn(&lrp->args); + + status = nfs4_proc_layoutreturn(lrp, flags); +out: + dprintk("<-- %s status: %d\n", __func__, status); + return status; +} + +/* Return true if layoutreturn is needed */ +static bool +pnfs_layout_need_return(struct pnfs_layout_hdr *lo) +{ + if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) + return false; + return pnfs_mark_layout_stateid_return(lo, &lo->plh_return_segs, + lo->plh_return_iomode, + lo->plh_return_seq) != EBUSY; +} + +static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct inode *inode= lo->plh_inode; + + if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) + return; + spin_lock(&inode->i_lock); + if (pnfs_layout_need_return(lo)) { + const struct cred *cred; + nfs4_stateid stateid; + enum pnfs_iomode iomode; + bool send; + + send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode); + spin_unlock(&inode->i_lock); + if (send) { + /* Send an async layoutreturn so we dont deadlock */ + pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, + PNFS_FL_LAYOUTRETURN_ASYNC); + } + } else + spin_unlock(&inode->i_lock); +} + +/* + * Initiates a LAYOUTRETURN(FILE), and removes the pnfs_layout_hdr + * when the layout segment list is empty. + * + * Note that a pnfs_layout_hdr can exist with an empty layout segment + * list when LAYOUTGET has failed, or when LAYOUTGET succeeded, but the + * deviceid is marked invalid. + */ +int +_pnfs_return_layout(struct inode *ino) +{ + struct pnfs_layout_hdr *lo = NULL; + struct nfs_inode *nfsi = NFS_I(ino); + struct pnfs_layout_range range = { + .iomode = IOMODE_ANY, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + LIST_HEAD(tmp_list); + const struct cred *cred; + nfs4_stateid stateid; + int status = 0; + bool send, valid_layout; + + dprintk("NFS: %s for inode %llu\n", __func__, ino->i_ino); + + spin_lock(&ino->i_lock); + lo = nfsi->layout; + if (!lo) { + spin_unlock(&ino->i_lock); + dprintk("NFS: %s no layout to return\n", __func__); + goto out; + } + /* Reference matched in nfs4_layoutreturn_release */ + pnfs_get_layout_hdr(lo); + /* Is there an outstanding layoutreturn ? */ + if (test_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) { + spin_unlock(&ino->i_lock); + if (wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, + TASK_UNINTERRUPTIBLE)) + goto out_put_layout_hdr; + spin_lock(&ino->i_lock); + } + valid_layout = pnfs_layout_is_valid(lo); + pnfs_clear_layoutcommit(ino, &tmp_list); + pnfs_mark_matching_lsegs_return(lo, &tmp_list, &range, 0); + + + /* Don't send a LAYOUTRETURN if list was initially empty */ + if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags) || + !valid_layout) { + spin_unlock(&ino->i_lock); + dprintk("NFS: %s no layout segments to return\n", __func__); + goto out_wait_layoutreturn; + } + + send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, NULL); + spin_unlock(&ino->i_lock); + if (NFS_SERVER(ino)->pnfs_curr_ld->return_range) + NFS_SERVER(ino)->pnfs_curr_ld->return_range(lo, &range); + if (send) + status = pnfs_send_layoutreturn(lo, &stateid, &cred, IOMODE_ANY, + 0); +out_wait_layoutreturn: + wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, TASK_UNINTERRUPTIBLE); +out_put_layout_hdr: + pnfs_free_lseg_list(&tmp_list); + pnfs_put_layout_hdr(lo); +out: + dprintk("<-- %s status: %d\n", __func__, status); + return status; +} + +int +pnfs_commit_and_return_layout(struct inode *inode) +{ + struct pnfs_layout_hdr *lo; + int ret; + + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (lo == NULL) { + spin_unlock(&inode->i_lock); + return 0; + } + pnfs_get_layout_hdr(lo); + /* Block new layoutgets and read/write to ds */ + lo->plh_block_lgets++; + spin_unlock(&inode->i_lock); + filemap_fdatawait(inode->i_mapping); + ret = pnfs_layoutcommit_inode(inode, true); + if (ret == 0) + ret = _pnfs_return_layout(inode); + spin_lock(&inode->i_lock); + lo->plh_block_lgets--; + spin_unlock(&inode->i_lock); + pnfs_put_layout_hdr(lo); + return ret; +} + +static int pnfs_layout_return_on_reboot(struct pnfs_layout_hdr *lo) +{ + struct inode *inode = lo->plh_inode; + const struct cred *cred; + + spin_lock(&inode->i_lock); + if (!pnfs_layout_is_valid(lo)) { + spin_unlock(&inode->i_lock); + return 0; + } + cred = get_cred(lo->plh_lc_cred); + pnfs_get_layout_hdr(lo); + spin_unlock(&inode->i_lock); + + return pnfs_send_layoutreturn(lo, &zero_stateid, &cred, IOMODE_ANY, + PNFS_FL_LAYOUTRETURN_PRIVILEGED); +} + +bool pnfs_roc(struct inode *ino, struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, const struct cred *cred, + bool sync) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct nfs_open_context *ctx; + struct nfs4_state *state; + struct pnfs_layout_hdr *lo; + struct pnfs_layout_segment *lseg, *next; + const struct cred *lc_cred; + nfs4_stateid stateid; + enum pnfs_iomode iomode = 0; + bool layoutreturn = false, roc = false; + bool skip_read; + + if (!nfs_have_layout(ino)) + return false; +retry: + rcu_read_lock(); + spin_lock(&ino->i_lock); + lo = nfsi->layout; + if (!lo || !pnfs_layout_is_valid(lo) || + test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags)) { + lo = NULL; + goto out_noroc; + } + + /* no roc if we hold a delegation */ + skip_read = false; + if (nfs4_check_delegation(ino, FMODE_READ)) { + if (nfs4_check_delegation(ino, FMODE_WRITE)) { + lo = NULL; + goto out_noroc; + } + skip_read = true; + } + + list_for_each_entry_rcu(ctx, &nfsi->open_files, list) { + state = ctx->state; + if (state == NULL) + continue; + /* Don't return layout if there is open file state */ + if (state->state & FMODE_WRITE) { + lo = NULL; + goto out_noroc; + } + if (state->state & FMODE_READ) + skip_read = true; + } + + if (skip_read) { + bool writes = false; + + list_for_each_entry(lseg, &lo->plh_segs, pls_list) { + if (lseg->pls_range.iomode != IOMODE_READ) { + writes = true; + break; + } + } + if (!writes) { + lo = NULL; + goto out_noroc; + } + } + + pnfs_get_layout_hdr(lo); + if (test_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) { + if (!sync) { + pnfs_set_plh_return_info( + lo, skip_read ? IOMODE_RW : IOMODE_ANY, 0); + goto out_noroc; + } + spin_unlock(&ino->i_lock); + rcu_read_unlock(); + wait_on_bit(&lo->plh_flags, NFS_LAYOUT_RETURN, + TASK_UNINTERRUPTIBLE); + pnfs_put_layout_hdr(lo); + goto retry; + } + + list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) { + if (skip_read && lseg->pls_range.iomode == IOMODE_READ) + continue; + /* If we are sending layoutreturn, invalidate all valid lsegs */ + if (!test_and_clear_bit(NFS_LSEG_ROC, &lseg->pls_flags)) + continue; + /* + * Note: mark lseg for return so pnfs_layout_remove_lseg + * doesn't invalidate the layout for us. + */ + set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags); + if (!mark_lseg_invalid(lseg, &lo->plh_return_segs)) + continue; + pnfs_set_plh_return_info(lo, lseg->pls_range.iomode, 0); + } + + if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) + goto out_noroc; + + /* ROC in two conditions: + * 1. there are ROC lsegs + * 2. we don't send layoutreturn + */ + /* lo ref dropped in pnfs_roc_release() */ + layoutreturn = pnfs_prepare_layoutreturn(lo, &stateid, &lc_cred, &iomode); + /* If the creds don't match, we can't compound the layoutreturn */ + if (!layoutreturn || cred_fscmp(cred, lc_cred) != 0) + goto out_noroc; + + roc = layoutreturn; + pnfs_init_layoutreturn_args(args, lo, &stateid, iomode); + res->lrs_present = 0; + layoutreturn = false; + put_cred(lc_cred); + +out_noroc: + spin_unlock(&ino->i_lock); + rcu_read_unlock(); + pnfs_layoutcommit_inode(ino, sync); + if (roc) { + struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld; + if (ld->prepare_layoutreturn) + ld->prepare_layoutreturn(args); + pnfs_put_layout_hdr(lo); + return true; + } + if (layoutreturn) + pnfs_send_layoutreturn(lo, &stateid, &lc_cred, iomode, 0); + pnfs_put_layout_hdr(lo); + return false; +} + +int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp, + struct nfs4_layoutreturn_res **respp, int *ret) +{ + struct nfs4_layoutreturn_args *arg = *argpp; + int retval = -EAGAIN; + + if (!arg) + return 0; + /* Handle Layoutreturn errors */ + switch (*ret) { + case 0: + retval = 0; + break; + case -NFS4ERR_NOMATCHING_LAYOUT: + /* Was there an RPC level error? If not, retry */ + if (task->tk_rpc_status == 0) + break; + /* + * Is there a fatal network level error? + * If so release the layout, but flag the error. + */ + if ((task->tk_rpc_status == -ENETDOWN || + task->tk_rpc_status == -ENETUNREACH) && + task->tk_flags & RPC_TASK_NETUNREACH_FATAL) { + *ret = 0; + (*respp)->lrs_present = 0; + retval = -EIO; + break; + } + /* If the call was not sent, let caller handle it */ + if (!RPC_WAS_SENT(task)) + return 0; + switch (task->tk_rpc_status) { + default: + /* + * Defer the layoutreturn if it was due + * to the server being down. + */ + *ret = -NFS4ERR_NOMATCHING_LAYOUT; + break; + case -EACCES: + case -EIO: + case -EKEYEXPIRED: + case -ERESTARTSYS: + case -EINTR: + /* Don't retry */ + *ret = 0; + break; + } + (*respp)->lrs_present = 0; + retval = 0; + break; + case -NFS4ERR_DELAY: + /* Let the caller handle the retry */ + *ret = -NFS4ERR_NOMATCHING_LAYOUT; + return 0; + case -NFS4ERR_OLD_STATEID: + if (!nfs4_layout_refresh_old_stateid(&arg->stateid, + &arg->range, arg->inode)) + break; + *ret = -NFS4ERR_NOMATCHING_LAYOUT; + return -EAGAIN; + } + *argpp = NULL; + *respp = NULL; + return retval; +} + +void pnfs_roc_release(struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, int ret) +{ + struct pnfs_layout_hdr *lo = args->layout; + struct inode *inode = args->inode; + const nfs4_stateid *res_stateid = NULL; + struct nfs4_xdr_opaque_data *ld_private = args->ld_private; + LIST_HEAD(freeme); + + switch (ret) { + case -NFS4ERR_BADSESSION: + case -NFS4ERR_DEADSESSION: + case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case -NFS4ERR_NOMATCHING_LAYOUT: + spin_lock(&inode->i_lock); + pnfs_layoutreturn_retry_later_locked(lo, &args->stateid, + &args->range, &freeme); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&freeme); + break; + case 0: + if (res->lrs_present) + res_stateid = &res->stateid; + fallthrough; + default: + pnfs_layoutreturn_free_lsegs(lo, &args->stateid, &args->range, + res_stateid); + } + trace_nfs4_layoutreturn_on_close(args->inode, &args->stateid, ret); + if (ld_private && ld_private->ops && ld_private->ops->free) + ld_private->ops->free(ld_private); + pnfs_put_layout_hdr(lo); +} + +bool pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct pnfs_layout_hdr *lo; + bool sleep = false; + + /* we might not have grabbed lo reference. so need to check under + * i_lock */ + spin_lock(&ino->i_lock); + lo = nfsi->layout; + if (lo && test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) { + rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); + sleep = true; + } + spin_unlock(&ino->i_lock); + return sleep; +} + +/* + * Compare two layout segments for sorting into layout cache. + * We want to preferentially return RW over RO layouts, so ensure those + * are seen first. + */ +static s64 +pnfs_lseg_range_cmp(const struct pnfs_layout_range *l1, + const struct pnfs_layout_range *l2) +{ + s64 d; + + /* high offset > low offset */ + d = l1->offset - l2->offset; + if (d) + return d; + + /* short length > long length */ + d = l2->length - l1->length; + if (d) + return d; + + /* read > read/write */ + return (int)(l1->iomode == IOMODE_READ) - (int)(l2->iomode == IOMODE_READ); +} + +static bool +pnfs_lseg_range_is_after(const struct pnfs_layout_range *l1, + const struct pnfs_layout_range *l2) +{ + return pnfs_lseg_range_cmp(l1, l2) > 0; +} + +static bool +pnfs_lseg_no_merge(struct pnfs_layout_segment *lseg, + struct pnfs_layout_segment *old) +{ + return false; +} + +void +pnfs_generic_layout_insert_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg, + bool (*is_after)(const struct pnfs_layout_range *, + const struct pnfs_layout_range *), + bool (*do_merge)(struct pnfs_layout_segment *, + struct pnfs_layout_segment *), + struct list_head *free_me) +{ + struct pnfs_layout_segment *lp, *tmp; + + dprintk("%s:Begin\n", __func__); + + list_for_each_entry_safe(lp, tmp, &lo->plh_segs, pls_list) { + if (test_bit(NFS_LSEG_VALID, &lp->pls_flags) == 0) + continue; + if (do_merge(lseg, lp)) { + mark_lseg_invalid(lp, free_me); + continue; + } + if (is_after(&lseg->pls_range, &lp->pls_range)) + continue; + list_add_tail(&lseg->pls_list, &lp->pls_list); + dprintk("%s: inserted lseg %p " + "iomode %d offset %llu length %llu before " + "lp %p iomode %d offset %llu length %llu\n", + __func__, lseg, lseg->pls_range.iomode, + lseg->pls_range.offset, lseg->pls_range.length, + lp, lp->pls_range.iomode, lp->pls_range.offset, + lp->pls_range.length); + goto out; + } + list_add_tail(&lseg->pls_list, &lo->plh_segs); + dprintk("%s: inserted lseg %p " + "iomode %d offset %llu length %llu at tail\n", + __func__, lseg, lseg->pls_range.iomode, + lseg->pls_range.offset, lseg->pls_range.length); +out: + pnfs_get_layout_hdr(lo); + + dprintk("%s:Return\n", __func__); +} +EXPORT_SYMBOL_GPL(pnfs_generic_layout_insert_lseg); + +static void +pnfs_layout_insert_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg, + struct list_head *free_me) +{ + struct inode *inode = lo->plh_inode; + struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; + + if (ld->add_lseg != NULL) + ld->add_lseg(lo, lseg, free_me); + else + pnfs_generic_layout_insert_lseg(lo, lseg, + pnfs_lseg_range_is_after, + pnfs_lseg_no_merge, + free_me); +} + +static struct pnfs_layout_hdr * +alloc_init_layout_hdr(struct inode *ino, + struct nfs_open_context *ctx, + gfp_t gfp_flags) +{ + struct pnfs_layout_hdr *lo; + + lo = pnfs_alloc_layout_hdr(ino, gfp_flags); + if (!lo) + return NULL; + refcount_set(&lo->plh_refcount, 1); + INIT_LIST_HEAD(&lo->plh_layouts); + INIT_LIST_HEAD(&lo->plh_segs); + INIT_LIST_HEAD(&lo->plh_return_segs); + INIT_LIST_HEAD(&lo->plh_bulk_destroy); + lo->plh_inode = ino; + lo->plh_lc_cred = get_cred(ctx->cred); + lo->plh_flags |= 1 << NFS_LAYOUT_INVALID_STID; + return lo; +} + +static struct pnfs_layout_hdr * +pnfs_find_alloc_layout(struct inode *ino, + struct nfs_open_context *ctx, + gfp_t gfp_flags) + __releases(&ino->i_lock) + __acquires(&ino->i_lock) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct pnfs_layout_hdr *new = NULL; + + dprintk("%s Begin ino=%p layout=%p\n", __func__, ino, nfsi->layout); + + if (nfsi->layout != NULL) + goto out_existing; + spin_unlock(&ino->i_lock); + new = alloc_init_layout_hdr(ino, ctx, gfp_flags); + spin_lock(&ino->i_lock); + + if (likely(nfsi->layout == NULL)) { /* Won the race? */ + nfsi->layout = new; + return new; + } else if (new != NULL) + pnfs_free_layout_hdr(new); +out_existing: + pnfs_get_layout_hdr(nfsi->layout); + return nfsi->layout; +} + +/* + * iomode matching rules: + * iomode lseg strict match + * iomode + * ----- ----- ------ ----- + * ANY READ N/A true + * ANY RW N/A true + * RW READ N/A false + * RW RW N/A true + * READ READ N/A true + * READ RW true false + * READ RW false true + */ +static bool +pnfs_lseg_range_match(const struct pnfs_layout_range *ls_range, + const struct pnfs_layout_range *range, + bool strict_iomode) +{ + struct pnfs_layout_range range1; + + if ((range->iomode == IOMODE_RW && + ls_range->iomode != IOMODE_RW) || + (range->iomode != ls_range->iomode && + strict_iomode) || + !pnfs_lseg_range_intersecting(ls_range, range)) + return false; + + /* range1 covers only the first byte in the range */ + range1 = *range; + range1.length = 1; + return pnfs_lseg_range_contained(ls_range, &range1); +} + +/* + * lookup range in layout + */ +static struct pnfs_layout_segment * +pnfs_find_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_range *range, + bool strict_iomode) +{ + struct pnfs_layout_segment *lseg, *ret = NULL; + + dprintk("%s:Begin\n", __func__); + + list_for_each_entry(lseg, &lo->plh_segs, pls_list) { + if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags) && + pnfs_lseg_range_match(&lseg->pls_range, range, + strict_iomode)) { + ret = pnfs_get_lseg(lseg); + break; + } + } + + dprintk("%s:Return lseg %p ref %d\n", + __func__, ret, ret ? refcount_read(&ret->pls_refcount) : 0); + return ret; +} + +/* + * Use mdsthreshold hints set at each OPEN to determine if I/O should go + * to the MDS or over pNFS + * + * The nfs_inode read_io and write_io fields are cumulative counters reset + * when there are no layout segments. Note that in pnfs_update_layout iomode + * is set to IOMODE_READ for a READ request, and set to IOMODE_RW for a + * WRITE request. + * + * A return of true means use MDS I/O. + * + * From rfc 5661: + * If a file's size is smaller than the file size threshold, data accesses + * SHOULD be sent to the metadata server. If an I/O request has a length that + * is below the I/O size threshold, the I/O SHOULD be sent to the metadata + * server. If both file size and I/O size are provided, the client SHOULD + * reach or exceed both thresholds before sending its read or write + * requests to the data server. + */ +static bool pnfs_within_mdsthreshold(struct nfs_open_context *ctx, + struct inode *ino, int iomode) +{ + struct nfs4_threshold *t = ctx->mdsthreshold; + struct nfs_inode *nfsi = NFS_I(ino); + loff_t fsize = i_size_read(ino); + bool size = false, size_set = false, io = false, io_set = false, ret = false; + + if (t == NULL) + return ret; + + dprintk("%s bm=0x%x rd_sz=%llu wr_sz=%llu rd_io=%llu wr_io=%llu\n", + __func__, t->bm, t->rd_sz, t->wr_sz, t->rd_io_sz, t->wr_io_sz); + + switch (iomode) { + case IOMODE_READ: + if (t->bm & THRESHOLD_RD) { + dprintk("%s fsize %llu\n", __func__, fsize); + size_set = true; + if (fsize < t->rd_sz) + size = true; + } + if (t->bm & THRESHOLD_RD_IO) { + dprintk("%s nfsi->read_io %llu\n", __func__, + nfsi->read_io); + io_set = true; + if (nfsi->read_io < t->rd_io_sz) + io = true; + } + break; + case IOMODE_RW: + if (t->bm & THRESHOLD_WR) { + dprintk("%s fsize %llu\n", __func__, fsize); + size_set = true; + if (fsize < t->wr_sz) + size = true; + } + if (t->bm & THRESHOLD_WR_IO) { + dprintk("%s nfsi->write_io %llu\n", __func__, + nfsi->write_io); + io_set = true; + if (nfsi->write_io < t->wr_io_sz) + io = true; + } + break; + } + if (size_set && io_set) { + if (size && io) + ret = true; + } else if (size || io) + ret = true; + + dprintk("<-- %s size %d io %d ret %d\n", __func__, size, io, ret); + return ret; +} + +static int pnfs_prepare_to_retry_layoutget(struct pnfs_layout_hdr *lo) +{ + /* + * send layoutcommit as it can hold up layoutreturn due to lseg + * reference + */ + pnfs_layoutcommit_inode(lo->plh_inode, false); + return wait_on_bit_action(&lo->plh_flags, NFS_LAYOUT_RETURN, + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); +} + +static void nfs_layoutget_begin(struct pnfs_layout_hdr *lo) +{ + atomic_inc(&lo->plh_outstanding); +} + +static void nfs_layoutget_end(struct pnfs_layout_hdr *lo) +{ + if (atomic_dec_and_test(&lo->plh_outstanding) && + test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags)) { + smp_mb__after_atomic(); + wake_up_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN); + } +} + +static bool pnfs_is_first_layoutget(struct pnfs_layout_hdr *lo) +{ + return test_bit(NFS_LAYOUT_FIRST_LAYOUTGET, &lo->plh_flags); +} + +static void pnfs_clear_first_layoutget(struct pnfs_layout_hdr *lo) +{ + unsigned long *bitlock = &lo->plh_flags; + + clear_bit_unlock(NFS_LAYOUT_FIRST_LAYOUTGET, bitlock); + smp_mb__after_atomic(); + wake_up_bit(bitlock, NFS_LAYOUT_FIRST_LAYOUTGET); +} + +static void _add_to_server_list(struct pnfs_layout_hdr *lo, + struct nfs_server *server) +{ + if (!test_and_set_bit(NFS_LAYOUT_HASHED, &lo->plh_flags)) { + struct nfs_client *clp = server->nfs_client; + + /* The lo must be on the clp list if there is any + * chance of a CB_LAYOUTRECALL(FILE) coming in. + */ + spin_lock(&clp->cl_lock); + list_add_tail_rcu(&lo->plh_layouts, &server->layouts); + spin_unlock(&clp->cl_lock); + } +} + +/* + * Layout segment is retreived from the server if not cached. + * The appropriate layout segment is referenced and returned to the caller. + */ +struct pnfs_layout_segment * +pnfs_update_layout(struct inode *ino, + struct nfs_open_context *ctx, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + bool strict_iomode, + gfp_t gfp_flags) +{ + struct pnfs_layout_range arg = { + .iomode = iomode, + .offset = pos, + .length = count, + }; + unsigned pg_offset; + struct nfs_server *server = NFS_SERVER(ino); + struct nfs_client *clp = server->nfs_client; + struct pnfs_layout_hdr *lo = NULL; + struct pnfs_layout_segment *lseg = NULL; + struct nfs4_layoutget *lgp; + nfs4_stateid stateid; + struct nfs4_exception exception = { + .inode = ino, + }; + unsigned long giveup = jiffies + (clp->cl_lease_time << 1); + bool first; + + if (!pnfs_enabled_sb(NFS_SERVER(ino))) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_NO_PNFS); + goto out; + } + + if (pnfs_within_mdsthreshold(ctx, ino, iomode)) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_MDSTHRESH); + goto out; + } + +lookup_again: + if (!nfs4_valid_open_stateid(ctx->state)) { + trace_pnfs_update_layout(ino, pos, count, + iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_INVALID_OPEN); + lseg = ERR_PTR(-EIO); + goto out; + } + + lseg = ERR_PTR(nfs4_client_recover_expired_lease(clp)); + if (IS_ERR(lseg)) + goto out; + first = false; + spin_lock(&ino->i_lock); + lo = pnfs_find_alloc_layout(ino, ctx, gfp_flags); + if (lo == NULL) { + spin_unlock(&ino->i_lock); + lseg = ERR_PTR(-ENOMEM); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_NOMEM); + goto out; + } + + /* Do we even need to bother with this? */ + if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags)) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_BULK_RECALL); + dprintk("%s matches recall, use MDS\n", __func__); + goto out_unlock; + } + + /* if LAYOUTGET already failed once we don't try again */ + if (pnfs_layout_io_test_failed(lo, iomode)) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_IO_TEST_FAIL); + goto out_unlock; + } + + /* + * If the layout segment list is empty, but there are outstanding + * layoutget calls, then they might be subject to a layoutrecall. + */ + if (test_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags) && + atomic_read(&lo->plh_outstanding) != 0) { + spin_unlock(&ino->i_lock); + lseg = ERR_PTR(wait_on_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN, + TASK_KILLABLE)); + if (IS_ERR(lseg)) + goto out_put_layout_hdr; + pnfs_put_layout_hdr(lo); + goto lookup_again; + } + + /* + * Because we free lsegs when sending LAYOUTRETURN, we need to wait + * for LAYOUTRETURN. + */ + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) { + spin_unlock(&ino->i_lock); + dprintk("%s wait for layoutreturn\n", __func__); + lseg = ERR_PTR(pnfs_prepare_to_retry_layoutget(lo)); + if (!IS_ERR(lseg)) { + dprintk("%s retrying\n", __func__); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, + lseg, + PNFS_UPDATE_LAYOUT_RETRY); + pnfs_put_layout_hdr(lo); + goto lookup_again; + } + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_RETURN); + goto out_put_layout_hdr; + } + + lseg = pnfs_find_lseg(lo, &arg, strict_iomode); + if (lseg) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_FOUND_CACHED); + goto out_unlock; + } + + /* + * Choose a stateid for the LAYOUTGET. If we don't have a layout + * stateid, or it has been invalidated, then we must use the open + * stateid. + */ + if (test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) { + int status; + + /* + * The first layoutget for the file. Need to serialize per + * RFC 5661 Errata 3208. + */ + if (test_and_set_bit(NFS_LAYOUT_FIRST_LAYOUTGET, + &lo->plh_flags)) { + spin_unlock(&ino->i_lock); + lseg = ERR_PTR(wait_on_bit(&lo->plh_flags, + NFS_LAYOUT_FIRST_LAYOUTGET, + TASK_KILLABLE)); + if (IS_ERR(lseg)) + goto out_put_layout_hdr; + pnfs_put_layout_hdr(lo); + dprintk("%s retrying\n", __func__); + goto lookup_again; + } + + spin_unlock(&ino->i_lock); + first = true; + status = nfs4_select_rw_stateid(ctx->state, + iomode == IOMODE_RW ? FMODE_WRITE : FMODE_READ, + NULL, &stateid, NULL); + if (status != 0) { + lseg = ERR_PTR(status); + trace_pnfs_update_layout(ino, pos, count, + iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_INVALID_OPEN); + nfs4_schedule_stateid_recovery(server, ctx->state); + pnfs_clear_first_layoutget(lo); + pnfs_put_layout_hdr(lo); + goto lookup_again; + } + spin_lock(&ino->i_lock); + } else { + nfs4_stateid_copy(&stateid, &lo->plh_stateid); + } + + if (pnfs_layoutgets_blocked(lo)) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_BLOCKED); + goto out_unlock; + } + nfs_layoutget_begin(lo); + spin_unlock(&ino->i_lock); + + _add_to_server_list(lo, server); + + pg_offset = arg.offset & ~PAGE_MASK; + if (pg_offset) { + arg.offset -= pg_offset; + arg.length += pg_offset; + } + if (arg.length != NFS4_MAX_UINT64) + arg.length = PAGE_ALIGN(arg.length); + + lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &stateid, &arg, gfp_flags); + if (!lgp) { + lseg = ERR_PTR(-ENOMEM); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, NULL, + PNFS_UPDATE_LAYOUT_NOMEM); + nfs_layoutget_end(lo); + goto out_put_layout_hdr; + } + + lgp->lo = lo; + pnfs_get_layout_hdr(lo); + + lseg = nfs4_proc_layoutget(lgp, &exception); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET); + nfs_layoutget_end(lo); + if (IS_ERR(lseg)) { + switch(PTR_ERR(lseg)) { + case -EBUSY: + if (time_after(jiffies, giveup)) + lseg = NULL; + break; + case -ERECALLCONFLICT: + case -EAGAIN: + break; + case -ENODATA: + /* The server returned NFS4ERR_LAYOUTUNAVAILABLE */ + pnfs_layout_set_fail_bit( + lo, pnfs_iomode_to_fail_bit(iomode)); + lseg = NULL; + goto out_put_layout_hdr; + default: + if (!nfs_error_is_fatal(PTR_ERR(lseg))) { + pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + lseg = NULL; + } + goto out_put_layout_hdr; + } + if (lseg) { + if (!exception.retry) + goto out_put_layout_hdr; + if (first) + pnfs_clear_first_layoutget(lo); + trace_pnfs_update_layout(ino, pos, count, + iomode, lo, lseg, PNFS_UPDATE_LAYOUT_RETRY); + pnfs_put_layout_hdr(lo); + goto lookup_again; + } + } else { + pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + } + +out_put_layout_hdr: + if (first) + pnfs_clear_first_layoutget(lo); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_EXIT); + pnfs_put_layout_hdr(lo); +out: + dprintk("%s: inode %s/%llu pNFS layout segment %s for " + "(%s, offset: %llu, length: %llu)\n", + __func__, ino->i_sb->s_id, + (unsigned long long)ino->i_ino, + IS_ERR_OR_NULL(lseg) ? "not found" : "found", + iomode==IOMODE_RW ? "read/write" : "read-only", + (unsigned long long)pos, + (unsigned long long)count); + return lseg; +out_unlock: + spin_unlock(&ino->i_lock); + goto out_put_layout_hdr; +} +EXPORT_SYMBOL_GPL(pnfs_update_layout); + +static bool +pnfs_sanity_check_layout_range(struct pnfs_layout_range *range) +{ + switch (range->iomode) { + case IOMODE_READ: + case IOMODE_RW: + break; + default: + return false; + } + if (range->offset == NFS4_MAX_UINT64) + return false; + if (range->length == 0) + return false; + if (range->length != NFS4_MAX_UINT64 && + range->length > NFS4_MAX_UINT64 - range->offset) + return false; + return true; +} + +static struct pnfs_layout_hdr * +_pnfs_grab_empty_layout(struct inode *ino, struct nfs_open_context *ctx) +{ + struct pnfs_layout_hdr *lo; + + spin_lock(&ino->i_lock); + lo = pnfs_find_alloc_layout(ino, ctx, nfs_io_gfp_mask()); + if (!lo) + goto out_unlock; + if (!test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) + goto out_unlock; + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + goto out_unlock; + if (pnfs_layoutgets_blocked(lo)) + goto out_unlock; + if (test_and_set_bit(NFS_LAYOUT_FIRST_LAYOUTGET, &lo->plh_flags)) + goto out_unlock; + nfs_layoutget_begin(lo); + spin_unlock(&ino->i_lock); + _add_to_server_list(lo, NFS_SERVER(ino)); + return lo; + +out_unlock: + spin_unlock(&ino->i_lock); + pnfs_put_layout_hdr(lo); + return NULL; +} + +static void _lgopen_prepare_attached(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct inode *ino = data->dentry->d_inode; + struct pnfs_layout_range rng = { + .iomode = (data->o_arg.fmode & FMODE_WRITE) ? + IOMODE_RW: IOMODE_READ, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + struct nfs4_layoutget *lgp; + struct pnfs_layout_hdr *lo; + + /* Heuristic: don't send layoutget if we have cached data */ + if (rng.iomode == IOMODE_READ && + (i_size_read(ino) == 0 || ino->i_mapping->nrpages != 0)) + return; + + lo = _pnfs_grab_empty_layout(ino, ctx); + if (!lo) + return; + lgp = pnfs_alloc_init_layoutget_args(ino, ctx, ¤t_stateid, &rng, + nfs_io_gfp_mask()); + if (!lgp) { + pnfs_clear_first_layoutget(lo); + nfs_layoutget_end(lo); + pnfs_put_layout_hdr(lo); + return; + } + lgp->lo = lo; + data->lgp = lgp; + data->o_arg.lg_args = &lgp->args; + data->o_res.lg_res = &lgp->res; +} + +static void _lgopen_prepare_floating(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct inode *ino = data->dentry->d_inode; + struct pnfs_layout_range rng = { + .iomode = (data->o_arg.fmode & FMODE_WRITE) ? + IOMODE_RW: IOMODE_READ, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + struct nfs4_layoutget *lgp; + + lgp = pnfs_alloc_init_layoutget_args(ino, ctx, ¤t_stateid, &rng, + nfs_io_gfp_mask()); + if (!lgp) + return; + data->lgp = lgp; + data->o_arg.lg_args = &lgp->args; + data->o_res.lg_res = &lgp->res; +} + +void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct nfs_server *server = NFS_SERVER(data->dir->d_inode); + + if (!(pnfs_enabled_sb(server) && + server->pnfs_curr_ld->flags & PNFS_LAYOUTGET_ON_OPEN)) + return; + /* Could check on max_ops, but currently hardcoded high enough */ + if (!nfs_server_capable(data->dir->d_inode, NFS_CAP_LGOPEN)) + return; + if (data->lgp) + return; + if (data->state) + _lgopen_prepare_attached(data, ctx); + else + _lgopen_prepare_floating(data, ctx); +} + +void pnfs_parse_lgopen(struct inode *ino, struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx) +{ + struct pnfs_layout_hdr *lo; + struct pnfs_layout_segment *lseg; + struct nfs_server *srv = NFS_SERVER(ino); + u32 iomode; + + if (!lgp) + return; + dprintk("%s: entered with status %i\n", __func__, lgp->res.status); + if (lgp->res.status) { + switch (lgp->res.status) { + default: + break; + /* + * Halt lgopen attempts if the server doesn't recognise + * the "current stateid" value, the layout type, or the + * layoutget operation as being valid. + * Also if it complains about too many ops in the compound + * or of the request/reply being too big. + */ + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_NOTSUPP: + case -NFS4ERR_REP_TOO_BIG: + case -NFS4ERR_REP_TOO_BIG_TO_CACHE: + case -NFS4ERR_REQ_TOO_BIG: + case -NFS4ERR_TOO_MANY_OPS: + case -NFS4ERR_UNKNOWN_LAYOUTTYPE: + srv->caps &= ~NFS_CAP_LGOPEN; + } + return; + } + if (!lgp->lo) { + lo = _pnfs_grab_empty_layout(ino, ctx); + if (!lo) + return; + lgp->lo = lo; + } else + lo = lgp->lo; + + lseg = pnfs_layout_process(lgp); + if (!IS_ERR(lseg)) { + iomode = lgp->args.range.iomode; + pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + pnfs_put_lseg(lseg); + } +} + +void nfs4_lgopen_release(struct nfs4_layoutget *lgp) +{ + if (lgp != NULL) { + if (lgp->lo) { + pnfs_clear_first_layoutget(lgp->lo); + nfs_layoutget_end(lgp->lo); + } + pnfs_layoutget_free(lgp); + } +} + +struct pnfs_layout_segment * +pnfs_layout_process(struct nfs4_layoutget *lgp) +{ + struct pnfs_layout_hdr *lo = lgp->lo; + struct nfs4_layoutget_res *res = &lgp->res; + struct pnfs_layout_segment *lseg; + struct inode *ino = lo->plh_inode; + LIST_HEAD(free_me); + + if (!pnfs_sanity_check_layout_range(&res->range)) + return ERR_PTR(-EINVAL); + + /* Inject layout blob into I/O device driver */ + lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res, lgp->gfp_flags); + if (IS_ERR_OR_NULL(lseg)) { + if (!lseg) + lseg = ERR_PTR(-ENOMEM); + + dprintk("%s: Could not allocate layout: error %ld\n", + __func__, PTR_ERR(lseg)); + return lseg; + } + + pnfs_init_lseg(lo, lseg, &res->range, &res->stateid); + + spin_lock(&ino->i_lock); + if (pnfs_layoutgets_blocked(lo)) { + dprintk("%s forget reply due to state\n", __func__); + goto out_forget; + } + + if (test_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags) && + !pnfs_is_first_layoutget(lo)) + goto out_forget; + + if (nfs4_stateid_match_other(&lo->plh_stateid, &res->stateid)) { + /* existing state ID, make sure the sequence number matches. */ + if (pnfs_layout_stateid_blocked(lo, &res->stateid)) { + if (!pnfs_layout_is_valid(lo)) + lo->plh_barrier = 0; + dprintk("%s forget reply due to sequence\n", __func__); + goto out_forget; + } + pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, false); + } else if (pnfs_layout_is_valid(lo)) { + /* + * We got an entirely new state ID. Mark all segments for the + * inode invalid, and retry the layoutget + */ + struct pnfs_layout_range range = { + .iomode = IOMODE_ANY, + .length = NFS4_MAX_UINT64, + }; + pnfs_mark_matching_lsegs_return(lo, &free_me, &range, 0); + goto out_forget; + } else { + /* We have a completely new layout */ + pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, true); + } + + pnfs_get_lseg(lseg); + pnfs_layout_insert_lseg(lo, lseg, &free_me); + + + if (res->return_on_close) + set_bit(NFS_LSEG_ROC, &lseg->pls_flags); + + spin_unlock(&ino->i_lock); + pnfs_free_lseg_list(&free_me); + return lseg; + +out_forget: + spin_unlock(&ino->i_lock); + lseg->pls_layout = lo; + NFS_SERVER(ino)->pnfs_curr_ld->free_lseg(lseg); + return ERR_PTR(-EAGAIN); +} + +/** + * pnfs_mark_matching_lsegs_return - Free or return matching layout segments + * @lo: pointer to layout header + * @tmp_list: list header to be used with pnfs_free_lseg_list() + * @return_range: describe layout segment ranges to be returned + * @seq: stateid seqid to match + * + * This function is mainly intended for use by layoutrecall. It attempts + * to free the layout segment immediately, or else to mark it for return + * as soon as its reference count drops to zero. + * + * Returns + * - 0: a layoutreturn needs to be scheduled. + * - EBUSY: there are layout segment that are still in use. + * - ENOENT: there are no layout segments that need to be returned. + */ +int +pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo, + struct list_head *tmp_list, + const struct pnfs_layout_range *return_range, + u32 seq) +{ + struct pnfs_layout_segment *lseg, *next; + struct nfs_server *server = NFS_SERVER(lo->plh_inode); + int remaining = 0; + + dprintk("%s:Begin lo %p\n", __func__, lo); + + assert_spin_locked(&lo->plh_inode->i_lock); + + if (test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) + tmp_list = &lo->plh_return_segs; + + list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) + if (pnfs_match_lseg_recall(lseg, return_range, seq)) { + dprintk("%s: marking lseg %p iomode %d " + "offset %llu length %llu\n", __func__, + lseg, lseg->pls_range.iomode, + lseg->pls_range.offset, + lseg->pls_range.length); + if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) + tmp_list = &lo->plh_return_segs; + if (mark_lseg_invalid(lseg, tmp_list)) + continue; + remaining++; + set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags); + pnfs_lseg_cancel_io(server, lseg); + } + + if (remaining) { + pnfs_set_plh_return_info(lo, return_range->iomode, seq); + return -EBUSY; + } + + if (!list_empty(&lo->plh_return_segs)) { + pnfs_set_plh_return_info(lo, return_range->iomode, seq); + return 0; + } + + return -ENOENT; +} + +static void +pnfs_mark_layout_for_return(struct inode *inode, + const struct pnfs_layout_range *range) +{ + struct pnfs_layout_hdr *lo; + bool return_now = false; + + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (!pnfs_layout_is_valid(lo)) { + spin_unlock(&inode->i_lock); + return; + } + pnfs_set_plh_return_info(lo, range->iomode, 0); + /* + * mark all matching lsegs so that we are sure to have no live + * segments at hand when sending layoutreturn. See pnfs_put_lseg() + * for how it works. + */ + if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, range, 0) != -EBUSY) { + const struct cred *cred; + nfs4_stateid stateid; + enum pnfs_iomode iomode; + + return_now = pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode); + spin_unlock(&inode->i_lock); + if (return_now) + pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, + PNFS_FL_LAYOUTRETURN_ASYNC); + } else { + spin_unlock(&inode->i_lock); + nfs_commit_inode(inode, 0); + } +} + +void pnfs_error_mark_layout_for_return(struct inode *inode, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_layout_range range = { + .iomode = lseg->pls_range.iomode, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + + pnfs_mark_layout_for_return(inode, &range); +} +EXPORT_SYMBOL_GPL(pnfs_error_mark_layout_for_return); + +static bool +pnfs_layout_can_be_returned(struct pnfs_layout_hdr *lo) +{ + return pnfs_layout_is_valid(lo) && + !test_bit(NFS_LAYOUT_INODE_FREEING, &lo->plh_flags) && + !test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags); +} + +static struct pnfs_layout_segment * +pnfs_find_first_lseg(struct pnfs_layout_hdr *lo, + const struct pnfs_layout_range *range, + enum pnfs_iomode iomode) +{ + struct pnfs_layout_segment *lseg; + + list_for_each_entry(lseg, &lo->plh_segs, pls_list) { + if (!test_bit(NFS_LSEG_VALID, &lseg->pls_flags)) + continue; + if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags)) + continue; + if (lseg->pls_range.iomode != iomode && iomode != IOMODE_ANY) + continue; + if (pnfs_lseg_range_intersecting(&lseg->pls_range, range)) + return lseg; + } + return NULL; +} + +/* Find open file states whose mode matches that of the range */ +static bool +pnfs_should_return_unused_layout(struct pnfs_layout_hdr *lo, + const struct pnfs_layout_range *range) +{ + struct list_head *head; + struct nfs_open_context *ctx; + fmode_t mode = 0; + + if (!pnfs_layout_can_be_returned(lo) || + !pnfs_find_first_lseg(lo, range, range->iomode)) + return false; + + head = &NFS_I(lo->plh_inode)->open_files; + list_for_each_entry_rcu(ctx, head, list) { + if (ctx->state) + mode |= ctx->state->state & (FMODE_READ|FMODE_WRITE); + } + + switch (range->iomode) { + default: + break; + case IOMODE_READ: + mode &= ~FMODE_WRITE; + break; + case IOMODE_RW: + if (pnfs_find_first_lseg(lo, range, IOMODE_READ)) + mode &= ~FMODE_READ; + } + return mode == 0; +} + +static int pnfs_layout_return_unused_byserver(struct nfs_server *server, + void *data) +{ + const struct pnfs_layout_range *range = data; + const struct cred *cred; + struct pnfs_layout_hdr *lo; + struct inode *inode; + nfs4_stateid stateid; + enum pnfs_iomode iomode; + +restart: + rcu_read_lock(); + list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) { + inode = lo->plh_inode; + if (!inode || !pnfs_layout_can_be_returned(lo) || + test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) + continue; + spin_lock(&inode->i_lock); + if (!lo->plh_inode || + !pnfs_should_return_unused_layout(lo, range)) { + spin_unlock(&inode->i_lock); + continue; + } + pnfs_get_layout_hdr(lo); + pnfs_set_plh_return_info(lo, range->iomode, 0); + if (pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs, + range, 0) != 0 || + !pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode)) { + spin_unlock(&inode->i_lock); + rcu_read_unlock(); + pnfs_put_layout_hdr(lo); + cond_resched(); + goto restart; + } + spin_unlock(&inode->i_lock); + rcu_read_unlock(); + pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, + PNFS_FL_LAYOUTRETURN_ASYNC); + pnfs_put_layout_hdr(lo); + cond_resched(); + goto restart; + } + rcu_read_unlock(); + return 0; +} + +void +pnfs_layout_return_unused_byclid(struct nfs_client *clp, + enum pnfs_iomode iomode) +{ + struct pnfs_layout_range range = { + .iomode = iomode, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + + nfs_client_for_each_server(clp, pnfs_layout_return_unused_byserver, + &range); +} + +/* Check if we have we have a valid layout but if there isn't an intersection + * between the request and the pgio->pg_lseg, put this pgio->pg_lseg away. + */ +void +pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req) +{ + if (pgio->pg_lseg == NULL || + (test_bit(NFS_LSEG_VALID, &pgio->pg_lseg->pls_flags) && + pnfs_lseg_request_intersecting(pgio->pg_lseg, req))) + return; + pnfs_put_lseg(pgio->pg_lseg); + pgio->pg_lseg = NULL; +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_check_layout); + +void +pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +{ + u64 rd_size; + + pnfs_generic_pg_check_layout(pgio, req); + if (pgio->pg_lseg == NULL) { + if (pgio->pg_dreq == NULL) + rd_size = i_size_read(pgio->pg_inode) - req_offset(req); + else + rd_size = nfs_dreq_bytes_left(pgio->pg_dreq, + req_offset(req)); + + pgio->pg_lseg = + pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), + req_offset(req), rd_size, + IOMODE_READ, false, + nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + return; + } + } + /* If no lseg, fall back to read through mds */ + if (pgio->pg_lseg == NULL) + nfs_pageio_reset_read_mds(pgio); + +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_read); + +void +pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, u64 wb_size) +{ + pnfs_generic_pg_check_layout(pgio, req); + if (pgio->pg_lseg == NULL) { + pgio->pg_lseg = + pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req), + req_offset(req), wb_size, IOMODE_RW, + false, nfs_io_gfp_mask()); + if (IS_ERR(pgio->pg_lseg)) { + pgio->pg_error = PTR_ERR(pgio->pg_lseg); + pgio->pg_lseg = NULL; + return; + } + } + /* If no lseg, fall back to write through mds */ + if (pgio->pg_lseg == NULL) + nfs_pageio_reset_write_mds(pgio); +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_write); + +void +pnfs_generic_pg_cleanup(struct nfs_pageio_descriptor *desc) +{ + if (desc->pg_lseg) { + pnfs_put_lseg(desc->pg_lseg); + desc->pg_lseg = NULL; + } +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_cleanup); + +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +size_t +pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, + struct nfs_page *prev, struct nfs_page *req) +{ + unsigned int size; + u64 seg_end, req_start, seg_left; + + size = nfs_generic_pg_test(pgio, prev, req); + if (!size) + return 0; + + /* + * 'size' contains the number of bytes left in the current page (up + * to the original size asked for in @req->wb_bytes). + * + * Calculate how many bytes are left in the layout segment + * and if there are less bytes than 'size', return that instead. + * + * Please also note that 'end_offset' is actually the offset of the + * first byte that lies outside the pnfs_layout_range. FIXME? + * + */ + if (pgio->pg_lseg) { + seg_end = pnfs_end_offset(pgio->pg_lseg->pls_range.offset, + pgio->pg_lseg->pls_range.length); + req_start = req_offset(req); + + /* start of request is past the last byte of this segment */ + if (req_start >= seg_end) + return 0; + + /* adjust 'size' iff there are fewer bytes left in the + * segment than what nfs_generic_pg_test returned */ + seg_left = seg_end - req_start; + if (seg_left < size) + size = (unsigned int)seg_left; + } + + return size; +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_test); + +int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *hdr) +{ + struct nfs_pageio_descriptor pgio; + + /* Resend all requests through the MDS */ + nfs_pageio_init_write(&pgio, hdr->inode, FLUSH_STABLE, true, + hdr->completion_ops); + return nfs_pageio_resend(&pgio, hdr); +} +EXPORT_SYMBOL_GPL(pnfs_write_done_resend_to_mds); + +static void pnfs_ld_handle_write_error(struct nfs_pgio_header *hdr) +{ + + dprintk("pnfs write error = %d\n", hdr->pnfs_error); + if (NFS_SERVER(hdr->inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_ERROR) { + pnfs_return_layout(hdr->inode); + } + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) + hdr->task.tk_status = pnfs_write_done_resend_to_mds(hdr); +} + +/* + * Called by non rpc-based layout drivers + */ +void pnfs_ld_write_done(struct nfs_pgio_header *hdr) +{ + if (likely(!hdr->pnfs_error)) { + pnfs_set_layoutcommit(hdr->inode, hdr->lseg, + hdr->mds_offset + hdr->res.count); + hdr->mds_ops->rpc_call_done(&hdr->task, hdr); + } + trace_nfs4_pnfs_write(hdr, hdr->pnfs_error); + if (unlikely(hdr->pnfs_error)) + pnfs_ld_handle_write_error(hdr); + hdr->mds_ops->rpc_release(hdr); +} +EXPORT_SYMBOL_GPL(pnfs_ld_write_done); + +static void +pnfs_write_through_mds(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + list_splice_tail_init(&hdr->pages, &mirror->pg_list); + nfs_pageio_reset_write_mds(desc); + mirror->pg_recoalesce = 1; + } + hdr->completion_ops->completion(hdr); +} + +static enum pnfs_try_status +pnfs_try_to_write_data(struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops, + struct pnfs_layout_segment *lseg, + int how) +{ + struct inode *inode = hdr->inode; + enum pnfs_try_status trypnfs; + struct nfs_server *nfss = NFS_SERVER(inode); + + hdr->mds_ops = call_ops; + + dprintk("%s: Writing ino:%llu %u@%llu (how %d)\n", __func__, + inode->i_ino, hdr->args.count, hdr->args.offset, how); + trypnfs = nfss->pnfs_curr_ld->write_pagelist(hdr, how); + if (trypnfs != PNFS_NOT_ATTEMPTED) + nfs_inc_stats(inode, NFSIOS_PNFS_WRITE); + dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs); + return trypnfs; +} + +static void +pnfs_do_write(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr, int how) +{ + const struct rpc_call_ops *call_ops = desc->pg_rpc_callops; + struct pnfs_layout_segment *lseg = desc->pg_lseg; + enum pnfs_try_status trypnfs; + + trypnfs = pnfs_try_to_write_data(hdr, call_ops, lseg, how); + switch (trypnfs) { + case PNFS_NOT_ATTEMPTED: + pnfs_write_through_mds(desc, hdr); + break; + case PNFS_ATTEMPTED: + break; + case PNFS_TRY_AGAIN: + /* cleanup hdr and prepare to redo pnfs */ + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + list_splice_init(&hdr->pages, &mirror->pg_list); + mirror->pg_recoalesce = 1; + } + hdr->mds_ops->rpc_release(hdr); + } +} + +static void pnfs_writehdr_free(struct nfs_pgio_header *hdr) +{ + pnfs_put_lseg(hdr->lseg); + nfs_pgio_header_free(hdr); +} + +int +pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) +{ + struct nfs_pgio_header *hdr; + int ret; + + hdr = nfs_pgio_header_alloc(desc->pg_rw_ops); + if (!hdr) { + desc->pg_error = -ENOMEM; + return desc->pg_error; + } + nfs_pgheader_init(desc, hdr, pnfs_writehdr_free); + + hdr->lseg = pnfs_get_lseg(desc->pg_lseg); + ret = nfs_generic_pgio(desc, hdr); + if (!ret) + pnfs_do_write(desc, hdr, desc->pg_ioflags); + + return ret; +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_writepages); + +int pnfs_read_done_resend_to_mds(struct nfs_pgio_header *hdr) +{ + struct nfs_pageio_descriptor pgio; + + /* Resend all requests through the MDS */ + nfs_pageio_init_read(&pgio, hdr->inode, true, hdr->completion_ops); + return nfs_pageio_resend(&pgio, hdr); +} +EXPORT_SYMBOL_GPL(pnfs_read_done_resend_to_mds); + +static void pnfs_ld_handle_read_error(struct nfs_pgio_header *hdr) +{ + dprintk("pnfs read error = %d\n", hdr->pnfs_error); + if (NFS_SERVER(hdr->inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_ERROR) { + pnfs_return_layout(hdr->inode); + } + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) + hdr->task.tk_status = pnfs_read_done_resend_to_mds(hdr); +} + +/* + * Called by non rpc-based layout drivers + */ +void pnfs_ld_read_done(struct nfs_pgio_header *hdr) +{ + if (likely(!hdr->pnfs_error)) + hdr->mds_ops->rpc_call_done(&hdr->task, hdr); + trace_nfs4_pnfs_read(hdr, hdr->pnfs_error); + if (unlikely(hdr->pnfs_error)) + pnfs_ld_handle_read_error(hdr); + hdr->mds_ops->rpc_release(hdr); +} +EXPORT_SYMBOL_GPL(pnfs_ld_read_done); + +static void +pnfs_read_through_mds(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + list_splice_tail_init(&hdr->pages, &mirror->pg_list); + nfs_pageio_reset_read_mds(desc); + mirror->pg_recoalesce = 1; + } + hdr->completion_ops->completion(hdr); +} + +/* + * Call the appropriate parallel I/O subsystem read function. + */ +static enum pnfs_try_status +pnfs_try_to_read_data(struct nfs_pgio_header *hdr, + const struct rpc_call_ops *call_ops, + struct pnfs_layout_segment *lseg) +{ + struct inode *inode = hdr->inode; + struct nfs_server *nfss = NFS_SERVER(inode); + enum pnfs_try_status trypnfs; + + hdr->mds_ops = call_ops; + + dprintk("%s: Reading ino:%llu %u@%llu\n", + __func__, inode->i_ino, hdr->args.count, hdr->args.offset); + + trypnfs = nfss->pnfs_curr_ld->read_pagelist(hdr); + if (trypnfs != PNFS_NOT_ATTEMPTED) + nfs_inc_stats(inode, NFSIOS_PNFS_READ); + dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs); + return trypnfs; +} + +/* Resend all requests through pnfs. */ +void pnfs_read_resend_pnfs(struct nfs_pgio_header *hdr, + unsigned int mirror_idx) +{ + struct nfs_pageio_descriptor pgio; + + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + /* Prevent deadlocks with layoutreturn! */ + pnfs_put_lseg(hdr->lseg); + hdr->lseg = NULL; + + nfs_pageio_init_read(&pgio, hdr->inode, false, + hdr->completion_ops); + pgio.pg_mirror_idx = mirror_idx; + hdr->task.tk_status = nfs_pageio_resend(&pgio, hdr); + } +} +EXPORT_SYMBOL_GPL(pnfs_read_resend_pnfs); + +static void +pnfs_do_read(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr) +{ + const struct rpc_call_ops *call_ops = desc->pg_rpc_callops; + struct pnfs_layout_segment *lseg = desc->pg_lseg; + enum pnfs_try_status trypnfs; + + trypnfs = pnfs_try_to_read_data(hdr, call_ops, lseg); + switch (trypnfs) { + case PNFS_NOT_ATTEMPTED: + pnfs_read_through_mds(desc, hdr); + break; + case PNFS_ATTEMPTED: + break; + case PNFS_TRY_AGAIN: + /* cleanup hdr and prepare to redo pnfs */ + if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { + struct nfs_pgio_mirror *mirror = nfs_pgio_current_mirror(desc); + list_splice_init(&hdr->pages, &mirror->pg_list); + mirror->pg_recoalesce = 1; + } + hdr->mds_ops->rpc_release(hdr); + } +} + +static void pnfs_readhdr_free(struct nfs_pgio_header *hdr) +{ + pnfs_put_lseg(hdr->lseg); + nfs_pgio_header_free(hdr); +} + +int +pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc) +{ + struct nfs_pgio_header *hdr; + int ret; + + hdr = nfs_pgio_header_alloc(desc->pg_rw_ops); + if (!hdr) { + desc->pg_error = -ENOMEM; + return desc->pg_error; + } + nfs_pgheader_init(desc, hdr, pnfs_readhdr_free); + hdr->lseg = pnfs_get_lseg(desc->pg_lseg); + ret = nfs_generic_pgio(desc, hdr); + if (!ret) + pnfs_do_read(desc, hdr); + return ret; +} +EXPORT_SYMBOL_GPL(pnfs_generic_pg_readpages); + +static void pnfs_clear_layoutcommitting(struct inode *inode) +{ + unsigned long *bitlock = &NFS_I(inode)->flags; + + clear_bit_unlock(NFS_INO_LAYOUTCOMMITTING, bitlock); + smp_mb__after_atomic(); + wake_up_bit(bitlock, NFS_INO_LAYOUTCOMMITTING); +} + +/* + * There can be multiple RW segments. + */ +static void pnfs_list_write_lseg(struct inode *inode, struct list_head *listp) +{ + struct pnfs_layout_segment *lseg; + + list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list) { + if (lseg->pls_range.iomode == IOMODE_RW && + test_and_clear_bit(NFS_LSEG_LAYOUTCOMMIT, &lseg->pls_flags)) + list_add(&lseg->pls_lc_list, listp); + } +} + +static void pnfs_list_write_lseg_done(struct inode *inode, struct list_head *listp) +{ + struct pnfs_layout_segment *lseg, *tmp; + + /* Matched by references in pnfs_set_layoutcommit */ + list_for_each_entry_safe(lseg, tmp, listp, pls_lc_list) { + list_del_init(&lseg->pls_lc_list); + pnfs_put_lseg(lseg); + } + + pnfs_clear_layoutcommitting(inode); +} + +void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg) +{ + pnfs_layout_io_set_failed(lseg->pls_layout, lseg->pls_range.iomode); +} +EXPORT_SYMBOL_GPL(pnfs_set_lo_fail); + +void +pnfs_set_layoutcommit(struct inode *inode, struct pnfs_layout_segment *lseg, + loff_t end_pos) +{ + struct nfs_inode *nfsi = NFS_I(inode); + bool mark_as_dirty = false; + + spin_lock(&inode->i_lock); + if (!test_and_set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) { + nfsi->layout->plh_lwb = end_pos; + mark_as_dirty = true; + dprintk("%s: Set layoutcommit for inode %llu ", + __func__, inode->i_ino); + } else if (end_pos > nfsi->layout->plh_lwb) + nfsi->layout->plh_lwb = end_pos; + if (!test_and_set_bit(NFS_LSEG_LAYOUTCOMMIT, &lseg->pls_flags)) { + /* references matched in nfs4_layoutcommit_release */ + pnfs_get_lseg(lseg); + } + spin_unlock(&inode->i_lock); + dprintk("%s: lseg %p end_pos %llu\n", + __func__, lseg, nfsi->layout->plh_lwb); + + /* if pnfs_layoutcommit_inode() runs between inode locks, the next one + * will be a noop because NFS_INO_LAYOUTCOMMIT will not be set */ + if (mark_as_dirty) + mark_inode_dirty_sync(inode); +} +EXPORT_SYMBOL_GPL(pnfs_set_layoutcommit); + +void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data) +{ + struct nfs_server *nfss = NFS_SERVER(data->args.inode); + + if (nfss->pnfs_curr_ld->cleanup_layoutcommit) + nfss->pnfs_curr_ld->cleanup_layoutcommit(data); + pnfs_list_write_lseg_done(data->args.inode, &data->lseg_list); +} + +/* + * For the LAYOUT4_NFSV4_1_FILES layout type, NFS_DATA_SYNC WRITEs and + * NFS_UNSTABLE WRITEs with a COMMIT to data servers must store enough + * data to disk to allow the server to recover the data if it crashes. + * LAYOUTCOMMIT is only needed when the NFL4_UFLG_COMMIT_THRU_MDS flag + * is off, and a COMMIT is sent to a data server, or + * if WRITEs to a data server return NFS_DATA_SYNC. + */ +int +pnfs_layoutcommit_inode(struct inode *inode, bool sync) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; + struct nfs4_layoutcommit_data *data; + struct nfs_inode *nfsi = NFS_I(inode); + loff_t end_pos; + int status; + bool mark_as_dirty = false; + + if (!pnfs_layoutcommit_outstanding(inode)) + return 0; + + dprintk("--> %s inode %llu\n", __func__, inode->i_ino); + + status = -EAGAIN; + if (test_and_set_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags)) { + if (!sync) + goto out; + status = wait_on_bit_lock_action(&nfsi->flags, + NFS_INO_LAYOUTCOMMITTING, + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + if (status) + goto out; + } + + status = -ENOMEM; + /* Note kzalloc ensures data->res.seq_res.sr_slot == NULL */ + data = kzalloc_obj(*data, nfs_io_gfp_mask()); + if (!data) + goto clear_layoutcommitting; + + status = 0; + spin_lock(&inode->i_lock); + if (!test_and_clear_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) + goto out_unlock; + + INIT_LIST_HEAD(&data->lseg_list); + pnfs_list_write_lseg(inode, &data->lseg_list); + + end_pos = nfsi->layout->plh_lwb; + + nfs4_stateid_copy(&data->args.stateid, &nfsi->layout->plh_stateid); + data->cred = get_cred(nfsi->layout->plh_lc_cred); + spin_unlock(&inode->i_lock); + + data->args.inode = inode; + nfs_fattr_init(&data->fattr); + data->args.bitmask = NFS_SERVER(inode)->cache_consistency_bitmask; + data->res.fattr = &data->fattr; + if (end_pos != 0) + data->args.lastbytewritten = end_pos - 1; + else + data->args.lastbytewritten = U64_MAX; + data->res.server = NFS_SERVER(inode); + + if (ld->prepare_layoutcommit) { + status = ld->prepare_layoutcommit(&data->args); + if (status) { + if (status != -ENOSPC) + put_cred(data->cred); + spin_lock(&inode->i_lock); + set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags); + if (end_pos > nfsi->layout->plh_lwb) + nfsi->layout->plh_lwb = end_pos; + if (status != -ENOSPC) + goto out_unlock; + spin_unlock(&inode->i_lock); + mark_as_dirty = true; + } + } + + + status = nfs4_proc_layoutcommit(data, sync); +out: + if (status || mark_as_dirty) + mark_inode_dirty_sync(inode); + dprintk("<-- %s status %d\n", __func__, status); + return status; +out_unlock: + spin_unlock(&inode->i_lock); + kfree(data); +clear_layoutcommitting: + pnfs_clear_layoutcommitting(inode); + goto out; +} +EXPORT_SYMBOL_GPL(pnfs_layoutcommit_inode); + +int +pnfs_generic_sync(struct inode *inode, bool datasync) +{ + return pnfs_layoutcommit_inode(inode, true); +} +EXPORT_SYMBOL_GPL(pnfs_generic_sync); + +struct nfs4_threshold *pnfs_mdsthreshold_alloc(void) +{ + struct nfs4_threshold *thp; + + thp = kzalloc_obj(*thp, nfs_io_gfp_mask()); + if (!thp) { + dprintk("%s mdsthreshold allocation failed\n", __func__); + return NULL; + } + return thp; +} + +#if IS_ENABLED(CONFIG_NFS_V4_2) +int +pnfs_report_layoutstat(struct inode *inode, gfp_t gfp_flags) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; + struct nfs_server *server = NFS_SERVER(inode); + struct nfs_inode *nfsi = NFS_I(inode); + struct nfs42_layoutstat_data *data; + struct pnfs_layout_hdr *hdr; + int status = 0; + + if (!pnfs_enabled_sb(server) || !ld->prepare_layoutstats) + goto out; + + if (!nfs_server_capable(inode, NFS_CAP_LAYOUTSTATS)) + goto out; + + if (test_and_set_bit(NFS_INO_LAYOUTSTATS, &nfsi->flags)) + goto out; + + spin_lock(&inode->i_lock); + if (!NFS_I(inode)->layout) { + spin_unlock(&inode->i_lock); + goto out_clear_layoutstats; + } + hdr = NFS_I(inode)->layout; + pnfs_get_layout_hdr(hdr); + spin_unlock(&inode->i_lock); + + data = kzalloc_obj(*data, gfp_flags); + if (!data) { + status = -ENOMEM; + goto out_put; + } + + data->args.fh = NFS_FH(inode); + data->args.inode = inode; + status = ld->prepare_layoutstats(&data->args); + if (status) + goto out_free; + + status = nfs42_proc_layoutstats_generic(NFS_SERVER(inode), data); + +out: + dprintk("%s returns %d\n", __func__, status); + return status; + +out_free: + kfree(data); +out_put: + pnfs_put_layout_hdr(hdr); +out_clear_layoutstats: + smp_mb__before_atomic(); + clear_bit(NFS_INO_LAYOUTSTATS, &nfsi->flags); + smp_mb__after_atomic(); + goto out; +} +EXPORT_SYMBOL_GPL(pnfs_report_layoutstat); +#endif + +unsigned int layoutstats_timer; +module_param(layoutstats_timer, uint, 0644); +EXPORT_SYMBOL_GPL(layoutstats_timer); diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h new file mode 100644 index 000000000..eb39859c2 --- /dev/null +++ b/fs/nfs/pnfs.h @@ -0,0 +1,928 @@ +/* + * pNFS client data structures. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#ifndef FS_NFS_PNFS_H +#define FS_NFS_PNFS_H + +#include +#include +#include +#include + +struct nfs4_exception; +struct nfs4_opendata; + +enum { + NFS_LSEG_VALID = 0, /* cleared when lseg is recalled/returned */ + NFS_LSEG_ROC, /* roc bit received from server */ + NFS_LSEG_LAYOUTCOMMIT, /* layoutcommit bit set for layoutcommit */ + NFS_LSEG_LAYOUTRETURN, /* layoutreturn bit set for layoutreturn */ + NFS_LSEG_UNAVAILABLE, /* unavailable bit set for temporary problem */ +}; + +/* Individual ip address */ +struct nfs4_pnfs_ds_addr { + struct sockaddr_storage da_addr; + size_t da_addrlen; + struct list_head da_node; /* nfs4_pnfs_dev_hlist dev_dslist */ + char *da_remotestr; /* human readable addr+port */ + const char *da_netid; + int da_transport; +}; + +struct nfs4_pnfs_ds { + struct list_head ds_node; /* nfs4_pnfs_dev_hlist dev_dslist */ + char *ds_remotestr; /* comma sep list of addrs */ + struct list_head ds_addrs; + const struct net *ds_net; + struct nfs_client *ds_clp; + refcount_t ds_count; + unsigned long ds_state; +#define NFS4DS_CONNECTING 0 /* ds is establishing connection */ +}; + +struct pnfs_layout_segment { + struct list_head pls_list; + struct list_head pls_lc_list; + struct list_head pls_commits; + struct pnfs_layout_range pls_range; + refcount_t pls_refcount; + u32 pls_seq; + unsigned long pls_flags; + struct pnfs_layout_hdr *pls_layout; +}; + +enum pnfs_try_status { + PNFS_ATTEMPTED = 0, + PNFS_NOT_ATTEMPTED = 1, + PNFS_TRY_AGAIN = 2, +}; + +#if IS_ENABLED(CONFIG_NFS_V4) + +#define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4" + +/* + * Default data server connection timeout and retrans vaules. + * Set by module parameters dataserver_timeo and dataserver_retrans. + */ +#define NFS4_DEF_DS_TIMEO 600 /* in tenths of a second */ +#define NFS4_DEF_DS_RETRANS 5 +#define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ) + +enum { + NFS_LAYOUT_RO_FAILED = 0, /* get ro layout failed stop trying */ + NFS_LAYOUT_RW_FAILED, /* get rw layout failed stop trying */ + NFS_LAYOUT_BULK_RECALL, /* bulk recall affecting layout */ + NFS_LAYOUT_RETURN, /* layoutreturn in progress */ + NFS_LAYOUT_RETURN_LOCK, /* Serialise layoutreturn */ + NFS_LAYOUT_RETURN_REQUESTED, /* Return this layout ASAP */ + NFS_LAYOUT_INVALID_STID, /* layout stateid id is invalid */ + NFS_LAYOUT_FIRST_LAYOUTGET, /* Serialize first layoutget */ + NFS_LAYOUT_INODE_FREEING, /* The inode is being freed */ + NFS_LAYOUT_HASHED, /* The layout visible */ + NFS_LAYOUT_DRAIN, +}; + +enum layoutdriver_policy_flags { + /* Should the pNFS client commit and return the layout upon truncate to + * a smaller size */ + PNFS_LAYOUTRET_ON_SETATTR = 1 << 0, + PNFS_LAYOUTRET_ON_ERROR = 1 << 1, + PNFS_READ_WHOLE_PAGE = 1 << 2, + PNFS_LAYOUTGET_ON_OPEN = 1 << 3, +}; + +enum pnfs_layout_destroy_mode { + PNFS_LAYOUT_INVALIDATE = 0, + PNFS_LAYOUT_BULK_RETURN, + PNFS_LAYOUT_FILE_BULK_RETURN, +}; + +struct nfs4_deviceid_node; + +/* Per-layout driver specific registration structure */ +struct pnfs_layoutdriver_type { + struct list_head pnfs_tblid; + const u32 id; + const char *name; + struct module *owner; + unsigned flags; + unsigned max_layoutget_response; + + int (*set_layoutdriver) (struct nfs_server *, const struct nfs_fh *); + int (*clear_layoutdriver) (struct nfs_server *); + + struct pnfs_layout_hdr * (*alloc_layout_hdr) (struct inode *inode, gfp_t gfp_flags); + void (*free_layout_hdr) (struct pnfs_layout_hdr *); + + struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags); + void (*free_lseg) (struct pnfs_layout_segment *lseg); + void (*add_lseg) (struct pnfs_layout_hdr *layoutid, + struct pnfs_layout_segment *lseg, + struct list_head *free_me); + + void (*return_range) (struct pnfs_layout_hdr *lo, + struct pnfs_layout_range *range); + + /* test for nfs page cache coalescing */ + const struct nfs_pageio_ops *pg_read_ops; + const struct nfs_pageio_ops *pg_write_ops; + + struct pnfs_ds_commit_info *(*get_ds_info) (struct inode *inode); + + int (*sync)(struct inode *inode, bool datasync); + + /* + * Return PNFS_ATTEMPTED to indicate the layout code has attempted + * I/O, else return PNFS_NOT_ATTEMPTED to fall back to normal NFS + */ + enum pnfs_try_status (*read_pagelist)(struct nfs_pgio_header *); + enum pnfs_try_status (*write_pagelist)(struct nfs_pgio_header *, int); + + void (*free_deviceid_node) (struct nfs4_deviceid_node *); + struct nfs4_deviceid_node * (*alloc_deviceid_node) + (struct nfs_server *server, struct pnfs_device *pdev, + gfp_t gfp_flags); + + int (*prepare_layoutreturn) (struct nfs4_layoutreturn_args *); + + void (*cleanup_layoutcommit) (struct nfs4_layoutcommit_data *data); + int (*prepare_layoutcommit) (struct nfs4_layoutcommit_args *args); + int (*prepare_layoutstats) (struct nfs42_layoutstat_args *args); + + void (*cancel_io)(struct pnfs_layout_segment *lseg); +}; + +struct pnfs_commit_ops { + void (*setup_ds_info)(struct pnfs_ds_commit_info *, + struct pnfs_layout_segment *); + void (*release_ds_info)(struct pnfs_ds_commit_info *, + struct inode *inode); + int (*commit_pagelist)(struct inode *inode, + struct list_head *mds_pages, + int how, + struct nfs_commit_info *cinfo); + void (*mark_request_commit) (struct nfs_page *req, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx); + void (*clear_request_commit) (struct nfs_page *req, + struct nfs_commit_info *cinfo); + int (*scan_commit_lists) (struct nfs_commit_info *cinfo, + int max); + void (*recover_commit_reqs) (struct list_head *list, + struct nfs_commit_info *cinfo); +}; + +struct pnfs_layout_hdr { + refcount_t plh_refcount; + atomic_t plh_outstanding; /* number of RPCs out */ + struct list_head plh_layouts; /* other client layouts */ + struct list_head plh_bulk_destroy; + struct list_head plh_segs; /* layout segments list */ + struct list_head plh_return_segs; /* invalid layout segments */ + unsigned long plh_block_lgets; /* block LAYOUTGET if >0 */ + unsigned long plh_retry_timestamp; + unsigned long plh_flags; + nfs4_stateid plh_stateid; + u32 plh_barrier; /* ignore lower seqids */ + u32 plh_return_seq; + enum pnfs_iomode plh_return_iomode; + loff_t plh_lwb; /* last write byte for layoutcommit */ + const struct cred *plh_lc_cred; /* layoutcommit cred */ + struct inode *plh_inode; + struct rcu_head plh_rcu; +}; + +struct pnfs_device { + struct nfs4_deviceid dev_id; + unsigned int layout_type; + unsigned int mincount; + unsigned int maxcount; /* gdia_maxcount */ + struct page **pages; + unsigned int pgbase; + unsigned int pglen; /* reply buffer length */ + unsigned char nocache : 1;/* May not be cached */ +}; + +#define NFS4_PNFS_GETDEVLIST_MAXNUM 16 + +struct pnfs_devicelist { + unsigned int eof; + unsigned int num_devs; + struct nfs4_deviceid dev_id[NFS4_PNFS_GETDEVLIST_MAXNUM]; +}; + +extern int pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *); +extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *); +extern const struct pnfs_layoutdriver_type *pnfs_find_layoutdriver(u32 id); +extern void pnfs_put_layoutdriver(const struct pnfs_layoutdriver_type *ld); + +/* nfs4proc.c */ +#define PNFS_FL_LAYOUTRETURN_ASYNC (1U << 0) +#define PNFS_FL_LAYOUTRETURN_PRIVILEGED (1U << 1) + +extern size_t max_response_pages(struct nfs_server *server); +extern int nfs4_proc_getdeviceinfo(struct nfs_server *server, + struct pnfs_device *dev, + const struct cred *cred); +extern struct pnfs_layout_segment * +nfs4_proc_layoutget(struct nfs4_layoutget *lgp, + struct nfs4_exception *exception); +extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, + unsigned int flags); + +/* pnfs.c */ +void pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo); +void pnfs_put_lseg(struct pnfs_layout_segment *lseg); + +void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, struct nfs_fsinfo *); +void unset_pnfs_layoutdriver(struct nfs_server *); +void pnfs_generic_pg_check_layout(struct nfs_pageio_descriptor *pgio, struct nfs_page *req); +void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *); +int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc); +void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, u64 wb_size); +void pnfs_generic_pg_cleanup(struct nfs_pageio_descriptor *); +int pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc); +size_t pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, + struct nfs_page *prev, struct nfs_page *req); +void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg); +struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp); +void pnfs_layoutget_free(struct nfs4_layoutget *lgp); +void pnfs_free_lseg_list(struct list_head *tmp_list); +void pnfs_destroy_layout(struct nfs_inode *); +void pnfs_destroy_layout_final(struct nfs_inode *); +void pnfs_destroy_all_layouts(struct nfs_client *); +int pnfs_layout_destroy_byfsid(struct nfs_client *clp, struct nfs_fsid *fsid, + enum pnfs_layout_destroy_mode mode); +int pnfs_layout_destroy_byclid(struct nfs_client *clp, + enum pnfs_layout_destroy_mode mode); +bool nfs4_layout_refresh_old_stateid(nfs4_stateid *dst, + struct pnfs_layout_range *dst_range, + struct inode *inode); +void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo); +void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, + const nfs4_stateid *new, + const struct cred *cred, + bool update_barrier); +int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, + struct list_head *tmp_list, + const struct pnfs_layout_range *recall_range, + u32 seq); +int pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo, + struct list_head *tmp_list, + const struct pnfs_layout_range *recall_range, + u32 seq); +int pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo, + struct list_head *lseg_list); +bool pnfs_roc(struct inode *ino, struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, const struct cred *cred, + bool sync); +int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp, + struct nfs4_layoutreturn_res **respp, int *ret); +void pnfs_roc_release(struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, + int ret); +bool pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task); +void pnfs_set_layoutcommit(struct inode *, struct pnfs_layout_segment *, loff_t); +void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data); +int pnfs_layoutcommit_inode(struct inode *inode, bool sync); +int pnfs_generic_sync(struct inode *inode, bool datasync); +int pnfs_nfs_generic_sync(struct inode *inode, bool datasync); +int _pnfs_return_layout(struct inode *); +int pnfs_commit_and_return_layout(struct inode *); +void pnfs_ld_write_done(struct nfs_pgio_header *); +void pnfs_ld_read_done(struct nfs_pgio_header *); +void pnfs_read_resend_pnfs(struct nfs_pgio_header *, unsigned int mirror_idx); +struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino, + struct nfs_open_context *ctx, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + bool strict_iomode, + gfp_t gfp_flags); +void pnfs_layoutreturn_retry_later(struct pnfs_layout_hdr *lo, + const nfs4_stateid *arg_stateid, + const struct pnfs_layout_range *range); +void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo, + const nfs4_stateid *arg_stateid, + const struct pnfs_layout_range *range, + const nfs4_stateid *stateid); + +void pnfs_generic_layout_insert_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg, + bool (*is_after)(const struct pnfs_layout_range *lseg_range, + const struct pnfs_layout_range *old), + bool (*do_merge)(struct pnfs_layout_segment *lseg, + struct pnfs_layout_segment *old), + struct list_head *free_me); + +void nfs4_deviceid_mark_client_invalid(struct nfs_client *clp); +int pnfs_read_done_resend_to_mds(struct nfs_pgio_header *); +int pnfs_write_done_resend_to_mds(struct nfs_pgio_header *); +struct nfs4_threshold *pnfs_mdsthreshold_alloc(void); +void pnfs_error_mark_layout_for_return(struct inode *inode, + struct pnfs_layout_segment *lseg); +void pnfs_layout_return_unused_byclid(struct nfs_client *clp, + enum pnfs_iomode iomode); +int pnfs_layout_handle_reboot(struct nfs_client *clp); + +/* nfs4_deviceid_flags */ +enum { + NFS_DEVICEID_INVALID = 0, /* set when MDS clientid recalled */ + NFS_DEVICEID_UNAVAILABLE, /* device temporarily unavailable */ + NFS_DEVICEID_NOCACHE, /* device may not be cached */ +}; + +/* pnfs_dev.c */ +struct nfs4_deviceid_node { + struct hlist_node node; + struct hlist_node tmpnode; + const struct pnfs_layoutdriver_type *ld; + const struct nfs_client *nfs_client; + unsigned long flags; + unsigned long timestamp_unavailable; + struct nfs4_deviceid deviceid; + struct rcu_head rcu; + atomic_t ref; +}; + +struct nfs4_deviceid_node * +nfs4_find_get_deviceid(struct nfs_server *server, + const struct nfs4_deviceid *id, const struct cred *cred, + gfp_t gfp_mask); +void nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *); +void nfs4_init_deviceid_node(struct nfs4_deviceid_node *, struct nfs_server *, + const struct nfs4_deviceid *); +bool nfs4_put_deviceid_node(struct nfs4_deviceid_node *); +void nfs4_mark_deviceid_available(struct nfs4_deviceid_node *node); +void nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node); +bool nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node); +void nfs4_deviceid_purge_client(const struct nfs_client *); + +/* pnfs_nfs.c */ +struct pnfs_commit_array *pnfs_alloc_commit_array(size_t n, gfp_t gfp_flags); +void pnfs_free_commit_array(struct pnfs_commit_array *p); +struct pnfs_commit_array *pnfs_add_commit_array(struct pnfs_ds_commit_info *, + struct pnfs_commit_array *, + struct pnfs_layout_segment *); + +void pnfs_generic_ds_cinfo_release_lseg(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg); +void pnfs_generic_ds_cinfo_destroy(struct pnfs_ds_commit_info *fl_cinfo); + +void pnfs_generic_clear_request_commit(struct nfs_page *req, + struct nfs_commit_info *cinfo); +void pnfs_generic_commit_release(void *calldata); +void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data); +void pnfs_generic_rw_release(void *data); +void pnfs_generic_recover_commit_reqs(struct list_head *dst, + struct nfs_commit_info *cinfo); +int pnfs_generic_commit_pagelist(struct inode *inode, + struct list_head *mds_pages, + int how, + struct nfs_commit_info *cinfo, + int (*initiate_commit)(struct nfs_commit_data *data, + int how)); +int pnfs_generic_scan_commit_lists(struct nfs_commit_info *cinfo, int max); +void pnfs_generic_write_commit_done(struct rpc_task *task, void *data); +void nfs4_pnfs_ds_put(struct nfs4_pnfs_ds *ds); +struct nfs4_pnfs_ds *nfs4_pnfs_ds_add(const struct net *net, + struct list_head *dsaddrs, + gfp_t gfp_flags); +void nfs4_pnfs_v3_ds_connect_unload(void); +int nfs4_pnfs_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds, + struct nfs4_deviceid_node *devid, unsigned int timeo, + unsigned int retrans, u32 version, u32 minor_version); +struct nfs4_pnfs_ds_addr *nfs4_decode_mp_ds_addr(struct net *net, + struct xdr_stream *xdr, + gfp_t gfp_flags); +void pnfs_layout_mark_request_commit(struct nfs_page *req, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx); +void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx); +void pnfs_parse_lgopen(struct inode *ino, struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx); +void nfs4_lgopen_release(struct nfs4_layoutget *lgp); + +static inline bool nfs_have_layout(struct inode *inode) +{ + return NFS_I(inode)->layout != NULL; +} + +static inline bool pnfs_layout_is_valid(const struct pnfs_layout_hdr *lo) +{ + return test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags) == 0; +} + +static inline struct nfs4_deviceid_node * +nfs4_get_deviceid(struct nfs4_deviceid_node *d) +{ + atomic_inc(&d->ref); + return d; +} + +static inline struct pnfs_layout_segment * +pnfs_get_lseg(struct pnfs_layout_segment *lseg) +{ + if (lseg) { + refcount_inc(&lseg->pls_refcount); + smp_mb__after_atomic(); + } + return lseg; +} + +static inline bool +pnfs_is_valid_lseg(struct pnfs_layout_segment *lseg) +{ + return test_bit(NFS_LSEG_VALID, &lseg->pls_flags) != 0; +} + +/* Return true if a layout driver is being used for this mountpoint */ +static inline int pnfs_enabled_sb(struct nfs_server *nfss) +{ + return nfss->pnfs_curr_ld != NULL; +} + +static inline int +pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how, + struct nfs_commit_info *cinfo) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + + if (fl_cinfo == NULL || fl_cinfo->ncommitting == 0) + return PNFS_NOT_ATTEMPTED; + return fl_cinfo->ops->commit_pagelist(inode, mds_pages, how, cinfo); +} + +static inline struct pnfs_ds_commit_info * +pnfs_get_ds_info(struct inode *inode) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; + + if (ld == NULL || ld->get_ds_info == NULL) + return NULL; + return ld->get_ds_info(inode); +} + +static inline void +pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode) +{ + struct pnfs_ds_commit_info *inode_cinfo = pnfs_get_ds_info(inode); + if (inode_cinfo != NULL) + fl_cinfo->ops = inode_cinfo->ops; +} + +static inline void +pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo) +{ + INIT_LIST_HEAD(&fl_cinfo->commits); + fl_cinfo->ops = NULL; +} + +static inline void +pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode) +{ + if (fl_cinfo->ops != NULL && fl_cinfo->ops->release_ds_info != NULL) + fl_cinfo->ops->release_ds_info(fl_cinfo, inode); +} + +static inline void +pnfs_generic_mark_devid_invalid(struct nfs4_deviceid_node *node) +{ + set_bit(NFS_DEVICEID_INVALID, &node->flags); +} + +static inline bool +pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, u32 ds_commit_idx) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + + if (!lseg || !fl_cinfo->ops || !fl_cinfo->ops->mark_request_commit) + return false; + fl_cinfo->ops->mark_request_commit(req, lseg, cinfo, ds_commit_idx); + return true; +} + +static inline bool +pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + + if (!fl_cinfo || !fl_cinfo->ops || !fl_cinfo->ops->clear_request_commit) + return false; + fl_cinfo->ops->clear_request_commit(req, cinfo); + return true; +} + +static inline int +pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo, + int max) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + + if (!fl_cinfo || fl_cinfo->nwritten == 0) + return 0; + return fl_cinfo->ops->scan_commit_lists(cinfo, max); +} + +static inline void +pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + + if (fl_cinfo && fl_cinfo->nwritten != 0) + fl_cinfo->ops->recover_commit_reqs(head, cinfo); +} + +/* Should the pNFS client commit and return the layout upon a setattr */ +static inline bool +pnfs_ld_layoutret_on_setattr(struct inode *inode) +{ + if (!pnfs_enabled_sb(NFS_SERVER(inode))) + return false; + return NFS_SERVER(inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_SETATTR; +} + +static inline bool +pnfs_ld_read_whole_page(struct inode *inode) +{ + if (!pnfs_enabled_sb(NFS_SERVER(inode))) + return false; + return NFS_SERVER(inode)->pnfs_curr_ld->flags & PNFS_READ_WHOLE_PAGE; +} + +static inline int +pnfs_sync_inode(struct inode *inode, bool datasync) +{ + if (!pnfs_enabled_sb(NFS_SERVER(inode))) + return 0; + return NFS_SERVER(inode)->pnfs_curr_ld->sync(inode, datasync); +} + +static inline bool +pnfs_layoutcommit_outstanding(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + return test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags) != 0 || + test_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags) != 0; +} + +static inline int pnfs_return_layout(struct inode *ino) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct nfs_server *nfss = NFS_SERVER(ino); + + if (pnfs_enabled_sb(nfss) && nfsi->layout) { + set_bit(NFS_LAYOUT_RETURN_REQUESTED, &nfsi->layout->plh_flags); + return _pnfs_return_layout(ino); + } + + return 0; +} + +static inline bool +pnfs_use_threshold(struct nfs4_threshold **dst, struct nfs4_threshold *src, + struct nfs_server *nfss) +{ + return (dst && src && src->bm != 0 && nfss->pnfs_curr_ld && + nfss->pnfs_curr_ld->id == src->l_type); +} + +static inline u64 +pnfs_calc_offset_end(u64 offset, u64 len) +{ + if (len == NFS4_MAX_UINT64 || len >= NFS4_MAX_UINT64 - offset) + return NFS4_MAX_UINT64; + return offset + len - 1; +} + +static inline u64 +pnfs_calc_offset_length(u64 offset, u64 end) +{ + if (end == NFS4_MAX_UINT64 || end <= offset) + return NFS4_MAX_UINT64; + return 1 + end - offset; +} + +static inline void +pnfs_copy_range(struct pnfs_layout_range *dst, + const struct pnfs_layout_range *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline u64 +pnfs_end_offset(u64 start, u64 len) +{ + if (NFS4_MAX_UINT64 - start <= len) + return NFS4_MAX_UINT64; + return start + len; +} + +/* + * Are 2 ranges intersecting? + * start1 end1 + * [----------------------------------) + * start2 end2 + * [----------------) + */ +static inline bool +pnfs_is_range_intersecting(u64 start1, u64 end1, u64 start2, u64 end2) +{ + return (end1 == NFS4_MAX_UINT64 || start2 < end1) && + (end2 == NFS4_MAX_UINT64 || start1 < end2); +} + +static inline bool +pnfs_lseg_range_intersecting(const struct pnfs_layout_range *l1, + const struct pnfs_layout_range *l2) +{ + u64 end1 = pnfs_end_offset(l1->offset, l1->length); + u64 end2 = pnfs_end_offset(l2->offset, l2->length); + + return pnfs_is_range_intersecting(l1->offset, end1, l2->offset, end2); +} + +static inline bool +pnfs_lseg_request_intersecting(struct pnfs_layout_segment *lseg, struct nfs_page *req) +{ + u64 seg_last = pnfs_end_offset(lseg->pls_range.offset, lseg->pls_range.length); + u64 req_last = req_offset(req) + req->wb_bytes; + + return pnfs_is_range_intersecting(lseg->pls_range.offset, seg_last, + req_offset(req), req_last); +} + +static inline void pnfs_lseg_cancel_io(struct nfs_server *server, + struct pnfs_layout_segment *lseg) +{ + if (server->pnfs_curr_ld->cancel_io) + server->pnfs_curr_ld->cancel_io(lseg); +} + +extern unsigned int layoutstats_timer; + +#ifdef NFS_DEBUG +void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id); +#else +static inline void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id) +{ +} + +#endif /* NFS_DEBUG */ +#else /* CONFIG_NFS_V4 */ + +static inline bool nfs_have_layout(struct inode *inode) +{ + return false; +} + +static inline void pnfs_destroy_all_layouts(struct nfs_client *clp) +{ +} + +static inline void pnfs_destroy_layout(struct nfs_inode *nfsi) +{ +} + +static inline void pnfs_destroy_layout_final(struct nfs_inode *nfsi) +{ +} + +static inline int pnfs_layout_handle_reboot(struct nfs_client *clp) +{ + return 0; +} + +static inline struct pnfs_layout_segment * +pnfs_get_lseg(struct pnfs_layout_segment *lseg) +{ + return NULL; +} + +static inline void pnfs_put_lseg(struct pnfs_layout_segment *lseg) +{ +} + +static inline int pnfs_return_layout(struct inode *ino) +{ + return 0; +} + +static inline int pnfs_commit_and_return_layout(struct inode *inode) +{ + return 0; +} + +static inline bool +pnfs_ld_layoutret_on_setattr(struct inode *inode) +{ + return false; +} + +static inline bool +pnfs_ld_read_whole_page(struct inode *inode) +{ + return false; +} + +static inline int +pnfs_sync_inode(struct inode *inode, bool datasync) +{ + return 0; +} + +static inline bool +pnfs_layoutcommit_outstanding(struct inode *inode) +{ + return false; +} + +static inline bool pnfs_roc(struct inode *ino, + struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, + const struct cred *cred, bool sync) +{ + return false; +} + +static inline int +pnfs_roc_done(struct rpc_task *task, + struct nfs4_layoutreturn_args **argpp, + struct nfs4_layoutreturn_res **respp, + int *ret) +{ + return 0; +} + +static inline void +pnfs_roc_release(struct nfs4_layoutreturn_args *args, + struct nfs4_layoutreturn_res *res, + int ret) +{ +} + +static inline bool +pnfs_wait_on_layoutreturn(struct inode *ino, struct rpc_task *task) +{ + return false; +} + +static inline void set_pnfs_layoutdriver(struct nfs_server *s, + const struct nfs_fh *mntfh, + struct nfs_fsinfo *fsinfo) +{ +} + +static inline void unset_pnfs_layoutdriver(struct nfs_server *s) +{ +} + +static inline int +pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how, + struct nfs_commit_info *cinfo) +{ + return PNFS_NOT_ATTEMPTED; +} + +static inline struct pnfs_ds_commit_info * +pnfs_get_ds_info(struct inode *inode) +{ + return NULL; +} + +static inline void +pnfs_init_ds_commit_info_ops(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode) +{ +} + +static inline void +pnfs_init_ds_commit_info(struct pnfs_ds_commit_info *fl_cinfo) +{ +} + +static inline void +pnfs_release_ds_info(struct pnfs_ds_commit_info *fl_cinfo, struct inode *inode) +{ +} + +static inline bool +pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, u32 ds_commit_idx) +{ + return false; +} + +static inline bool +pnfs_clear_request_commit(struct nfs_page *req, struct nfs_commit_info *cinfo) +{ + return false; +} + +static inline int +pnfs_scan_commit_lists(struct inode *inode, struct nfs_commit_info *cinfo, + int max) +{ + return 0; +} + +static inline void +pnfs_recover_commit_reqs(struct list_head *head, struct nfs_commit_info *cinfo) +{ +} + +static inline int pnfs_layoutcommit_inode(struct inode *inode, bool sync) +{ + return 0; +} + +static inline bool +pnfs_use_threshold(struct nfs4_threshold **dst, struct nfs4_threshold *src, + struct nfs_server *nfss) +{ + return false; +} + +static inline struct nfs4_threshold *pnfs_mdsthreshold_alloc(void) +{ + return NULL; +} + +static inline void nfs4_pnfs_v3_ds_connect_unload(void) +{ +} + +static inline bool nfs4_layout_refresh_old_stateid(nfs4_stateid *dst, + struct pnfs_layout_range *dst_range, + struct inode *inode) +{ + return false; +} + +static inline void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ +} + +static inline void pnfs_parse_lgopen(struct inode *ino, + struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx) +{ +} + +static inline void nfs4_lgopen_release(struct nfs4_layoutget *lgp) +{ +} + +static inline bool pnfs_layout_is_valid(const struct pnfs_layout_hdr *lo) +{ + return false; +} + +#endif /* CONFIG_NFS_V4 */ + +#if IS_ENABLED(CONFIG_NFS_V4_2) +int pnfs_report_layoutstat(struct inode *inode, gfp_t gfp_flags); +#else +static inline int +pnfs_report_layoutstat(struct inode *inode, gfp_t gfp_flags) +{ + return 0; +} +#endif + +#endif /* FS_NFS_PNFS_H */ diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c new file mode 100644 index 000000000..274abdd6d --- /dev/null +++ b/fs/nfs/pnfs_dev.c @@ -0,0 +1,381 @@ +/* + * Device operations for the pnfs client. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand + * Garth Goodson + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#include +#include +#include "nfs4session.h" +#include "internal.h" +#include "pnfs.h" + +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS + +/* + * Device ID RCU cache. A device ID is unique per server and layout type. + */ +#define NFS4_DEVICE_ID_HASH_BITS 5 +#define NFS4_DEVICE_ID_HASH_SIZE (1 << NFS4_DEVICE_ID_HASH_BITS) +#define NFS4_DEVICE_ID_HASH_MASK (NFS4_DEVICE_ID_HASH_SIZE - 1) + + +static struct hlist_head nfs4_deviceid_cache[NFS4_DEVICE_ID_HASH_SIZE]; +static DEFINE_SPINLOCK(nfs4_deviceid_lock); + +#ifdef NFS_DEBUG +void +nfs4_print_deviceid(const struct nfs4_deviceid *id) +{ + u32 *p = (u32 *)id; + + dprintk("%s: device id= [%x%x%x%x]\n", __func__, + p[0], p[1], p[2], p[3]); +} +EXPORT_SYMBOL_GPL(nfs4_print_deviceid); +#endif + +static inline u32 +nfs4_deviceid_hash(const struct nfs4_deviceid *id) +{ + unsigned char *cptr = (unsigned char *)id->data; + unsigned int nbytes = NFS4_DEVICEID4_SIZE; + u32 x = 0; + + while (nbytes--) { + x *= 37; + x += *cptr++; + } + return x & NFS4_DEVICE_ID_HASH_MASK; +} + +static struct nfs4_deviceid_node * +_lookup_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id, + long hash) +{ + struct nfs4_deviceid_node *d; + + hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[hash], node) + if (d->ld == ld && d->nfs_client == clp && + !memcmp(&d->deviceid, id, sizeof(*id))) { + if (atomic_read(&d->ref)) + return d; + else + continue; + } + return NULL; +} + +static struct nfs4_deviceid_node * +nfs4_get_device_info(struct nfs_server *server, + const struct nfs4_deviceid *dev_id, + const struct cred *cred, gfp_t gfp_flags) +{ + struct nfs4_deviceid_node *d = NULL; + struct pnfs_device *pdev = NULL; + struct page **pages = NULL; + u32 max_resp_sz; + int max_pages; + int rc, i; + + /* + * Use the session max response size as the basis for setting + * GETDEVICEINFO's maxcount + */ + max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz; + max_pages = nfs_page_array_len(0, max_resp_sz); + dprintk("%s: server %p max_resp_sz %u max_pages %d\n", + __func__, server, max_resp_sz, max_pages); + + pdev = kzalloc_obj(*pdev, gfp_flags); + if (!pdev) + return NULL; + + pages = kzalloc_objs(struct page *, max_pages, gfp_flags); + if (!pages) + goto out_free_pdev; + + for (i = 0; i < max_pages; i++) { + pages[i] = alloc_page(gfp_flags); + if (!pages[i]) + goto out_free_pages; + } + + memcpy(&pdev->dev_id, dev_id, sizeof(*dev_id)); + pdev->layout_type = server->pnfs_curr_ld->id; + pdev->pages = pages; + pdev->pgbase = 0; + pdev->pglen = max_resp_sz; + pdev->mincount = 0; + pdev->maxcount = max_resp_sz - nfs41_maxgetdevinfo_overhead; + + rc = nfs4_proc_getdeviceinfo(server, pdev, cred); + dprintk("%s getdevice info returns %d\n", __func__, rc); + if (rc) + goto out_free_pages; + + /* + * Found new device, need to decode it and then add it to the + * list of known devices for this mountpoint. + */ + d = server->pnfs_curr_ld->alloc_deviceid_node(server, pdev, + gfp_flags); + if (d && pdev->nocache) + set_bit(NFS_DEVICEID_NOCACHE, &d->flags); + +out_free_pages: + while (--i >= 0) + __free_page(pages[i]); + kfree(pages); +out_free_pdev: + kfree(pdev); + dprintk("<-- %s d %p\n", __func__, d); + return d; +} + +/* + * Lookup a deviceid in cache and get a reference count on it if found + * + * @clp nfs_client associated with deviceid + * @id deviceid to look up + */ +static struct nfs4_deviceid_node * +__nfs4_find_get_deviceid(struct nfs_server *server, + const struct nfs4_deviceid *id, long hash) +{ + struct nfs4_deviceid_node *d; + + rcu_read_lock(); + d = _lookup_deviceid(server->pnfs_curr_ld, server->nfs_client, id, + hash); + if (d != NULL && !atomic_inc_not_zero(&d->ref)) + d = NULL; + rcu_read_unlock(); + return d; +} + +struct nfs4_deviceid_node * +nfs4_find_get_deviceid(struct nfs_server *server, + const struct nfs4_deviceid *id, const struct cred *cred, + gfp_t gfp_mask) +{ + long hash = nfs4_deviceid_hash(id); + struct nfs4_deviceid_node *d, *new; + + d = __nfs4_find_get_deviceid(server, id, hash); + if (d) + goto found; + + new = nfs4_get_device_info(server, id, cred, gfp_mask); + if (!new) { + trace_nfs4_find_deviceid(server, id, -ENOENT); + return new; + } + + spin_lock(&nfs4_deviceid_lock); + d = __nfs4_find_get_deviceid(server, id, hash); + if (d) { + spin_unlock(&nfs4_deviceid_lock); + server->pnfs_curr_ld->free_deviceid_node(new); + } else { + atomic_inc(&new->ref); + hlist_add_head_rcu(&new->node, &nfs4_deviceid_cache[hash]); + spin_unlock(&nfs4_deviceid_lock); + d = new; + } +found: + trace_nfs4_find_deviceid(server, id, 0); + return d; +} +EXPORT_SYMBOL_GPL(nfs4_find_get_deviceid); + +/* + * Remove a deviceid from cache + * + * @clp nfs_client associated with deviceid + * @id the deviceid to unhash + * + * @ret the unhashed node, if found and dereferenced to zero, NULL otherwise. + */ +void +nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id) +{ + struct nfs4_deviceid_node *d; + + spin_lock(&nfs4_deviceid_lock); + rcu_read_lock(); + d = _lookup_deviceid(ld, clp, id, nfs4_deviceid_hash(id)); + rcu_read_unlock(); + if (!d) { + spin_unlock(&nfs4_deviceid_lock); + return; + } + hlist_del_init_rcu(&d->node); + clear_bit(NFS_DEVICEID_NOCACHE, &d->flags); + spin_unlock(&nfs4_deviceid_lock); + + /* balance the initial ref set in pnfs_insert_deviceid */ + nfs4_put_deviceid_node(d); +} +EXPORT_SYMBOL_GPL(nfs4_delete_deviceid); + +void +nfs4_init_deviceid_node(struct nfs4_deviceid_node *d, struct nfs_server *server, + const struct nfs4_deviceid *id) +{ + INIT_HLIST_NODE(&d->node); + INIT_HLIST_NODE(&d->tmpnode); + d->ld = server->pnfs_curr_ld; + d->nfs_client = server->nfs_client; + d->flags = 0; + d->deviceid = *id; + atomic_set(&d->ref, 1); +} +EXPORT_SYMBOL_GPL(nfs4_init_deviceid_node); + +/* + * Dereference a deviceid node and delete it when its reference count drops + * to zero. + * + * @d deviceid node to put + * + * return true iff the node was deleted + * Note that since the test for d->ref == 0 is sufficient to establish + * that the node is no longer hashed in the global device id cache. + */ +bool +nfs4_put_deviceid_node(struct nfs4_deviceid_node *d) +{ + if (test_bit(NFS_DEVICEID_NOCACHE, &d->flags)) { + if (atomic_add_unless(&d->ref, -1, 2)) + return false; + nfs4_delete_deviceid(d->ld, d->nfs_client, &d->deviceid); + } + if (!atomic_dec_and_test(&d->ref)) + return false; + trace_nfs4_deviceid_free(d->nfs_client, &d->deviceid); + d->ld->free_deviceid_node(d); + return true; +} +EXPORT_SYMBOL_GPL(nfs4_put_deviceid_node); + +void +nfs4_mark_deviceid_available(struct nfs4_deviceid_node *node) +{ + if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) { + clear_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags); + smp_mb__after_atomic(); + } +} +EXPORT_SYMBOL_GPL(nfs4_mark_deviceid_available); + +void +nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node) +{ + node->timestamp_unavailable = jiffies; + smp_mb__before_atomic(); + set_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags); + smp_mb__after_atomic(); +} +EXPORT_SYMBOL_GPL(nfs4_mark_deviceid_unavailable); + +bool +nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node) +{ + if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) { + unsigned long start, end; + + end = jiffies; + start = end - PNFS_DEVICE_RETRY_TIMEOUT; + if (time_in_range(node->timestamp_unavailable, start, end)) + return true; + clear_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags); + smp_mb__after_atomic(); + } + return false; +} +EXPORT_SYMBOL_GPL(nfs4_test_deviceid_unavailable); + +static void +_deviceid_purge_client(const struct nfs_client *clp, long hash) +{ + struct nfs4_deviceid_node *d; + HLIST_HEAD(tmp); + + spin_lock(&nfs4_deviceid_lock); + rcu_read_lock(); + hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[hash], node) + if (d->nfs_client == clp && atomic_read(&d->ref)) { + hlist_del_init_rcu(&d->node); + hlist_add_head(&d->tmpnode, &tmp); + clear_bit(NFS_DEVICEID_NOCACHE, &d->flags); + } + rcu_read_unlock(); + spin_unlock(&nfs4_deviceid_lock); + + if (hlist_empty(&tmp)) + return; + + while (!hlist_empty(&tmp)) { + d = hlist_entry(tmp.first, struct nfs4_deviceid_node, tmpnode); + hlist_del(&d->tmpnode); + nfs4_put_deviceid_node(d); + } +} + +void +nfs4_deviceid_purge_client(const struct nfs_client *clp) +{ + long h; + + if (!(clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_MDS)) + return; + for (h = 0; h < NFS4_DEVICE_ID_HASH_SIZE; h++) + _deviceid_purge_client(clp, h); +} + +/* + * Stop use of all deviceids associated with an nfs_client + */ +void +nfs4_deviceid_mark_client_invalid(struct nfs_client *clp) +{ + struct nfs4_deviceid_node *d; + int i; + + rcu_read_lock(); + for (i = 0; i < NFS4_DEVICE_ID_HASH_SIZE; i ++){ + hlist_for_each_entry_rcu(d, &nfs4_deviceid_cache[i], node) + if (d->nfs_client == clp) + set_bit(NFS_DEVICEID_INVALID, &d->flags); + } + rcu_read_unlock(); +} diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c new file mode 100644 index 000000000..648c95b78 --- /dev/null +++ b/fs/nfs/pnfs_nfs.c @@ -0,0 +1,1225 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Common NFS I/O operations for the pnfs file based + * layout drivers. + * + * Copyright (c) 2014, Primary Data, Inc. All rights reserved. + * + * Tom Haynes + */ + +#include +#include +#include +#include + +#include "nfs4session.h" +#include "internal.h" +#include "pnfs.h" +#include "netns.h" +#include "nfs4trace.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS + +void pnfs_generic_rw_release(void *data) +{ + struct nfs_pgio_header *hdr = data; + + nfs_put_client(hdr->ds_clp); + hdr->mds_ops->rpc_release(data); +} +EXPORT_SYMBOL_GPL(pnfs_generic_rw_release); + +/* Fake up some data that will cause nfs_commit_release to retry the writes. */ +void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data) +{ + struct nfs_writeverf *verf = data->res.verf; + + data->task.tk_status = 0; + memset(&verf->verifier, 0, sizeof(verf->verifier)); + verf->committed = NFS_UNSTABLE; +} +EXPORT_SYMBOL_GPL(pnfs_generic_prepare_to_resend_writes); + +void pnfs_generic_write_commit_done(struct rpc_task *task, void *data) +{ + struct nfs_commit_data *wdata = data; + + /* Note this may cause RPC to be resent */ + wdata->mds_ops->rpc_call_done(task, data); +} +EXPORT_SYMBOL_GPL(pnfs_generic_write_commit_done); + +void pnfs_generic_commit_release(void *calldata) +{ + struct nfs_commit_data *data = calldata; + + data->completion_ops->completion(data); + pnfs_put_lseg(data->lseg); + nfs_put_client(data->ds_clp); + nfs_commitdata_release(data); +} +EXPORT_SYMBOL_GPL(pnfs_generic_commit_release); + +static struct pnfs_layout_segment * +pnfs_free_bucket_lseg(struct pnfs_commit_bucket *bucket) +{ + if (list_empty(&bucket->committing) && list_empty(&bucket->written)) { + struct pnfs_layout_segment *freeme = bucket->lseg; + bucket->lseg = NULL; + return freeme; + } + return NULL; +} + +/* The generic layer is about to remove the req from the commit list. + * If this will make the bucket empty, it will need to put the lseg reference. + * Note this must be called holding nfsi->commit_mutex + */ +void +pnfs_generic_clear_request_commit(struct nfs_page *req, + struct nfs_commit_info *cinfo) +{ + struct pnfs_commit_bucket *bucket = NULL; + + if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags)) + goto out; + cinfo->ds->nwritten--; + if (list_is_singular(&req->wb_list)) + bucket = list_first_entry(&req->wb_list, + struct pnfs_commit_bucket, written); +out: + nfs_request_remove_commit_list(req, cinfo); + if (bucket) + pnfs_put_lseg(pnfs_free_bucket_lseg(bucket)); +} +EXPORT_SYMBOL_GPL(pnfs_generic_clear_request_commit); + +struct pnfs_commit_array * +pnfs_alloc_commit_array(size_t n, gfp_t gfp_flags) +{ + struct pnfs_commit_array *p; + struct pnfs_commit_bucket *b; + + p = kmalloc_flex(*p, buckets, n, gfp_flags); + if (!p) + return NULL; + p->nbuckets = n; + INIT_LIST_HEAD(&p->cinfo_list); + INIT_LIST_HEAD(&p->lseg_list); + p->lseg = NULL; + for (b = &p->buckets[0]; n != 0; b++, n--) { + INIT_LIST_HEAD(&b->written); + INIT_LIST_HEAD(&b->committing); + b->lseg = NULL; + b->direct_verf.committed = NFS_INVALID_STABLE_HOW; + } + return p; +} +EXPORT_SYMBOL_GPL(pnfs_alloc_commit_array); + +void +pnfs_free_commit_array(struct pnfs_commit_array *p) +{ + kfree_rcu(p, rcu); +} +EXPORT_SYMBOL_GPL(pnfs_free_commit_array); + +static struct pnfs_commit_array * +pnfs_find_commit_array_by_lseg(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_commit_array *array; + + list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) { + if (array->lseg == lseg) + return array; + } + return NULL; +} + +struct pnfs_commit_array * +pnfs_add_commit_array(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_commit_array *new, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_commit_array *array; + + array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg); + if (array) + return array; + new->lseg = lseg; + refcount_set(&new->refcount, 1); + list_add_rcu(&new->cinfo_list, &fl_cinfo->commits); + list_add(&new->lseg_list, &lseg->pls_commits); + return new; +} +EXPORT_SYMBOL_GPL(pnfs_add_commit_array); + +static struct pnfs_commit_array * +pnfs_lookup_commit_array(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_commit_array *array; + + rcu_read_lock(); + array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg); + if (!array) { + rcu_read_unlock(); + fl_cinfo->ops->setup_ds_info(fl_cinfo, lseg); + rcu_read_lock(); + array = pnfs_find_commit_array_by_lseg(fl_cinfo, lseg); + } + rcu_read_unlock(); + return array; +} + +static void +pnfs_release_commit_array_locked(struct pnfs_commit_array *array) +{ + list_del_rcu(&array->cinfo_list); + list_del(&array->lseg_list); + pnfs_free_commit_array(array); +} + +static void +pnfs_put_commit_array_locked(struct pnfs_commit_array *array) +{ + if (refcount_dec_and_test(&array->refcount)) + pnfs_release_commit_array_locked(array); +} + +static void +pnfs_put_commit_array(struct pnfs_commit_array *array, struct inode *inode) +{ + if (refcount_dec_and_lock(&array->refcount, &inode->i_lock)) { + pnfs_release_commit_array_locked(array); + spin_unlock(&inode->i_lock); + } +} + +static struct pnfs_commit_array * +pnfs_get_commit_array(struct pnfs_commit_array *array) +{ + if (refcount_inc_not_zero(&array->refcount)) + return array; + return NULL; +} + +static void +pnfs_remove_and_free_commit_array(struct pnfs_commit_array *array) +{ + array->lseg = NULL; + list_del_init(&array->lseg_list); + pnfs_put_commit_array_locked(array); +} + +void +pnfs_generic_ds_cinfo_release_lseg(struct pnfs_ds_commit_info *fl_cinfo, + struct pnfs_layout_segment *lseg) +{ + struct pnfs_commit_array *array, *tmp; + + list_for_each_entry_safe(array, tmp, &lseg->pls_commits, lseg_list) + pnfs_remove_and_free_commit_array(array); +} +EXPORT_SYMBOL_GPL(pnfs_generic_ds_cinfo_release_lseg); + +void +pnfs_generic_ds_cinfo_destroy(struct pnfs_ds_commit_info *fl_cinfo) +{ + struct pnfs_commit_array *array, *tmp; + + list_for_each_entry_safe(array, tmp, &fl_cinfo->commits, cinfo_list) + pnfs_remove_and_free_commit_array(array); +} +EXPORT_SYMBOL_GPL(pnfs_generic_ds_cinfo_destroy); + +/* + * Locks the nfs_page requests for commit and moves them to + * @bucket->committing. + */ +static int +pnfs_bucket_scan_ds_commit_list(struct pnfs_commit_bucket *bucket, + struct nfs_commit_info *cinfo, + int max) +{ + struct list_head *src = &bucket->written; + struct list_head *dst = &bucket->committing; + int ret; + + lockdep_assert_held(&NFS_I(cinfo->inode)->commit_mutex); + ret = nfs_scan_commit_list(src, dst, cinfo, max); + if (ret) { + cinfo->ds->nwritten -= ret; + cinfo->ds->ncommitting += ret; + } + return ret; +} + +static int pnfs_bucket_scan_array(struct nfs_commit_info *cinfo, + struct pnfs_commit_bucket *buckets, + unsigned int nbuckets, + int max) +{ + unsigned int i; + int rv = 0, cnt; + + for (i = 0; i < nbuckets && max != 0; i++) { + cnt = pnfs_bucket_scan_ds_commit_list(&buckets[i], cinfo, max); + rv += cnt; + max -= cnt; + } + return rv; +} + +/* Move reqs from written to committing lists, returning count + * of number moved. + */ +int pnfs_generic_scan_commit_lists(struct nfs_commit_info *cinfo, int max) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + struct pnfs_commit_array *array; + int rv = 0, cnt; + + rcu_read_lock(); + list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) { + if (!array->lseg || !pnfs_get_commit_array(array)) + continue; + rcu_read_unlock(); + cnt = pnfs_bucket_scan_array(cinfo, array->buckets, + array->nbuckets, max); + rcu_read_lock(); + pnfs_put_commit_array(array, cinfo->inode); + rv += cnt; + max -= cnt; + if (!max) + break; + } + rcu_read_unlock(); + return rv; +} +EXPORT_SYMBOL_GPL(pnfs_generic_scan_commit_lists); + +static unsigned int +pnfs_bucket_recover_commit_reqs(struct list_head *dst, + struct pnfs_commit_bucket *buckets, + unsigned int nbuckets, + struct nfs_commit_info *cinfo) +{ + struct pnfs_commit_bucket *b; + struct pnfs_layout_segment *freeme; + unsigned int nwritten, ret = 0; + unsigned int i; + +restart: + for (i = 0, b = buckets; i < nbuckets; i++, b++) { + nwritten = nfs_scan_commit_list(&b->written, dst, cinfo, 0); + if (!nwritten) + continue; + ret += nwritten; + freeme = pnfs_free_bucket_lseg(b); + if (freeme) { + pnfs_put_lseg(freeme); + goto restart; + } + } + return ret; +} + +/* Pull everything off the committing lists and dump into @dst. */ +void pnfs_generic_recover_commit_reqs(struct list_head *dst, + struct nfs_commit_info *cinfo) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + struct pnfs_commit_array *array; + unsigned int nwritten; + + lockdep_assert_held(&NFS_I(cinfo->inode)->commit_mutex); + rcu_read_lock(); + list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) { + if (!array->lseg || !pnfs_get_commit_array(array)) + continue; + rcu_read_unlock(); + nwritten = pnfs_bucket_recover_commit_reqs(dst, + array->buckets, + array->nbuckets, + cinfo); + rcu_read_lock(); + pnfs_put_commit_array(array, cinfo->inode); + fl_cinfo->nwritten -= nwritten; + } + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(pnfs_generic_recover_commit_reqs); + +static struct pnfs_layout_segment * +pnfs_bucket_get_committing(struct list_head *head, + struct pnfs_commit_bucket *bucket, + struct nfs_commit_info *cinfo) +{ + struct pnfs_layout_segment *lseg; + struct list_head *pos; + + list_for_each(pos, &bucket->committing) + cinfo->ds->ncommitting--; + list_splice_init(&bucket->committing, head); + lseg = pnfs_free_bucket_lseg(bucket); + if (!lseg) + lseg = pnfs_get_lseg(bucket->lseg); + return lseg; +} + +static struct nfs_commit_data * +pnfs_bucket_fetch_commitdata(struct pnfs_commit_bucket *bucket, + struct nfs_commit_info *cinfo) +{ + struct nfs_commit_data *data = nfs_commitdata_alloc(); + + if (!data) + return NULL; + data->lseg = pnfs_bucket_get_committing(&data->pages, bucket, cinfo); + return data; +} + +static void pnfs_generic_retry_commit(struct pnfs_commit_bucket *buckets, + unsigned int nbuckets, + struct nfs_commit_info *cinfo, + unsigned int idx) +{ + struct pnfs_commit_bucket *bucket; + struct pnfs_layout_segment *freeme; + LIST_HEAD(pages); + + for (bucket = buckets; idx < nbuckets; bucket++, idx++) { + if (list_empty(&bucket->committing)) + continue; + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + freeme = pnfs_bucket_get_committing(&pages, bucket, cinfo); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + nfs_retry_commit(&pages, freeme, cinfo, idx); + pnfs_put_lseg(freeme); + } +} + +static unsigned int +pnfs_bucket_alloc_ds_commits(struct list_head *list, + struct pnfs_commit_bucket *buckets, + unsigned int nbuckets, + struct nfs_commit_info *cinfo) +{ + struct pnfs_commit_bucket *bucket; + struct nfs_commit_data *data; + unsigned int i; + unsigned int nreq = 0; + + for (i = 0, bucket = buckets; i < nbuckets; i++, bucket++) { + if (list_empty(&bucket->committing)) + continue; + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + if (!list_empty(&bucket->committing)) { + data = pnfs_bucket_fetch_commitdata(bucket, cinfo); + if (!data) + goto out_error; + data->ds_commit_index = i; + list_add_tail(&data->list, list); + nreq++; + } + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + } + return nreq; +out_error: + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + /* Clean up on error */ + pnfs_generic_retry_commit(buckets, nbuckets, cinfo, i); + return nreq; +} + +static unsigned int +pnfs_alloc_ds_commits_list(struct list_head *list, + struct pnfs_ds_commit_info *fl_cinfo, + struct nfs_commit_info *cinfo) +{ + struct pnfs_commit_array *array; + unsigned int ret = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(array, &fl_cinfo->commits, cinfo_list) { + if (!array->lseg || !pnfs_get_commit_array(array)) + continue; + rcu_read_unlock(); + ret += pnfs_bucket_alloc_ds_commits(list, array->buckets, + array->nbuckets, cinfo); + rcu_read_lock(); + pnfs_put_commit_array(array, cinfo->inode); + } + rcu_read_unlock(); + return ret; +} + +/* This follows nfs_commit_list pretty closely */ +int +pnfs_generic_commit_pagelist(struct inode *inode, struct list_head *mds_pages, + int how, struct nfs_commit_info *cinfo, + int (*initiate_commit)(struct nfs_commit_data *data, + int how)) +{ + struct pnfs_ds_commit_info *fl_cinfo = cinfo->ds; + struct nfs_commit_data *data, *tmp; + LIST_HEAD(list); + unsigned int nreq = 0; + + if (!list_empty(mds_pages)) { + data = nfs_commitdata_alloc(); + if (!data) { + nfs_retry_commit(mds_pages, NULL, cinfo, -1); + return -ENOMEM; + } + data->ds_commit_index = -1; + list_splice_init(mds_pages, &data->pages); + list_add_tail(&data->list, &list); + nreq++; + } + + nreq += pnfs_alloc_ds_commits_list(&list, fl_cinfo, cinfo); + if (nreq == 0) + goto out; + + list_for_each_entry_safe(data, tmp, &list, list) { + list_del(&data->list); + if (data->ds_commit_index < 0) { + nfs_init_commit(data, NULL, NULL, cinfo); + nfs_initiate_commit(NFS_CLIENT(inode), data, + NFS_PROTO(data->inode), + data->mds_ops, how, + RPC_TASK_CRED_NOREF, NULL); + } else { + nfs_init_commit(data, NULL, data->lseg, cinfo); + initiate_commit(data, how); + } + } +out: + return PNFS_ATTEMPTED; +} +EXPORT_SYMBOL_GPL(pnfs_generic_commit_pagelist); + +/* + * Data server cache + * + * Data servers can be mapped to different device ids, but should + * never be shared between net namespaces. + * + * nfs4_pnfs_ds reference counting: + * - set to 1 on allocation + * - incremented when a device id maps a data server already in the cache. + * - decremented when deviceid is removed from the cache. + */ + +/* Debug routines */ +static void +print_ds(struct nfs4_pnfs_ds *ds) +{ + if (ds == NULL) { + printk(KERN_WARNING "%s NULL device\n", __func__); + return; + } + printk(KERN_WARNING " ds %s\n" + " ref count %d\n" + " client %p\n" + " cl_exchange_flags %x\n", + ds->ds_remotestr, + refcount_read(&ds->ds_count), ds->ds_clp, + ds->ds_clp ? ds->ds_clp->cl_exchange_flags : 0); +} + +static bool +same_sockaddr(struct sockaddr *addr1, struct sockaddr *addr2) +{ + struct sockaddr_in *a, *b; + struct sockaddr_in6 *a6, *b6; + + if (addr1->sa_family != addr2->sa_family) + return false; + + switch (addr1->sa_family) { + case AF_INET: + a = (struct sockaddr_in *)addr1; + b = (struct sockaddr_in *)addr2; + + if (a->sin_addr.s_addr == b->sin_addr.s_addr && + a->sin_port == b->sin_port) + return true; + break; + + case AF_INET6: + a6 = (struct sockaddr_in6 *)addr1; + b6 = (struct sockaddr_in6 *)addr2; + + /* LINKLOCAL addresses must have matching scope_id */ + if (ipv6_addr_src_scope(&a6->sin6_addr) == + IPV6_ADDR_SCOPE_LINKLOCAL && + a6->sin6_scope_id != b6->sin6_scope_id) + return false; + + if (ipv6_addr_equal(&a6->sin6_addr, &b6->sin6_addr) && + a6->sin6_port == b6->sin6_port) + return true; + break; + + default: + dprintk("%s: unhandled address family: %u\n", + __func__, addr1->sa_family); + return false; + } + + return false; +} + +/* + * Checks if 'dsaddrs1' contains a subset of 'dsaddrs2'. If it does, + * declare a match. + */ +static bool +_same_data_server_addrs_locked(const struct list_head *dsaddrs1, + const struct list_head *dsaddrs2) +{ + struct nfs4_pnfs_ds_addr *da1, *da2; + struct sockaddr *sa1, *sa2; + bool match = false; + + list_for_each_entry(da1, dsaddrs1, da_node) { + sa1 = (struct sockaddr *)&da1->da_addr; + match = false; + list_for_each_entry(da2, dsaddrs2, da_node) { + sa2 = (struct sockaddr *)&da2->da_addr; + match = same_sockaddr(sa1, sa2); + if (match) + break; + } + if (!match) + break; + } + return match; +} + +/* + * Lookup DS by addresses. nfs4_ds_cache_lock is held + */ +static struct nfs4_pnfs_ds * +_data_server_lookup_locked(const struct nfs_net *nn, const struct list_head *dsaddrs) +{ + struct nfs4_pnfs_ds *ds; + + list_for_each_entry(ds, &nn->nfs4_data_server_cache, ds_node) + if (_same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs)) + return ds; + return NULL; +} + +static struct nfs4_pnfs_ds_addr *nfs4_pnfs_ds_addr_alloc(gfp_t gfp_flags) +{ + struct nfs4_pnfs_ds_addr *da = kzalloc_obj(*da, gfp_flags); + if (da) + INIT_LIST_HEAD(&da->da_node); + return da; +} + +static void nfs4_pnfs_ds_addr_free(struct nfs4_pnfs_ds_addr *da) +{ + kfree(da->da_remotestr); + kfree(da->da_netid); + kfree(da); +} + +static void destroy_ds(struct nfs4_pnfs_ds *ds) +{ + struct nfs4_pnfs_ds_addr *da; + + dprintk("--> %s\n", __func__); + ifdebug(FACILITY) + print_ds(ds); + + nfs_put_client(ds->ds_clp); + + while (!list_empty(&ds->ds_addrs)) { + da = list_first_entry(&ds->ds_addrs, + struct nfs4_pnfs_ds_addr, + da_node); + list_del_init(&da->da_node); + nfs4_pnfs_ds_addr_free(da); + } + + kfree(ds->ds_remotestr); + kfree(ds); +} + +void nfs4_pnfs_ds_put(struct nfs4_pnfs_ds *ds) +{ + struct nfs_net *nn = net_generic(ds->ds_net, nfs_net_id); + + if (refcount_dec_and_lock(&ds->ds_count, &nn->nfs4_data_server_lock)) { + list_del_init(&ds->ds_node); + spin_unlock(&nn->nfs4_data_server_lock); + destroy_ds(ds); + } +} +EXPORT_SYMBOL_GPL(nfs4_pnfs_ds_put); + +/* + * Create a string with a human readable address and port to avoid + * complicated setup around many dprinks. + */ +static char * +nfs4_pnfs_remotestr(struct list_head *dsaddrs, gfp_t gfp_flags) +{ + struct nfs4_pnfs_ds_addr *da; + char *remotestr; + size_t len; + char *p; + + len = 3; /* '{', '}' and eol */ + list_for_each_entry(da, dsaddrs, da_node) { + len += strlen(da->da_remotestr) + 1; /* string plus comma */ + } + + remotestr = kzalloc(len, gfp_flags); + if (!remotestr) + return NULL; + + p = remotestr; + *(p++) = '{'; + len--; + list_for_each_entry(da, dsaddrs, da_node) { + size_t ll = strlen(da->da_remotestr); + + if (ll > len) + goto out_err; + + memcpy(p, da->da_remotestr, ll); + p += ll; + len -= ll; + + if (len < 1) + goto out_err; + (*p++) = ','; + len--; + } + if (len < 2) + goto out_err; + *(p++) = '}'; + *p = '\0'; + return remotestr; +out_err: + kfree(remotestr); + return NULL; +} + +/* + * Given a list of multipath struct nfs4_pnfs_ds_addr, add it to ds cache if + * uncached and return cached struct nfs4_pnfs_ds. + */ +struct nfs4_pnfs_ds * +nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_flags) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + struct nfs4_pnfs_ds *tmp_ds, *ds = NULL; + char *remotestr; + + if (list_empty(dsaddrs)) { + dprintk("%s: no addresses defined\n", __func__); + goto out; + } + + ds = kzalloc_obj(*ds, gfp_flags); + if (!ds) + goto out; + + /* this is only used for debugging, so it's ok if its NULL */ + remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags); + + spin_lock(&nn->nfs4_data_server_lock); + tmp_ds = _data_server_lookup_locked(nn, dsaddrs); + if (tmp_ds == NULL) { + INIT_LIST_HEAD(&ds->ds_addrs); + list_splice_init(dsaddrs, &ds->ds_addrs); + ds->ds_remotestr = remotestr; + refcount_set(&ds->ds_count, 1); + INIT_LIST_HEAD(&ds->ds_node); + ds->ds_net = net; + ds->ds_clp = NULL; + list_add(&ds->ds_node, &nn->nfs4_data_server_cache); + dprintk("%s add new data server %s\n", __func__, + ds->ds_remotestr); + } else { + kfree(remotestr); + kfree(ds); + refcount_inc(&tmp_ds->ds_count); + dprintk("%s data server %s found, inc'ed ds_count to %d\n", + __func__, tmp_ds->ds_remotestr, + refcount_read(&tmp_ds->ds_count)); + ds = tmp_ds; + } + spin_unlock(&nn->nfs4_data_server_lock); +out: + return ds; +} +EXPORT_SYMBOL_GPL(nfs4_pnfs_ds_add); + +static int nfs4_wait_ds_connect(struct nfs4_pnfs_ds *ds) +{ + might_sleep(); + return wait_on_bit(&ds->ds_state, NFS4DS_CONNECTING, TASK_KILLABLE); +} + +static void nfs4_clear_ds_conn_bit(struct nfs4_pnfs_ds *ds) +{ + smp_mb__before_atomic(); + clear_and_wake_up_bit(NFS4DS_CONNECTING, &ds->ds_state); +} + +static struct nfs_client *(*get_v3_ds_connect)( + struct nfs_server *mds_srv, + const struct sockaddr_storage *ds_addr, + int ds_addrlen, + int ds_proto, + unsigned int ds_timeo, + unsigned int ds_retrans); + +static bool load_v3_ds_connect(void) +{ + if (!get_v3_ds_connect) { + get_v3_ds_connect = symbol_request(nfs3_set_ds_client); + WARN_ON_ONCE(!get_v3_ds_connect); + } + + return(get_v3_ds_connect != NULL); +} + +void nfs4_pnfs_v3_ds_connect_unload(void) +{ + if (get_v3_ds_connect) { + symbol_put(nfs3_set_ds_client); + get_v3_ds_connect = NULL; + } +} + +static int _nfs4_pnfs_v3_ds_connect(struct nfs_server *mds_srv, + struct nfs4_pnfs_ds *ds, + unsigned int timeo, + unsigned int retrans) +{ + struct nfs_client *clp = ERR_PTR(-EIO); + struct nfs_client *mds_clp = mds_srv->nfs_client; + enum xprtsec_policies xprtsec_policy = mds_clp->cl_xprtsec.policy; + struct nfs4_pnfs_ds_addr *da; + unsigned long connect_timeout = timeo * (retrans + 1) * HZ / 10; + int ds_proto; + int status = 0; + + dprintk("--> %s DS %s\n", __func__, ds->ds_remotestr); + + if (!load_v3_ds_connect()) + return -EPROTONOSUPPORT; + + list_for_each_entry(da, &ds->ds_addrs, da_node) { + dprintk("%s: DS %s: trying address %s\n", + __func__, ds->ds_remotestr, da->da_remotestr); + + if (!IS_ERR(clp)) { + struct xprt_create xprt_args = { + .ident = da->da_transport, + .net = clp->cl_net, + .dstaddr = (struct sockaddr *)&da->da_addr, + .addrlen = da->da_addrlen, + .servername = clp->cl_hostname, + .connect_timeout = connect_timeout, + .reconnect_timeout = connect_timeout, + .xprtsec = clp->cl_xprtsec, + }; + + if (xprt_args.ident == XPRT_TRANSPORT_TCP && + clp->cl_proto == XPRT_TRANSPORT_TCP_TLS) + xprt_args.ident = XPRT_TRANSPORT_TCP_TLS; + + if (xprt_args.ident != clp->cl_proto) + continue; + if (xprt_args.dstaddr->sa_family != + clp->cl_addr.ss_family) + continue; + /* Add this address as an alias */ + rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args, + rpc_clnt_test_and_add_xprt, NULL); + continue; + } + + ds_proto = da->da_transport; + if (ds_proto == XPRT_TRANSPORT_TCP && + xprtsec_policy != RPC_XPRTSEC_NONE) + ds_proto = XPRT_TRANSPORT_TCP_TLS; + + clp = get_v3_ds_connect(mds_srv, &da->da_addr, da->da_addrlen, + ds_proto, timeo, retrans); + if (IS_ERR(clp)) + continue; + clp->cl_rpcclient->cl_softerr = 0; + clp->cl_rpcclient->cl_softrtry = 0; + } + + if (IS_ERR(clp)) { + status = PTR_ERR(clp); + goto out; + } + + smp_wmb(); + WRITE_ONCE(ds->ds_clp, clp); + dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr); +out: + return status; +} + +static int _nfs4_pnfs_v4_ds_connect(struct nfs_server *mds_srv, + struct nfs4_pnfs_ds *ds, + unsigned int timeo, + unsigned int retrans, + u32 minor_version) +{ + struct nfs_client *clp = ERR_PTR(-EIO); + struct nfs_client *mds_clp = mds_srv->nfs_client; + enum xprtsec_policies xprtsec_policy = mds_clp->cl_xprtsec.policy; + struct nfs4_pnfs_ds_addr *da; + int ds_proto; + int status = 0; + + dprintk("--> %s DS %s\n", __func__, ds->ds_remotestr); + + list_for_each_entry(da, &ds->ds_addrs, da_node) { + char servername[48]; + + dprintk("%s: DS %s: trying address %s\n", + __func__, ds->ds_remotestr, da->da_remotestr); + + if (!IS_ERR(clp) && clp->cl_mvops->session_trunk) { + struct xprt_create xprt_args = { + .ident = da->da_transport, + .net = clp->cl_net, + .dstaddr = (struct sockaddr *)&da->da_addr, + .addrlen = da->da_addrlen, + .servername = clp->cl_hostname, + .xprtsec = clp->cl_xprtsec, + }; + struct nfs4_add_xprt_data xprtdata = { + .clp = clp, + }; + struct rpc_add_xprt_test rpcdata = { + .add_xprt_test = clp->cl_mvops->session_trunk, + .data = &xprtdata, + }; + + if (xprt_args.ident == XPRT_TRANSPORT_TCP && + clp->cl_proto == XPRT_TRANSPORT_TCP_TLS) { + struct sockaddr *addr = + (struct sockaddr *)&da->da_addr; + struct sockaddr_in *sin = + (struct sockaddr_in *)&da->da_addr; + struct sockaddr_in6 *sin6 = + (struct sockaddr_in6 *)&da->da_addr; + + /* for NFS with TLS we need to supply a correct + * servername of the trunked transport, not the + * servername of the main transport stored in + * clp->cl_hostname. And set the protocol to + * indicate to use TLS + */ + servername[0] = '\0'; + switch(addr->sa_family) { + case AF_INET: + snprintf(servername, sizeof(servername), + "%pI4", &sin->sin_addr.s_addr); + break; + case AF_INET6: + snprintf(servername, sizeof(servername), + "%pI6", &sin6->sin6_addr); + break; + default: + /* do not consider this address */ + continue; + } + xprt_args.ident = XPRT_TRANSPORT_TCP_TLS; + xprt_args.servername = servername; + } + if (xprt_args.ident != clp->cl_proto) + continue; + if (xprt_args.dstaddr->sa_family != + clp->cl_addr.ss_family) + continue; + + /** + * Test this address for session trunking and + * add as an alias + */ + xprtdata.cred = nfs4_get_clid_cred(clp); + rpc_clnt_add_xprt(clp->cl_rpcclient, &xprt_args, + rpc_clnt_setup_test_and_add_xprt, + &rpcdata); + if (xprtdata.cred) + put_cred(xprtdata.cred); + } else { + ds_proto = da->da_transport; + if (ds_proto == XPRT_TRANSPORT_TCP && + xprtsec_policy != RPC_XPRTSEC_NONE) + ds_proto = XPRT_TRANSPORT_TCP_TLS; + + clp = nfs4_set_ds_client(mds_srv, &da->da_addr, + da->da_addrlen, ds_proto, + timeo, retrans, minor_version); + if (IS_ERR(clp)) + continue; + + status = nfs4_init_ds_session(clp, + mds_srv->nfs_client->cl_lease_time); + if (status) { + nfs_put_client(clp); + clp = ERR_PTR(-EIO); + continue; + } + } + } + + if (IS_ERR(clp)) { + status = PTR_ERR(clp); + goto out; + } + + smp_wmb(); + WRITE_ONCE(ds->ds_clp, clp); + dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr); +out: + return status; +} + +/* + * Create an rpc connection to the nfs4_pnfs_ds data server. + * Currently only supports IPv4 and IPv6 addresses. + * If connection fails, make devid unavailable and return a -errno. + */ +int nfs4_pnfs_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds, + struct nfs4_deviceid_node *devid, unsigned int timeo, + unsigned int retrans, u32 version, u32 minor_version) +{ + int err; + + do { + err = nfs4_wait_ds_connect(ds); + if (err || ds->ds_clp) + goto out; + if (nfs4_test_deviceid_unavailable(devid)) { + err = -ENODEV; + goto out; + } + } while (test_and_set_bit(NFS4DS_CONNECTING, &ds->ds_state) != 0); + + if (ds->ds_clp) + goto connect_done; + + switch (version) { + case 3: + err = _nfs4_pnfs_v3_ds_connect(mds_srv, ds, timeo, retrans); + break; + case 4: + err = _nfs4_pnfs_v4_ds_connect(mds_srv, ds, timeo, retrans, + minor_version); + break; + default: + dprintk("%s: unsupported DS version %d\n", __func__, version); + err = -EPROTONOSUPPORT; + } + +connect_done: + nfs4_clear_ds_conn_bit(ds); +out: + /* + * At this point the ds->ds_clp should be ready, but it might have + * hit an error. + */ + if (!err) { + if (!ds->ds_clp || !nfs_client_init_is_complete(ds->ds_clp)) { + WARN_ON_ONCE(ds->ds_clp || + !nfs4_test_deviceid_unavailable(devid)); + err = -EINVAL; + } else + err = nfs_client_init_status(ds->ds_clp); + } + + trace_pnfs_ds_connect(ds->ds_remotestr, err); + return err; +} +EXPORT_SYMBOL_GPL(nfs4_pnfs_ds_connect); + +/* + * Currently only supports ipv4, ipv6 and one multi-path address. + */ +struct nfs4_pnfs_ds_addr * +nfs4_decode_mp_ds_addr(struct net *net, struct xdr_stream *xdr, gfp_t gfp_flags) +{ + struct nfs4_pnfs_ds_addr *da = NULL; + char *buf, *portstr; + __be16 port; + ssize_t nlen, rlen; + int tmp[2]; + char *netid; + size_t len; + char *startsep = ""; + char *endsep = ""; + + + /* r_netid */ + nlen = xdr_stream_decode_string_dup(xdr, &netid, XDR_MAX_NETOBJ, + gfp_flags); + if (unlikely(nlen <= 0)) + goto out_err; + + /* r_addr: ip/ip6addr with port in dec octets - see RFC 5665 */ + /* port is ".ABC.DEF", 8 chars max */ + rlen = xdr_stream_decode_string_dup(xdr, &buf, INET6_ADDRSTRLEN + + IPV6_SCOPE_ID_LEN + 8, gfp_flags); + if (unlikely(rlen <= 0)) + goto out_free_netid; + + /* replace port '.' with '-' */ + portstr = strrchr(buf, '.'); + if (!portstr) { + dprintk("%s: Failed finding expected dot in port\n", + __func__); + goto out_free_buf; + } + *portstr = '-'; + + /* find '.' between address and port */ + portstr = strrchr(buf, '.'); + if (!portstr) { + dprintk("%s: Failed finding expected dot between address and " + "port\n", __func__); + goto out_free_buf; + } + *portstr = '\0'; + + da = nfs4_pnfs_ds_addr_alloc(gfp_flags); + if (unlikely(!da)) + goto out_free_buf; + + if (!rpc_pton(net, buf, portstr-buf, (struct sockaddr *)&da->da_addr, + sizeof(da->da_addr))) { + dprintk("%s: error parsing address %s\n", __func__, buf); + goto out_free_da; + } + + portstr++; + sscanf(portstr, "%d-%d", &tmp[0], &tmp[1]); + port = htons((tmp[0] << 8) | (tmp[1])); + + switch (da->da_addr.ss_family) { + case AF_INET: + ((struct sockaddr_in *)&da->da_addr)->sin_port = port; + da->da_addrlen = sizeof(struct sockaddr_in); + break; + + case AF_INET6: + ((struct sockaddr_in6 *)&da->da_addr)->sin6_port = port; + da->da_addrlen = sizeof(struct sockaddr_in6); + startsep = "["; + endsep = "]"; + break; + + default: + dprintk("%s: unsupported address family: %u\n", + __func__, da->da_addr.ss_family); + goto out_free_da; + } + + da->da_transport = xprt_find_transport_ident(netid); + if (da->da_transport < 0) { + dprintk("%s: ERROR: unknown r_netid \"%s\"\n", + __func__, netid); + goto out_free_da; + } + + da->da_netid = netid; + + /* save human readable address */ + len = strlen(startsep) + strlen(buf) + strlen(endsep) + 7; + da->da_remotestr = kzalloc(len, gfp_flags); + + /* NULL is ok, only used for dprintk */ + if (da->da_remotestr) + snprintf(da->da_remotestr, len, "%s%s%s:%u", startsep, + buf, endsep, ntohs(port)); + + dprintk("%s: Parsed DS addr %s\n", __func__, da->da_remotestr); + kfree(buf); + return da; + +out_free_da: + kfree(da); +out_free_buf: + dprintk("%s: Error parsing DS addr: %s\n", __func__, buf); + kfree(buf); +out_free_netid: + kfree(netid); +out_err: + return NULL; +} +EXPORT_SYMBOL_GPL(nfs4_decode_mp_ds_addr); + +void +pnfs_layout_mark_request_commit(struct nfs_page *req, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx) +{ + struct list_head *list; + struct pnfs_commit_array *array; + struct pnfs_commit_bucket *bucket; + + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + array = pnfs_lookup_commit_array(cinfo->ds, lseg); + if (!array || !pnfs_is_valid_lseg(lseg)) + goto out_resched; + bucket = &array->buckets[ds_commit_idx]; + list = &bucket->written; + /* Non-empty buckets hold a reference on the lseg. That ref + * is normally transferred to the COMMIT call and released + * there. It could also be released if the last req is pulled + * off due to a rewrite, in which case it will be done in + * pnfs_common_clear_request_commit + */ + if (!bucket->lseg) + bucket->lseg = pnfs_get_lseg(lseg); + set_bit(PG_COMMIT_TO_DS, &req->wb_flags); + cinfo->ds->nwritten++; + + nfs_request_add_commit_list_locked(req, list, cinfo); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + nfs_folio_mark_unstable(req, cinfo); + return; +out_resched: + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + cinfo->completion_ops->resched_write(cinfo, req); +} +EXPORT_SYMBOL_GPL(pnfs_layout_mark_request_commit); + +int +pnfs_nfs_generic_sync(struct inode *inode, bool datasync) +{ + int ret; + + if (!pnfs_layoutcommit_outstanding(inode)) + return 0; + ret = nfs_commit_inode(inode, FLUSH_SYNC); + if (ret < 0) + return ret; + if (datasync) + return 0; + return pnfs_layoutcommit_inode(inode, true); +} +EXPORT_SYMBOL_GPL(pnfs_nfs_generic_sync); + diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c new file mode 100644 index 000000000..03c2c1f31 --- /dev/null +++ b/fs/nfs/proc.c @@ -0,0 +1,781 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/proc.c + * + * Copyright (C) 1992, 1993, 1994 Rick Sladkey + * + * OS-independent nfs remote procedure call functions + * + * Tuned by Alan Cox for >3K buffers + * so at last we can have decent(ish) throughput off a + * Sun server. + * + * Coding optimized and cleaned up by Florian La Roche. + * Note: Error returns are optimized for NFS_OK, which isn't translated via + * nfs_stat_to_errno(), but happens to be already the right return code. + * + * Also, the code currently doesn't check the size of the packet, when + * it decodes the packet. + * + * Feel free to fix it and mail me the diffs if it worries you. + * + * Completely rewritten to support the new RPC call interface; + * rewrote and moved the entire XDR stuff to xdr.c + * --Olaf Kirch June 1996 + * + * The code below initializes all auto variables explicitly, otherwise + * it will fail to work as a module (gcc generates a memset call for an + * incomplete struct). + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_PROC + +/* + * Bare-bones access to getattr: this is for nfs_read_super. + */ +static int +nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + struct nfs_fattr *fattr = info->fattr; + struct nfs2_fsstat fsinfo; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_GETATTR], + .rpc_argp = fhandle, + .rpc_resp = fattr, + }; + int status; + + dprintk("%s: call getattr\n", __func__); + nfs_fattr_init(fattr); + status = rpc_call_sync(server->client, &msg, 0); + /* Retry with default authentication if different */ + if (status && server->nfs_client->cl_rpcclient != server->client) + status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0); + dprintk("%s: reply getattr: %d\n", __func__, status); + if (status) + return status; + dprintk("%s: call statfs\n", __func__); + msg.rpc_proc = &nfs_procedures[NFSPROC_STATFS]; + msg.rpc_resp = &fsinfo; + status = rpc_call_sync(server->client, &msg, 0); + /* Retry with default authentication if different */ + if (status && server->nfs_client->cl_rpcclient != server->client) + status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0); + dprintk("%s: reply statfs: %d\n", __func__, status); + if (status) + return status; + info->rtmax = NFS_MAXDATA; + info->rtpref = fsinfo.tsize; + info->rtmult = fsinfo.bsize; + info->wtmax = NFS_MAXDATA; + info->wtpref = fsinfo.tsize; + info->wtmult = fsinfo.bsize; + info->dtpref = fsinfo.tsize; + info->maxfilesize = 0x7FFFFFFF; + info->lease_time = 0; + info->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED; + info->xattr_support = 0; + return 0; +} + +/* + * One function for each procedure in the NFS protocol. + */ +static int +nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fattr *fattr, struct inode *inode) +{ + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_GETATTR], + .rpc_argp = fhandle, + .rpc_resp = fattr, + }; + int status; + unsigned short task_flags = 0; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (inode && (server->flags & NFS_MOUNT_SOFTREVAL)) + task_flags |= RPC_TASK_TIMEOUT; + + dprintk("NFS call getattr\n"); + nfs_fattr_init(fattr); + status = rpc_call_sync(server->client, &msg, task_flags); + dprintk("NFS reply getattr: %d\n", status); + return status; +} + +static int +nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, + struct iattr *sattr) +{ + struct inode *inode = d_inode(dentry); + struct nfs_sattrargs arg = { + .fh = NFS_FH(inode), + .sattr = sattr + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_SETATTR], + .rpc_argp = &arg, + .rpc_resp = fattr, + }; + int status; + + /* Mask out the non-modebit related stuff from attr->ia_mode */ + sattr->ia_mode &= S_IALLUGO; + + dprintk("NFS call setattr\n"); + if (sattr->ia_valid & ATTR_FILE) + msg.rpc_cred = nfs_file_cred(sattr->ia_file); + nfs_fattr_init(fattr); + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + if (status == 0) + nfs_setattr_update_inode(inode, sattr, fattr); + dprintk("NFS reply setattr: %d\n", status); + return status; +} + +static int +nfs_proc_lookup(struct inode *dir, struct dentry *dentry, const struct qstr *name, + struct nfs_fh *fhandle, struct nfs_fattr *fattr) +{ + struct nfs_diropargs arg = { + .fh = NFS_FH(dir), + .name = name->name, + .len = name->len + }; + struct nfs_diropok res = { + .fh = fhandle, + .fattr = fattr + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_LOOKUP], + .rpc_argp = &arg, + .rpc_resp = &res, + }; + int status; + unsigned short task_flags = 0; + + /* Is this is an attribute revalidation, subject to softreval? */ + if (nfs_lookup_is_soft_revalidate(dentry)) + task_flags |= RPC_TASK_TIMEOUT; + + dprintk("NFS call lookup %pd2\n", dentry); + nfs_fattr_init(fattr); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, task_flags); + dprintk("NFS reply lookup: %d\n", status); + return status; +} + +static int nfs_proc_readlink(struct inode *inode, struct page *page, + unsigned int pgbase, unsigned int pglen) +{ + struct nfs_readlinkargs args = { + .fh = NFS_FH(inode), + .pgbase = pgbase, + .pglen = pglen, + .pages = &page + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_READLINK], + .rpc_argp = &args, + }; + int status; + + dprintk("NFS call readlink\n"); + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + dprintk("NFS reply readlink: %d\n", status); + return status; +} + +struct nfs_createdata { + struct nfs_createargs arg; + struct nfs_diropok res; + struct nfs_fh fhandle; + struct nfs_fattr fattr; +}; + +static struct nfs_createdata *nfs_alloc_createdata(struct inode *dir, + struct dentry *dentry, struct iattr *sattr) +{ + struct nfs_createdata *data; + + data = kmalloc_obj(*data); + + if (data != NULL) { + data->arg.fh = NFS_FH(dir); + data->arg.name = dentry->d_name.name; + data->arg.len = dentry->d_name.len; + data->arg.sattr = sattr; + nfs_fattr_init(&data->fattr); + data->fhandle.size = 0; + data->res.fh = &data->fhandle; + data->res.fattr = &data->fattr; + } + return data; +}; + +static void nfs_free_createdata(const struct nfs_createdata *data) +{ + kfree(data); +} + +static int +nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, + int flags) +{ + struct nfs_createdata *data; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_CREATE], + }; + int status = -ENOMEM; + + dprintk("NFS call create %pd\n", dentry); + data = nfs_alloc_createdata(dir, dentry, sattr); + if (data == NULL) + goto out; + msg.rpc_argp = &data->arg; + msg.rpc_resp = &data->res; + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + if (status == 0) + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); + nfs_free_createdata(data); +out: + dprintk("NFS reply create: %d\n", status); + return status; +} + +/* + * In NFSv2, mknod is grafted onto the create call. + */ +static int +nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, + dev_t rdev) +{ + struct nfs_createdata *data; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_CREATE], + }; + umode_t mode; + int status = -ENOMEM; + + dprintk("NFS call mknod %pd\n", dentry); + + mode = sattr->ia_mode; + if (S_ISFIFO(mode)) { + sattr->ia_mode = (mode & ~S_IFMT) | S_IFCHR; + sattr->ia_valid &= ~ATTR_SIZE; + } else if (S_ISCHR(mode) || S_ISBLK(mode)) { + sattr->ia_valid |= ATTR_SIZE; + sattr->ia_size = new_encode_dev(rdev);/* get out your barf bag */ + } + + data = nfs_alloc_createdata(dir, dentry, sattr); + if (data == NULL) + goto out; + msg.rpc_argp = &data->arg; + msg.rpc_resp = &data->res; + + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + + if (status == -EINVAL && S_ISFIFO(mode)) { + sattr->ia_mode = mode; + nfs_fattr_init(data->res.fattr); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + } + if (status == 0) + status = nfs_instantiate(dentry, data->res.fh, data->res.fattr); + nfs_free_createdata(data); +out: + dprintk("NFS reply mknod: %d\n", status); + return status; +} + +static int +nfs_proc_remove(struct inode *dir, struct dentry *dentry) +{ + struct nfs_removeargs arg = { + .fh = NFS_FH(dir), + .name = dentry->d_name, + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_REMOVE], + .rpc_argp = &arg, + }; + int status; + + dprintk("NFS call remove %pd2\n",dentry); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + + dprintk("NFS reply remove: %d\n", status); + return status; +} + +static void +nfs_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) +{ + msg->rpc_proc = &nfs_procedures[NFSPROC_REMOVE]; +} + +static void nfs_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data) +{ + rpc_call_start(task); +} + +static int nfs_proc_unlink_done(struct rpc_task *task, struct inode *dir) +{ + nfs_mark_for_revalidate(dir); + return 1; +} + +static void +nfs_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry, + struct inode *same_parent) +{ + msg->rpc_proc = &nfs_procedures[NFSPROC_RENAME]; +} + +static void nfs_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data) +{ + rpc_call_start(task); +} + +static int +nfs_proc_rename_done(struct rpc_task *task, struct inode *old_dir, + struct inode *new_dir) +{ + nfs_mark_for_revalidate(old_dir); + nfs_mark_for_revalidate(new_dir); + return 1; +} + +static int +nfs_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) +{ + struct nfs_linkargs arg = { + .fromfh = NFS_FH(inode), + .tofh = NFS_FH(dir), + .toname = name->name, + .tolen = name->len + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_LINK], + .rpc_argp = &arg, + }; + int status; + + dprintk("NFS call link %s\n", name->name); + status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); + nfs_mark_for_revalidate(inode); + nfs_mark_for_revalidate(dir); + dprintk("NFS reply link: %d\n", status); + return status; +} + +static int +nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct folio *folio, + unsigned int len, struct iattr *sattr) +{ + struct page *page = &folio->page; + struct nfs_fh *fh; + struct nfs_fattr *fattr; + struct nfs_symlinkargs arg = { + .fromfh = NFS_FH(dir), + .fromname = dentry->d_name.name, + .fromlen = dentry->d_name.len, + .pages = &page, + .pathlen = len, + .sattr = sattr + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_SYMLINK], + .rpc_argp = &arg, + }; + int status = -ENAMETOOLONG; + + dprintk("NFS call symlink %pd\n", dentry); + + if (len > NFS2_MAXPATHLEN) + goto out; + + fh = nfs_alloc_fhandle(); + fattr = nfs_alloc_fattr(); + status = -ENOMEM; + if (fh == NULL || fattr == NULL) + goto out_free; + + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + + /* + * V2 SYMLINK requests don't return any attributes. Setting the + * filehandle size to zero indicates to nfs_instantiate that it + * should fill in the data with a LOOKUP call on the wire. + */ + if (status == 0) + status = nfs_instantiate(dentry, fh, fattr); + +out_free: + nfs_free_fattr(fattr); + nfs_free_fhandle(fh); +out: + dprintk("NFS reply symlink: %d\n", status); + return status; +} + +static struct dentry * +nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) +{ + struct nfs_createdata *data; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_MKDIR], + }; + struct dentry *alias = NULL; + int status = -ENOMEM; + + dprintk("NFS call mkdir %pd\n", dentry); + data = nfs_alloc_createdata(dir, dentry, sattr); + if (data == NULL) + goto out; + msg.rpc_argp = &data->arg; + msg.rpc_resp = &data->res; + + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + if (status == 0) { + alias = nfs_add_or_obtain(dentry, data->res.fh, data->res.fattr); + status = PTR_ERR_OR_ZERO(alias); + } else + alias = ERR_PTR(status); + nfs_free_createdata(data); +out: + dprintk("NFS reply mkdir: %d\n", status); + return alias; +} + +static int +nfs_proc_rmdir(struct inode *dir, const struct qstr *name) +{ + struct nfs_diropargs arg = { + .fh = NFS_FH(dir), + .name = name->name, + .len = name->len + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_RMDIR], + .rpc_argp = &arg, + }; + int status; + + dprintk("NFS call rmdir %s\n", name->name); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); + dprintk("NFS reply rmdir: %d\n", status); + return status; +} + +/* + * The READDIR implementation is somewhat hackish - we pass a temporary + * buffer to the encode function, which installs it in the receive + * the receive iovec. The decode function just parses the reply to make + * sure it is syntactically correct; the entries itself are decoded + * from nfs_readdir by calling the decode_entry function directly. + */ +static int nfs_proc_readdir(struct nfs_readdir_arg *nr_arg, + struct nfs_readdir_res *nr_res) +{ + struct inode *dir = d_inode(nr_arg->dentry); + struct nfs_readdirargs arg = { + .fh = NFS_FH(dir), + .cookie = nr_arg->cookie, + .count = nr_arg->page_len, + .pages = nr_arg->pages, + }; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_READDIR], + .rpc_argp = &arg, + .rpc_cred = nr_arg->cred, + }; + int status; + + dprintk("NFS call readdir %llu\n", (unsigned long long)nr_arg->cookie); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nr_res->verf[0] = nr_res->verf[1] = 0; + + nfs_invalidate_atime(dir); + + dprintk("NFS reply readdir: %d\n", status); + return status; +} + +static int +nfs_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsstat *stat) +{ + struct nfs2_fsstat fsinfo; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_STATFS], + .rpc_argp = fhandle, + .rpc_resp = &fsinfo, + }; + int status; + + dprintk("NFS call statfs\n"); + nfs_fattr_init(stat->fattr); + status = rpc_call_sync(server->client, &msg, 0); + dprintk("NFS reply statfs: %d\n", status); + if (status) + goto out; + stat->tbytes = (u64)fsinfo.blocks * fsinfo.bsize; + stat->fbytes = (u64)fsinfo.bfree * fsinfo.bsize; + stat->abytes = (u64)fsinfo.bavail * fsinfo.bsize; + stat->tfiles = 0; + stat->ffiles = 0; + stat->afiles = 0; +out: + return status; +} + +static int +nfs_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + struct nfs2_fsstat fsinfo; + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_STATFS], + .rpc_argp = fhandle, + .rpc_resp = &fsinfo, + }; + int status; + + dprintk("NFS call fsinfo\n"); + nfs_fattr_init(info->fattr); + status = rpc_call_sync(server->client, &msg, 0); + dprintk("NFS reply fsinfo: %d\n", status); + if (status) + goto out; + info->rtmax = NFS_MAXDATA; + info->rtpref = fsinfo.tsize; + info->rtmult = fsinfo.bsize; + info->wtmax = NFS_MAXDATA; + info->wtpref = fsinfo.tsize; + info->wtmult = fsinfo.bsize; + info->dtpref = fsinfo.tsize; + info->maxfilesize = 0x7FFFFFFF; + info->lease_time = 0; +out: + return status; +} + +static int +nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_pathconf *info) +{ + info->max_link = 0; + info->max_namelen = NFS2_MAXNAMLEN; + info->case_preserving = true; + return 0; +} + +static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + struct inode *inode = hdr->inode; + + nfs_invalidate_atime(inode); + if (task->tk_status >= 0) { + nfs_refresh_inode(inode, hdr->res.fattr); + /* Emulate the eof flag, which isn't normally needed in NFSv2 + * as it is guaranteed to always return the file attributes + */ + if ((hdr->res.count == 0 && hdr->args.count > 0) || + hdr->args.offset + hdr->res.count >= hdr->res.fattr->size) + hdr->res.eof = 1; + } + return 0; +} + +static void nfs_proc_read_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg) +{ + msg->rpc_proc = &nfs_procedures[NFSPROC_READ]; +} + +static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + rpc_call_start(task); + return 0; +} + +static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ + if (task->tk_status >= 0) { + hdr->res.count = hdr->args.count; + nfs_writeback_update_inode(hdr); + } + return 0; +} + +static void nfs_proc_write_setup(struct nfs_pgio_header *hdr, + struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + /* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */ + hdr->args.stable = NFS_FILE_SYNC; + msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE]; +} + +static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) +{ + BUG(); +} + +static void +nfs_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) +{ + BUG(); +} + +static int +nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl) +{ + struct inode *inode = file_inode(filp); + + return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl, NULL); +} + +/* Helper functions for NFS lock bounds checking */ +#define NFS_LOCK32_OFFSET_MAX ((__s32)0x7fffffffUL) +static int nfs_lock_check_bounds(const struct file_lock *fl) +{ + __s32 start, end; + + start = (__s32)fl->fl_start; + if ((loff_t)start != fl->fl_start) + goto out_einval; + + if (fl->fl_end != OFFSET_MAX) { + end = (__s32)fl->fl_end; + if ((loff_t)end != fl->fl_end) + goto out_einval; + } else + end = NFS_LOCK32_OFFSET_MAX; + + if (start < 0 || start > end) + goto out_einval; + return 0; +out_einval: + return -EINVAL; +} + +static int nfs_have_delegation(struct inode *inode, fmode_t type, int flags) +{ + return 0; +} + +static void nfs_return_delegation(struct inode *inode) +{ + if (S_ISREG(inode->i_mode)) + nfs_wb_all(inode); +} + +static const struct inode_operations nfs_dir_inode_operations = { + .create = nfs_create, + .lookup = nfs_lookup, + .atomic_open = nfs_atomic_open_v23, + .link = nfs_link, + .unlink = nfs_unlink, + .symlink = nfs_symlink, + .mkdir = nfs_mkdir, + .rmdir = nfs_rmdir, + .mknod = nfs_mknod, + .rename = nfs_rename, + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +}; + +static const struct inode_operations nfs_file_inode_operations = { + .permission = nfs_permission, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +}; + +const struct nfs_rpc_ops nfs_v2_clientops = { + .version = 2, /* protocol version */ + .dentry_ops = &nfs_dentry_operations, + .dir_inode_ops = &nfs_dir_inode_operations, + .file_inode_ops = &nfs_file_inode_operations, + .file_ops = &nfs_file_operations, + .getroot = nfs_proc_get_root, + .submount = nfs_submount, + .try_get_tree = nfs_try_get_tree, + .getattr = nfs_proc_getattr, + .setattr = nfs_proc_setattr, + .lookup = nfs_proc_lookup, + .access = NULL, /* access */ + .readlink = nfs_proc_readlink, + .create = nfs_proc_create, + .remove = nfs_proc_remove, + .unlink_setup = nfs_proc_unlink_setup, + .unlink_rpc_prepare = nfs_proc_unlink_rpc_prepare, + .unlink_done = nfs_proc_unlink_done, + .rename_setup = nfs_proc_rename_setup, + .rename_rpc_prepare = nfs_proc_rename_rpc_prepare, + .rename_done = nfs_proc_rename_done, + .link = nfs_proc_link, + .symlink = nfs_proc_symlink, + .mkdir = nfs_proc_mkdir, + .rmdir = nfs_proc_rmdir, + .readdir = nfs_proc_readdir, + .mknod = nfs_proc_mknod, + .statfs = nfs_proc_statfs, + .fsinfo = nfs_proc_fsinfo, + .pathconf = nfs_proc_pathconf, + .decode_dirent = nfs2_decode_dirent, + .pgio_rpc_prepare = nfs_proc_pgio_rpc_prepare, + .read_setup = nfs_proc_read_setup, + .read_done = nfs_read_done, + .write_setup = nfs_proc_write_setup, + .write_done = nfs_write_done, + .commit_setup = nfs_proc_commit_setup, + .commit_rpc_prepare = nfs_proc_commit_rpc_prepare, + .lock = nfs_proc_lock, + .lock_check_bounds = nfs_lock_check_bounds, + .close_context = nfs_close_context, + .have_delegation = nfs_have_delegation, + .return_delegation = nfs_return_delegation, + .alloc_client = nfs_alloc_client, + .init_client = nfs_init_client, + .free_client = nfs_free_client, + .create_server = nfs_create_server, + .clone_server = nfs_clone_server, +}; diff --git a/fs/nfs/read.c b/fs/nfs/read.c new file mode 100644 index 000000000..2b70bd2b9 --- /dev/null +++ b/fs/nfs/read.c @@ -0,0 +1,499 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/read.c + * + * Block I/O for NFS + * + * Partial copy of Linus' read cache modifications to fs/nfs/file.c + * modified for async RPC by okir@monad.swb.de + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "nfs4_fs.h" +#include "internal.h" +#include "iostat.h" +#include "fscache.h" +#include "pnfs.h" +#include "nfstrace.h" +#include "delegation.h" + +#define NFSDBG_FACILITY NFSDBG_PAGECACHE + +const struct nfs_pgio_completion_ops nfs_async_read_completion_ops; +static const struct nfs_rw_ops nfs_rw_read_ops; + +static struct kmem_cache *nfs_rdata_cachep; + +static struct nfs_pgio_header *nfs_readhdr_alloc(void) +{ + struct nfs_pgio_header *p = kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL); + + if (p) + p->rw_mode = FMODE_READ; + return p; +} + +static void nfs_readhdr_free(struct nfs_pgio_header *rhdr) +{ + kfree(rhdr->res.scratch); + kmem_cache_free(nfs_rdata_cachep, rhdr); +} + +static int nfs_return_empty_folio(struct folio *folio) +{ + folio_zero_segment(folio, 0, folio_size(folio)); + folio_mark_uptodate(folio); + if (nfs_netfs_folio_unlock(folio)) + folio_unlock(folio); + return 0; +} + +void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, + struct inode *inode, bool force_mds, + const struct nfs_pgio_completion_ops *compl_ops) +{ + struct nfs_server *server = NFS_SERVER(inode); + const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops; + +#if IS_ENABLED(CONFIG_NFS_V4) + if (server->pnfs_curr_ld && !force_mds) + pg_ops = server->pnfs_curr_ld->pg_read_ops; +#endif /* CONFIG_NFS_V4 */ + nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_read_ops, + server->rsize, 0); +} +EXPORT_SYMBOL_GPL(nfs_pageio_init_read); + +void nfs_pageio_complete_read(struct nfs_pageio_descriptor *pgio) +{ + struct nfs_pgio_mirror *pgm; + unsigned long npages; + + nfs_pageio_complete(pgio); + + /* It doesn't make sense to do mirrored reads! */ + WARN_ON_ONCE(pgio->pg_mirror_count != 1); + + pgm = &pgio->pg_mirrors[0]; + NFS_I(pgio->pg_inode)->read_io += pgm->pg_bytes_written; + npages = (pgm->pg_bytes_written + PAGE_SIZE - 1) >> PAGE_SHIFT; + nfs_add_stats(pgio->pg_inode, NFSIOS_READPAGES, npages); +} + + +void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio) +{ + struct nfs_pgio_mirror *mirror; + + if (pgio->pg_ops && pgio->pg_ops->pg_cleanup) + pgio->pg_ops->pg_cleanup(pgio); + + pgio->pg_ops = &nfs_pgio_rw_ops; + + /* read path should never have more than one mirror */ + WARN_ON_ONCE(pgio->pg_mirror_count != 1); + + mirror = &pgio->pg_mirrors[0]; + mirror->pg_bsize = NFS_SERVER(pgio->pg_inode)->rsize; +} +EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds); + +bool nfs_read_alloc_scratch(struct nfs_pgio_header *hdr, size_t size) +{ + WARN_ON(hdr->res.scratch != NULL); + hdr->res.scratch = kmalloc(size, GFP_KERNEL); + return hdr->res.scratch != NULL; +} +EXPORT_SYMBOL_GPL(nfs_read_alloc_scratch); + +static void nfs_readpage_release(struct nfs_page *req, int error) +{ + struct folio *folio = nfs_page_to_folio(req); + + if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) + if (nfs_netfs_folio_unlock(folio)) + folio_unlock(folio); + + nfs_release_request(req); +} + +static void nfs_page_group_set_uptodate(struct nfs_page *req) +{ + bool uptodate = false; + + nfs_page_group_lock(req); + if (!test_bit(PG_READ_FAILED, &req->wb_head->wb_flags) && + nfs_page_group_sync_on_bit_locked(req, PG_UPTODATE)) + uptodate = true; + nfs_page_group_unlock(req); + + if (uptodate) + folio_mark_uptodate(nfs_page_to_folio(req)); +} + +static void nfs_page_group_mark_read_failed(struct nfs_page *req) +{ + struct nfs_page *tmp; + + nfs_page_group_lock(req); + set_bit(PG_READ_FAILED, &req->wb_head->wb_flags); + tmp = req; + do { + clear_bit(PG_UPTODATE, &tmp->wb_flags); + tmp = tmp->wb_this_page; + } while (tmp != req); + nfs_page_group_unlock(req); +} + +static void nfs_read_completion(struct nfs_pgio_header *hdr) +{ + unsigned long bytes = 0; + int error; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) + goto out; + while (!list_empty(&hdr->pages)) { + struct nfs_page *req = nfs_list_entry(hdr->pages.next); + struct folio *folio = nfs_page_to_folio(req); + unsigned long start = req->wb_pgbase; + unsigned long end = req->wb_pgbase + req->wb_bytes; + + if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) { + /* note: regions of the page not covered by a + * request are zeroed in nfs_read_add_folio + */ + if (bytes > hdr->good_bytes) { + /* nothing in this request was good, so zero + * the full extent of the request */ + folio_zero_segment(folio, start, end); + + } else if (hdr->good_bytes - bytes < req->wb_bytes) { + /* part of this request has good bytes, but + * not all. zero the bad bytes */ + start += hdr->good_bytes - bytes; + WARN_ON(start < req->wb_pgbase); + folio_zero_segment(folio, start, end); + } + } + error = 0; + bytes += req->wb_bytes; + if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) { + if (bytes <= hdr->good_bytes) + nfs_page_group_set_uptodate(req); + else { + nfs_page_group_mark_read_failed(req); + error = hdr->error; + xchg(&nfs_req_openctx(req)->error, error); + } + } else + nfs_page_group_set_uptodate(req); + nfs_list_remove_request(req); + nfs_readpage_release(req, error); + } + nfs_netfs_read_completion(hdr); + +out: + hdr->release(hdr); +} + +static void nfs_initiate_read(struct nfs_pgio_header *hdr, + struct rpc_message *msg, + const struct nfs_rpc_ops *rpc_ops, + struct rpc_task_setup *task_setup_data, int how) +{ + rpc_ops->read_setup(hdr, msg); + nfs_netfs_initiate_read(hdr); + trace_nfs_initiate_read(hdr); +} + +static void +nfs_async_read_error(struct list_head *head, int error) +{ + struct nfs_page *req; + + while (!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_readpage_release(req, error); + } +} + +const struct nfs_pgio_completion_ops nfs_async_read_completion_ops = { + .error_cleanup = nfs_async_read_error, + .completion = nfs_read_completion, +}; + +/* + * This is the callback from RPC telling us whether a reply was + * received or some error occurred (timeout or socket shutdown). + */ +static int nfs_readpage_done(struct rpc_task *task, + struct nfs_pgio_header *hdr, + struct inode *inode) +{ + int status = NFS_PROTO(inode)->read_done(task, hdr); + if (status != 0) + return status; + + nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count); + trace_nfs_readpage_done(task, hdr); + + if (task->tk_status == -ESTALE) { + nfs_set_inode_stale(inode); + nfs_mark_for_revalidate(inode); + } + return 0; +} + +static void nfs_readpage_retry(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs_pgio_args *argp = &hdr->args; + struct nfs_pgio_res *resp = &hdr->res; + + /* This is a short read! */ + nfs_inc_stats(hdr->inode, NFSIOS_SHORTREAD); + trace_nfs_readpage_short(task, hdr); + + /* Has the server at least made some progress? */ + if (resp->count == 0) { + nfs_set_pgio_error(hdr, -EIO, argp->offset); + return; + } + + /* For non rpc-based layout drivers, retry-through-MDS */ + if (!task->tk_ops) { + hdr->pnfs_error = -EAGAIN; + return; + } + + /* Yes, so retry the read at the end of the hdr */ + hdr->mds_offset += resp->count; + argp->offset += resp->count; + argp->pgbase += resp->count; + argp->count -= resp->count; + resp->count = 0; + resp->eof = 0; + rpc_restart_call_prepare(task); +} + +static void nfs_readpage_result(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + if (hdr->res.eof) { + loff_t pos = hdr->args.offset + hdr->res.count; + unsigned int new = pos - hdr->io_start; + + if (hdr->good_bytes > new) { + hdr->good_bytes = new; + set_bit(NFS_IOHDR_EOF, &hdr->flags); + clear_bit(NFS_IOHDR_ERROR, &hdr->flags); + } + } else if (hdr->res.count < hdr->args.count) + nfs_readpage_retry(task, hdr); +} + +int nfs_read_add_folio(struct nfs_pageio_descriptor *pgio, + struct nfs_open_context *ctx, + struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + struct nfs_server *server = NFS_SERVER(inode); + size_t fsize = folio_size(folio); + unsigned int rsize = server->rsize; + struct nfs_page *new; + unsigned int len, aligned_len; + int error; + + len = nfs_folio_length(folio); + if (len == 0) + return nfs_return_empty_folio(folio); + + aligned_len = min_t(unsigned int, ALIGN(len, rsize), fsize); + + new = nfs_page_create_from_folio(ctx, folio, 0, aligned_len); + if (IS_ERR(new)) { + error = PTR_ERR(new); + if (nfs_netfs_folio_unlock(folio)) + folio_unlock(folio); + goto out; + } + + if (len < fsize) + folio_zero_segment(folio, len, fsize); + if (!nfs_pageio_add_request(pgio, new)) { + nfs_list_remove_request(new); + error = pgio->pg_error; + nfs_readpage_release(new, error); + goto out; + } + return 0; +out: + return error; +} + +/* + * Actually read a folio over the wire. + */ +static int nfs_do_read_folio(struct file *file, struct folio *folio) +{ + struct inode *inode = file_inode(file); + struct nfs_pageio_descriptor pgio; + struct nfs_open_context *ctx; + int ret; + + ctx = get_nfs_open_context(nfs_file_open_context(file)); + + xchg(&ctx->error, 0); + nfs_pageio_init_read(&pgio, inode, false, + &nfs_async_read_completion_ops); + + ret = nfs_read_add_folio(&pgio, ctx, folio); + if (ret) + goto out_put; + + nfs_pageio_complete_read(&pgio); + nfs_update_delegated_atime(inode); + if (pgio.pg_error < 0) { + ret = pgio.pg_error; + goto out_put; + } + + ret = folio_wait_locked_killable(folio); + if (!folio_test_uptodate(folio) && !ret) + ret = xchg(&ctx->error, 0); + +out_put: + put_nfs_open_context(ctx); + return ret; +} + +/* + * Synchronously read a folio. + * + * This is not heavily used as most users to try an asynchronous + * large read through ->readahead first. + */ +int nfs_read_folio(struct file *file, struct folio *folio) +{ + struct inode *inode = file_inode(file); + loff_t pos = folio_pos(folio); + size_t len = folio_size(folio); + int ret; + + trace_nfs_aop_readpage(inode, pos, len); + nfs_inc_stats(inode, NFSIOS_VFSREADPAGE); + task_io_account_read(len); + + /* + * Try to flush any pending writes to the file.. + * + * NOTE! Because we own the folio lock, there cannot + * be any new pending writes generated at this point + * for this folio (other folios can be written to). + */ + ret = nfs_wb_folio(inode, folio); + if (ret) + goto out_unlock; + if (folio_test_uptodate(folio)) + goto out_unlock; + + ret = -ESTALE; + if (NFS_STALE(inode)) + goto out_unlock; + + ret = nfs_netfs_read_folio(file, folio); + if (ret) + ret = nfs_do_read_folio(file, folio); +out: + trace_nfs_aop_readpage_done(inode, pos, len, ret); + return ret; +out_unlock: + folio_unlock(folio); + goto out; +} + +void nfs_readahead(struct readahead_control *ractl) +{ + struct nfs_pageio_descriptor pgio; + struct nfs_open_context *ctx; + unsigned int nr_pages = readahead_count(ractl); + struct file *file = ractl->file; + struct inode *inode = ractl->mapping->host; + struct folio *folio; + int ret; + + trace_nfs_aop_readahead(inode, readahead_pos(ractl), nr_pages); + nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); + task_io_account_read(readahead_length(ractl)); + + ret = -ESTALE; + if (NFS_STALE(inode)) + goto out; + + ret = nfs_netfs_readahead(ractl); + if (!ret) + goto out; + + if (file == NULL) { + ret = -EBADF; + ctx = nfs_find_open_context(inode, NULL, FMODE_READ); + if (ctx == NULL) + goto out; + } else + ctx = get_nfs_open_context(nfs_file_open_context(file)); + + nfs_pageio_init_read(&pgio, inode, false, + &nfs_async_read_completion_ops); + + while ((folio = readahead_folio(ractl)) != NULL) { + ret = nfs_read_add_folio(&pgio, ctx, folio); + if (ret) + break; + } + + nfs_pageio_complete_read(&pgio); + nfs_update_delegated_atime(inode); + + put_nfs_open_context(ctx); +out: + trace_nfs_aop_readahead_done(inode, nr_pages, ret); +} + +int __init nfs_init_readpagecache(void) +{ + nfs_rdata_cachep = kmem_cache_create("nfs_read_data", + sizeof(struct nfs_pgio_header), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (nfs_rdata_cachep == NULL) + return -ENOMEM; + + return 0; +} + +void nfs_destroy_readpagecache(void) +{ + kmem_cache_destroy(nfs_rdata_cachep); +} + +static const struct nfs_rw_ops nfs_rw_read_ops = { + .rw_alloc_header = nfs_readhdr_alloc, + .rw_free_header = nfs_readhdr_free, + .rw_done = nfs_readpage_done, + .rw_result = nfs_readpage_result, + .rw_initiate = nfs_initiate_read, +}; diff --git a/fs/nfs/super.c b/fs/nfs/super.c new file mode 100644 index 000000000..cb19f1540 --- /dev/null +++ b/fs/nfs/super.c @@ -0,0 +1,1463 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/super.c + * + * Copyright (C) 1992 Rick Sladkey + * + * nfs superblock handling functions + * + * Modularised by Alan Cox , while hacking some + * experimental NFS changes. Modularisation taken straight from SYS5 fs. + * + * Change to nfs_read_super() to permit NFS mounts to multi-homed hosts. + * J.S.Peatfield@damtp.cam.ac.uk + * + * Split from inode.c by David Howells + * + * - superblocks are indexed on server only - all inodes, dentries, etc. associated with a + * particular server are held in the same superblock + * - NFS superblocks can have several effective roots to the dentry tree + * - directory type roots are spliced into the tree when a path from one root reaches the root + * of another (see nfs_lookup()) + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "iostat.h" +#include "internal.h" +#include "fscache.h" +#include "nfs4session.h" +#include "pnfs.h" +#include "nfs.h" +#include "netns.h" +#include "sysfs.h" +#include "nfs4idmap.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +const struct super_operations nfs_sops = { + .alloc_inode = nfs_alloc_inode, + .free_inode = nfs_free_inode, + .write_inode = nfs_write_inode, + .drop_inode = nfs_drop_inode, + .statfs = nfs_statfs, + .evict_inode = nfs_evict_inode, + .umount_begin = nfs_umount_begin, + .show_options = nfs_show_options, + .show_devname = nfs_show_devname, + .show_path = nfs_show_path, + .show_stats = nfs_show_stats, +}; +EXPORT_SYMBOL_GPL(nfs_sops); + +#ifdef CONFIG_NFS_V4_2 +static const struct nfs_ssc_client_ops nfs_ssc_clnt_ops_tbl = { + .sco_sb_deactive = nfs_sb_deactive, +}; +#endif + +#if IS_ENABLED(CONFIG_NFS_V4) +static int __init register_nfs4_fs(void) +{ + return register_filesystem(&nfs4_fs_type); +} + +static void unregister_nfs4_fs(void) +{ + unregister_filesystem(&nfs4_fs_type); +} +#else +static int __init register_nfs4_fs(void) +{ + return 0; +} + +static void unregister_nfs4_fs(void) +{ +} +#endif + +#ifdef CONFIG_NFS_V4_2 +static void nfs_ssc_register_ops(void) +{ + nfs_ssc_register(&nfs_ssc_clnt_ops_tbl); +} + +static void nfs_ssc_unregister_ops(void) +{ + nfs_ssc_unregister(&nfs_ssc_clnt_ops_tbl); +} +#endif /* CONFIG_NFS_V4_2 */ + +static struct shrinker *acl_shrinker; + +/* + * Register the NFS filesystems + */ +int __init register_nfs_fs(void) +{ + int ret; + + ret = register_filesystem(&nfs_fs_type); + if (ret < 0) + goto error_0; + + ret = register_nfs4_fs(); + if (ret < 0) + goto error_1; + + ret = nfs_register_sysctl(); + if (ret < 0) + goto error_2; + + acl_shrinker = shrinker_alloc(0, "nfs-acl"); + if (!acl_shrinker) { + ret = -ENOMEM; + goto error_3; + } + + acl_shrinker->count_objects = nfs_access_cache_count; + acl_shrinker->scan_objects = nfs_access_cache_scan; + + shrinker_register(acl_shrinker); + +#ifdef CONFIG_NFS_V4_2 + nfs_ssc_register_ops(); +#endif + return 0; +error_3: + nfs_unregister_sysctl(); +error_2: + unregister_nfs4_fs(); +error_1: + unregister_filesystem(&nfs_fs_type); +error_0: + return ret; +} + +/* + * Unregister the NFS filesystems + */ +void __exit unregister_nfs_fs(void) +{ + shrinker_free(acl_shrinker); + nfs_unregister_sysctl(); + unregister_nfs4_fs(); +#ifdef CONFIG_NFS_V4_2 + nfs_ssc_unregister_ops(); +#endif + unregister_filesystem(&nfs_fs_type); +} + +bool nfs_sb_active(struct super_block *sb) +{ + struct nfs_server *server = NFS_SB(sb); + + if (!atomic_inc_not_zero(&sb->s_active)) + return false; + if (atomic_inc_return(&server->active) != 1) + atomic_dec(&sb->s_active); + return true; +} +EXPORT_SYMBOL_GPL(nfs_sb_active); + +void nfs_sb_deactive(struct super_block *sb) +{ + struct nfs_server *server = NFS_SB(sb); + + if (atomic_dec_and_test(&server->active)) + deactivate_super(sb); +} +EXPORT_SYMBOL_GPL(nfs_sb_deactive); + +int nfs_client_for_each_server(struct nfs_client *clp, + int (*fn)(struct nfs_server *server, void *data), void *data) +{ + struct nfs_server *server, *last = NULL; + int ret = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (!(server->super && nfs_sb_active(server->super))) + continue; + rcu_read_unlock(); + if (last) + nfs_sb_deactive(last->super); + last = server; + ret = fn(server, data); + if (ret) + goto out; + cond_resched(); + rcu_read_lock(); + } + rcu_read_unlock(); +out: + if (last) + nfs_sb_deactive(last->super); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_client_for_each_server); + +/* + * Deliver file system statistics to userspace + */ +int nfs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + struct nfs_server *server = NFS_SB(dentry->d_sb); + unsigned char blockbits; + unsigned long blockres; + struct nfs_fh *fh = NFS_FH(d_inode(dentry)); + struct nfs_fsstat res; + int error = -ENOMEM; + + res.fattr = nfs_alloc_fattr(); + if (res.fattr == NULL) + goto out_err; + + error = server->nfs_client->rpc_ops->statfs(server, fh, &res); + if (unlikely(error == -ESTALE)) { + struct dentry *pd_dentry; + + pd_dentry = dget_parent(dentry); + nfs_zap_caches(d_inode(pd_dentry)); + dput(pd_dentry); + } + nfs_free_fattr(res.fattr); + if (error < 0) + goto out_err; + + buf->f_type = NFS_SUPER_MAGIC; + + /* + * Current versions of glibc do not correctly handle the + * case where f_frsize != f_bsize. Eventually we want to + * report the value of wtmult in this field. + */ + buf->f_frsize = dentry->d_sb->s_blocksize; + + /* + * On most *nix systems, f_blocks, f_bfree, and f_bavail + * are reported in units of f_frsize. Linux hasn't had + * an f_frsize field in its statfs struct until recently, + * thus historically Linux's sys_statfs reports these + * fields in units of f_bsize. + */ + buf->f_bsize = dentry->d_sb->s_blocksize; + blockbits = dentry->d_sb->s_blocksize_bits; + blockres = (1 << blockbits) - 1; + buf->f_blocks = (res.tbytes + blockres) >> blockbits; + buf->f_bfree = (res.fbytes + blockres) >> blockbits; + buf->f_bavail = (res.abytes + blockres) >> blockbits; + + buf->f_files = res.tfiles; + buf->f_ffree = res.afiles; + + buf->f_namelen = server->namelen; + + return 0; + + out_err: + dprintk("%s: statfs error = %d\n", __func__, -error); + return error; +} +EXPORT_SYMBOL_GPL(nfs_statfs); + +/* + * Map the security flavour number to a name + */ +static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour) +{ + static const struct { + rpc_authflavor_t flavour; + const char *str; + } sec_flavours[NFS_AUTH_INFO_MAX_FLAVORS] = { + /* update NFS_AUTH_INFO_MAX_FLAVORS when this list changes! */ + { RPC_AUTH_NULL, "null" }, + { RPC_AUTH_UNIX, "sys" }, + { RPC_AUTH_GSS_KRB5, "krb5" }, + { RPC_AUTH_GSS_KRB5I, "krb5i" }, + { RPC_AUTH_GSS_KRB5P, "krb5p" }, + { RPC_AUTH_GSS_LKEY, "lkey" }, + { RPC_AUTH_GSS_LKEYI, "lkeyi" }, + { RPC_AUTH_GSS_LKEYP, "lkeyp" }, + { RPC_AUTH_GSS_SPKM, "spkm" }, + { RPC_AUTH_GSS_SPKMI, "spkmi" }, + { RPC_AUTH_GSS_SPKMP, "spkmp" }, + { UINT_MAX, "unknown" } + }; + int i; + + for (i = 0; sec_flavours[i].flavour != UINT_MAX; i++) { + if (sec_flavours[i].flavour == flavour) + break; + } + return sec_flavours[i].str; +} + +static void nfs_show_mountd_netid(struct seq_file *m, struct nfs_server *nfss, + int showdefaults) +{ + struct sockaddr *sap = (struct sockaddr *) &nfss->mountd_address; + char *proto = NULL; + + switch (sap->sa_family) { + case AF_INET: + switch (nfss->mountd_protocol) { + case IPPROTO_UDP: + proto = RPCBIND_NETID_UDP; + break; + case IPPROTO_TCP: + proto = RPCBIND_NETID_TCP; + break; + } + break; + case AF_INET6: + switch (nfss->mountd_protocol) { + case IPPROTO_UDP: + proto = RPCBIND_NETID_UDP6; + break; + case IPPROTO_TCP: + proto = RPCBIND_NETID_TCP6; + break; + } + break; + } + if (proto || showdefaults) + seq_printf(m, ",mountproto=%s", proto ?: "auto"); +} + +static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss, + int showdefaults) +{ + struct sockaddr *sap = (struct sockaddr *)&nfss->mountd_address; + + if (nfss->flags & NFS_MOUNT_LEGACY_INTERFACE) + return; + + switch (sap->sa_family) { + case AF_INET: { + struct sockaddr_in *sin = (struct sockaddr_in *)sap; + seq_printf(m, ",mountaddr=%pI4", &sin->sin_addr.s_addr); + break; + } + case AF_INET6: { + struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap; + seq_printf(m, ",mountaddr=%pI6c", &sin6->sin6_addr); + break; + } + default: + if (showdefaults) + seq_puts(m, ",mountaddr=unspecified"); + } + + if (nfss->mountd_version || showdefaults) + seq_printf(m, ",mountvers=%u", nfss->mountd_version); + if ((nfss->mountd_port && + nfss->mountd_port != (unsigned short)NFS_UNSPEC_PORT) || + showdefaults) + seq_printf(m, ",mountport=%u", nfss->mountd_port); + + nfs_show_mountd_netid(m, nfss, showdefaults); +} + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_show_nfsv4_options(struct seq_file *m, struct nfs_server *nfss, + int showdefaults) +{ + struct nfs_client *clp = nfss->nfs_client; + + seq_printf(m, ",clientaddr=%s", clp->cl_ipaddr); +} +#else +static void nfs_show_nfsv4_options(struct seq_file *m, struct nfs_server *nfss, + int showdefaults) +{ +} +#endif + +static void nfs_show_nfs_version(struct seq_file *m, + unsigned int version, + unsigned int minorversion) +{ + seq_printf(m, ",vers=%u", version); + if (version == 4) + seq_printf(m, ".%u", minorversion); +} + +/* + * Describe the mount options in force on this server representation + */ +static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, + int showdefaults) +{ + static const struct proc_nfs_info { + int flag; + const char *str; + const char *nostr; + } nfs_info[] = { + { NFS_MOUNT_SOFT, ",soft", "" }, + { NFS_MOUNT_SOFTERR, ",softerr", "" }, + { NFS_MOUNT_SOFTREVAL, ",softreval", "" }, + { NFS_MOUNT_POSIX, ",posix", "" }, + { NFS_MOUNT_NOCTO, ",nocto", "" }, + { NFS_MOUNT_NOAC, ",noac", "" }, + { NFS_MOUNT_NONLM, ",nolock", "" }, + { NFS_MOUNT_NOACL, ",noacl", "" }, + { NFS_MOUNT_NORDIRPLUS, ",nordirplus", "" }, + { NFS_MOUNT_FORCE_RDIRPLUS, ",rdirplus=force", "" }, + { NFS_MOUNT_UNSHARED, ",nosharecache", "" }, + { NFS_MOUNT_NORESVPORT, ",noresvport", "" }, + { NFS_MOUNT_NETUNREACH_FATAL, + ",fatal_neterrors=ENETDOWN:ENETUNREACH", + ",fatal_neterrors=none" }, + { 0, NULL, NULL } + }; + const struct proc_nfs_info *nfs_infop; + struct nfs_client *clp = nfss->nfs_client; + u32 version = clp->rpc_ops->version; + int local_flock, local_fcntl; + + nfs_show_nfs_version(m, version, clp->cl_minorversion); + seq_printf(m, ",rsize=%u", nfss->rsize); + seq_printf(m, ",wsize=%u", nfss->wsize); + if (nfss->bsize != 0) + seq_printf(m, ",bsize=%u", nfss->bsize); + seq_printf(m, ",namlen=%u", nfss->namelen); + if (nfss->acregmin != NFS_DEF_ACREGMIN*HZ || showdefaults) + seq_printf(m, ",acregmin=%u", nfss->acregmin/HZ); + if (nfss->acregmax != NFS_DEF_ACREGMAX*HZ || showdefaults) + seq_printf(m, ",acregmax=%u", nfss->acregmax/HZ); + if (nfss->acdirmin != NFS_DEF_ACDIRMIN*HZ || showdefaults) + seq_printf(m, ",acdirmin=%u", nfss->acdirmin/HZ); + if (nfss->acdirmax != NFS_DEF_ACDIRMAX*HZ || showdefaults) + seq_printf(m, ",acdirmax=%u", nfss->acdirmax/HZ); + if (!(nfss->flags & (NFS_MOUNT_SOFT|NFS_MOUNT_SOFTERR))) + seq_puts(m, ",hard"); + for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) { + if (nfss->flags & nfs_infop->flag) + seq_puts(m, nfs_infop->str); + else + seq_puts(m, nfs_infop->nostr); + } + rcu_read_lock(); + seq_printf(m, ",proto=%s", + rpc_peeraddr2str(nfss->client, RPC_DISPLAY_NETID)); + rcu_read_unlock(); + if (clp->cl_nconnect > 0) + seq_printf(m, ",nconnect=%u", clp->cl_nconnect); + if (version == 4) { + if (clp->cl_max_connect > 1) + seq_printf(m, ",max_connect=%u", clp->cl_max_connect); + if (nfss->port != NFS_PORT) + seq_printf(m, ",port=%u", nfss->port); + } else + if (nfss->port) + seq_printf(m, ",port=%u", nfss->port); + + seq_printf(m, ",timeo=%lu", 10U * nfss->client->cl_timeout->to_initval / HZ); + seq_printf(m, ",retrans=%u", nfss->client->cl_timeout->to_retries); + seq_printf(m, ",sec=%s", nfs_pseudoflavour_to_name(nfss->client->cl_auth->au_flavor)); + switch (clp->cl_xprtsec.policy) { + case RPC_XPRTSEC_TLS_ANON: + seq_puts(m, ",xprtsec=tls"); + break; + case RPC_XPRTSEC_TLS_X509: + seq_puts(m, ",xprtsec=mtls"); + break; + default: + break; + } + if (clp->cl_xprtsec.cert_serial) + seq_puts(m, ",cert_serial="); + if (clp->cl_xprtsec.privkey_serial) + seq_puts(m, ",privkey_serial="); + + if (version != 4) + nfs_show_mountd_options(m, nfss, showdefaults); + else + nfs_show_nfsv4_options(m, nfss, showdefaults); + + if (nfss->options & NFS_OPTION_FSCACHE) { +#ifdef CONFIG_NFS_FSCACHE + if (nfss->fscache_uniq) + seq_printf(m, ",fsc=%s", nfss->fscache_uniq); + else + seq_puts(m, ",fsc"); +#else + seq_puts(m, ",fsc"); +#endif + } + + if (nfss->options & NFS_OPTION_MIGRATION) + seq_puts(m, ",migration"); + + if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) { + if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) + seq_puts(m, ",lookupcache=none"); + else + seq_puts(m, ",lookupcache=pos"); + } + + local_flock = nfss->flags & NFS_MOUNT_LOCAL_FLOCK; + local_fcntl = nfss->flags & NFS_MOUNT_LOCAL_FCNTL; + + if (!local_flock && !local_fcntl) + seq_puts(m, ",local_lock=none"); + else if (local_flock && local_fcntl) + seq_puts(m, ",local_lock=all"); + else if (local_flock) + seq_puts(m, ",local_lock=flock"); + else + seq_puts(m, ",local_lock=posix"); + + if (nfss->flags & NFS_MOUNT_NO_ALIGNWRITE) + seq_puts(m, ",noalignwrite"); + + if (nfss->flags & NFS_MOUNT_WRITE_EAGER) { + if (nfss->flags & NFS_MOUNT_WRITE_WAIT) + seq_puts(m, ",write=wait"); + else + seq_puts(m, ",write=eager"); + } +} + +/* + * Describe the mount options on this VFS mountpoint + */ +int nfs_show_options(struct seq_file *m, struct dentry *root) +{ + struct nfs_server *nfss = NFS_SB(root->d_sb); + + nfs_show_mount_options(m, nfss, 0); + + rcu_read_lock(); + seq_printf(m, ",addr=%s", + rpc_peeraddr2str(nfss->nfs_client->cl_rpcclient, + RPC_DISPLAY_ADDR)); + rcu_read_unlock(); + + return 0; +} +EXPORT_SYMBOL_GPL(nfs_show_options); + +#if IS_ENABLED(CONFIG_NFS_V4) +static void show_lease(struct seq_file *m, struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + unsigned long expire; + + seq_printf(m, ",lease_time=%ld", clp->cl_lease_time / HZ); + expire = clp->cl_last_renewal + clp->cl_lease_time; + seq_printf(m, ",lease_expired=%ld", + time_after(expire, jiffies) ? 0 : (jiffies - expire) / HZ); +} + +static void show_sessions(struct seq_file *m, struct nfs_server *server) +{ + if (nfs4_has_session(server->nfs_client)) + seq_puts(m, ",sessions"); +} + +static void show_pnfs(struct seq_file *m, struct nfs_server *server) +{ + seq_printf(m, ",pnfs="); + if (server->pnfs_curr_ld) + seq_printf(m, "%s", server->pnfs_curr_ld->name); + else + seq_printf(m, "not configured"); +} + +static void show_implementation_id(struct seq_file *m, struct nfs_server *nfss) +{ + if (nfss->nfs_client && nfss->nfs_client->cl_implid) { + struct nfs41_impl_id *impl_id = nfss->nfs_client->cl_implid; + seq_printf(m, "\n\timpl_id:\tname='%s',domain='%s'," + "date='%llu,%u'", + impl_id->name, impl_id->domain, + impl_id->date.seconds, impl_id->date.nseconds); + } +} +#else /* CONFIG_NFS_V4 */ +static void show_implementation_id(struct seq_file *m, struct nfs_server *nfss) +{ +} +#endif /* CONFIG_NFS_V4 */ + +int nfs_show_devname(struct seq_file *m, struct dentry *root) +{ + char *page = kmalloc(PAGE_SIZE, GFP_KERNEL); + char *devname, *dummy; + int err = 0; + if (!page) + return -ENOMEM; + devname = nfs_path(&dummy, root, page, PAGE_SIZE, 0); + if (IS_ERR(devname)) + err = PTR_ERR(devname); + else + seq_escape(m, devname, " \t\n\\"); + kfree(page); + return err; +} +EXPORT_SYMBOL_GPL(nfs_show_devname); + +int nfs_show_path(struct seq_file *m, struct dentry *dentry) +{ + seq_puts(m, "/"); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_show_path); + +/* + * Present statistical information for this VFS mountpoint + */ +int nfs_show_stats(struct seq_file *m, struct dentry *root) +{ + int i, cpu; + struct nfs_server *nfss = NFS_SB(root->d_sb); + struct rpc_auth *auth = nfss->client->cl_auth; + struct nfs_iostats totals = { }; + + seq_printf(m, "statvers=%s", NFS_IOSTAT_VERS); + + /* + * Display all mount option settings + */ + seq_puts(m, "\n\topts:\t"); + seq_puts(m, sb_rdonly(root->d_sb) ? "ro" : "rw"); + seq_puts(m, root->d_sb->s_flags & SB_SYNCHRONOUS ? ",sync" : ""); + seq_puts(m, root->d_sb->s_flags & SB_NOATIME ? ",noatime" : ""); + seq_puts(m, root->d_sb->s_flags & SB_NODIRATIME ? ",nodiratime" : ""); + nfs_show_mount_options(m, nfss, 1); + + seq_printf(m, "\n\tage:\t%lu", (jiffies - nfss->mount_time) / HZ); + + show_implementation_id(m, nfss); + + seq_puts(m, "\n\tcaps:\t"); + seq_printf(m, "caps=0x%x", nfss->caps); + seq_printf(m, ",wtmult=%u", nfss->wtmult); + seq_printf(m, ",dtsize=%u", nfss->dtsize); + seq_printf(m, ",bsize=%u", nfss->bsize); + seq_printf(m, ",namlen=%u", nfss->namelen); + +#if IS_ENABLED(CONFIG_NFS_V4) + if (nfss->nfs_client->rpc_ops->version == 4) { + seq_puts(m, "\n\tnfsv4:\t"); + seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); + seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); + seq_printf(m, ",bm2=0x%x", nfss->attr_bitmask[2]); + seq_printf(m, ",acl=0x%x", nfss->acl_bitmask); + show_sessions(m, nfss); + show_pnfs(m, nfss); + show_lease(m, nfss); + } +#endif + + /* + * Display security flavor in effect for this mount + */ + seq_printf(m, "\n\tsec:\tflavor=%u", auth->au_ops->au_flavor); + if (auth->au_flavor) + seq_printf(m, ",pseudoflavor=%u", auth->au_flavor); + + /* + * Display superblock I/O counters + */ + for_each_possible_cpu(cpu) { + struct nfs_iostats *stats; + + preempt_disable(); + stats = per_cpu_ptr(nfss->io_stats, cpu); + + for (i = 0; i < __NFSIOS_COUNTSMAX; i++) + totals.events[i] += stats->events[i]; + for (i = 0; i < __NFSIOS_BYTESMAX; i++) + totals.bytes[i] += stats->bytes[i]; + + preempt_enable(); + } + + seq_puts(m, "\n\tevents:\t"); + for (i = 0; i < __NFSIOS_COUNTSMAX; i++) + seq_printf(m, "%lu ", totals.events[i]); + seq_puts(m, "\n\tbytes:\t"); + for (i = 0; i < __NFSIOS_BYTESMAX; i++) + seq_printf(m, "%Lu ", totals.bytes[i]); + seq_putc(m, '\n'); + + rpc_clnt_show_stats(m, nfss->client); + + return 0; +} +EXPORT_SYMBOL_GPL(nfs_show_stats); + +/* + * Begin unmount by attempting to remove all automounted mountpoints we added + * in response to xdev traversals and referrals + */ +void nfs_umount_begin(struct super_block *sb) +{ + struct nfs_server *server; + struct rpc_clnt *rpc; + + server = NFS_SB(sb); + /* -EIO all pending I/O */ + rpc = server->client_acl; + if (!IS_ERR(rpc)) + rpc_killall_tasks(rpc); + rpc = server->client; + if (!IS_ERR(rpc)) + rpc_killall_tasks(rpc); +} +EXPORT_SYMBOL_GPL(nfs_umount_begin); + +/* + * Return true if 'match' is in auth_info or auth_info is empty. + * Return false otherwise. + */ +bool nfs_auth_info_match(const struct nfs_auth_info *auth_info, + rpc_authflavor_t match) +{ + int i; + + if (!auth_info->flavor_len) + return true; + + for (i = 0; i < auth_info->flavor_len; i++) { + if (auth_info->flavors[i] == match) + return true; + } + return false; +} +EXPORT_SYMBOL_GPL(nfs_auth_info_match); + +/* + * Ensure that a specified authtype in ctx->auth_info is supported by + * the server. Returns 0 and sets ctx->selected_flavor if it's ok, and + * -EACCES if not. + */ +static int nfs_verify_authflavors(struct nfs_fs_context *ctx, + rpc_authflavor_t *server_authlist, + unsigned int count) +{ + rpc_authflavor_t flavor = RPC_AUTH_MAXFLAVOR; + bool found_auth_null = false; + unsigned int i; + + /* + * If the sec= mount option is used, the specified flavor or AUTH_NULL + * must be in the list returned by the server. + * + * AUTH_NULL has a special meaning when it's in the server list - it + * means that the server will ignore the rpc creds, so any flavor + * can be used but still use the sec= that was specified. + * + * Note also that the MNT procedure in MNTv1 does not return a list + * of supported security flavors. In this case, nfs_mount() fabricates + * a security flavor list containing just AUTH_NULL. + */ + for (i = 0; i < count; i++) { + flavor = server_authlist[i]; + + if (nfs_auth_info_match(&ctx->auth_info, flavor)) + goto out; + + if (flavor == RPC_AUTH_NULL) + found_auth_null = true; + } + + if (found_auth_null) { + flavor = ctx->auth_info.flavors[0]; + goto out; + } + + dfprintk(MOUNT, + "NFS: specified auth flavors not supported by server\n"); + return -EACCES; + +out: + ctx->selected_flavor = flavor; + dfprintk(MOUNT, "NFS: using auth flavor %u\n", ctx->selected_flavor); + return 0; +} + +/* + * Use the remote server's MOUNT service to request the NFS file handle + * corresponding to the provided path. + */ +static int nfs_request_mount(struct fs_context *fc, + struct nfs_fh *root_fh, + rpc_authflavor_t *server_authlist, + unsigned int *server_authlist_len) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct nfs_mount_request request = { + .sap = &ctx->mount_server._address, + .dirpath = ctx->nfs_server.export_path, + .protocol = ctx->mount_server.protocol, + .fh = root_fh, + .noresvport = ctx->flags & NFS_MOUNT_NORESVPORT, + .auth_flav_len = server_authlist_len, + .auth_flavs = server_authlist, + .net = fc->net_ns, + }; + int status; + + if (ctx->mount_server.version == 0) { + switch (ctx->version) { + default: + ctx->mount_server.version = NFS_MNT3_VERSION; + break; + case 2: + ctx->mount_server.version = NFS_MNT_VERSION; + } + } + request.version = ctx->mount_server.version; + + if (ctx->mount_server.hostname) + request.hostname = ctx->mount_server.hostname; + else + request.hostname = ctx->nfs_server.hostname; + + /* + * Construct the mount server's address. + */ + if (ctx->mount_server.address.sa_family == AF_UNSPEC) { + memcpy(request.sap, &ctx->nfs_server._address, + ctx->nfs_server.addrlen); + ctx->mount_server.addrlen = ctx->nfs_server.addrlen; + } + request.salen = ctx->mount_server.addrlen; + nfs_set_port(request.sap, &ctx->mount_server.port, 0); + + /* + * Now ask the mount server to map our export path + * to a file handle. + */ + if ((request.protocol == XPRT_TRANSPORT_UDP) == + !(ctx->flags & NFS_MOUNT_TCP)) + /* + * NFS protocol and mount protocol are both UDP or neither UDP + * so timeouts are compatible. Use NFS timeouts for MOUNT + */ + status = nfs_mount(&request, ctx->timeo, ctx->retrans); + else + status = nfs_mount(&request, NFS_UNSPEC_TIMEO, NFS_UNSPEC_RETRANS); + if (status != 0) { + dfprintk(MOUNT, "NFS: unable to mount server %s, error %d\n", + request.hostname, status); + return status; + } + + return 0; +} + +static struct nfs_server *nfs_try_mount_request(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + int status; + unsigned int i; + bool tried_auth_unix = false; + bool auth_null_in_list = false; + struct nfs_server *server = ERR_PTR(-EACCES); + rpc_authflavor_t authlist[NFS_MAX_SECFLAVORS]; + unsigned int authlist_len = ARRAY_SIZE(authlist); + + /* make sure 'nolock'/'lock' override the 'local_lock' mount option */ + if (ctx->lock_status) { + if (ctx->lock_status == NFS_LOCK_NOLOCK) { + ctx->flags |= NFS_MOUNT_NONLM; + ctx->flags |= (NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } else { + ctx->flags &= ~NFS_MOUNT_NONLM; + ctx->flags &= ~(NFS_MOUNT_LOCAL_FLOCK | NFS_MOUNT_LOCAL_FCNTL); + } + } + status = nfs_request_mount(fc, ctx->mntfh, authlist, &authlist_len); + if (status) + return ERR_PTR(status); + + /* + * Was a sec= authflavor specified in the options? First, verify + * whether the server supports it, and then just try to use it if so. + */ + if (ctx->auth_info.flavor_len > 0) { + status = nfs_verify_authflavors(ctx, authlist, authlist_len); + dfprintk(MOUNT, "NFS: using auth flavor %u\n", + ctx->selected_flavor); + if (status) + return ERR_PTR(status); + return ctx->nfs_mod->rpc_ops->create_server(fc); + } + + /* + * No sec= option was provided. RFC 2623, section 2.7 suggests we + * SHOULD prefer the flavor listed first. However, some servers list + * AUTH_NULL first. Avoid ever choosing AUTH_NULL. + */ + for (i = 0; i < authlist_len; ++i) { + rpc_authflavor_t flavor; + struct rpcsec_gss_info info; + + flavor = authlist[i]; + switch (flavor) { + case RPC_AUTH_UNIX: + tried_auth_unix = true; + break; + case RPC_AUTH_NULL: + auth_null_in_list = true; + continue; + default: + if (rpcauth_get_gssinfo(flavor, &info) != 0) + continue; + break; + } + dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", flavor); + ctx->selected_flavor = flavor; + server = ctx->nfs_mod->rpc_ops->create_server(fc); + if (!IS_ERR(server)) + return server; + } + + /* + * Nothing we tried so far worked. At this point, give up if we've + * already tried AUTH_UNIX or if the server's list doesn't contain + * AUTH_NULL + */ + if (tried_auth_unix || !auth_null_in_list) + return server; + + /* Last chance! Try AUTH_UNIX */ + dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", RPC_AUTH_UNIX); + ctx->selected_flavor = RPC_AUTH_UNIX; + return ctx->nfs_mod->rpc_ops->create_server(fc); +} + +int nfs_try_get_tree(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + + if (ctx->need_mount) + ctx->server = nfs_try_mount_request(fc); + else + ctx->server = ctx->nfs_mod->rpc_ops->create_server(fc); + + return nfs_get_tree_common(fc); +} +EXPORT_SYMBOL_GPL(nfs_try_get_tree); + + +#define NFS_REMOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \ + | NFS_MOUNT_SECURE \ + | NFS_MOUNT_TCP \ + | NFS_MOUNT_VER3 \ + | NFS_MOUNT_KERBEROS \ + | NFS_MOUNT_NONLM \ + | NFS_MOUNT_BROKEN_SUID \ + | NFS_MOUNT_STRICTLOCK \ + | NFS_MOUNT_LEGACY_INTERFACE) + +#define NFS_MOUNT_CMP_FLAGMASK (NFS_REMOUNT_CMP_FLAGMASK & \ + ~(NFS_MOUNT_UNSHARED | NFS_MOUNT_NORESVPORT)) + +static int +nfs_compare_remount_data(struct nfs_server *nfss, + struct nfs_fs_context *ctx) +{ + if ((ctx->flags ^ nfss->flags) & NFS_REMOUNT_CMP_FLAGMASK || + ctx->rsize != nfss->rsize || + ctx->wsize != nfss->wsize || + ctx->version != nfss->nfs_client->rpc_ops->version || + ctx->minorversion != nfss->nfs_client->cl_minorversion || + ctx->retrans != nfss->client->cl_timeout->to_retries || + !nfs_auth_info_match(&ctx->auth_info, nfss->client->cl_auth->au_flavor) || + ctx->acregmin != nfss->acregmin / HZ || + ctx->acregmax != nfss->acregmax / HZ || + ctx->acdirmin != nfss->acdirmin / HZ || + ctx->acdirmax != nfss->acdirmax / HZ || + ctx->timeo != (10U * nfss->client->cl_timeout->to_initval / HZ) || + (ctx->options & NFS_OPTION_FSCACHE) != (nfss->options & NFS_OPTION_FSCACHE) || + ctx->nfs_server.port != nfss->port || + ctx->nfs_server.addrlen != nfss->nfs_client->cl_addrlen || + !rpc_cmp_addr((struct sockaddr *)&ctx->nfs_server.address, + (struct sockaddr *)&nfss->nfs_client->cl_addr)) + return -EINVAL; + + return 0; +} + +int nfs_reconfigure(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct super_block *sb = fc->root->d_sb; + struct nfs_server *nfss = sb->s_fs_info; + int ret; + + sync_filesystem(sb); + + /* + * Userspace mount programs that send binary options generally send + * them populated with default values. We have no way to know which + * ones were explicitly specified. Fall back to legacy behavior and + * just return success. + */ + if (ctx->skip_reconfig_option_check) + return 0; + + /* + * noac is a special case. It implies -o sync, but that's not + * necessarily reflected in the mtab options. reconfigure_super + * will clear SB_SYNCHRONOUS if -o sync wasn't specified in the + * remount options, so we have to explicitly reset it. + */ + if (ctx->flags & NFS_MOUNT_NOAC) { + fc->sb_flags |= SB_SYNCHRONOUS; + fc->sb_flags_mask |= SB_SYNCHRONOUS; + } + + /* compare new mount options with old ones */ + ret = nfs_compare_remount_data(nfss, ctx); + if (ret) + return ret; + + return nfs_probe_server(nfss, NFS_FH(d_inode(fc->root))); +} +EXPORT_SYMBOL_GPL(nfs_reconfigure); + +/* + * Finish setting up an NFS superblock + */ +static void nfs_fill_super(struct super_block *sb, struct nfs_fs_context *ctx) +{ + struct nfs_server *server = NFS_SB(sb); + + sb->s_blocksize_bits = 0; + sb->s_blocksize = 0; + sb->s_xattr = server->nfs_client->cl_nfs_mod->xattr; + sb->s_op = server->nfs_client->cl_nfs_mod->sops; + if (server->bsize) + sb->s_blocksize = + nfs_block_size(server->bsize, &sb->s_blocksize_bits); + + switch (server->nfs_client->rpc_ops->version) { + case 2: + sb->s_time_gran = 1000; + sb->s_time_min = 0; + sb->s_time_max = U32_MAX; + break; + case 3: + /* + * The VFS shouldn't apply the umask to mode bits. + * We will do so ourselves when necessary. + */ + sb->s_flags |= SB_POSIXACL; + sb->s_time_gran = 1; + sb->s_time_min = 0; + sb->s_time_max = U32_MAX; + sb->s_export_op = &nfs_export_ops; + break; + case 4: + sb->s_iflags |= SB_I_NOUMASK; + sb->s_time_gran = 1; + sb->s_time_min = S64_MIN; + sb->s_time_max = S64_MAX; + if (server->caps & NFS_CAP_ATOMIC_OPEN_V1) + sb->s_export_op = &nfs_export_ops; + break; + } + + sb->s_magic = NFS_SUPER_MAGIC; + + /* We probably want something more informative here */ + snprintf(sb->s_id, sizeof(sb->s_id), + "%u:%u", MAJOR(sb->s_dev), MINOR(sb->s_dev)); + + if (sb->s_blocksize == 0) + sb->s_blocksize = nfs_block_bits(server->wsize, + &sb->s_blocksize_bits); + + nfs_super_set_maxbytes(sb, server->maxfilesize); + nfs_sysfs_move_server_to_sb(sb); + server->has_sec_mnt_opts = ctx->has_sec_mnt_opts; +} + +static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b, + const struct fs_context *fc) +{ + const struct nfs_server *a = s->s_fs_info; + const struct rpc_clnt *clnt_a = a->client; + const struct rpc_clnt *clnt_b = b->client; + + if ((s->s_flags & NFS_SB_MASK) != (fc->sb_flags & NFS_SB_MASK)) + goto Ebusy; + if (a->nfs_client != b->nfs_client) + goto Ebusy; + if ((a->flags ^ b->flags) & NFS_MOUNT_CMP_FLAGMASK) + goto Ebusy; + if (a->wsize != b->wsize) + goto Ebusy; + if (a->rsize != b->rsize) + goto Ebusy; + if (a->acregmin != b->acregmin) + goto Ebusy; + if (a->acregmax != b->acregmax) + goto Ebusy; + if (a->acdirmin != b->acdirmin) + goto Ebusy; + if (a->acdirmax != b->acdirmax) + goto Ebusy; + if (clnt_a->cl_auth->au_flavor != clnt_b->cl_auth->au_flavor) + goto Ebusy; + return 1; +Ebusy: + return 0; +} + +static int nfs_set_super(struct super_block *s, struct fs_context *fc) +{ + struct nfs_server *server = fc->s_fs_info; + int ret; + + set_default_d_op(s, server->nfs_client->rpc_ops->dentry_ops); + ret = set_anon_super(s, server); + if (ret == 0) + server->s_dev = s->s_dev; + return ret; +} + +static int nfs_compare_super_address(struct nfs_server *server1, + struct nfs_server *server2) +{ + struct rpc_xprt *xprt1, *xprt2; + struct sockaddr *sap1, *sap2; + + rcu_read_lock(); + + xprt1 = rcu_dereference(server1->client->cl_xprt); + xprt2 = rcu_dereference(server2->client->cl_xprt); + + if (!net_eq(xprt1->xprt_net, xprt2->xprt_net)) + goto out_unlock; + + rcu_read_unlock(); + + sap1 = (struct sockaddr *)&server1->nfs_client->cl_addr; + sap2 = (struct sockaddr *)&server2->nfs_client->cl_addr; + + if (sap1->sa_family != sap2->sa_family) + return 0; + + switch (sap1->sa_family) { + case AF_INET: { + struct sockaddr_in *sin1 = (struct sockaddr_in *)sap1; + struct sockaddr_in *sin2 = (struct sockaddr_in *)sap2; + if (sin1->sin_addr.s_addr != sin2->sin_addr.s_addr) + return 0; + if (sin1->sin_port != sin2->sin_port) + return 0; + break; + } + case AF_INET6: { + struct sockaddr_in6 *sin1 = (struct sockaddr_in6 *)sap1; + struct sockaddr_in6 *sin2 = (struct sockaddr_in6 *)sap2; + if (!ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr)) + return 0; + if (sin1->sin6_port != sin2->sin6_port) + return 0; + break; + } + default: + return 0; + } + + return 1; + +out_unlock: + rcu_read_unlock(); + return 0; +} + +static int nfs_compare_userns(const struct nfs_server *old, + const struct nfs_server *new) +{ + const struct user_namespace *oldns = &init_user_ns; + const struct user_namespace *newns = &init_user_ns; + + if (old->client && old->client->cl_cred) + oldns = old->client->cl_cred->user_ns; + if (new->client && new->client->cl_cred) + newns = new->client->cl_cred->user_ns; + if (oldns != newns) + return 0; + return 1; +} + +static int nfs_compare_super(struct super_block *sb, struct fs_context *fc) +{ + struct nfs_server *server = fc->s_fs_info, *old = NFS_SB(sb); + + if (!nfs_compare_super_address(old, server)) + return 0; + /* Note: NFS_MOUNT_UNSHARED == NFS4_MOUNT_UNSHARED */ + if (old->flags & NFS_MOUNT_UNSHARED) + return 0; + if (memcmp(&old->fsid, &server->fsid, sizeof(old->fsid)) != 0) + return 0; + if (!nfs_compare_userns(old, server)) + return 0; + if ((old->has_sec_mnt_opts || fc->security) && + security_sb_mnt_opts_compat(sb, fc->security)) + return 0; + return nfs_compare_mount_options(sb, server, fc); +} + +#ifdef CONFIG_NFS_FSCACHE +static int nfs_get_cache_cookie(struct super_block *sb, + struct nfs_fs_context *ctx) +{ + struct nfs_server *nfss = NFS_SB(sb); + char *uniq = NULL; + int ulen = 0; + + nfss->fscache = NULL; + + if (!ctx) + return 0; + + if (ctx->clone_data.sb) { + struct nfs_server *mnt_s = NFS_SB(ctx->clone_data.sb); + if (!(mnt_s->options & NFS_OPTION_FSCACHE)) + return 0; + if (mnt_s->fscache_uniq) { + uniq = mnt_s->fscache_uniq; + ulen = strlen(uniq); + } + } else { + if (!(ctx->options & NFS_OPTION_FSCACHE)) + return 0; + if (ctx->fscache_uniq) { + uniq = ctx->fscache_uniq; + ulen = strlen(ctx->fscache_uniq); + } + } + + return nfs_fscache_get_super_cookie(sb, uniq, ulen); +} +#else +static int nfs_get_cache_cookie(struct super_block *sb, + struct nfs_fs_context *ctx) +{ + return 0; +} +#endif + +int nfs_get_tree_common(struct fs_context *fc) +{ + struct nfs_fs_context *ctx = nfs_fc2context(fc); + struct super_block *s; + int (*compare_super)(struct super_block *, struct fs_context *) = nfs_compare_super; + struct nfs_server *server = ctx->server; + int error; + + ctx->server = NULL; + if (IS_ERR(server)) + return PTR_ERR(server); + + if (server->flags & NFS_MOUNT_UNSHARED) + compare_super = NULL; + + /* -o noac implies -o sync */ + if (server->flags & NFS_MOUNT_NOAC) + fc->sb_flags |= SB_SYNCHRONOUS; + + /* Get a superblock - note that we may end up sharing one that already exists */ + fc->s_fs_info = server; + s = sget_fc(fc, compare_super, nfs_set_super); + fc->s_fs_info = NULL; + if (IS_ERR(s)) { + error = PTR_ERR(s); + nfs_errorf(fc, "NFS: Couldn't get superblock"); + goto out_err_nosb; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } else { + error = super_setup_bdi_name(s, "%u:%u", MAJOR(server->s_dev), + MINOR(server->s_dev)); + if (error) + goto error_splat_super; + s->s_bdi->io_pages = server->rpages; + server->super = s; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + nfs_fill_super(s, ctx); + error = nfs_get_cache_cookie(s, ctx); + if (error < 0) + goto error_splat_super; + } + + error = nfs_get_root(s, fc); + if (error < 0) { + nfs_errorf(fc, "NFS: Couldn't get root dentry"); + goto error_splat_super; + } + + s->s_flags |= SB_ACTIVE; + error = 0; + +out: + return error; + +out_err_nosb: + nfs_free_server(server); + goto out; +error_splat_super: + deactivate_locked_super(s); + goto out; +} + +/* + * Destroy an NFS superblock + */ +void nfs_kill_super(struct super_block *s) +{ + struct nfs_server *server = NFS_SB(s); + + nfs_sysfs_move_sb_to_server(server); + kill_anon_super(s); + + nfs_fscache_release_super_cookie(s); + + nfs_free_server(server); +} +EXPORT_SYMBOL_GPL(nfs_kill_super); + +#if IS_ENABLED(CONFIG_NFS_V4) + +/* + * NFS v4 module parameters need to stay in the + * NFS client for backwards compatibility + */ +unsigned int nfs_callback_set_tcpport; +unsigned short nfs_callback_nr_threads; +/* Default cache timeout is 10 minutes */ +unsigned int nfs_idmap_cache_timeout = 600; +/* Turn off NFSv4 uid/gid mapping when using AUTH_SYS */ +bool nfs4_disable_idmapping = true; +unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE; +unsigned short max_session_cb_slots = NFS4_DEF_CB_SLOT_TABLE_SIZE; +unsigned short send_implementation_id = 1; +char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN] = ""; +bool recover_lost_locks = false; +short nfs_delay_retrans = -1; + +EXPORT_SYMBOL_GPL(nfs_callback_nr_threads); +EXPORT_SYMBOL_GPL(nfs_callback_set_tcpport); +EXPORT_SYMBOL_GPL(nfs_idmap_cache_timeout); +EXPORT_SYMBOL_GPL(nfs4_disable_idmapping); +EXPORT_SYMBOL_GPL(max_session_slots); +EXPORT_SYMBOL_GPL(max_session_cb_slots); +EXPORT_SYMBOL_GPL(send_implementation_id); +EXPORT_SYMBOL_GPL(nfs4_client_id_uniquifier); +EXPORT_SYMBOL_GPL(recover_lost_locks); +EXPORT_SYMBOL_GPL(nfs_delay_retrans); + +#define NFS_CALLBACK_MAXPORTNR (65535U) + +static int param_set_portnr(const char *val, const struct kernel_param *kp) +{ + unsigned long num; + int ret; + + if (!val) + return -EINVAL; + ret = kstrtoul(val, 0, &num); + if (ret || num > NFS_CALLBACK_MAXPORTNR) + return -EINVAL; + *((unsigned int *)kp->arg) = num; + return 0; +} +static const struct kernel_param_ops param_ops_portnr = { + .set = param_set_portnr, + .get = param_get_uint, +}; +#define param_check_portnr(name, p) __param_check(name, p, unsigned int) + +module_param_named(callback_tcpport, nfs_callback_set_tcpport, portnr, 0644); +module_param_named(callback_nr_threads, nfs_callback_nr_threads, ushort, 0644); +MODULE_PARM_DESC(callback_nr_threads, "Number of threads that will be " + "assigned to the NFSv4 callback channels."); +module_param(nfs_idmap_cache_timeout, int, 0644); +module_param(nfs4_disable_idmapping, bool, 0644); +module_param_string(nfs4_unique_id, nfs4_client_id_uniquifier, + NFS4_CLIENT_ID_UNIQ_LEN, 0600); +MODULE_PARM_DESC(nfs4_disable_idmapping, + "Turn off NFSv4 idmapping when using 'sec=sys'"); +module_param(max_session_slots, ushort, 0644); +MODULE_PARM_DESC(max_session_slots, "Maximum number of outstanding NFSv4.1 " + "requests the client will negotiate"); +module_param(max_session_cb_slots, ushort, 0644); +MODULE_PARM_DESC(max_session_cb_slots, "Maximum number of parallel NFSv4.1 " + "callbacks the client will process for a given server"); +module_param(send_implementation_id, ushort, 0644); +MODULE_PARM_DESC(send_implementation_id, + "Send implementation ID with NFSv4.1 exchange_id"); +MODULE_PARM_DESC(nfs4_unique_id, "nfs_client_id4 uniquifier string"); + +module_param(recover_lost_locks, bool, 0644); +MODULE_PARM_DESC(recover_lost_locks, + "If the server reports that a lock might be lost, " + "try to recover it risking data corruption."); + +module_param_named(delay_retrans, nfs_delay_retrans, short, 0644); +MODULE_PARM_DESC(delay_retrans, + "Unless negative, specifies the number of times the NFSv4 " + "client retries a request before returning an EAGAIN error, " + "after a reply of NFS4ERR_DELAY from the server."); +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c new file mode 100644 index 000000000..74a072896 --- /dev/null +++ b/fs/nfs/symlink.c @@ -0,0 +1,80 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/symlink.c + * + * Copyright (C) 1992 Rick Sladkey + * + * Optimization changes Copyright (C) 1994 Florian La Roche + * + * Jun 7 1999, cache symlink lookups in the page cache. -DaveM + * + * nfs symlink handling code + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" + +/* Symlink caching in the page cache is even more simplistic + * and straight-forward than readdir caching. + */ + +static int nfs_symlink_filler(struct file *file, struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + int error; + + error = NFS_PROTO(inode)->readlink(inode, &folio->page, 0, PAGE_SIZE); + folio_end_read(folio, error == 0); + return error; +} + +static const char *nfs_get_link(struct dentry *dentry, + struct inode *inode, + struct delayed_call *done) +{ + struct folio *folio; + void *err; + + if (!dentry) { + err = ERR_PTR(nfs_revalidate_mapping_rcu(inode)); + if (err) + return err; + folio = filemap_get_folio(inode->i_mapping, 0); + if (IS_ERR(folio)) + return ERR_PTR(-ECHILD); + if (!folio_test_uptodate(folio)) { + folio_put(folio); + return ERR_PTR(-ECHILD); + } + } else { + err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping)); + if (err) + return err; + folio = read_cache_folio(&inode->i_data, 0, nfs_symlink_filler, + NULL); + if (IS_ERR(folio)) + return ERR_CAST(folio); + } + set_delayed_call(done, page_put_link, folio); + return folio_address(folio); +} + +/* + * symlinks can't do much... + */ +const struct inode_operations nfs_symlink_inode_operations = { + .get_link = nfs_get_link, + .getattr = nfs_getattr, + .setattr = nfs_setattr, + .fileattr_get = nfs_fileattr_get, +}; diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c new file mode 100644 index 000000000..f579df0e8 --- /dev/null +++ b/fs/nfs/sysctl.c @@ -0,0 +1,46 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/sysctl.c + * + * Sysctl interface to NFS parameters + */ +#include +#include +#include +#include +#include +#include +#include + +static struct ctl_table_header *nfs_callback_sysctl_table; + +static const struct ctl_table nfs_cb_sysctls[] = { + { + .procname = "nfs_mountpoint_timeout", + .data = &nfs_mountpoint_expiry_timeout, + .maxlen = sizeof(nfs_mountpoint_expiry_timeout), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .procname = "nfs_congestion_kb", + .data = &nfs_congestion_kb, + .maxlen = sizeof(nfs_congestion_kb), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +}; + +int nfs_register_sysctl(void) +{ + nfs_callback_sysctl_table = register_sysctl("fs/nfs", nfs_cb_sysctls); + if (nfs_callback_sysctl_table == NULL) + return -ENOMEM; + return 0; +} + +void nfs_unregister_sysctl(void) +{ + unregister_sysctl_table(nfs_callback_sysctl_table); + nfs_callback_sysctl_table = NULL; +} diff --git a/fs/nfs/sysfs.c b/fs/nfs/sysfs.c new file mode 100644 index 000000000..3a197252a --- /dev/null +++ b/fs/nfs/sysfs.c @@ -0,0 +1,474 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2019 Hammerspace Inc + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "nfs4_fs.h" +#include "netns.h" +#include "sysfs.h" + +static struct kset *nfs_kset; + +static void nfs_kset_release(struct kobject *kobj) +{ + struct kset *kset = container_of(kobj, struct kset, kobj); + kfree(kset); +} + +static const struct kobj_ns_type_operations *nfs_netns_object_child_ns_type( + const struct kobject *kobj) +{ + return &net_ns_type_operations; +} + +static struct kobj_type nfs_kset_type = { + .release = nfs_kset_release, + .sysfs_ops = &kobj_sysfs_ops, + .child_ns_type = nfs_netns_object_child_ns_type, +}; + +int nfs_sysfs_init(void) +{ + int ret; + + nfs_kset = kzalloc_obj(*nfs_kset); + if (!nfs_kset) + return -ENOMEM; + + ret = kobject_set_name(&nfs_kset->kobj, "nfs"); + if (ret) { + kfree(nfs_kset); + return ret; + } + + nfs_kset->kobj.parent = fs_kobj; + nfs_kset->kobj.ktype = &nfs_kset_type; + nfs_kset->kobj.kset = NULL; + + ret = kset_register(nfs_kset); + if (ret) { + kfree(nfs_kset); + return ret; + } + + return 0; +} + +void nfs_sysfs_exit(void) +{ + kset_unregister(nfs_kset); +} + +static ssize_t nfs_netns_identifier_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + ssize_t ret; + + rcu_read_lock(); + ret = sysfs_emit(buf, "%s\n", rcu_dereference(c->identifier)); + rcu_read_unlock(); + return ret; +} + +/* Strip trailing '\n' */ +static size_t nfs_string_strip(const char *c, size_t len) +{ + while (len > 0 && c[len-1] == '\n') + --len; + return len; +} + +static ssize_t nfs_netns_identifier_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + const char *old; + char *p; + size_t len; + + len = nfs_string_strip(buf, min_t(size_t, count, CONTAINER_ID_MAXLEN)); + if (!len) + return 0; + p = kmemdup_nul(buf, len, GFP_KERNEL); + if (!p) + return -ENOMEM; + old = rcu_dereference_protected(xchg(&c->identifier, (char __rcu *)p), 1); + if (old) { + synchronize_rcu(); + kfree(old); + } + return count; +} + +static void nfs_netns_client_release(struct kobject *kobj) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + + kfree(rcu_dereference_raw(c->identifier)); +} + +static const struct ns_common *nfs_netns_client_namespace(const struct kobject *kobj) +{ + return to_ns_common(container_of(kobj, struct nfs_netns_client, + kobject)->net); +} + +static struct kobj_attribute nfs_netns_client_id = __ATTR(identifier, + 0644, nfs_netns_identifier_show, nfs_netns_identifier_store); + +static struct attribute *nfs_netns_client_attrs[] = { + &nfs_netns_client_id.attr, + NULL, +}; +ATTRIBUTE_GROUPS(nfs_netns_client); + +static struct kobj_type nfs_netns_client_type = { + .release = nfs_netns_client_release, + .default_groups = nfs_netns_client_groups, + .sysfs_ops = &kobj_sysfs_ops, + .namespace = nfs_netns_client_namespace, +}; + +static void nfs_netns_object_release(struct kobject *kobj) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + nfs_net_kobj); + kfree(c); +} + +static const struct ns_common *nfs_netns_namespace(const struct kobject *kobj) +{ + return to_ns_common(container_of(kobj, struct nfs_netns_client, + nfs_net_kobj)->net); +} + +static struct kobj_type nfs_netns_object_type = { + .release = nfs_netns_object_release, + .sysfs_ops = &kobj_sysfs_ops, + .namespace = nfs_netns_namespace, +}; + +static struct nfs_netns_client *nfs_netns_client_alloc(struct kobject *parent, + struct net *net) +{ + struct nfs_netns_client *p; + + p = kzalloc_obj(*p); + if (p) { + p->net = net; + p->kobject.kset = nfs_kset; + p->nfs_net_kobj.kset = nfs_kset; + + if (kobject_init_and_add(&p->nfs_net_kobj, &nfs_netns_object_type, + parent, "net") != 0) { + kobject_put(&p->nfs_net_kobj); + return NULL; + } + + if (kobject_init_and_add(&p->kobject, &nfs_netns_client_type, + &p->nfs_net_kobj, "nfs_client") == 0) + return p; + + kobject_put(&p->kobject); + kobject_put(&p->nfs_net_kobj); + } + return NULL; +} + +void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net) +{ + struct nfs_netns_client *clp; + + clp = nfs_netns_client_alloc(&nfs_kset->kobj, net); + if (clp) { + netns->nfs_client = clp; + kobject_uevent(&clp->kobject, KOBJ_ADD); + } +} + +void nfs_netns_sysfs_destroy(struct nfs_net *netns) +{ + struct nfs_netns_client *clp = netns->nfs_client; + + if (clp) { + kobject_uevent(&clp->kobject, KOBJ_REMOVE); + kobject_del(&clp->kobject); + kobject_put(&clp->kobject); + kobject_del(&clp->nfs_net_kobj); + kobject_put(&clp->nfs_net_kobj); + netns->nfs_client = NULL; + } +} + +static bool shutdown_match_client(const struct rpc_task *task, const void *data) +{ + return true; +} + +static void shutdown_client(struct rpc_clnt *clnt) +{ + clnt->cl_shutdown = 1; + rpc_cancel_tasks(clnt, -EIO, shutdown_match_client, NULL); +} + +/* + * Shut down the nfs_client only once all the superblocks + * have been shut down. + */ +static void shutdown_nfs_client(struct nfs_client *clp) +{ + struct nfs_server *server; + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (!(server->flags & NFS_MOUNT_SHUTDOWN)) { + rcu_read_unlock(); + return; + } + } + rcu_read_unlock(); + nfs_mark_client_ready(clp, -EIO); + shutdown_client(clp->cl_rpcclient); +} + +static ssize_t +shutdown_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct nfs_server *server = container_of(kobj, struct nfs_server, kobj); + bool shutdown = server->flags & NFS_MOUNT_SHUTDOWN; + return sysfs_emit(buf, "%d\n", shutdown); +} + +static ssize_t +shutdown_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct nfs_server *server; + int ret, val; + + server = container_of(kobj, struct nfs_server, kobj); + + ret = kstrtoint(buf, 0, &val); + if (ret) + return ret; + + if (val != 1) + return -EINVAL; + + /* already shut down? */ + if (server->flags & NFS_MOUNT_SHUTDOWN) + goto out; + + server->flags |= NFS_MOUNT_SHUTDOWN; + shutdown_client(server->client); + + if (!IS_ERR(server->client_acl)) + shutdown_client(server->client_acl); + + if (server->nlm_host) + nlmclnt_shutdown_rpc_clnt(server->nlm_host); +out: + shutdown_nfs_client(server->nfs_client); + return count; +} + +static struct kobj_attribute nfs_sysfs_attr_shutdown = __ATTR_RW(shutdown); + +#if IS_ENABLED(CONFIG_NFS_V4) +static ssize_t +implid_domain_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct nfs_server *server = container_of(kobj, struct nfs_server, kobj); + struct nfs41_impl_id *impl_id = server->nfs_client->cl_implid; + + if (!impl_id || strlen(impl_id->domain) == 0) + return 0; //sysfs_emit(buf, ""); + return sysfs_emit(buf, "%s\n", impl_id->domain); +} + +static struct kobj_attribute nfs_sysfs_attr_implid_domain = __ATTR_RO(implid_domain); + + +static ssize_t +implid_name_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct nfs_server *server = container_of(kobj, struct nfs_server, kobj); + struct nfs41_impl_id *impl_id = server->nfs_client->cl_implid; + + if (!impl_id || strlen(impl_id->name) == 0) + return 0; //sysfs_emit(buf, ""); + return sysfs_emit(buf, "%s\n", impl_id->name); +} + +static struct kobj_attribute nfs_sysfs_attr_implid_name = __ATTR_RO(implid_name); + +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ + +#define RPC_CLIENT_NAME_SIZE 64 + +void nfs_sysfs_link_rpc_client(struct nfs_server *server, + struct rpc_clnt *clnt, const char *uniq) +{ + char name[RPC_CLIENT_NAME_SIZE]; + int ret; + + strscpy(name, clnt->cl_program->name, sizeof(name)); + strncat(name, uniq ? uniq : "", sizeof(name) - strlen(name) - 1); + strncat(name, "_client", sizeof(name) - strlen(name) - 1); + + ret = sysfs_create_link_nowarn(&server->kobj, + &clnt->cl_sysfs->kobject, name); + if (ret < 0) + pr_warn("NFS: can't create link to %s in sysfs (%d)\n", + name, ret); +} +EXPORT_SYMBOL_GPL(nfs_sysfs_link_rpc_client); + +static void nfs_sysfs_sb_release(struct kobject *kobj) +{ + /* no-op: why? see lib/kobject.c kobject_cleanup() */ +} + +static const struct ns_common *nfs_netns_server_namespace(const struct kobject *kobj) +{ + return to_ns_common(container_of(kobj, struct nfs_server, + kobj)->nfs_client->cl_net); +} + +static struct kobj_type nfs_sb_ktype = { + .release = nfs_sysfs_sb_release, + .sysfs_ops = &kobj_sysfs_ops, + .namespace = nfs_netns_server_namespace, + .child_ns_type = nfs_netns_object_child_ns_type, +}; + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_sysfs_add_nfsv41_server(struct nfs_server *server) +{ + int ret; + + if (!server->nfs_client->cl_implid) + return; + + ret = sysfs_create_file_ns(&server->kobj, &nfs_sysfs_attr_implid_domain.attr, + nfs_netns_server_namespace(&server->kobj)); + if (ret < 0) + pr_warn("NFS: sysfs_create_file_ns for server-%d failed (%d)\n", + server->s_sysfs_id, ret); + + ret = sysfs_create_file_ns(&server->kobj, &nfs_sysfs_attr_implid_name.attr, + nfs_netns_server_namespace(&server->kobj)); + if (ret < 0) + pr_warn("NFS: sysfs_create_file_ns for server-%d failed (%d)\n", + server->s_sysfs_id, ret); +} +#else /* CONFIG_NFS_V4 */ +static inline void nfs_sysfs_add_nfsv41_server(struct nfs_server *server) +{ +} +#endif /* CONFIG_NFS_V4 */ + +#if IS_ENABLED(CONFIG_NFS_LOCALIO) + +static ssize_t +localio_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct nfs_server *server = container_of(kobj, struct nfs_server, kobj); + bool localio = nfs_server_is_local(server->nfs_client); + return sysfs_emit(buf, "%d\n", localio); +} + +static struct kobj_attribute nfs_sysfs_attr_localio = __ATTR_RO(localio); + +static void nfs_sysfs_add_nfs_localio_server(struct nfs_server *server) +{ + int ret = sysfs_create_file_ns(&server->kobj, &nfs_sysfs_attr_localio.attr, + nfs_netns_server_namespace(&server->kobj)); + if (ret < 0) + pr_warn("NFS: sysfs_create_file_ns for server-%d failed (%d)\n", + server->s_sysfs_id, ret); +} +#else +static inline void nfs_sysfs_add_nfs_localio_server(struct nfs_server *server) +{ +} +#endif /* IS_ENABLED(CONFIG_NFS_LOCALIO) */ + +void nfs_sysfs_add_server(struct nfs_server *server) +{ + int ret; + + ret = kobject_init_and_add(&server->kobj, &nfs_sb_ktype, + &nfs_kset->kobj, "server-%d", server->s_sysfs_id); + if (ret < 0) { + pr_warn("NFS: nfs sysfs add server-%d failed (%d)\n", + server->s_sysfs_id, ret); + return; + } + ret = sysfs_create_file_ns(&server->kobj, &nfs_sysfs_attr_shutdown.attr, + nfs_netns_server_namespace(&server->kobj)); + if (ret < 0) + pr_warn("NFS: sysfs_create_file_ns for server-%d failed (%d)\n", + server->s_sysfs_id, ret); + + nfs_sysfs_add_nfsv41_server(server); + nfs_sysfs_add_nfs_localio_server(server); +} +EXPORT_SYMBOL_GPL(nfs_sysfs_add_server); + +void nfs_sysfs_move_server_to_sb(struct super_block *s) +{ + struct nfs_server *server = s->s_fs_info; + int ret; + + ret = kobject_rename(&server->kobj, s->s_id); + if (ret < 0) + pr_warn("NFS: rename sysfs %s failed (%d)\n", + server->kobj.name, ret); +} + +void nfs_sysfs_move_sb_to_server(struct nfs_server *server) +{ + const char *s; + int ret = -ENOMEM; + + s = kasprintf(GFP_KERNEL, "server-%d", server->s_sysfs_id); + if (s) { + ret = kobject_rename(&server->kobj, s); + kfree(s); + } + if (ret < 0) + pr_warn("NFS: rename sysfs %s failed (%d)\n", + server->kobj.name, ret); +} + +/* unlink, not dec-ref */ +void nfs_sysfs_remove_server(struct nfs_server *server) +{ + kobject_del(&server->kobj); +} diff --git a/fs/nfs/sysfs.h b/fs/nfs/sysfs.h new file mode 100644 index 000000000..c5d1990ca --- /dev/null +++ b/fs/nfs/sysfs.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2019 Hammerspace Inc + */ + +#ifndef __NFS_SYSFS_H +#define __NFS_SYSFS_H + +#define CONTAINER_ID_MAXLEN (64) + +struct nfs_netns_client { + struct kobject kobject; + struct kobject nfs_net_kobj; + struct net *net; + const char __rcu *identifier; +}; + +extern struct kobject *nfs_net_kobj; + +extern int nfs_sysfs_init(void); +extern void nfs_sysfs_exit(void); + +void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net); +void nfs_netns_sysfs_destroy(struct nfs_net *netns); + +void nfs_sysfs_link_rpc_client(struct nfs_server *server, + struct rpc_clnt *clnt, const char *sysfs_prefix); +void nfs_sysfs_add_server(struct nfs_server *s); +void nfs_sysfs_move_server_to_sb(struct super_block *s); +void nfs_sysfs_move_sb_to_server(struct nfs_server *s); +void nfs_sysfs_remove_server(struct nfs_server *s); + +#endif diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c new file mode 100644 index 000000000..b57cfaa4d --- /dev/null +++ b/fs/nfs/unlink.c @@ -0,0 +1,532 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/nfs/unlink.c + * + * nfs sillydelete handling + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "internal.h" +#include "nfs4_fs.h" +#include "iostat.h" +#include "delegation.h" + +#include "nfstrace.h" + +/** + * nfs_free_unlinkdata - release data from a sillydelete operation. + * @data: pointer to unlink structure. + */ +static void +nfs_free_unlinkdata(struct nfs_unlinkdata *data) +{ + put_cred(data->cred); + kfree(data->args.name.name); + kfree(data); +} + +/** + * nfs_async_unlink_done - Sillydelete post-processing + * @task: rpc_task of the sillydelete + * @calldata: pointer to nfs_unlinkdata + * + * Do the directory attribute update. + */ +static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) +{ + struct nfs_unlinkdata *data = calldata; + struct inode *dir = d_inode(data->dentry->d_parent); + + trace_nfs_sillyrename_unlink(data, task->tk_status); + if (!NFS_PROTO(dir)->unlink_done(task, dir)) + rpc_restart_call_prepare(task); +} + +/** + * nfs_async_unlink_release - Release the sillydelete data. + * @calldata: struct nfs_unlinkdata to release + * + * We need to call nfs_put_unlinkdata as a 'tk_release' task since the + * rpc_task would be freed too. + */ +static void nfs_async_unlink_release(void *calldata) +{ + struct nfs_unlinkdata *data = calldata; + struct dentry *dentry = data->dentry; + struct super_block *sb = dentry->d_sb; + + up_read_non_owner(&NFS_I(d_inode(dentry->d_parent))->rmdir_sem); + d_lookup_done(dentry); + nfs_free_unlinkdata(data); + dput(dentry); + nfs_sb_deactive(sb); +} + +static void nfs_unlink_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_unlinkdata *data = calldata; + struct inode *dir = d_inode(data->dentry->d_parent); + NFS_PROTO(dir)->unlink_rpc_prepare(task, data); +} + +static const struct rpc_call_ops nfs_unlink_ops = { + .rpc_call_done = nfs_async_unlink_done, + .rpc_release = nfs_async_unlink_release, + .rpc_call_prepare = nfs_unlink_prepare, +}; + +static void nfs_do_call_unlink(struct inode *inode, struct nfs_unlinkdata *data) +{ + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_message = &msg, + .callback_ops = &nfs_unlink_ops, + .callback_data = data, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + struct rpc_task *task; + struct inode *dir = d_inode(data->dentry->d_parent); + + if (nfs_server_capable(inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + nfs_sb_active(dir->i_sb); + data->args.fh = NFS_FH(dir); + nfs_fattr_init(data->res.dir_attr); + + NFS_PROTO(dir)->unlink_setup(&msg, data->dentry, inode); + + task_setup_data.rpc_client = NFS_CLIENT(dir); + task = rpc_run_task(&task_setup_data); + if (!IS_ERR(task)) + rpc_put_task_async(task); +} + +static int nfs_call_unlink(struct dentry *dentry, struct inode *inode, struct nfs_unlinkdata *data) +{ + struct inode *dir = d_inode(dentry->d_parent); + struct dentry *alias; + + down_read_non_owner(&NFS_I(dir)->rmdir_sem); + alias = d_alloc_parallel(dentry->d_parent, &data->args.name); + if (IS_ERR(alias)) { + up_read_non_owner(&NFS_I(dir)->rmdir_sem); + return 0; + } + if (!d_in_lookup(alias)) { + int ret; + void *devname_garbage = NULL; + + /* + * Hey, we raced with lookup... See if we need to transfer + * the sillyrename information to the aliased dentry. + */ + spin_lock(&alias->d_lock); + if (d_really_is_positive(alias) && + !nfs_compare_fh(NFS_FH(inode), NFS_FH(d_inode(alias))) && + !(alias->d_flags & DCACHE_NFSFS_RENAMED)) { + devname_garbage = alias->d_fsdata; + alias->d_fsdata = data; + alias->d_flags |= DCACHE_NFSFS_RENAMED; + ret = 1; + } else + ret = 0; + spin_unlock(&alias->d_lock); + dput(alias); + up_read_non_owner(&NFS_I(dir)->rmdir_sem); + /* + * If we'd displaced old cached devname, free it. At that + * point dentry is definitely not a root, so we won't need + * that anymore. + */ + kfree(devname_garbage); + return ret; + } + data->dentry = alias; + nfs_do_call_unlink(inode, data); + return 1; +} + +/** + * nfs_async_unlink - asynchronous unlinking of a file + * @dentry: parent directory of dentry + * @name: name of dentry to unlink + */ +static int +nfs_async_unlink(struct dentry *dentry, const struct qstr *name) +{ + struct nfs_unlinkdata *data; + int status = -ENOMEM; + void *devname_garbage = NULL; + + data = kzalloc_obj(*data); + if (data == NULL) + goto out; + data->args.name.name = kstrdup(name->name, GFP_KERNEL); + if (!data->args.name.name) + goto out_free; + data->args.name.len = name->len; + + data->cred = get_current_cred(); + data->res.dir_attr = &data->dir_attr; + + status = -EBUSY; + spin_lock(&dentry->d_lock); + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + goto out_unlock; + dentry->d_flags |= DCACHE_NFSFS_RENAMED; + devname_garbage = dentry->d_fsdata; + dentry->d_fsdata = data; + spin_unlock(&dentry->d_lock); + /* + * If we'd displaced old cached devname, free it. At that + * point dentry is definitely not a root, so we won't need + * that anymore. + */ + kfree(devname_garbage); + return 0; +out_unlock: + spin_unlock(&dentry->d_lock); + put_cred(data->cred); + kfree(data->args.name.name); +out_free: + kfree(data); +out: + return status; +} + +/** + * nfs_complete_unlink - Initialize completion of the sillydelete + * @dentry: dentry to delete + * @inode: inode + * + * Since we're most likely to be called by dentry_iput(), we + * only use the dentry to find the sillydelete. We then copy the name + * into the qstr. + */ +void +nfs_complete_unlink(struct dentry *dentry, struct inode *inode) +{ + struct nfs_unlinkdata *data; + + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_NFSFS_RENAMED; + data = dentry->d_fsdata; + dentry->d_fsdata = NULL; + spin_unlock(&dentry->d_lock); + + NFS_PROTO(inode)->return_delegation(inode); + + if (NFS_STALE(inode) || !nfs_call_unlink(dentry, inode, data)) + nfs_free_unlinkdata(data); +} + +/* Cancel a queued async unlink. Called when a sillyrename run fails. */ +static void +nfs_cancel_async_unlink(struct dentry *dentry) +{ + spin_lock(&dentry->d_lock); + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { + struct nfs_unlinkdata *data = dentry->d_fsdata; + + dentry->d_flags &= ~DCACHE_NFSFS_RENAMED; + dentry->d_fsdata = NULL; + spin_unlock(&dentry->d_lock); + nfs_free_unlinkdata(data); + return; + } + spin_unlock(&dentry->d_lock); +} + +/** + * nfs_async_rename_done - Sillyrename post-processing + * @task: rpc_task of the sillyrename + * @calldata: nfs_renamedata for the sillyrename + * + * Do the directory attribute updates and the d_move + */ +static void nfs_async_rename_done(struct rpc_task *task, void *calldata) +{ + struct nfs_renamedata *data = calldata; + struct inode *old_dir = data->old_dir; + struct inode *new_dir = data->new_dir; + struct dentry *old_dentry = data->old_dentry; + + trace_nfs_async_rename_done(old_dir, old_dentry, + new_dir, data->new_dentry, task->tk_status); + if (!NFS_PROTO(old_dir)->rename_done(task, old_dir, new_dir)) { + rpc_restart_call_prepare(task); + return; + } + + if (data->complete) + data->complete(task, data); +} + +/** + * nfs_async_rename_release - Release the sillyrename data. + * @calldata: the struct nfs_renamedata to be released + */ +static void nfs_async_rename_release(void *calldata) +{ + struct nfs_renamedata *data = calldata; + struct super_block *sb = data->old_dir->i_sb; + + if (d_really_is_positive(data->old_dentry)) + nfs_mark_for_revalidate(d_inode(data->old_dentry)); + + /* The result of the rename is unknown. Play it safe by + * forcing a new lookup */ + if (data->cancelled) { + spin_lock(&data->old_dir->i_lock); + nfs_force_lookup_revalidate(data->old_dir); + spin_unlock(&data->old_dir->i_lock); + if (data->new_dir != data->old_dir) { + spin_lock(&data->new_dir->i_lock); + nfs_force_lookup_revalidate(data->new_dir); + spin_unlock(&data->new_dir->i_lock); + } + } + + dput(data->old_dentry); + dput(data->new_dentry); + iput(data->old_dir); + iput(data->new_dir); + nfs_sb_deactive(sb); + put_cred(data->cred); + kfree(data); +} + +static void nfs_rename_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_renamedata *data = calldata; + NFS_PROTO(data->old_dir)->rename_rpc_prepare(task, data); +} + +static const struct rpc_call_ops nfs_rename_ops = { + .rpc_call_done = nfs_async_rename_done, + .rpc_release = nfs_async_rename_release, + .rpc_call_prepare = nfs_rename_prepare, +}; + +/** + * nfs_async_rename - perform an asynchronous rename operation + * @old_dir: directory that currently holds the dentry to be renamed + * @new_dir: target directory for the rename + * @old_dentry: original dentry to be renamed + * @new_dentry: dentry to which the old_dentry should be renamed + * @complete: Function to run on successful completion + * + * It's expected that valid references to the dentries and inodes are held + */ +struct rpc_task * +nfs_async_rename(struct inode *old_dir, struct inode *new_dir, + struct dentry *old_dentry, struct dentry *new_dentry, + void (*complete)(struct rpc_task *, struct nfs_renamedata *)) +{ + struct nfs_renamedata *data; + struct rpc_message msg = { }; + struct rpc_task_setup task_setup_data = { + .rpc_message = &msg, + .callback_ops = &nfs_rename_ops, + .workqueue = nfsiod_workqueue, + .rpc_client = NFS_CLIENT(old_dir), + .flags = RPC_TASK_ASYNC | RPC_TASK_CRED_NOREF, + }; + + if (nfs_server_capable(old_dir, NFS_CAP_MOVEABLE) && + nfs_server_capable(new_dir, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + data = kzalloc_obj(*data); + if (data == NULL) + return ERR_PTR(-ENOMEM); + task_setup_data.task = &data->task; + task_setup_data.callback_data = data; + + data->cred = get_current_cred(); + + msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + msg.rpc_cred = data->cred; + + /* set up nfs_renamedata */ + data->old_dir = old_dir; + ihold(old_dir); + data->new_dir = new_dir; + ihold(new_dir); + data->old_dentry = dget(old_dentry); + data->new_dentry = dget(new_dentry); + nfs_fattr_init(&data->old_fattr); + nfs_fattr_init(&data->new_fattr); + data->complete = complete; + + /* set up nfs_renameargs */ + data->args.old_dir = NFS_FH(old_dir); + data->args.old_name = &old_dentry->d_name; + data->args.new_dir = NFS_FH(new_dir); + data->args.new_name = &new_dentry->d_name; + + /* set up nfs_renameres */ + data->res.old_fattr = &data->old_fattr; + data->res.new_fattr = &data->new_fattr; + + nfs_sb_active(old_dir->i_sb); + + NFS_PROTO(data->old_dir)->rename_setup(&msg, old_dentry, new_dentry, + old_dir == new_dir ? old_dir : NULL); + + return rpc_run_task(&task_setup_data); +} + +/* + * Perform tasks needed when a sillyrename is done such as cancelling the + * queued async unlink if it failed. + */ +static void +nfs_complete_sillyrename(struct rpc_task *task, struct nfs_renamedata *data) +{ + struct dentry *dentry = data->old_dentry; + + if (task->tk_status != 0) { + nfs_cancel_async_unlink(dentry); + return; + } +} + +#define SILLYNAME_PREFIX ".nfs" +#define SILLYNAME_PREFIX_LEN ((unsigned)sizeof(SILLYNAME_PREFIX) - 1) +#define SILLYNAME_FILEID_LEN ((unsigned)sizeof(u64) << 1) +#define SILLYNAME_COUNTER_LEN ((unsigned)sizeof(unsigned int) << 1) +#define SILLYNAME_LEN (SILLYNAME_PREFIX_LEN + \ + SILLYNAME_FILEID_LEN + \ + SILLYNAME_COUNTER_LEN) + +/** + * nfs_sillyrename - Perform a silly-rename of a dentry + * @dir: inode of directory that contains dentry + * @dentry: dentry to be sillyrenamed + * + * NFSv2/3 is stateless and the server doesn't know when the client is + * holding a file open. To prevent application problems when a file is + * unlinked while it's still open, the client performs a "silly-rename". + * That is, it renames the file to a hidden file in the same directory, + * and only performs the unlink once the last reference to it is put. + * + * The final cleanup is done during dentry_iput. + * + * (Note: NFSv4 is stateful, and has opens, so in theory an NFSv4 server + * could take responsibility for keeping open files referenced. The server + * would also need to ensure that opened-but-deleted files were kept over + * reboots. However, we may not assume a server does so. (RFC 5661 + * does provide an OPEN4_RESULT_PRESERVE_UNLINKED flag that a server can + * use to advertise that it does this; some day we may take advantage of + * it.)) + */ +int +nfs_sillyrename(struct inode *dir, struct dentry *dentry) +{ + static unsigned int sillycounter; + unsigned char silly[SILLYNAME_LEN + 1]; + unsigned long long fileid; + struct dentry *sdentry; + struct inode *inode = d_inode(dentry); + struct rpc_task *task; + int error = -EBUSY; + + dfprintk(VFS, "NFS: silly-rename(%pd2, ct=%d)\n", + dentry, d_count(dentry)); + nfs_inc_stats(dir, NFSIOS_SILLYRENAME); + + /* + * We don't allow a dentry to be silly-renamed twice. + */ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + goto out; + + fileid = d_inode(dentry)->i_ino; + + sdentry = NULL; + do { + dput(sdentry); + sillycounter++; + scnprintf(silly, sizeof(silly), + SILLYNAME_PREFIX "%0*llx%0*x", + SILLYNAME_FILEID_LEN, fileid, + SILLYNAME_COUNTER_LEN, sillycounter); + + dfprintk(VFS, "NFS: trying to rename %pd to %s\n", + dentry, silly); + + sdentry = lookup_noperm(&QSTR(silly), dentry->d_parent); + /* + * N.B. Better to return EBUSY here ... it could be + * dangerous to delete the file while it's in use. + */ + if (IS_ERR(sdentry)) + goto out; + } while (d_inode(sdentry) != NULL); /* need negative lookup */ + + ihold(inode); + + /* queue unlink first. Can't do this from rpc_release as it + * has to allocate memory + */ + error = nfs_async_unlink(dentry, &sdentry->d_name); + if (error) + goto out_dput; + + /* run the rename task, undo unlink if it fails */ + task = nfs_async_rename(dir, dir, dentry, sdentry, + nfs_complete_sillyrename); + if (IS_ERR(task)) { + error = -EBUSY; + nfs_cancel_async_unlink(dentry); + goto out_dput; + } + + /* wait for the RPC task to complete, unless a SIGKILL intervenes */ + error = rpc_wait_for_completion_task(task); + if (error == 0) + error = task->tk_status; + switch (error) { + case 0: + /* The rename succeeded */ + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + spin_lock(&inode->i_lock); + NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); + spin_unlock(&inode->i_lock); + d_move(dentry, sdentry); + break; + case -ERESTARTSYS: + /* The result of the rename is unknown. Play it safe by + * forcing a new lookup */ + d_drop(dentry); + d_drop(sdentry); + } + rpc_put_task(task); +out_dput: + iput(inode); + dput(sdentry); +out: + return error; +} diff --git a/fs/nfs/write.c b/fs/nfs/write.c new file mode 100644 index 000000000..d2b03ceae --- /dev/null +++ b/fs/nfs/write.c @@ -0,0 +1,2212 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/fs/nfs/write.c + * + * Write file data over NFS. + * + * Copyright (C) 1996, 1997, Olaf Kirch + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include "delegation.h" +#include "internal.h" +#include "iostat.h" +#include "nfs4_fs.h" +#include "fscache.h" +#include "pnfs.h" + +#include "nfstrace.h" + +#define NFSDBG_FACILITY NFSDBG_PAGECACHE + +#define MIN_POOL_WRITE (32) +#define MIN_POOL_COMMIT (4) + +struct nfs_io_completion { + void (*complete)(void *data); + void *data; + struct kref refcount; +}; + +/* + * Local function declarations + */ +static void nfs_redirty_request(struct nfs_page *req); +static const struct rpc_call_ops nfs_commit_ops; +static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops; +static const struct nfs_commit_completion_ops nfs_commit_completion_ops; +static const struct nfs_rw_ops nfs_rw_write_ops; +static void nfs_inode_remove_request(struct nfs_page *req); +static void nfs_clear_request_commit(struct nfs_commit_info *cinfo, + struct nfs_page *req); +static void nfs_init_cinfo_from_inode(struct nfs_commit_info *cinfo, + struct inode *inode); + +static struct kmem_cache *nfs_wdata_cachep; +static mempool_t *nfs_wdata_mempool; +static struct kmem_cache *nfs_cdata_cachep; +static mempool_t *nfs_commit_mempool; + +struct nfs_commit_data *nfs_commitdata_alloc(void) +{ + struct nfs_commit_data *p; + + p = kmem_cache_zalloc(nfs_cdata_cachep, nfs_io_gfp_mask()); + if (!p) { + p = mempool_alloc(nfs_commit_mempool, GFP_NOWAIT); + if (!p) + return NULL; + memset(p, 0, sizeof(*p)); + } + INIT_LIST_HEAD(&p->pages); + return p; +} +EXPORT_SYMBOL_GPL(nfs_commitdata_alloc); + +void nfs_commit_free(struct nfs_commit_data *p) +{ + mempool_free(p, nfs_commit_mempool); +} +EXPORT_SYMBOL_GPL(nfs_commit_free); + +static struct nfs_pgio_header *nfs_writehdr_alloc(void) +{ + struct nfs_pgio_header *p; + + p = kmem_cache_zalloc(nfs_wdata_cachep, nfs_io_gfp_mask()); + if (!p) { + p = mempool_alloc(nfs_wdata_mempool, GFP_NOWAIT); + if (!p) + return NULL; + memset(p, 0, sizeof(*p)); + } + p->rw_mode = FMODE_WRITE; + return p; +} + +static void nfs_writehdr_free(struct nfs_pgio_header *hdr) +{ + mempool_free(hdr, nfs_wdata_mempool); +} + +static struct nfs_io_completion *nfs_io_completion_alloc(gfp_t gfp_flags) +{ + return kmalloc_obj(struct nfs_io_completion, gfp_flags); +} + +static void nfs_io_completion_init(struct nfs_io_completion *ioc, + void (*complete)(void *), void *data) +{ + ioc->complete = complete; + ioc->data = data; + kref_init(&ioc->refcount); +} + +static void nfs_io_completion_release(struct kref *kref) +{ + struct nfs_io_completion *ioc = container_of(kref, + struct nfs_io_completion, refcount); + ioc->complete(ioc->data); + kfree(ioc); +} + +static void nfs_io_completion_get(struct nfs_io_completion *ioc) +{ + if (ioc != NULL) + kref_get(&ioc->refcount); +} + +static void nfs_io_completion_put(struct nfs_io_completion *ioc) +{ + if (ioc != NULL) + kref_put(&ioc->refcount, nfs_io_completion_release); +} + +static void +nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode) +{ + if (!test_and_set_bit(PG_INODE_REF, &req->wb_flags)) { + kref_get(&req->wb_kref); + atomic_long_inc(&NFS_I(inode)->nrequests); + } +} + +static void nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) +{ + if (test_and_clear_bit(PG_REMOVE, &req->wb_flags)) + nfs_page_set_inode_ref(req, inode); +} + +/** + * nfs_folio_find_head_request - find head request associated with a folio + * @folio: pointer to folio + * + * must be called while holding the inode lock. + * + * returns matching head request with reference held, or NULL if not found. + */ +static struct nfs_page *nfs_folio_find_head_request(struct folio *folio) +{ + struct address_space *mapping = folio->mapping; + struct nfs_page *req; + + if (!folio_test_private(folio)) + return NULL; + spin_lock(&mapping->i_private_lock); + req = folio->private; + if (req) { + WARN_ON_ONCE(req->wb_head != req); + kref_get(&req->wb_kref); + } + spin_unlock(&mapping->i_private_lock); + return req; +} + +/* Adjust the file length if we're writing beyond the end */ +static void nfs_grow_file(struct folio *folio, unsigned int offset, + unsigned int count) +{ + struct inode *inode = folio->mapping->host; + loff_t end, i_size; + pgoff_t end_index; + + spin_lock(&inode->i_lock); + i_size = i_size_read(inode); + end_index = ((i_size - 1) >> folio_shift(folio)) << folio_order(folio); + if (i_size > 0 && folio->index < end_index) + goto out; + end = folio_pos(folio) + (loff_t)offset + (loff_t)count; + if (i_size >= end) + goto out; + trace_nfs_size_grow(inode, end); + i_size_write(inode, end); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; + nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); +out: + /* Atomically update timestamps if they are delegated to us. */ + nfs_update_delegated_mtime_locked(inode); + spin_unlock(&inode->i_lock); + nfs_fscache_invalidate(inode, 0); +} + +/* A writeback failed: mark the page as bad, and invalidate the page cache */ +static void nfs_set_pageerror(struct address_space *mapping) +{ + struct inode *inode = mapping->host; + + nfs_zap_mapping(mapping->host, mapping); + /* Force file size revalidation */ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_SIZE); + spin_unlock(&inode->i_lock); +} + +static void nfs_mapping_set_error(struct folio *folio, int error) +{ + struct address_space *mapping = folio->mapping; + + filemap_set_wb_err(mapping, error); + if (mapping->host) + errseq_set(&mapping->host->i_sb->s_wb_err, + error == -ENOSPC ? -ENOSPC : -EIO); + nfs_set_pageerror(mapping); +} + +/* + * nfs_page_covers_folio + * @req: struct nfs_page + * + * Return true if the request covers the whole folio. + * Note that the caller should ensure all subrequests have been joined + */ +static bool nfs_page_group_covers_page(struct nfs_page *req) +{ + unsigned int len = nfs_folio_length(nfs_page_to_folio(req)); + + return req->wb_pgbase == 0 && req->wb_bytes == len; +} + +/* We can set the PG_uptodate flag if we see that a write request + * covers the full page. + */ +static void nfs_mark_uptodate(struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + + if (folio_test_uptodate(folio)) + return; + if (!nfs_page_group_covers_page(req)) + return; + folio_mark_uptodate(folio); +} + +static int wb_priority(struct writeback_control *wbc) +{ + int ret = 0; + + if (wbc->sync_mode == WB_SYNC_ALL) + ret = FLUSH_COND_STABLE; + return ret; +} + +/* + * NFS congestion control + */ + +int nfs_congestion_kb; + +#define NFS_CONGESTION_ON_THRESH (nfs_congestion_kb >> (PAGE_SHIFT-10)) +#define NFS_CONGESTION_OFF_THRESH \ + (NFS_CONGESTION_ON_THRESH - (NFS_CONGESTION_ON_THRESH >> 2)) + +static void nfs_folio_set_writeback(struct folio *folio) +{ + struct nfs_server *nfss = NFS_SERVER(folio->mapping->host); + + folio_start_writeback(folio); + if (atomic_long_inc_return(&nfss->writeback) > NFS_CONGESTION_ON_THRESH) + nfss->write_congested = 1; +} + +static void nfs_folio_end_writeback(struct folio *folio) +{ + struct nfs_server *nfss = NFS_SERVER(folio->mapping->host); + + folio_end_writeback_no_dropbehind(folio); + if (atomic_long_dec_return(&nfss->writeback) < + NFS_CONGESTION_OFF_THRESH) { + nfss->write_congested = 0; + wake_up_all(&nfss->write_congestion_wait); + } +} + +static void nfs_page_end_writeback(struct nfs_page *req) +{ + if (nfs_page_group_sync_on_bit(req, PG_WB_END)) { + nfs_unlock_request(req); + nfs_folio_end_writeback(nfs_page_to_folio(req)); + } else + nfs_unlock_request(req); +} + +/* + * nfs_destroy_unlinked_subrequests - destroy recently unlinked subrequests + * + * @destroy_list - request list (using wb_this_page) terminated by @old_head + * @old_head - the old head of the list + * + * All subrequests must be locked and removed from all lists, so at this point + * they are only "active" in this function, and possibly in nfs_wait_on_request + * with a reference held by some other context. + */ +static void +nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list, + struct nfs_page *old_head, + struct inode *inode) +{ + while (destroy_list) { + struct nfs_page *subreq = destroy_list; + + destroy_list = (subreq->wb_this_page == old_head) ? + NULL : subreq->wb_this_page; + + /* Note: lock subreq in order to change subreq->wb_head */ + nfs_page_set_headlock(subreq); + WARN_ON_ONCE(old_head != subreq->wb_head); + + /* make sure old group is not used */ + subreq->wb_this_page = subreq; + subreq->wb_head = subreq; + + clear_bit(PG_REMOVE, &subreq->wb_flags); + + /* Note: races with nfs_page_group_destroy() */ + if (!kref_read(&subreq->wb_kref)) { + /* Check if we raced with nfs_page_group_destroy() */ + if (test_and_clear_bit(PG_TEARDOWN, &subreq->wb_flags)) { + nfs_page_clear_headlock(subreq); + nfs_free_request(subreq); + } else + nfs_page_clear_headlock(subreq); + continue; + } + nfs_page_clear_headlock(subreq); + + nfs_release_request(old_head); + + if (test_and_clear_bit(PG_INODE_REF, &subreq->wb_flags)) { + nfs_release_request(subreq); + atomic_long_dec(&NFS_I(inode)->nrequests); + } + + /* subreq is now totally disconnected from page group or any + * write / commit lists. last chance to wake any waiters */ + nfs_unlock_and_release_request(subreq); + } +} + +/* + * nfs_join_page_group - destroy subrequests of the head req + * @head: the page used to lookup the "page group" of nfs_page structures + * @inode: Inode to which the request belongs. + * + * This function joins all sub requests to the head request by first + * locking all requests in the group, cancelling any pending operations + * and finally updating the head request to cover the whole range covered by + * the (former) group. All subrequests are removed from any write or commit + * lists, unlinked from the group and destroyed. + */ +void nfs_join_page_group(struct nfs_page *head, struct nfs_commit_info *cinfo, + struct inode *inode) +{ + struct nfs_page *subreq; + struct nfs_page *destroy_list = NULL; + unsigned int pgbase, off, bytes; + + pgbase = head->wb_pgbase; + bytes = head->wb_bytes; + off = head->wb_offset; + for (subreq = head->wb_this_page; subreq != head; + subreq = subreq->wb_this_page) { + /* Subrequests should always form a contiguous range */ + if (pgbase > subreq->wb_pgbase) { + off -= pgbase - subreq->wb_pgbase; + bytes += pgbase - subreq->wb_pgbase; + pgbase = subreq->wb_pgbase; + } + bytes = max(subreq->wb_pgbase + subreq->wb_bytes + - pgbase, bytes); + } + + /* Set the head request's range to cover the former page group */ + head->wb_pgbase = pgbase; + head->wb_bytes = bytes; + head->wb_offset = off; + + /* Now that all requests are locked, make sure they aren't on any list. + * Commit list removal accounting is done after locks are dropped */ + subreq = head; + do { + nfs_clear_request_commit(cinfo, subreq); + subreq = subreq->wb_this_page; + } while (subreq != head); + + /* unlink subrequests from head, destroy them later */ + if (head->wb_this_page != head) { + /* destroy list will be terminated by head */ + destroy_list = head->wb_this_page; + head->wb_this_page = head; + } + + nfs_destroy_unlinked_subrequests(destroy_list, head, inode); +} + +/** + * nfs_wait_on_request - Wait for a request to complete. + * @req: request to wait upon. + * + * Interruptible by fatal signals only. + * The user is responsible for holding a count on the request. + */ +static int nfs_wait_on_request(struct nfs_page *req) +{ + if (!test_bit(PG_BUSY, &req->wb_flags)) + return 0; + set_bit(PG_CONTENDED2, &req->wb_flags); + smp_mb__after_atomic(); + return wait_on_bit_io(&req->wb_flags, PG_BUSY, + TASK_UNINTERRUPTIBLE); +} + +/* + * nfs_unroll_locks - unlock all newly locked reqs and wait on @req + * @head: head request of page group, must be holding head lock + * @req: request that couldn't lock and needs to wait on the req bit lock + * + * This is a helper function for nfs_lock_and_join_requests + * returns 0 on success, < 0 on error. + */ +static void +nfs_unroll_locks(struct nfs_page *head, struct nfs_page *req) +{ + struct nfs_page *tmp; + + /* relinquish all the locks successfully grabbed this run */ + for (tmp = head->wb_this_page ; tmp != req; tmp = tmp->wb_this_page) { + if (!kref_read(&tmp->wb_kref)) + continue; + nfs_unlock_and_release_request(tmp); + } +} + +/* + * nfs_page_group_lock_subreq - try to lock a subrequest + * @head: head request of page group + * @subreq: request to lock + * + * This is a helper function for nfs_lock_and_join_requests which + * must be called with the head request and page group both locked. + * On error, it returns with the page group unlocked. + */ +static int +nfs_page_group_lock_subreq(struct nfs_page *head, struct nfs_page *subreq) +{ + int ret; + + if (!kref_get_unless_zero(&subreq->wb_kref)) + return 0; + while (!nfs_lock_request(subreq)) { + nfs_page_group_unlock(head); + ret = nfs_wait_on_request(subreq); + if (!ret) + ret = nfs_page_group_lock(head); + if (ret < 0) { + nfs_unroll_locks(head, subreq); + nfs_release_request(subreq); + return ret; + } + } + return 0; +} + +/* + * nfs_lock_and_join_requests - join all subreqs to the head req + * @folio: the folio used to lookup the "page group" of nfs_page structures + * + * This function joins all sub requests to the head request by first + * locking all requests in the group, cancelling any pending operations + * and finally updating the head request to cover the whole range covered by + * the (former) group. All subrequests are removed from any write or commit + * lists, unlinked from the group and destroyed. + * + * Returns a locked, referenced pointer to the head request - which after + * this call is guaranteed to be the only request associated with the page. + * Returns NULL if no requests are found for @folio, or a ERR_PTR if an + * error was encountered. + */ +static struct nfs_page *nfs_lock_and_join_requests(struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + struct nfs_page *head, *subreq; + struct nfs_commit_info cinfo; + int ret; + + /* + * A reference is taken only on the head request which acts as a + * reference to the whole page group - the group will not be destroyed + * until the head reference is released. + */ +retry: + head = nfs_folio_find_head_request(folio); + if (!head) + return NULL; + + while (!nfs_lock_request(head)) { + ret = nfs_wait_on_request(head); + if (ret < 0) { + nfs_release_request(head); + return ERR_PTR(ret); + } + } + + ret = nfs_page_group_lock(head); + if (ret < 0) + goto out_unlock; + + /* Ensure that nobody removed the request before we locked it */ + if (head != folio->private) { + nfs_page_group_unlock(head); + nfs_unlock_and_release_request(head); + goto retry; + } + + nfs_cancel_remove_inode(head, inode); + + /* lock each request in the page group */ + for (subreq = head->wb_this_page; + subreq != head; + subreq = subreq->wb_this_page) { + ret = nfs_page_group_lock_subreq(head, subreq); + if (ret < 0) + goto out_unlock; + } + + nfs_page_group_unlock(head); + + nfs_init_cinfo_from_inode(&cinfo, inode); + nfs_join_page_group(head, &cinfo, inode); + return head; + +out_unlock: + nfs_unlock_and_release_request(head); + return ERR_PTR(ret); +} + +static void nfs_write_error(struct nfs_page *req, int error) +{ + trace_nfs_write_error(nfs_page_to_inode(req), req, error); + nfs_mapping_set_error(nfs_page_to_folio(req), error); + nfs_inode_remove_request(req); + nfs_page_end_writeback(req); + nfs_release_request(req); +} + +/* + * Find an associated nfs write request, and prepare to flush it out + * May return an error if the user signalled nfs_wait_on_request(). + */ +static int nfs_do_writepage(struct folio *folio, struct writeback_control *wbc, + struct nfs_pageio_descriptor *pgio) +{ + struct nfs_page *req; + int ret; + + nfs_pageio_cond_complete(pgio, folio->index); + + req = nfs_lock_and_join_requests(folio); + if (!req) + return 0; + if (IS_ERR(req)) + return PTR_ERR(req); + + trace_nfs_do_writepage(req); + nfs_folio_set_writeback(folio); + WARN_ON_ONCE(test_bit(PG_CLEAN, &req->wb_flags)); + + /* If there is a fatal error that covers this write, just exit */ + ret = pgio->pg_error; + if (nfs_error_is_fatal_on_server(ret)) + goto out_launder; + + if (!nfs_pageio_add_request(pgio, req)) { + ret = pgio->pg_error; + /* + * Remove the problematic req upon fatal errors on the server + */ + if (nfs_error_is_fatal_on_server(ret)) + goto out_launder; + folio_redirty_for_writepage(wbc, folio); + nfs_redirty_request(req); + pgio->pg_error = 0; + return ret; + } + + nfs_add_stats(folio->mapping->host, NFSIOS_WRITEPAGES, 1); + return 0; + +out_launder: + nfs_write_error(req, ret); + return 0; +} + +/* + * Write an mmapped page to the server. + */ +static int nfs_writepage_locked(struct folio *folio, + struct writeback_control *wbc) +{ + struct nfs_pageio_descriptor pgio; + struct inode *inode = folio->mapping->host; + int err; + + nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); + nfs_pageio_init_write(&pgio, inode, 0, false, + &nfs_async_write_completion_ops); + err = nfs_do_writepage(folio, wbc, &pgio); + pgio.pg_error = 0; + nfs_pageio_complete(&pgio); + return err; +} + +static void nfs_io_completion_commit(void *inode) +{ + nfs_commit_inode(inode, 0); +} + +int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) +{ + struct inode *inode = mapping->host; + struct nfs_pageio_descriptor pgio; + struct nfs_io_completion *ioc = NULL; + unsigned int mntflags = NFS_SERVER(inode)->flags; + struct nfs_server *nfss = NFS_SERVER(inode); + int priority = 0; + int err; + + trace_nfs_writepages(inode, wbc->range_start, wbc->range_end - wbc->range_start); + + /* Wait with writeback until write congestion eases */ + if (wbc->sync_mode == WB_SYNC_NONE && nfss->write_congested) { + err = wait_event_killable(nfss->write_congestion_wait, + nfss->write_congested == 0); + if (err) + goto out_err; + } + + nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); + + if (!(mntflags & NFS_MOUNT_WRITE_EAGER) || wbc->for_kupdate || + wbc->for_background || wbc->for_sync) { + ioc = nfs_io_completion_alloc(GFP_KERNEL); + if (ioc) + nfs_io_completion_init(ioc, nfs_io_completion_commit, + inode); + priority = wb_priority(wbc); + } + + do { + struct folio *folio = NULL; + + nfs_pageio_init_write(&pgio, inode, priority, false, + &nfs_async_write_completion_ops); + pgio.pg_io_completion = ioc; + while ((folio = writeback_iter(mapping, wbc, folio, &err))) { + err = nfs_do_writepage(folio, wbc, &pgio); + folio_unlock(folio); + } + pgio.pg_error = 0; + nfs_pageio_complete(&pgio); + if (err == -EAGAIN && mntflags & NFS_MOUNT_SOFTERR) + break; + } while (err < 0 && !nfs_error_is_fatal(err)); + nfs_io_completion_put(ioc); + + if (err > 0) + err = 0; +out_err: + trace_nfs_writepages_done(inode, wbc->range_start, wbc->range_end - wbc->range_start, err); + return err; +} + +/* + * Insert a write request into an inode + */ +static void nfs_inode_add_request(struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + struct address_space *mapping = folio->mapping; + struct nfs_inode *nfsi = NFS_I(mapping->host); + + WARN_ON_ONCE(req->wb_this_page != req); + + /* Lock the request! */ + nfs_lock_request(req); + spin_lock(&mapping->i_private_lock); + set_bit(PG_MAPPED, &req->wb_flags); + folio_set_private(folio); + folio->private = req; + spin_unlock(&mapping->i_private_lock); + atomic_long_inc(&nfsi->nrequests); + /* this a head request for a page group - mark it as having an + * extra reference so sub groups can follow suit. + * This flag also informs pgio layer when to bump nrequests when + * adding subrequests. */ + WARN_ON(test_and_set_bit(PG_INODE_REF, &req->wb_flags)); + kref_get(&req->wb_kref); +} + +/* + * Remove a write request from an inode + */ +static void nfs_inode_remove_request(struct nfs_page *req) +{ + struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); + + nfs_page_group_lock(req); + if (nfs_page_group_sync_on_bit_locked(req, PG_REMOVE)) { + struct folio *folio = nfs_page_to_folio(req->wb_head); + struct address_space *mapping = folio->mapping; + + spin_lock(&mapping->i_private_lock); + if (likely(folio)) { + folio->private = NULL; + folio_clear_private(folio); + clear_bit(PG_MAPPED, &req->wb_head->wb_flags); + } + spin_unlock(&mapping->i_private_lock); + + folio_end_dropbehind(folio); + } + nfs_page_group_unlock(req); + + if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { + atomic_long_dec(&nfsi->nrequests); + nfs_release_request(req); + } +} + +static void nfs_mark_request_dirty(struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + if (folio) + filemap_dirty_folio(folio_mapping(folio), folio); +} + +/** + * nfs_request_add_commit_list_locked - add request to a commit list + * @req: pointer to a struct nfs_page + * @dst: commit list head + * @cinfo: holds list lock and accounting info + * + * This sets the PG_CLEAN bit, updates the cinfo count of + * number of outstanding requests requiring a commit as well as + * the MM page stats. + * + * The caller must hold NFS_I(cinfo->inode)->commit_mutex, and the + * nfs_page lock. + */ +void +nfs_request_add_commit_list_locked(struct nfs_page *req, struct list_head *dst, + struct nfs_commit_info *cinfo) +{ + set_bit(PG_CLEAN, &req->wb_flags); + nfs_list_add_request(req, dst); + atomic_long_inc(&cinfo->mds->ncommit); +} +EXPORT_SYMBOL_GPL(nfs_request_add_commit_list_locked); + +/** + * nfs_request_add_commit_list - add request to a commit list + * @req: pointer to a struct nfs_page + * @cinfo: holds list lock and accounting info + * + * This sets the PG_CLEAN bit, updates the cinfo count of + * number of outstanding requests requiring a commit as well as + * the MM page stats. + * + * The caller must _not_ hold the cinfo->lock, but must be + * holding the nfs_page lock. + */ +void +nfs_request_add_commit_list(struct nfs_page *req, struct nfs_commit_info *cinfo) +{ + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + nfs_request_add_commit_list_locked(req, &cinfo->mds->list, cinfo); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + nfs_folio_mark_unstable(req, cinfo); +} +EXPORT_SYMBOL_GPL(nfs_request_add_commit_list); + +/** + * nfs_request_remove_commit_list - Remove request from a commit list + * @req: pointer to a nfs_page + * @cinfo: holds list lock and accounting info + * + * This clears the PG_CLEAN bit, and updates the cinfo's count of + * number of outstanding requests requiring a commit + * It does not update the MM page stats. + * + * The caller _must_ hold the cinfo->lock and the nfs_page lock. + */ +void +nfs_request_remove_commit_list(struct nfs_page *req, + struct nfs_commit_info *cinfo) +{ + if (!test_and_clear_bit(PG_CLEAN, &(req)->wb_flags)) + return; + nfs_list_remove_request(req); + atomic_long_dec(&cinfo->mds->ncommit); +} +EXPORT_SYMBOL_GPL(nfs_request_remove_commit_list); + +static void nfs_init_cinfo_from_inode(struct nfs_commit_info *cinfo, + struct inode *inode) +{ + cinfo->inode = inode; + cinfo->mds = &NFS_I(inode)->commit_info; + cinfo->ds = pnfs_get_ds_info(inode); + cinfo->dreq = NULL; + cinfo->completion_ops = &nfs_commit_completion_ops; +} + +void nfs_init_cinfo(struct nfs_commit_info *cinfo, + struct inode *inode, + struct nfs_direct_req *dreq) +{ + if (dreq) + nfs_init_cinfo_from_dreq(cinfo, dreq); + else + nfs_init_cinfo_from_inode(cinfo, inode); +} +EXPORT_SYMBOL_GPL(nfs_init_cinfo); + +/* + * Add a request to the inode's commit list. + */ +void +nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, u32 ds_commit_idx) +{ + if (pnfs_mark_request_commit(req, lseg, cinfo, ds_commit_idx)) + return; + nfs_request_add_commit_list(req, cinfo); +} + +static void nfs_folio_clear_commit(struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + + if (folio) { + long nr = DIV_ROUND_UP(req->wb_bytes, PAGE_SIZE); + + node_stat_mod_folio(folio, NR_WRITEBACK, -nr); + bdi_wb_stat_mod(folio->mapping->host, WB_WRITEBACK, -nr); + } +} + +/* Called holding the request lock on @req */ +static void nfs_clear_request_commit(struct nfs_commit_info *cinfo, + struct nfs_page *req) +{ + if (test_bit(PG_CLEAN, &req->wb_flags)) { + struct nfs_open_context *ctx = nfs_req_openctx(req); + struct inode *inode = d_inode(ctx->dentry); + + mutex_lock(&NFS_I(inode)->commit_mutex); + if (!pnfs_clear_request_commit(req, cinfo)) { + nfs_request_remove_commit_list(req, cinfo); + } + mutex_unlock(&NFS_I(inode)->commit_mutex); + nfs_folio_clear_commit(req); + } +} + +int nfs_write_need_commit(struct nfs_pgio_header *hdr) +{ + if (hdr->verf.committed == NFS_DATA_SYNC) + return hdr->lseg == NULL; + return hdr->verf.committed != NFS_FILE_SYNC; +} + +static void nfs_async_write_init(struct nfs_pgio_header *hdr) +{ + nfs_io_completion_get(hdr->io_completion); +} + +static void nfs_write_completion(struct nfs_pgio_header *hdr) +{ + struct nfs_commit_info cinfo; + unsigned long bytes = 0; + + if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) + goto out; + nfs_init_cinfo_from_inode(&cinfo, hdr->inode); + while (!list_empty(&hdr->pages)) { + struct nfs_page *req = nfs_list_entry(hdr->pages.next); + + bytes += req->wb_bytes; + nfs_list_remove_request(req); + if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && + (hdr->good_bytes < bytes)) { + trace_nfs_comp_error(hdr->inode, req, hdr->error); + nfs_mapping_set_error(nfs_page_to_folio(req), + hdr->error); + goto remove_req; + } + if (nfs_write_need_commit(hdr)) { + struct nfs_open_context *ctx = + req->wb_lock_context->open_context; + + /* Reset wb_nio, since the write was successful. */ + req->wb_nio = 0; + memcpy(&req->wb_verf, &hdr->verf.verifier, sizeof(req->wb_verf)); + clear_bit(NFS_CONTEXT_WRITE_SYNC, &ctx->flags); + nfs_mark_request_commit(req, hdr->lseg, &cinfo, + hdr->ds_commit_idx); + goto next; + } +remove_req: + nfs_inode_remove_request(req); +next: + nfs_page_end_writeback(req); + nfs_release_request(req); + } +out: + nfs_io_completion_put(hdr->io_completion); + hdr->release(hdr); +} + +unsigned long +nfs_reqs_to_commit(struct nfs_commit_info *cinfo) +{ + return atomic_long_read(&cinfo->mds->ncommit); +} + +/* NFS_I(cinfo->inode)->commit_mutex held by caller */ +int +nfs_scan_commit_list(struct list_head *src, struct list_head *dst, + struct nfs_commit_info *cinfo, int max) +{ + struct nfs_page *req, *tmp; + int ret = 0; + + list_for_each_entry_safe(req, tmp, src, wb_list) { + kref_get(&req->wb_kref); + if (!nfs_lock_request(req)) { + nfs_release_request(req); + continue; + } + nfs_request_remove_commit_list(req, cinfo); + clear_bit(PG_COMMIT_TO_DS, &req->wb_flags); + nfs_list_add_request(req, dst); + ret++; + if ((ret == max) && !cinfo->dreq) + break; + cond_resched(); + } + return ret; +} +EXPORT_SYMBOL_GPL(nfs_scan_commit_list); + +/* + * nfs_scan_commit - Scan an inode for commit requests + * @inode: NFS inode to scan + * @dst: mds destination list + * @cinfo: mds and ds lists of reqs ready to commit + * + * Moves requests from the inode's 'commit' request list. + * The requests are *not* checked to ensure that they form a contiguous set. + */ +int +nfs_scan_commit(struct inode *inode, struct list_head *dst, + struct nfs_commit_info *cinfo) +{ + int ret = 0; + + if (!atomic_long_read(&cinfo->mds->ncommit)) + return 0; + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + if (atomic_long_read(&cinfo->mds->ncommit) > 0) { + const int max = INT_MAX; + + ret = nfs_scan_commit_list(&cinfo->mds->list, dst, + cinfo, max); + ret += pnfs_scan_commit_lists(inode, cinfo, max - ret); + } + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + return ret; +} + +/* + * Search for an existing write request, and attempt to update + * it to reflect a new dirty region on a given page. + * + * If the attempt fails, then the existing request is flushed out + * to disk. + */ +static struct nfs_page *nfs_try_to_update_request(struct folio *folio, + unsigned int offset, + unsigned int bytes) +{ + struct nfs_page *req; + unsigned int rqend; + unsigned int end; + int error; + + trace_nfs_try_to_update_request(folio_inode(folio), offset, bytes); + end = offset + bytes; + + req = nfs_lock_and_join_requests(folio); + if (IS_ERR_OR_NULL(req)) + goto out; + + rqend = req->wb_offset + req->wb_bytes; + /* + * Tell the caller to flush out the request if + * the offsets are non-contiguous. + * Note: nfs_flush_incompatible() will already + * have flushed out requests having wrong owners. + */ + if (offset > rqend || end < req->wb_offset) + goto out_flushme; + + /* Okay, the request matches. Update the region */ + if (offset < req->wb_offset) { + req->wb_offset = offset; + req->wb_pgbase = offset; + } + if (end > rqend) + req->wb_bytes = end - req->wb_offset; + else + req->wb_bytes = rqend - req->wb_offset; + req->wb_nio = 0; +out: + trace_nfs_try_to_update_request_done(folio_inode(folio), offset, bytes, + PTR_ERR_OR_ZERO(req)); + return req; +out_flushme: + /* + * Note: we mark the request dirty here because + * nfs_lock_and_join_requests() cannot preserve + * commit flags, so we have to replay the write. + */ + nfs_mark_request_dirty(req); + nfs_unlock_and_release_request(req); + error = nfs_wb_folio(folio->mapping->host, folio); + trace_nfs_try_to_update_request_done(folio_inode(folio), offset, bytes, error); + return (error < 0) ? ERR_PTR(error) : NULL; +} + +/* + * Try to update an existing write request, or create one if there is none. + * + * Note: Should always be called with the Page Lock held to prevent races + * if we have to add a new request. Also assumes that the caller has + * already called nfs_flush_incompatible() if necessary. + */ +static struct nfs_page *nfs_setup_write_request(struct nfs_open_context *ctx, + struct folio *folio, + unsigned int offset, + unsigned int bytes) +{ + struct nfs_page *req; + + req = nfs_try_to_update_request(folio, offset, bytes); + if (req != NULL) + goto out; + req = nfs_page_create_from_folio(ctx, folio, offset, bytes); + if (IS_ERR(req)) + goto out; + nfs_inode_add_request(req); +out: + return req; +} + +static int nfs_writepage_setup(struct nfs_open_context *ctx, + struct folio *folio, unsigned int offset, + unsigned int count) +{ + struct nfs_page *req; + + req = nfs_setup_write_request(ctx, folio, offset, count); + if (IS_ERR(req)) + return PTR_ERR(req); + trace_nfs_writepage_setup(req); + /* Update file length */ + nfs_grow_file(folio, offset, count); + nfs_mark_uptodate(req); + nfs_mark_request_dirty(req); + nfs_unlock_and_release_request(req); + return 0; +} + +int nfs_flush_incompatible(struct file *file, struct folio *folio) +{ + struct nfs_open_context *ctx = nfs_file_open_context(file); + struct nfs_lock_context *l_ctx; + struct file_lock_context *flctx = locks_inode_context(file_inode(file)); + struct nfs_page *req; + int do_flush, status; + /* + * Look for a request corresponding to this page. If there + * is one, and it belongs to another file, we flush it out + * before we try to copy anything into the page. Do this + * due to the lack of an ACCESS-type call in NFSv2. + * Also do the same if we find a request from an existing + * dropped page. + */ + do { + req = nfs_folio_find_head_request(folio); + if (req == NULL) + return 0; + l_ctx = req->wb_lock_context; + do_flush = nfs_page_to_folio(req) != folio || + !nfs_match_open_context(nfs_req_openctx(req), ctx); + if (l_ctx && flctx && + !(list_empty_careful(&flctx->flc_posix) && + list_empty_careful(&flctx->flc_flock))) { + do_flush |= l_ctx->lockowner != current->files; + } + nfs_release_request(req); + if (!do_flush) + return 0; + status = nfs_wb_folio(folio->mapping->host, folio); + } while (status == 0); + return status; +} + +/* + * Avoid buffered writes when a open context credential's key would + * expire soon. + * + * Returns -EACCES if the key will expire within RPC_KEY_EXPIRE_FAIL. + * + * Return 0 and set a credential flag which triggers the inode to flush + * and performs NFS_FILE_SYNC writes if the key will expired within + * RPC_KEY_EXPIRE_TIMEO. + */ +int +nfs_key_timeout_notify(struct file *filp, struct inode *inode) +{ + struct nfs_open_context *ctx = nfs_file_open_context(filp); + + if (nfs_ctx_key_to_expire(ctx, inode) && + !rcu_access_pointer(ctx->ll_cred)) + /* Already expired! */ + return -EACCES; + return 0; +} + +/* + * Test if the open context credential key is marked to expire soon. + */ +bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode) +{ + struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth; + struct rpc_cred *cred, *new, *old = NULL; + struct auth_cred acred = { + .cred = ctx->cred, + }; + bool ret = false; + + rcu_read_lock(); + cred = rcu_dereference(ctx->ll_cred); + if (cred && !(cred->cr_ops->crkey_timeout && + cred->cr_ops->crkey_timeout(cred))) + goto out; + rcu_read_unlock(); + + new = auth->au_ops->lookup_cred(auth, &acred, 0); + if (new == cred) { + put_rpccred(new); + return true; + } + if (IS_ERR_OR_NULL(new)) { + new = NULL; + ret = true; + } else if (new->cr_ops->crkey_timeout && + new->cr_ops->crkey_timeout(new)) + ret = true; + + rcu_read_lock(); + old = rcu_dereference_protected(xchg(&ctx->ll_cred, + RCU_INITIALIZER(new)), 1); +out: + rcu_read_unlock(); + put_rpccred(old); + return ret; +} + +/* + * If the page cache is marked as unsafe or invalid, then we can't rely on + * the PageUptodate() flag. In this case, we will need to turn off + * write optimisations that depend on the page contents being correct. + */ +static bool nfs_folio_write_uptodate(struct folio *folio, unsigned int pagelen) +{ + struct inode *inode = folio->mapping->host; + struct nfs_inode *nfsi = NFS_I(inode); + + if (nfs_have_delegated_attributes(inode)) + goto out; + if (nfsi->cache_validity & + (NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE)) + return false; + smp_rmb(); + if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags) && pagelen != 0) + return false; +out: + if (nfsi->cache_validity & NFS_INO_INVALID_DATA && pagelen != 0) + return false; + return folio_test_uptodate(folio) != 0; +} + +static bool +is_whole_file_wrlock(struct file_lock *fl) +{ + return fl->fl_start == 0 && fl->fl_end == OFFSET_MAX && + lock_is_write(fl); +} + +/* If we know the page is up to date, and we're not using byte range locks (or + * if we have the whole file locked for writing), it may be more efficient to + * extend the write to cover the entire page in order to avoid fragmentation + * inefficiencies. + * + * If the file is opened for synchronous writes then we can just skip the rest + * of the checks. + */ +static int nfs_can_extend_write(struct file *file, struct folio *folio, + unsigned int pagelen) +{ + struct inode *inode = file_inode(file); + struct file_lock_context *flctx = locks_inode_context(inode); + struct file_lock *fl; + int ret; + unsigned int mntflags = NFS_SERVER(inode)->flags; + + if (mntflags & NFS_MOUNT_NO_ALIGNWRITE) + return 0; + if (file->f_flags & O_DSYNC) + return 0; + if (!nfs_folio_write_uptodate(folio, pagelen)) + return 0; + if (nfs_have_write_delegation(inode)) + return 1; + if (!flctx || (list_empty_careful(&flctx->flc_flock) && + list_empty_careful(&flctx->flc_posix))) + return 1; + + /* Check to see if there are whole file write locks */ + ret = 0; + spin_lock(&flctx->flc_lock); + if (!list_empty(&flctx->flc_posix)) { + fl = list_first_entry(&flctx->flc_posix, struct file_lock, + c.flc_list); + if (is_whole_file_wrlock(fl)) + ret = 1; + } else if (!list_empty(&flctx->flc_flock)) { + fl = list_first_entry(&flctx->flc_flock, struct file_lock, + c.flc_list); + if (lock_is_write(fl)) + ret = 1; + } + spin_unlock(&flctx->flc_lock); + return ret; +} + +/* + * Update and possibly write a cached page of an NFS file. + * + * XXX: Keep an eye on generic_file_read to make sure it doesn't do bad + * things with a page scheduled for an RPC call (e.g. invalidate it). + */ +int nfs_update_folio(struct file *file, struct folio *folio, + unsigned int offset, unsigned int count) +{ + struct nfs_open_context *ctx = nfs_file_open_context(file); + struct address_space *mapping = folio->mapping; + struct inode *inode = mapping->host; + unsigned int pagelen = nfs_folio_length(folio); + int status = 0; + + nfs_inc_stats(inode, NFSIOS_VFSUPDATEPAGE); + + trace_nfs_update_folio(inode, offset, count); + + dprintk("NFS: nfs_update_folio(%pD2 %d@%lld)\n", file, count, + (long long)(folio_pos(folio) + offset)); + + if (!count) + goto out; + + if (nfs_can_extend_write(file, folio, pagelen)) { + unsigned int end = count + offset; + + offset = round_down(offset, PAGE_SIZE); + if (end < pagelen) + end = min(round_up(end, PAGE_SIZE), pagelen); + count = end - offset; + } + + status = nfs_writepage_setup(ctx, folio, offset, count); + if (status < 0) + nfs_set_pageerror(mapping); +out: + trace_nfs_update_folio_done(inode, offset, count, status); + dprintk("NFS: nfs_update_folio returns %d (isize %lld)\n", + status, (long long)i_size_read(inode)); + return status; +} + +static int flush_task_priority(int how) +{ + switch (how & (FLUSH_HIGHPRI|FLUSH_LOWPRI)) { + case FLUSH_HIGHPRI: + return RPC_PRIORITY_HIGH; + case FLUSH_LOWPRI: + return RPC_PRIORITY_LOW; + } + return RPC_PRIORITY_NORMAL; +} + +static void nfs_initiate_write(struct nfs_pgio_header *hdr, + struct rpc_message *msg, + const struct nfs_rpc_ops *rpc_ops, + struct rpc_task_setup *task_setup_data, int how) +{ + int priority = flush_task_priority(how); + + if (IS_SWAPFILE(hdr->inode)) + task_setup_data->flags |= RPC_TASK_SWAPPER; + task_setup_data->priority = priority; + rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client); + trace_nfs_initiate_write(hdr); +} + +/* If a nfs_flush_* function fails, it should remove reqs from @head and + * call this on each, which will prepare them to be retried on next + * writeback using standard nfs. + */ +static void nfs_redirty_request(struct nfs_page *req) +{ + struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); + + /* Bump the transmission count */ + req->wb_nio++; + nfs_mark_request_dirty(req); + atomic_long_inc(&nfsi->redirtied_pages); + nfs_page_end_writeback(req); + nfs_release_request(req); +} + +static void nfs_async_write_error(struct list_head *head, int error) +{ + struct nfs_page *req; + + while (!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + if (nfs_error_is_fatal_on_server(error)) + nfs_write_error(req, error); + else + nfs_redirty_request(req); + } +} + +static void nfs_async_write_reschedule_io(struct nfs_pgio_header *hdr) +{ + nfs_async_write_error(&hdr->pages, 0); +} + +static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops = { + .init_hdr = nfs_async_write_init, + .error_cleanup = nfs_async_write_error, + .completion = nfs_write_completion, + .reschedule_io = nfs_async_write_reschedule_io, +}; + +void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, + struct inode *inode, int ioflags, bool force_mds, + const struct nfs_pgio_completion_ops *compl_ops) +{ + struct nfs_server *server = NFS_SERVER(inode); + const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops; + +#if IS_ENABLED(CONFIG_NFS_V4) + if (server->pnfs_curr_ld && !force_mds) + pg_ops = server->pnfs_curr_ld->pg_write_ops; +#endif + nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_write_ops, + server->wsize, ioflags); +} +EXPORT_SYMBOL_GPL(nfs_pageio_init_write); + +void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio) +{ + struct nfs_pgio_mirror *mirror; + + if (pgio->pg_ops && pgio->pg_ops->pg_cleanup) + pgio->pg_ops->pg_cleanup(pgio); + + pgio->pg_ops = &nfs_pgio_rw_ops; + + nfs_pageio_stop_mirroring(pgio); + + mirror = &pgio->pg_mirrors[0]; + mirror->pg_bsize = NFS_SERVER(pgio->pg_inode)->wsize; +} +EXPORT_SYMBOL_GPL(nfs_pageio_reset_write_mds); + + +void nfs_commit_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_commit_data *data = calldata; + + NFS_PROTO(data->inode)->commit_rpc_prepare(task, data); +} + +static void nfs_writeback_check_extend(struct nfs_pgio_header *hdr, + struct nfs_fattr *fattr) +{ + struct nfs_pgio_args *argp = &hdr->args; + struct nfs_pgio_res *resp = &hdr->res; + u64 size = argp->offset + resp->count; + + if (!(fattr->valid & NFS_ATTR_FATTR_SIZE)) + fattr->size = size; + if (nfs_size_to_loff_t(fattr->size) < i_size_read(hdr->inode)) { + fattr->valid &= ~NFS_ATTR_FATTR_SIZE; + return; + } + if (size != fattr->size) + return; + /* Set attribute barrier */ + nfs_fattr_set_barrier(fattr); + /* ...and update size */ + fattr->valid |= NFS_ATTR_FATTR_SIZE; +} + +void nfs_writeback_update_inode(struct nfs_pgio_header *hdr) +{ + struct nfs_fattr *fattr = &hdr->fattr; + struct inode *inode = hdr->inode; + + if (nfs_have_delegated_mtime(inode)) { + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + spin_unlock(&inode->i_lock); + return; + } + + spin_lock(&inode->i_lock); + nfs_writeback_check_extend(hdr, fattr); + nfs_post_op_update_inode_force_wcc_locked(inode, fattr); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_writeback_update_inode); + +/* + * This function is called when the WRITE call is complete. + */ +static int nfs_writeback_done(struct rpc_task *task, + struct nfs_pgio_header *hdr, + struct inode *inode) +{ + int status; + + /* + * ->write_done will attempt to use post-op attributes to detect + * conflicting writes by other clients. A strict interpretation + * of close-to-open would allow us to continue caching even if + * another writer had changed the file, but some applications + * depend on tighter cache coherency when writing. + */ + status = NFS_PROTO(inode)->write_done(task, hdr); + if (status != 0) + return status; + + nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count); + trace_nfs_writeback_done(task, hdr); + + if (task->tk_status >= 0) { + enum nfs3_stable_how committed = hdr->res.verf->committed; + + if (committed == NFS_UNSTABLE) { + /* + * We have some uncommitted data on the server at + * this point, so ensure that we keep track of that + * fact irrespective of what later writes do. + */ + set_bit(NFS_IOHDR_UNSTABLE_WRITES, &hdr->flags); + } + + if (committed < hdr->args.stable) { + /* We tried a write call, but the server did not + * commit data to stable storage even though we + * requested it. + * Note: There is a known bug in Tru64 < 5.0 in which + * the server reports NFS_DATA_SYNC, but performs + * NFS_FILE_SYNC. We therefore implement this checking + * as a dprintk() in order to avoid filling syslog. + */ + static unsigned long complain; + + /* Note this will print the MDS for a DS write */ + if (time_before(complain, jiffies)) { + dprintk("NFS: faulty NFS server %s:" + " (committed = %d) != (stable = %d)\n", + NFS_SERVER(inode)->nfs_client->cl_hostname, + committed, hdr->args.stable); + complain = jiffies + 300 * HZ; + } + } + } + + /* Deal with the suid/sgid bit corner case */ + if (nfs_should_remove_suid(inode)) { + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MODE + | NFS_INO_REVAL_FORCED); + spin_unlock(&inode->i_lock); + } + return 0; +} + +/* + * This function is called when the WRITE call is complete. + */ +static void nfs_writeback_result(struct rpc_task *task, + struct nfs_pgio_header *hdr) +{ + struct nfs_pgio_args *argp = &hdr->args; + struct nfs_pgio_res *resp = &hdr->res; + + if (resp->count < argp->count && !list_empty(&hdr->pages)) { + static unsigned long complain; + struct nfs_open_context *ctx = + hdr->req->wb_lock_context->open_context; + + set_bit(NFS_CONTEXT_WRITE_SYNC, &ctx->flags); + /* This a short write! */ + nfs_inc_stats(hdr->inode, NFSIOS_SHORTWRITE); + + /* Has the server at least made some progress? */ + if (resp->count == 0) { + if (time_before(complain, jiffies)) { + printk(KERN_WARNING + "NFS: Server wrote zero bytes, expected %u.\n", + argp->count); + complain = jiffies + 300 * HZ; + } + nfs_set_pgio_error(hdr, -EIO, argp->offset); + task->tk_status = -EIO; + return; + } + + /* For non rpc-based layout drivers, retry-through-MDS */ + if (!task->tk_ops) { + hdr->pnfs_error = -EAGAIN; + return; + } + + /* Was this an NFSv2 write or an NFSv3 stable write? */ + if (resp->verf->committed != NFS_UNSTABLE) { + /* Resend from where the server left off */ + hdr->mds_offset += resp->count; + argp->offset += resp->count; + argp->pgbase += resp->count; + argp->count -= resp->count; + } else { + /* Resend as a stable write in order to avoid + * headaches in the case of a server crash. + */ + argp->stable = NFS_FILE_SYNC; + } + resp->count = 0; + resp->verf->committed = 0; + rpc_restart_call_prepare(task); + } +} + +static int wait_on_commit(struct nfs_mds_commit_info *cinfo) +{ + return wait_var_event_killable(&cinfo->rpcs_out, + !atomic_read(&cinfo->rpcs_out)); +} + +void nfs_commit_begin(struct nfs_mds_commit_info *cinfo) +{ + atomic_inc(&cinfo->rpcs_out); +} + +bool nfs_commit_end(struct nfs_mds_commit_info *cinfo) +{ + if (atomic_dec_and_test(&cinfo->rpcs_out)) { + wake_up_var(&cinfo->rpcs_out); + return true; + } + return false; +} + +void nfs_commitdata_release(struct nfs_commit_data *data) +{ + put_nfs_open_context(data->context); + nfs_commit_free(data); +} +EXPORT_SYMBOL_GPL(nfs_commitdata_release); + +int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, + const struct nfs_rpc_ops *nfs_ops, + const struct rpc_call_ops *call_ops, + int how, int flags, + struct nfsd_file *localio) +{ + struct rpc_task *task; + int priority = flush_task_priority(how); + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->cred, + }; + struct rpc_task_setup task_setup_data = { + .task = &data->task, + .rpc_client = clnt, + .rpc_message = &msg, + .callback_ops = call_ops, + .callback_data = data, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | flags, + .priority = priority, + }; + + if (nfs_server_capable(data->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + + /* Set up the initial task struct. */ + nfs_ops->commit_setup(data, &msg, &task_setup_data.rpc_client); + trace_nfs_initiate_commit(data); + + dprintk("NFS: initiated commit call\n"); + + if (localio) + return nfs_local_commit(localio, data, call_ops, how); + + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + if (how & FLUSH_SYNC) + rpc_wait_for_completion_task(task); + rpc_put_task(task); + return 0; +} +EXPORT_SYMBOL_GPL(nfs_initiate_commit); + +static loff_t nfs_get_lwb(struct list_head *head) +{ + loff_t lwb = 0; + struct nfs_page *req; + + list_for_each_entry(req, head, wb_list) + if (lwb < (req_offset(req) + req->wb_bytes)) + lwb = req_offset(req) + req->wb_bytes; + + return lwb; +} + +/* + * Set up the argument/result storage required for the RPC call. + */ +void nfs_init_commit(struct nfs_commit_data *data, + struct list_head *head, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo) +{ + struct nfs_page *first; + struct nfs_open_context *ctx; + struct inode *inode; + + /* Set up the RPC argument and reply structs + * NB: take care not to mess about with data->commit et al. */ + + if (head) + list_splice_init(head, &data->pages); + + first = nfs_list_entry(data->pages.next); + ctx = nfs_req_openctx(first); + inode = d_inode(ctx->dentry); + + data->inode = inode; + data->cred = ctx->cred; + data->lseg = lseg; /* reference transferred */ + /* only set lwb for pnfs commit */ + if (lseg) + data->lwb = nfs_get_lwb(&data->pages); + data->mds_ops = &nfs_commit_ops; + data->completion_ops = cinfo->completion_ops; + data->dreq = cinfo->dreq; + + data->args.fh = NFS_FH(data->inode); + /* Note: we always request a commit of the entire inode */ + data->args.offset = 0; + data->args.count = 0; + data->context = get_nfs_open_context(ctx); + data->res.fattr = &data->fattr; + data->res.verf = &data->verf; + nfs_fattr_init(&data->fattr); + nfs_commit_begin(cinfo->mds); +} +EXPORT_SYMBOL_GPL(nfs_init_commit); + +void nfs_retry_commit(struct list_head *page_list, + struct pnfs_layout_segment *lseg, + struct nfs_commit_info *cinfo, + u32 ds_commit_idx) +{ + struct nfs_page *req; + + while (!list_empty(page_list)) { + req = nfs_list_entry(page_list->next); + nfs_list_remove_request(req); + nfs_mark_request_commit(req, lseg, cinfo, ds_commit_idx); + nfs_folio_clear_commit(req); + nfs_unlock_and_release_request(req); + } +} +EXPORT_SYMBOL_GPL(nfs_retry_commit); + +static void nfs_commit_resched_write(struct nfs_commit_info *cinfo, + struct nfs_page *req) +{ + struct folio *folio = nfs_page_to_folio(req); + + filemap_dirty_folio(folio_mapping(folio), folio); +} + +/* + * Commit dirty pages + */ +static int +nfs_commit_list(struct inode *inode, struct list_head *head, int how, + struct nfs_commit_info *cinfo) +{ + struct nfs_commit_data *data; + struct nfsd_file *localio; + unsigned short task_flags = 0; + + /* another commit raced with us */ + if (list_empty(head)) + return 0; + + data = nfs_commitdata_alloc(); + if (!data) { + nfs_retry_commit(head, NULL, cinfo, -1); + return -ENOMEM; + } + + /* Set up the argument struct */ + nfs_init_commit(data, head, NULL, cinfo); + if (NFS_SERVER(inode)->nfs_client->cl_minorversion) + task_flags = RPC_TASK_MOVEABLE; + + localio = nfs_local_open_fh(NFS_SERVER(inode)->nfs_client, data->cred, + data->args.fh, &data->context->nfl, + data->context->mode); + return nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(inode), + data->mds_ops, how, + RPC_TASK_CRED_NOREF | task_flags, localio); +} + +/* + * COMMIT call returned + */ +static void nfs_commit_done(struct rpc_task *task, void *calldata) +{ + struct nfs_commit_data *data = calldata; + + /* Call the NFS version-specific code */ + NFS_PROTO(data->inode)->commit_done(task, data); + trace_nfs_commit_done(task, data); +} + +static void nfs_commit_release_pages(struct nfs_commit_data *data) +{ + const struct nfs_writeverf *verf = data->res.verf; + struct nfs_page *req; + int status = data->task.tk_status; + struct nfs_commit_info cinfo; + struct folio *folio; + + while (!list_empty(&data->pages)) { + req = nfs_list_entry(data->pages.next); + nfs_list_remove_request(req); + folio = nfs_page_to_folio(req); + nfs_folio_clear_commit(req); + + dprintk("NFS: commit (%s/%llu %d@%lld)", + nfs_req_openctx(req)->dentry->d_sb->s_id, + (unsigned long long)d_inode(nfs_req_openctx(req)->dentry)->i_ino, + req->wb_bytes, + (long long)req_offset(req)); + if (status < 0) { + if (folio) { + trace_nfs_commit_error(data->inode, req, + status); + nfs_mapping_set_error(folio, status); + nfs_inode_remove_request(req); + } + dprintk(", error = %d\n", status); + goto next; + } + + /* Okay, COMMIT succeeded, apparently. Check the verifier + * returned by the server against all stored verfs. */ + if (nfs_write_match_verf(verf, req)) { + /* We have a match */ + if (folio) + nfs_inode_remove_request(req); + dprintk(" OK\n"); + goto next; + } + /* We have a mismatch. Write the page again */ + dprintk(" mismatch\n"); + nfs_mark_request_dirty(req); + set_bit(NFS_CONTEXT_WRITE_SYNC, + &req->wb_lock_context->open_context->flags); + atomic_long_inc(&NFS_I(data->inode)->redirtied_pages); + next: + nfs_unlock_and_release_request(req); + /* Latency breaker */ + cond_resched(); + } + + nfs_init_cinfo(&cinfo, data->inode, data->dreq); + nfs_commit_end(cinfo.mds); +} + +static void nfs_commit_release(void *calldata) +{ + struct nfs_commit_data *data = calldata; + + data->completion_ops->completion(data); + nfs_commitdata_release(calldata); +} + +static const struct rpc_call_ops nfs_commit_ops = { + .rpc_call_prepare = nfs_commit_prepare, + .rpc_call_done = nfs_commit_done, + .rpc_release = nfs_commit_release, +}; + +static const struct nfs_commit_completion_ops nfs_commit_completion_ops = { + .completion = nfs_commit_release_pages, + .resched_write = nfs_commit_resched_write, +}; + +int nfs_generic_commit_list(struct inode *inode, struct list_head *head, + int how, struct nfs_commit_info *cinfo) +{ + int status; + + status = pnfs_commit_list(inode, head, how, cinfo); + if (status == PNFS_NOT_ATTEMPTED) + status = nfs_commit_list(inode, head, how, cinfo); + return status; +} + +static int __nfs_commit_inode(struct inode *inode, int how, + struct writeback_control *wbc) +{ + LIST_HEAD(head); + struct nfs_commit_info cinfo; + int may_wait = how & FLUSH_SYNC; + int ret, nscan; + + how &= ~FLUSH_SYNC; + nfs_init_cinfo_from_inode(&cinfo, inode); + nfs_commit_begin(cinfo.mds); + for (;;) { + ret = nscan = nfs_scan_commit(inode, &head, &cinfo); + if (ret <= 0) + break; + ret = nfs_generic_commit_list(inode, &head, how, &cinfo); + if (ret < 0) + break; + ret = 0; + if (wbc && wbc->sync_mode == WB_SYNC_NONE) { + if (nscan < wbc->nr_to_write) + wbc->nr_to_write -= nscan; + else + wbc->nr_to_write = 0; + } + if (nscan < INT_MAX) + break; + cond_resched(); + } + nfs_commit_end(cinfo.mds); + if (ret || !may_wait) + return ret; + return wait_on_commit(cinfo.mds); +} + +int nfs_commit_inode(struct inode *inode, int how) +{ + return __nfs_commit_inode(inode, how, NULL); +} +EXPORT_SYMBOL_GPL(nfs_commit_inode); + +int nfs_write_inode(struct inode *inode, struct writeback_control *wbc) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int flags = FLUSH_SYNC; + int ret = 0; + + if (wbc->sync_mode == WB_SYNC_NONE) { + /* no commits means nothing needs to be done */ + if (!atomic_long_read(&nfsi->commit_info.ncommit)) + goto check_requests_outstanding; + + /* Don't commit yet if this is a non-blocking flush and there + * are a lot of outstanding writes for this mapping. + */ + if (mapping_tagged(inode->i_mapping, PAGECACHE_TAG_WRITEBACK)) + goto out_mark_dirty; + + /* don't wait for the COMMIT response */ + flags = 0; + } + + ret = __nfs_commit_inode(inode, flags, wbc); + if (!ret) { + if (flags & FLUSH_SYNC) + return 0; + } else if (atomic_long_read(&nfsi->commit_info.ncommit)) + goto out_mark_dirty; + +check_requests_outstanding: + if (!atomic_read(&nfsi->commit_info.rpcs_out)) + return ret; +out_mark_dirty: + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_write_inode); + +/* + * Wrapper for filemap_write_and_wait_range() + * + * Needed for pNFS in order to ensure data becomes visible to the + * client. + */ +int nfs_filemap_write_and_wait_range(struct address_space *mapping, + loff_t lstart, loff_t lend) +{ + int ret; + + ret = filemap_write_and_wait_range(mapping, lstart, lend); + if (ret == 0) + ret = pnfs_sync_inode(mapping->host, true); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_filemap_write_and_wait_range); + +/* + * flush the inode to disk. + */ +int nfs_wb_all(struct inode *inode) +{ + int ret; + + trace_nfs_writeback_inode_enter(inode); + + ret = filemap_write_and_wait(inode->i_mapping); + if (ret) + goto out; + ret = nfs_commit_inode(inode, FLUSH_SYNC); + if (ret < 0) + goto out; + pnfs_sync_inode(inode, true); + ret = 0; + +out: + trace_nfs_writeback_inode_exit(inode, ret); + return ret; +} +EXPORT_SYMBOL_GPL(nfs_wb_all); + +int nfs_wb_folio_cancel(struct inode *inode, struct folio *folio) +{ + struct nfs_page *req; + int ret = 0; + + folio_wait_writeback(folio); + + /* blocking call to cancel all requests and join to a single (head) + * request */ + req = nfs_lock_and_join_requests(folio); + + if (IS_ERR(req)) { + ret = PTR_ERR(req); + } else if (req) { + /* all requests from this folio have been cancelled by + * nfs_lock_and_join_requests, so just remove the head + * request from the inode / page_private pointer and + * release it */ + nfs_inode_remove_request(req); + nfs_unlock_and_release_request(req); + folio_cancel_dirty(folio); + } + + return ret; +} + +/** + * nfs_wb_folio_reclaim - Write back all requests on one page + * @inode: pointer to page + * @folio: pointer to folio + * + * Assumes that the folio has been locked by the caller + */ +int nfs_wb_folio_reclaim(struct inode *inode, struct folio *folio) +{ + loff_t range_start = folio_pos(folio); + size_t len = folio_size(folio); + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, + .range_start = range_start, + .range_end = range_start + len - 1, + .for_sync = 1, + }; + int ret; + + if (folio_test_writeback(folio)) + return -EBUSY; + if (folio_clear_dirty_for_io(folio)) { + trace_nfs_writeback_folio_reclaim(inode, range_start, len); + ret = nfs_writepage_locked(folio, &wbc); + trace_nfs_writeback_folio_reclaim_done(inode, range_start, len, + ret); + return ret; + } + nfs_commit_inode(inode, 0); + return 0; +} + +/** + * nfs_wb_folio - Write back all requests on one page + * @inode: pointer to page + * @folio: pointer to folio + * + * Assumes that the folio has been locked by the caller, and will + * not unlock it. + */ +int nfs_wb_folio(struct inode *inode, struct folio *folio) +{ + loff_t range_start = folio_pos(folio); + size_t len = folio_size(folio); + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, + .range_start = range_start, + .range_end = range_start + len - 1, + }; + int ret; + + trace_nfs_writeback_folio(inode, range_start, len); + + for (;;) { + folio_wait_writeback(folio); + if (folio_clear_dirty_for_io(folio)) { + ret = nfs_writepage_locked(folio, &wbc); + if (ret < 0) + goto out_error; + continue; + } + ret = 0; + if (!folio_test_private(folio)) + break; + ret = nfs_commit_inode(inode, FLUSH_SYNC); + if (ret < 0) + goto out_error; + } +out_error: + trace_nfs_writeback_folio_done(inode, range_start, len, ret); + return ret; +} + +#ifdef CONFIG_MIGRATION +int nfs_migrate_folio(struct address_space *mapping, struct folio *dst, + struct folio *src, enum migrate_mode mode) +{ + /* + * If the private flag is set, the folio is currently associated with + * an in-progress read or write request. Don't try to migrate it. + * + * FIXME: we could do this in principle, but we'll need a way to ensure + * that we can safely release the inode reference while holding + * the folio lock. + */ + if (folio_test_private(src)) { + if (mode == MIGRATE_SYNC) + nfs_wb_folio(src->mapping->host, src); + if (folio_test_private(src)) + return -EBUSY; + } + + if (folio_test_private_2(src)) { /* [DEPRECATED] */ + if (mode == MIGRATE_ASYNC) + return -EBUSY; + folio_wait_private_2(src); + } + + return migrate_folio(mapping, dst, src, mode); +} +#endif + +int __init nfs_init_writepagecache(void) +{ + nfs_wdata_cachep = kmem_cache_create("nfs_write_data", + sizeof(struct nfs_pgio_header), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (nfs_wdata_cachep == NULL) + return -ENOMEM; + + nfs_wdata_mempool = mempool_create_slab_pool(MIN_POOL_WRITE, + nfs_wdata_cachep); + if (nfs_wdata_mempool == NULL) + goto out_destroy_write_cache; + + nfs_cdata_cachep = kmem_cache_create("nfs_commit_data", + sizeof(struct nfs_commit_data), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (nfs_cdata_cachep == NULL) + goto out_destroy_write_mempool; + + nfs_commit_mempool = mempool_create_slab_pool(MIN_POOL_COMMIT, + nfs_cdata_cachep); + if (nfs_commit_mempool == NULL) + goto out_destroy_commit_cache; + + /* + * NFS congestion size, scale with available memory. + * + * 64MB: 8192k + * 128MB: 11585k + * 256MB: 16384k + * 512MB: 23170k + * 1GB: 32768k + * 2GB: 46340k + * 4GB: 65536k + * 8GB: 92681k + * 16GB: 131072k + * + * This allows larger machines to have larger/more transfers. + * Limit the default to 256M + */ + nfs_congestion_kb = (16*int_sqrt(totalram_pages())) << (PAGE_SHIFT-10); + if (nfs_congestion_kb > 256*1024) + nfs_congestion_kb = 256*1024; + + return 0; + +out_destroy_commit_cache: + kmem_cache_destroy(nfs_cdata_cachep); +out_destroy_write_mempool: + mempool_destroy(nfs_wdata_mempool); +out_destroy_write_cache: + kmem_cache_destroy(nfs_wdata_cachep); + return -ENOMEM; +} + +void nfs_destroy_writepagecache(void) +{ + mempool_destroy(nfs_commit_mempool); + kmem_cache_destroy(nfs_cdata_cachep); + mempool_destroy(nfs_wdata_mempool); + kmem_cache_destroy(nfs_wdata_cachep); +} + +static const struct nfs_rw_ops nfs_rw_write_ops = { + .rw_alloc_header = nfs_writehdr_alloc, + .rw_free_header = nfs_writehdr_free, + .rw_done = nfs_writeback_done, + .rw_result = nfs_writeback_result, + .rw_initiate = nfs_initiate_write, +}; -- cgit v1.3.1