From 684b413b5483f57c890c171b9400076a0143b918 Mon Sep 17 00:00:00 2001 From: Niklas Cassel Date: Fri, 18 Sep 2026 16:06:42 +0200 Subject: virtio_blk: set the zone write granularity virtblk_read_zoned_limits() reads the write granularity that the device reports in virtio_blk_zoned_characteristics and assigns it to the physical block size and to io_min, but never to the limit that is named after it. queue_limits.zone_write_granularity is left at zero, so blk_validate_zoned_limits() raises it to the logical block size: if (lim->zone_write_granularity < lim->logical_block_size) lim->zone_write_granularity = lim->logical_block_size; A device that reports a granularity coarser than its logical block size, which is what the field exists to express, therefore has it silently reduced. A 512e host managed disk passed through to a guest reports a logical block size of 512 and a write granularity of 4096, and the guest ends up with a zone write granularity of 512. bio_split_alignment() returns lim->zone_write_granularity if it is non-zero and bio_split_io_at() may split a bio with as per bio_split_alignment(). This can real to the write getting rejected by the host drive, as the write is not aligned to the physical block size. zonefs also takes its block size from bdev_zone_write_granularity(), so it would incorrectly use 512 on a disk that requires 4096. sd_zbc_read_zones() sets the limit from the physical block size for the same reason. NVMe ZNS and null_blk leave it unset, but the fallback gives the right answer for them, as their write granularity is the logical block size. virtio carries a separate value that may exceed it. Set the zone write granularity from the value that the device reports. Fixes: 95bfec41bd3d ("virtio-blk: add support for zoned block devices") Signed-off-by: Niklas Cassel Reviewed-by: Stefan Hajnoczi Link: https://patch.msgid.link/20260918140641.2031075-2-cassel@kernel.org Signed-off-by: Jens Axboe --- fs/cachefiles/interface.c | 453 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 453 insertions(+) create mode 100644 fs/cachefiles/interface.c (limited to 'fs/cachefiles/interface.c') diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c new file mode 100644 index 000000000..50a000310 --- /dev/null +++ b/fs/cachefiles/interface.c @@ -0,0 +1,453 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* FS-Cache interface to CacheFiles + * + * Copyright (C) 2021 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +static atomic_t cachefiles_object_debug_id; + +/* + * Allocate a cache object record. + */ +static +struct cachefiles_object *cachefiles_alloc_object(struct fscache_cookie *cookie) +{ + struct fscache_volume *vcookie = cookie->volume; + struct cachefiles_volume *volume = vcookie->cache_priv; + struct cachefiles_object *object; + + _enter("{%s},%x,", vcookie->key, cookie->debug_id); + + object = kmem_cache_zalloc(cachefiles_object_jar, GFP_KERNEL); + if (!object) + return NULL; + + refcount_set(&object->ref, 1); + + spin_lock_init(&object->lock); + INIT_LIST_HEAD(&object->cache_link); + object->volume = volume; + object->debug_id = atomic_inc_return(&cachefiles_object_debug_id); + object->cookie = fscache_get_cookie(cookie, fscache_cookie_get_attach_object); + + fscache_count_object(vcookie->cache); + trace_cachefiles_ref(object->debug_id, cookie->debug_id, 1, + cachefiles_obj_new); + return object; +} + +/* + * Note that an object has been seen. + */ +void cachefiles_see_object(struct cachefiles_object *object, + enum cachefiles_obj_ref_trace why) +{ + trace_cachefiles_ref(object->debug_id, object->cookie->debug_id, + refcount_read(&object->ref), why); +} + +/* + * Increment the usage count on an object; + */ +struct cachefiles_object *cachefiles_grab_object(struct cachefiles_object *object, + enum cachefiles_obj_ref_trace why) +{ + int r; + + __refcount_inc(&object->ref, &r); + trace_cachefiles_ref(object->debug_id, object->cookie->debug_id, r, why); + return object; +} + +/* + * dispose of a reference to an object + */ +void cachefiles_put_object(struct cachefiles_object *object, + enum cachefiles_obj_ref_trace why) +{ + unsigned int object_debug_id = object->debug_id; + unsigned int cookie_debug_id = object->cookie->debug_id; + struct fscache_cache *cache; + bool done; + int r; + + done = __refcount_dec_and_test(&object->ref, &r); + trace_cachefiles_ref(object_debug_id, cookie_debug_id, r, why); + if (done) { + _debug("- kill object OBJ%x", object_debug_id); + + ASSERTCMP(object->file, ==, NULL); + + kfree(object->d_name); + cache = object->volume->cache->cache; + fscache_put_cookie(object->cookie, fscache_cookie_put_object); + object->cookie = NULL; + kmem_cache_free(cachefiles_object_jar, object); + fscache_uncount_object(cache); + } + + _leave(""); +} + +/* + * Adjust the size of a cache file if necessary to match the DIO size. We keep + * the EOF marker a multiple of DIO blocks so that we don't fall back to doing + * non-DIO for a partial block straddling the EOF, but we also have to be + * careful of someone expanding the file and accidentally accreting the + * padding. + */ +static int cachefiles_adjust_size(struct cachefiles_object *object) +{ + struct iattr newattrs; + struct file *file = object->file; + uint64_t ni_size; + loff_t oi_size; + int ret; + + ni_size = object->cookie->object_size; + ni_size = round_up(ni_size, CACHEFILES_DIO_BLOCK_SIZE); + + _enter("{OBJ%x},[%llu]", + object->debug_id, (unsigned long long) ni_size); + + if (!file) + return -ENOBUFS; + + oi_size = i_size_read(file_inode(file)); + if (oi_size == ni_size) + return 0; + + inode_lock(file_inode(file)); + + /* if there's an extension to a partial page at the end of the backing + * file, we need to discard the partial page so that we pick up new + * data after it */ + if (oi_size & ~PAGE_MASK && ni_size > oi_size) { + _debug("discard tail %llx", oi_size); + newattrs.ia_valid = ATTR_SIZE; + newattrs.ia_size = oi_size & PAGE_MASK; + ret = cachefiles_inject_remove_error(); + if (ret == 0) + ret = notify_change(&nop_mnt_idmap, file->f_path.dentry, + &newattrs, NULL); + if (ret < 0) + goto truncate_failed; + } + + newattrs.ia_valid = ATTR_SIZE; + newattrs.ia_size = ni_size; + ret = cachefiles_inject_write_error(); + if (ret == 0) + ret = notify_change(&nop_mnt_idmap, file->f_path.dentry, + &newattrs, NULL); + +truncate_failed: + inode_unlock(file_inode(file)); + + if (ret < 0) + trace_cachefiles_io_error(NULL, file_inode(file), ret, + cachefiles_trace_notify_change_error); + if (ret == -EIO) { + cachefiles_io_error_obj(object, "Size set failed"); + ret = -ENOBUFS; + } + + _leave(" = %d", ret); + return ret; +} + +/* + * Attempt to look up the nominated node in this cache + */ +static bool cachefiles_lookup_cookie(struct fscache_cookie *cookie) +{ + struct cachefiles_object *object; + struct cachefiles_cache *cache = cookie->volume->cache->cache_priv; + const struct cred *saved_cred; + bool success; + + object = cachefiles_alloc_object(cookie); + if (!object) + goto fail; + + _enter("{OBJ%x}", object->debug_id); + + if (!cachefiles_cook_key(object)) + goto fail_put; + + cookie->cache_priv = object; + + cachefiles_begin_secure(cache, &saved_cred); + + success = cachefiles_look_up_object(object); + if (!success) + goto fail_withdraw; + + cachefiles_see_object(object, cachefiles_obj_see_lookup_cookie); + + spin_lock(&cache->object_list_lock); + list_add(&object->cache_link, &cache->object_list); + spin_unlock(&cache->object_list_lock); + cachefiles_adjust_size(object); + + cachefiles_end_secure(cache, saved_cred); + _leave(" = t"); + return true; + +fail_withdraw: + cachefiles_end_secure(cache, saved_cred); + cachefiles_see_object(object, cachefiles_obj_see_lookup_failed); + fscache_caching_failed(cookie); + _debug("failed c=%08x o=%08x", cookie->debug_id, object->debug_id); + /* The caller holds an access count on the cookie, so we need them to + * drop it before we can withdraw the object. + */ + return false; + +fail_put: + cachefiles_put_object(object, cachefiles_obj_put_alloc_fail); +fail: + return false; +} + +/* + * Shorten the backing object to discard any dirty data and free up + * any unused granules. + */ +static bool cachefiles_shorten_object(struct cachefiles_object *object, + struct file *file, loff_t new_size) +{ + struct cachefiles_cache *cache = object->volume->cache; + struct inode *inode = file_inode(file); + loff_t i_size, dio_size; + int ret; + + dio_size = round_up(new_size, CACHEFILES_DIO_BLOCK_SIZE); + i_size = i_size_read(inode); + + trace_cachefiles_trunc(object, inode, i_size, dio_size, + cachefiles_trunc_shrink); + ret = cachefiles_inject_remove_error(); + if (ret == 0) + ret = vfs_truncate(&file->f_path, dio_size); + if (ret < 0) { + trace_cachefiles_io_error(object, file_inode(file), ret, + cachefiles_trace_trunc_error); + cachefiles_io_error_obj(object, "Trunc-to-size failed %d", ret); + cachefiles_remove_object_xattr(cache, object, file->f_path.dentry); + return false; + } + + if (new_size < dio_size) { + trace_cachefiles_trunc(object, inode, dio_size, new_size, + cachefiles_trunc_dio_adjust); + ret = cachefiles_inject_write_error(); + if (ret == 0) + ret = vfs_fallocate(file, FALLOC_FL_ZERO_RANGE, + new_size, dio_size - new_size); + if (ret < 0) { + trace_cachefiles_io_error(object, file_inode(file), ret, + cachefiles_trace_fallocate_error); + cachefiles_io_error_obj(object, "Trunc-to-dio-size failed %d", ret); + cachefiles_remove_object_xattr(cache, object, file->f_path.dentry); + return false; + } + } + + return true; +} + +/* + * Resize the backing object. + */ +static void cachefiles_resize_cookie(struct netfs_cache_resources *cres, + loff_t new_size) +{ + struct cachefiles_object *object = cachefiles_cres_object(cres); + struct cachefiles_cache *cache = object->volume->cache; + struct fscache_cookie *cookie = object->cookie; + const struct cred *saved_cred; + struct file *file = cachefiles_cres_file(cres); + loff_t old_size = cookie->object_size; + + _enter("%llu->%llu", old_size, new_size); + + if (new_size < old_size) { + cachefiles_begin_secure(cache, &saved_cred); + cachefiles_shorten_object(object, file, new_size); + cachefiles_end_secure(cache, saved_cred); + object->cookie->object_size = new_size; + return; + } + + /* The file is being expanded. We don't need to do anything + * particularly. cookie->initial_size doesn't change and so the point + * at which we have to download before doesn't change. + */ + cookie->object_size = new_size; +} + +/* + * Commit changes to the object as we drop it. + */ +static void cachefiles_commit_object(struct cachefiles_object *object, + struct cachefiles_cache *cache) +{ + bool update = false; + + if (test_and_clear_bit(FSCACHE_COOKIE_LOCAL_WRITE, &object->cookie->flags)) + update = true; + if (test_and_clear_bit(FSCACHE_COOKIE_NEEDS_UPDATE, &object->cookie->flags)) + update = true; + if (update) + cachefiles_set_object_xattr(object); + + if (test_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags)) + cachefiles_commit_tmpfile(cache, object); +} + +/* + * Finalise and object and close the VFS structs that we have. + */ +static void cachefiles_clean_up_object(struct cachefiles_object *object, + struct cachefiles_cache *cache) +{ + struct file *file; + + if (test_bit(FSCACHE_COOKIE_RETIRED, &object->cookie->flags)) { + if (!test_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags)) { + cachefiles_see_object(object, cachefiles_obj_see_clean_delete); + _debug("- inval object OBJ%x", object->debug_id); + cachefiles_delete_object(object, FSCACHE_OBJECT_WAS_RETIRED); + } else { + cachefiles_see_object(object, cachefiles_obj_see_clean_drop_tmp); + _debug("- inval object OBJ%x tmpfile", object->debug_id); + } + } else { + cachefiles_see_object(object, cachefiles_obj_see_clean_commit); + cachefiles_commit_object(object, cache); + } + + cachefiles_unmark_inode_in_use(object, object->file); + + spin_lock(&object->lock); + file = object->file; + object->file = NULL; + spin_unlock(&object->lock); + + if (file) + fput(file); +} + +/* + * Withdraw caching for a cookie. + */ +static void cachefiles_withdraw_cookie(struct fscache_cookie *cookie) +{ + struct cachefiles_object *object = cookie->cache_priv; + struct cachefiles_cache *cache = object->volume->cache; + const struct cred *saved_cred; + + _enter("o=%x", object->debug_id); + cachefiles_see_object(object, cachefiles_obj_see_withdraw_cookie); + + if (!list_empty(&object->cache_link)) { + spin_lock(&cache->object_list_lock); + cachefiles_see_object(object, cachefiles_obj_see_withdrawal); + list_del_init(&object->cache_link); + spin_unlock(&cache->object_list_lock); + } + + if (object->file) { + cachefiles_begin_secure(cache, &saved_cred); + cachefiles_clean_up_object(object, cache); + cachefiles_end_secure(cache, saved_cred); + } + + cookie->cache_priv = NULL; + cachefiles_put_object(object, cachefiles_obj_put_detach); +} + +/* + * Invalidate the storage associated with a cookie. + */ +static bool cachefiles_invalidate_cookie(struct fscache_cookie *cookie) +{ + struct cachefiles_object *object = cookie->cache_priv; + struct file *new_file, *old_file; + bool old_tmpfile; + + _enter("o=%x,[%llu]", object->debug_id, object->cookie->object_size); + + old_tmpfile = test_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags); + + if (!object->file) { + fscache_resume_after_invalidation(cookie); + _leave(" = t [light]"); + return true; + } + + new_file = cachefiles_create_tmpfile(object); + if (IS_ERR(new_file)) + goto failed; + + /* Substitute the VFS target */ + _debug("sub"); + spin_lock(&object->lock); + + old_file = object->file; + object->file = new_file; + object->content_info = CACHEFILES_CONTENT_NO_DATA; + set_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags); + set_bit(FSCACHE_COOKIE_NEEDS_UPDATE, &object->cookie->flags); + + spin_unlock(&object->lock); + _debug("subbed"); + + /* Allow I/O to take place again */ + fscache_resume_after_invalidation(cookie); + + if (old_file) { + if (!old_tmpfile) { + struct cachefiles_volume *volume = object->volume; + struct dentry *fan = volume->fanout[(u8)cookie->key_hash]; + struct dentry *obj; + + obj = start_removing_dentry(fan, old_file->f_path.dentry); + if (!IS_ERR(obj)) + cachefiles_bury_object(volume->cache, object, + fan, obj, + FSCACHE_OBJECT_INVALIDATED); + } + fput(old_file); + } + + _leave(" = t"); + return true; + +failed: + _leave(" = f"); + return false; +} + +const struct fscache_cache_ops cachefiles_cache_ops = { + .name = "cachefiles", + .acquire_volume = cachefiles_acquire_volume, + .free_volume = cachefiles_free_volume, + .lookup_cookie = cachefiles_lookup_cookie, + .withdraw_cookie = cachefiles_withdraw_cookie, + .invalidate_cookie = cachefiles_invalidate_cookie, + .begin_operation = cachefiles_begin_operation, + .resize_cookie = cachefiles_resize_cookie, + .prepare_to_write = cachefiles_prepare_to_write, +}; -- cgit v1.3.1