x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 179 insertions, 0 deletions

diff --git a/fs/afs/protocol_yfs.h b/fs/afs/protocol_yfs.h
new file mode 100644
index 000000000..b2f06c191
--- /dev/null
+++ b/fs/afs/protocol_yfs.h
@@ -0,0 +1,179 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/* YFS protocol bits
+ *
+ * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#define YFS_FS_SERVICE	2500
+#define YFS_CM_SERVICE	2501
+
+#define YFSCBMAX	1024
+
+enum YFS_CM_Operations {
+	YFSCBProbe		= 206,	/* probe client */
+	YFSCBGetLock		= 207,	/* get contents of CM lock table */
+	YFSCBXStatsVersion	= 209,	/* get version of extended statistics */
+	YFSCBGetXStats		= 210,	/* get contents of extended statistics data */
+	YFSCBInitCallBackState3	= 213,	/* initialise callback state, version 3 */
+	YFSCBProbeUuid		= 214,	/* check the client hasn't rebooted */
+	YFSCBGetServerPrefs	= 215,
+	YFSCBGetCellServDV	= 216,
+	YFSCBGetLocalCell	= 217,
+	YFSCBGetCacheConfig	= 218,
+	YFSCBGetCellByNum	= 65537,
+	YFSCBTellMeAboutYourself = 65538, /* get client capabilities */
+	YFSCBCallBack		= 64204,
+};
+
+enum YFS_FS_Operations {
+	YFSFETCHACL		= 64131, /* YFS Fetch file AFS3 ACL */
+	YFSFETCHSTATUS		= 64132, /* YFS Fetch file status */
+	YFSSTOREACL		= 64134, /* YFS Store file AFS3 ACL */
+	YFSSTORESTATUS		= 64135, /* YFS Store file status */
+	YFSREMOVEFILE		= 64136, /* YFS Remove a file */
+	YFSCREATEFILE		= 64137, /* YFS Create a file */
+	YFSRENAME		= 64138, /* YFS Rename or move a file or directory */
+	YFSSYMLINK		= 64139, /* YFS Create a symbolic link */
+	YFSLINK			= 64140, /* YFS Create a hard link */
+	YFSMAKEDIR		= 64141, /* YFS Create a directory */
+	YFSREMOVEDIR		= 64142, /* YFS Remove a directory */
+	YFSGETVOLUMESTATUS	= 64149, /* YFS Get volume status information */
+	YFSSETVOLUMESTATUS	= 64150, /* YFS Set volume status information */
+	YFSSETLOCK		= 64156, /* YFS Request a file lock */
+	YFSEXTENDLOCK		= 64157, /* YFS Extend a file lock */
+	YFSRELEASELOCK		= 64158, /* YFS Release a file lock */
+	YFSLOOKUP		= 64161, /* YFS lookup file in directory */
+	YFSFLUSHCPS		= 64165,
+	YFSFETCHOPAQUEACL	= 64168, /* YFS Fetch file YFS ACL */
+	YFSWHOAMI		= 64170,
+	YFSREMOVEACL		= 64171,
+	YFSREMOVEFILE2		= 64173,
+	YFSSTOREOPAQUEACL2	= 64174,
+	YFSRENAME_REPLACE	= 64176,
+	YFSRENAME_NOREPLACE	= 64177,
+	YFSRENAME_EXCHANGE	= 64187,
+	YFSINLINEBULKSTATUS	= 64536, /* YFS Fetch multiple file statuses with errors */
+	YFSFETCHDATA64		= 64537, /* YFS Fetch file data */
+	YFSSTOREDATA64		= 64538, /* YFS Store file data */
+	YFSUPDATESYMLINK	= 64540,
+};
+
+struct yfs_xdr_u64 {
+	__be32			msw;
+	__be32			lsw;
+} __packed;
+
+static inline u64 xdr_to_u64(const struct yfs_xdr_u64 x)
+{
+	return ((u64)ntohl(x.msw) << 32) | ntohl(x.lsw);
+}
+
+static inline struct yfs_xdr_u64 u64_to_xdr(const u64 x)
+{
+	return (struct yfs_xdr_u64){ .msw = htonl(x >> 32), .lsw = htonl(x) };
+}
+
+struct yfs_xdr_vnode {
+	struct yfs_xdr_u64	lo;
+	__be32			hi;
+	__be32			unique;
+} __packed;
+
+struct yfs_xdr_YFSFid {
+	struct yfs_xdr_u64	volume;
+	struct yfs_xdr_vnode	vnode;
+} __packed;
+
+
+struct yfs_xdr_YFSFetchStatus {
+	__be32			type;
+	__be32			nlink;
+	struct yfs_xdr_u64	size;
+	struct yfs_xdr_u64	data_version;
+	struct yfs_xdr_u64	author;
+	struct yfs_xdr_u64	owner;
+	struct yfs_xdr_u64	group;
+	__be32			mode;
+	__be32			caller_access;
+	__be32			anon_access;
+	struct yfs_xdr_vnode	parent;
+	__be32			data_access_protocol;
+	struct yfs_xdr_u64	mtime_client;
+	struct yfs_xdr_u64	mtime_server;
+	__be32			lock_count;
+	__be32			abort_code;
+} __packed;
+
+struct yfs_xdr_YFSCallBack {
+	__be32			version;
+	struct yfs_xdr_u64	expiration_time;
+	__be32			type;
+} __packed;
+
+struct yfs_xdr_YFSStoreStatus {
+	__be32			mask;
+	__be32			mode;
+	struct yfs_xdr_u64	mtime_client;
+	struct yfs_xdr_u64	owner;
+	struct yfs_xdr_u64	group;
+} __packed;
+
+struct yfs_xdr_RPCFlags {
+	__be32			rpc_flags;
+} __packed;
+
+struct yfs_xdr_YFSVolSync {
+	struct yfs_xdr_u64	vol_creation_date;
+	struct yfs_xdr_u64	vol_update_date;
+	struct yfs_xdr_u64	max_quota;
+	struct yfs_xdr_u64	blocks_in_use;
+	struct yfs_xdr_u64	blocks_avail;
+} __packed;
+
+enum yfs_volume_type {
+	yfs_volume_type_ro = 0,
+	yfs_volume_type_rw = 1,
+};
+
+#define yfs_FVSOnline		0x1
+#define yfs_FVSInservice	0x2
+#define yfs_FVSBlessed		0x4
+#define yfs_FVSNeedsSalvage	0x8
+
+struct yfs_xdr_YFSFetchVolumeStatus {
+	struct yfs_xdr_u64	vid;
+	struct yfs_xdr_u64	parent_id;
+	__be32			flags;
+	__be32			type;
+	struct yfs_xdr_u64	max_quota;
+	struct yfs_xdr_u64	blocks_in_use;
+	struct yfs_xdr_u64	part_blocks_avail;
+	struct yfs_xdr_u64	part_max_blocks;
+	struct yfs_xdr_u64	vol_copy_date;
+	struct yfs_xdr_u64	vol_backup_date;
+} __packed;
+
+struct yfs_xdr_YFSStoreVolumeStatus {
+	__be32			mask;
+	struct yfs_xdr_u64	min_quota;
+	struct yfs_xdr_u64	max_quota;
+	struct yfs_xdr_u64	file_quota;
+} __packed;
+
+enum yfs_lock_type {
+	yfs_LockNone		= -1,
+	yfs_LockRead		= 0,
+	yfs_LockWrite		= 1,
+	yfs_LockExtend		= 2,
+	yfs_LockRelease		= 3,
+	yfs_LockMandatoryRead	= 0x100,
+	yfs_LockMandatoryWrite	= 0x101,
+	yfs_LockMandatoryExtend	= 0x102,
+};
+
+/* RXYFS Viced Capability Flags */
+#define YFS_VICED_CAPABILITY_ERRORTRANS		0x0001 /* Deprecated v0.195 */
+#define YFS_VICED_CAPABILITY_64BITFILES		0x0002 /* Deprecated v0.195 */
+#define YFS_VICED_CAPABILITY_WRITELOCKACL	0x0004 /* Can lock a file even without lock perm */
+#define YFS_VICED_CAPABILITY_SANEACLS		0x0008 /* Deprecated v0.195 */