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, 122 insertions, 0 deletions

diff --git a/fs/smb/server/xattr.h b/fs/smb/server/xattr.h
new file mode 100644
index 000000000..505101a81
--- /dev/null
+++ b/fs/smb/server/xattr.h
@@ -0,0 +1,122 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ *   Copyright (C) 2021 Samsung Electronics Co., Ltd.
+ */
+
+#ifndef __XATTR_H__
+#define __XATTR_H__
+
+/*
+ * These are on-disk structures to store additional metadata into xattr to
+ * reproduce windows filesystem semantics. And they are encoded with NDR to
+ * compatible with samba's xattr meta format. The compatibility with samba
+ * is important because it can lose the information(file attribute,
+ * creation time, acls) about the existing files when switching between
+ * ksmbd and samba.
+ */
+
+/*
+ * Dos attribute flags used for what variable is valid.
+ */
+enum {
+	XATTR_DOSINFO_ATTRIB		= 0x00000001,
+	XATTR_DOSINFO_EA_SIZE		= 0x00000002,
+	XATTR_DOSINFO_SIZE		= 0x00000004,
+	XATTR_DOSINFO_ALLOC_SIZE	= 0x00000008,
+	XATTR_DOSINFO_CREATE_TIME	= 0x00000010,
+	XATTR_DOSINFO_CHANGE_TIME	= 0x00000020,
+	XATTR_DOSINFO_ITIME		= 0x00000040
+};
+
+/*
+ * Dos attribute structure which is compatible with samba's one.
+ * Storing it into the xattr named "DOSATTRIB" separately from inode
+ * allows ksmbd to faithfully reproduce windows filesystem semantics
+ * on top of a POSIX filesystem.
+ */
+struct xattr_dos_attrib {
+	__u16	version;	/* version 3 or version 4 */
+	__u32	flags;		/* valid flags */
+	__u32	attr;		/* Dos attribute */
+	__u32	ea_size;	/* EA size */
+	__u64	size;
+	__u64	alloc_size;
+	__u64	create_time;	/* File creation time */
+	__u64	change_time;	/* File change time */
+	__u64	itime;		/* Invented/Initial time */
+};
+
+/*
+ * Enumeration is used for computing posix acl hash.
+ */
+enum {
+	SMB_ACL_TAG_INVALID = 0,
+	SMB_ACL_USER,
+	SMB_ACL_USER_OBJ,
+	SMB_ACL_GROUP,
+	SMB_ACL_GROUP_OBJ,
+	SMB_ACL_OTHER,
+	SMB_ACL_MASK
+};
+
+#define SMB_ACL_READ			4
+#define SMB_ACL_WRITE			2
+#define SMB_ACL_EXECUTE			1
+
+struct xattr_acl_entry {
+	int type;
+	uid_t uid;
+	gid_t gid;
+	mode_t perm;
+};
+
+/*
+ * xattr_smb_acl structure is used for computing posix acl hash.
+ */
+struct xattr_smb_acl {
+	int count;
+	int next;
+	struct xattr_acl_entry entries[] __counted_by(count);
+};
+
+/* 64bytes hash in xattr_ntacl is computed with sha256 */
+#define XATTR_SD_HASH_TYPE_SHA256	0x1
+#define XATTR_SD_HASH_SIZE		64
+
+/*
+ * xattr_ntacl is used for storing ntacl and hashes.
+ * Hash is used for checking valid posix acl and ntacl in xattr.
+ */
+struct xattr_ntacl {
+	__u16	version; /* version 4*/
+	void	*sd_buf;
+	__u32	sd_size;
+	__u16	hash_type; /* hash type */
+	__u8	desc[10]; /* posix_acl description */
+	__u16	desc_len;
+	__u64	current_time;
+	__u8	hash[XATTR_SD_HASH_SIZE]; /* 64bytes hash for ntacl */
+	__u8	posix_acl_hash[XATTR_SD_HASH_SIZE]; /* 64bytes hash for posix acl */
+};
+
+/* DOS ATTRIBUTE XATTR PREFIX */
+#define DOS_ATTRIBUTE_PREFIX		"DOSATTRIB"
+#define DOS_ATTRIBUTE_PREFIX_LEN	(sizeof(DOS_ATTRIBUTE_PREFIX) - 1)
+#define XATTR_NAME_DOS_ATTRIBUTE	(XATTR_USER_PREFIX DOS_ATTRIBUTE_PREFIX)
+#define XATTR_NAME_DOS_ATTRIBUTE_LEN	\
+		(sizeof(XATTR_USER_PREFIX DOS_ATTRIBUTE_PREFIX) - 1)
+
+/* STREAM XATTR PREFIX */
+#define STREAM_PREFIX			"DosStream."
+#define STREAM_PREFIX_LEN		(sizeof(STREAM_PREFIX) - 1)
+#define XATTR_NAME_STREAM		(XATTR_USER_PREFIX STREAM_PREFIX)
+#define XATTR_NAME_STREAM_LEN		(sizeof(XATTR_NAME_STREAM) - 1)
+
+/* SECURITY DESCRIPTOR(NTACL) XATTR PREFIX */
+#define SD_PREFIX			"NTACL"
+#define SD_PREFIX_LEN	(sizeof(SD_PREFIX) - 1)
+#define XATTR_NAME_SD	(XATTR_SECURITY_PREFIX SD_PREFIX)
+#define XATTR_NAME_SD_LEN	\
+		(sizeof(XATTR_SECURITY_PREFIX SD_PREFIX) - 1)
+
+#endif /* __XATTR_H__ */