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, 137 insertions, 0 deletions

diff --git a/fs/orangefs/downcall.h b/fs/orangefs/downcall.h
new file mode 100644
index 000000000..ea2332e16
--- /dev/null
+++ b/fs/orangefs/downcall.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * (C) 2001 Clemson University and The University of Chicago
+ *
+ * See COPYING in top-level directory.
+ */
+
+/*
+ *  Definitions of downcalls used in Linux kernel module.
+ */
+
+#ifndef __DOWNCALL_H
+#define __DOWNCALL_H
+
+/*
+ * Sanitized the device-client core interaction
+ * for clean 32-64 bit usage
+ */
+struct orangefs_io_response {
+	__s64 amt_complete;
+};
+
+struct orangefs_lookup_response {
+	struct orangefs_object_kref refn;
+};
+
+struct orangefs_create_response {
+	struct orangefs_object_kref refn;
+};
+
+struct orangefs_symlink_response {
+	struct orangefs_object_kref refn;
+};
+
+struct orangefs_getattr_response {
+	struct ORANGEFS_sys_attr_s attributes;
+	char link_target[ORANGEFS_NAME_MAX];
+};
+
+struct orangefs_mkdir_response {
+	struct orangefs_object_kref refn;
+};
+
+struct orangefs_statfs_response {
+	__s64 block_size;
+	__s64 blocks_total;
+	__s64 blocks_avail;
+	__s64 files_total;
+	__s64 files_avail;
+};
+
+struct orangefs_fs_mount_response {
+	__s32 fs_id;
+	__s32 id;
+	struct orangefs_khandle root_khandle;
+};
+
+/* the getxattr response is the attribute value */
+struct orangefs_getxattr_response {
+	__s32 val_sz;
+	__s32 __pad1;
+	char val[ORANGEFS_MAX_XATTR_VALUELEN];
+};
+
+/* the listxattr response is an array of attribute names */
+struct orangefs_listxattr_response {
+	__s32 returned_count;
+	__s32 __pad1;
+	__u64 token;
+	char key[ORANGEFS_MAX_XATTR_LISTLEN * ORANGEFS_MAX_XATTR_NAMELEN];
+	__s32 keylen;
+	__s32 __pad2;
+	__s32 lengths[ORANGEFS_MAX_XATTR_LISTLEN];
+};
+
+struct orangefs_param_response {
+	union {
+		__s64 value64;
+		__s32 value32[2];
+	} u;
+};
+
+#define PERF_COUNT_BUF_SIZE 4096
+struct orangefs_perf_count_response {
+	char buffer[PERF_COUNT_BUF_SIZE];
+};
+
+#define FS_KEY_BUF_SIZE 4096
+struct orangefs_fs_key_response {
+	__s32 fs_keylen;
+	__s32 __pad1;
+	char fs_key[FS_KEY_BUF_SIZE];
+};
+
+/* 2.9.6 */
+struct orangefs_features_response {
+	__u64 features;
+};
+
+struct orangefs_downcall_s {
+	__s32 type;
+	__s32 status;
+	/* currently trailer is used only by readdir */
+	__s64 trailer_size;
+	char *trailer_buf;
+
+	union {
+		struct orangefs_io_response io;
+		struct orangefs_lookup_response lookup;
+		struct orangefs_create_response create;
+		struct orangefs_symlink_response sym;
+		struct orangefs_getattr_response getattr;
+		struct orangefs_mkdir_response mkdir;
+		struct orangefs_statfs_response statfs;
+		struct orangefs_fs_mount_response fs_mount;
+		struct orangefs_getxattr_response getxattr;
+		struct orangefs_listxattr_response listxattr;
+		struct orangefs_param_response param;
+		struct orangefs_perf_count_response perf_count;
+		struct orangefs_fs_key_response fs_key;
+		struct orangefs_features_response features;
+	} resp;
+};
+
+/*
+ * The readdir response comes in the trailer.  It is followed by the
+ * directory entries as described in dir.c.
+ */
+
+struct orangefs_readdir_response_s {
+	__u64 token;
+	__u64 directory_version;
+	__u32 __pad2;
+	__u32 orangefs_dirent_outcount;
+};
+
+#endif /* __DOWNCALL_H */