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, 53 insertions, 0 deletions

diff --git a/Documentation/admin-guide/cifs/introduction.rst b/Documentation/admin-guide/cifs/introduction.rst
new file mode 100644
index 000000000..ffc6e2564
--- /dev/null
+++ b/Documentation/admin-guide/cifs/introduction.rst
@@ -0,0 +1,53 @@
+============
+Introduction
+============
+
+  This is the client VFS module for the SMB3 NAS protocol as well
+  as for older dialects such as the Common Internet File System (CIFS)
+  protocol which was the successor to the Server Message Block
+  (SMB) protocol, the native file sharing mechanism for most early
+  PC operating systems. New and improved versions of CIFS are now
+  called SMB2 and SMB3. Use of SMB3 (and later, including SMB3.1.1
+  the most current dialect) is strongly preferred over using older
+  dialects like CIFS due to security reasons. All modern dialects,
+  including the most recent, SMB3.1.1, are supported by the CIFS VFS
+  module. The SMB3 protocol is implemented and supported by all major
+  file servers such as Windows (including Windows 2019 Server), as
+  well as by Samba (which provides excellent CIFS/SMB2/SMB3 server
+  support and tools for Linux and many other operating systems).
+  Apple systems also support SMB3 well, as do most Network Attached
+  Storage vendors, so this network filesystem client can mount to a
+  wide variety of systems. It also supports mounting to the cloud
+  (for example Microsoft Azure), including the necessary security
+  features.
+
+  The intent of this module is to provide the most advanced network
+  file system function for SMB3 compliant servers, including advanced
+  security features, excellent parallelized high performance i/o, better
+  POSIX compliance, secure per-user session establishment, encryption,
+  high performance safe distributed caching (leases/oplocks), optional packet
+  signing, large files, Unicode support and other internationalization
+  improvements. Since both Samba server and this filesystem client support the
+  CIFS Unix extensions, and the Linux client also supports SMB3 POSIX extensions,
+  the combination can provide a reasonable alternative to other network and
+  cluster file systems for fileserving in some Linux to Linux environments,
+  not just in Linux to Windows (or Linux to Mac) environments.
+
+  This filesystem has a mount utility (mount.cifs) and various user space
+  tools (including smbinfo and setcifsacl) that can be obtained from
+
+      https://git.samba.org/?p=cifs-utils.git
+
+  or
+
+      git://git.samba.org/cifs-utils.git
+
+  mount.cifs should be installed in the directory with the other mount helpers.
+
+  For more information on the module see the project wiki page at
+
+      https://wiki.samba.org/index.php/LinuxCIFS
+
+  and
+
+      https://wiki.samba.org/index.php/LinuxCIFS_utils