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, 70 insertions, 0 deletions

diff --git a/Documentation/input/devices/bcm5974.rst b/Documentation/input/devices/bcm5974.rst
new file mode 100644
index 000000000..4aca199b0
--- /dev/null
+++ b/Documentation/input/devices/bcm5974.rst
@@ -0,0 +1,70 @@
+.. include:: <isonum.txt>
+
+------------------------
+BCM5974 Driver (bcm5974)
+------------------------
+
+:Copyright: |copy| 2008-2009	Henrik Rydberg <rydberg@euromail.se>
+
+The USB initialization and package decoding was made by Scott Shawcroft as
+part of the touchd user-space driver project:
+
+:Copyright: |copy| 2008	Scott Shawcroft (scott.shawcroft@gmail.com)
+
+The BCM5974 driver is based on the appletouch driver:
+
+:Copyright: |copy| 2001-2004	Greg Kroah-Hartman (greg@kroah.com)
+:Copyright: |copy| 2005		Johannes Berg (johannes@sipsolutions.net)
+:Copyright: |copy| 2005		Stelian Pop (stelian@popies.net)
+:Copyright: |copy| 2005		Frank Arnold (frank@scirocco-5v-turbo.de)
+:Copyright: |copy| 2005		Peter Osterlund (petero2@telia.com)
+:Copyright: |copy| 2005		Michael Hanselmann (linux-kernel@hansmi.ch)
+:Copyright: |copy| 2006		Nicolas Boichat (nicolas@boichat.ch)
+
+This driver adds support for the multi-touch trackpad on the new Apple
+Macbook Air and Macbook Pro laptops. It replaces the appletouch driver on
+those computers, and integrates well with the synaptics driver of the Xorg
+system.
+
+Known to work on Macbook Air, Macbook Pro Penryn and the new unibody
+Macbook 5 and Macbook Pro 5.
+
+Usage
+-----
+
+The driver loads automatically for the supported usb device ids, and
+becomes available both as an event device (/dev/input/event*) and as a
+mouse via the mousedev driver (/dev/input/mice).
+
+USB Race
+--------
+
+The Apple multi-touch trackpads report both mouse and keyboard events via
+different interfaces of the same usb device. This creates a race condition
+with the HID driver, which, if not told otherwise, will find the standard
+HID mouse and keyboard, and claim the whole device. To remedy, the usb
+product id must be listed in the mouse_ignore list of the hid driver.
+
+Debug output
+------------
+
+To ease the development for new hardware version, verbose packet output can
+be switched on with the debug kernel module parameter. The range [1-9]
+yields different levels of verbosity. Example (as root)::
+
+    echo -n 9 > /sys/module/bcm5974/parameters/debug
+
+    tail -f /var/log/debug
+
+    echo -n 0 > /sys/module/bcm5974/parameters/debug
+
+Trivia
+------
+
+The driver was developed at the ubuntu forums in June 2008 [#f1]_, and now has
+a more permanent home at bitmath.org [#f2]_.
+
+.. Links
+
+.. [#f1] http://ubuntuforums.org/showthread.php?t=840040
+.. [#f2] http://bitmath.org/code/