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, 86 insertions, 0 deletions

diff --git a/Documentation/fb/internals.rst b/Documentation/fb/internals.rst
new file mode 100644
index 000000000..696b50aa7
--- /dev/null
+++ b/Documentation/fb/internals.rst
@@ -0,0 +1,86 @@
+=============================
+Frame Buffer device internals
+=============================
+
+This is a first start for some documentation about frame buffer device
+internals.
+
+Authors:
+
+- Geert Uytterhoeven <geert@linux-m68k.org>, 21 July 1998
+- James Simmons <jsimmons@user.sf.net>, Nov 26 2002
+
+--------------------------------------------------------------------------------
+
+Structures used by the frame buffer device API
+==============================================
+
+The following structures play a role in the game of frame buffer devices. They
+are defined in <linux/fb.h>.
+
+1. Outside the kernel (user space)
+
+  - struct fb_fix_screeninfo
+
+    Device independent unchangeable information about a frame buffer device and
+    a specific video mode. This can be obtained using the FBIOGET_FSCREENINFO
+    ioctl.
+
+  - struct fb_var_screeninfo
+
+    Device independent changeable information about a frame buffer device and a
+    specific video mode. This can be obtained using the FBIOGET_VSCREENINFO
+    ioctl, and updated with the FBIOPUT_VSCREENINFO ioctl. If you want to pan
+    the screen only, you can use the FBIOPAN_DISPLAY ioctl.
+
+  - struct fb_cmap
+
+    Device independent colormap information. You can get and set the colormap
+    using the FBIOGETCMAP and FBIOPUTCMAP ioctls.
+
+
+2. Inside the kernel
+
+  - struct fb_info
+
+    Generic information, API and low level information about a specific frame
+    buffer device instance (slot number, board address, ...).
+
+  - struct `par`
+
+    Device dependent information that uniquely defines the video mode for this
+    particular piece of hardware.
+
+
+Visuals used by the frame buffer device API
+===========================================
+
+
+Monochrome (FB_VISUAL_MONO01 and FB_VISUAL_MONO10)
+--------------------------------------------------
+Each pixel is either black or white.
+
+
+Pseudo color (FB_VISUAL_PSEUDOCOLOR and FB_VISUAL_STATIC_PSEUDOCOLOR)
+---------------------------------------------------------------------
+The whole pixel value is fed through a programmable lookup table that has one
+color (including red, green, and blue intensities) for each possible pixel
+value, and that color is displayed.
+
+
+True color (FB_VISUAL_TRUECOLOR)
+--------------------------------
+The pixel value is broken up into red, green, and blue fields.
+
+
+Direct color (FB_VISUAL_DIRECTCOLOR)
+------------------------------------
+The pixel value is broken up into red, green, and blue fields, each of which
+are looked up in separate red, green, and blue lookup tables.
+
+
+Grayscale displays
+------------------
+Grayscale and static grayscale are special variants of pseudo color and static
+pseudo color, where the red, green and blue components are always equal to
+each other.