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, 93 insertions, 0 deletions

diff --git a/tools/include/nolibc/compiler.h b/tools/include/nolibc/compiler.h
new file mode 100644
index 000000000..f2d7a81d0
--- /dev/null
+++ b/tools/include/nolibc/compiler.h
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: LGPL-2.1 OR MIT */
+/*
+ * NOLIBC compiler support header
+ * Copyright (C) 2023 Thomas Weißschuh <linux@weissschuh.net>
+ */
+#ifndef _NOLIBC_COMPILER_H
+#define _NOLIBC_COMPILER_H
+
+#if defined(__has_attribute)
+#  define __nolibc_has_attribute(attr) __has_attribute(attr)
+#else
+#  define __nolibc_has_attribute(attr) 0
+#endif
+
+#if defined(__has_feature)
+#  define __nolibc_has_feature(feature) __has_feature(feature)
+#else
+#  define __nolibc_has_feature(feature) 0
+#endif
+
+#define __nolibc_aligned(alignment) __attribute__((aligned(alignment)))
+#define __nolibc_aligned_as(type) __nolibc_aligned(__alignof__(type))
+
+#if __nolibc_has_attribute(naked)
+#  define __nolibc_entrypoint __attribute__((naked))
+#  define __nolibc_entrypoint_epilogue()
+#else
+#  define __nolibc_entrypoint __attribute__((optimize("Os", "omit-frame-pointer")))
+#  define __nolibc_entrypoint_epilogue() __builtin_unreachable()
+#endif /* __nolibc_has_attribute(naked) */
+
+#if defined(__SSP__) || defined(__SSP_STRONG__) || defined(__SSP_ALL__) || defined(__SSP_EXPLICIT__)
+
+#define _NOLIBC_STACKPROTECTOR
+
+#endif /* defined(__SSP__) ... */
+
+#if __nolibc_has_attribute(no_stack_protector)
+#  define __nolibc_no_stack_protector __attribute__((no_stack_protector))
+#else
+#  define __nolibc_no_stack_protector __attribute__((__optimize__("-fno-stack-protector")))
+#endif /* __nolibc_has_attribute(no_stack_protector) */
+
+#if __nolibc_has_attribute(__fallthrough__)
+#  define __nolibc_fallthrough do { } while (0); __attribute__((__fallthrough__))
+#else
+#  define __nolibc_fallthrough do { } while (0)
+#endif /* __nolibc_has_attribute(fallthrough) */
+
+#if defined(__STDC_VERSION__)
+#  define __nolibc_stdc_version __STDC_VERSION__
+#else
+#  define __nolibc_stdc_version 0
+#endif
+
+#define __nolibc_version(_major, _minor, _patch) ((_major) * 10000 + (_minor) * 100 + (_patch))
+
+#ifdef __GNUC__
+#  define __nolibc_gnuc_version \
+		__nolibc_version(__GNUC__, __GNUC_MINOR__, __GNUC_PATCHLEVEL__)
+#else
+#  define __nolibc_gnuc_version 0
+#endif /* __GNUC__ */
+
+#ifdef __clang__
+#  define __nolibc_clang_version \
+		__nolibc_version(__clang_major__, __clang_minor__, __clang_patchlevel__)
+#else
+#  define __nolibc_clang_version 0
+#endif /* __clang__ */
+
+#if __nolibc_stdc_version >= 201112L || \
+	__nolibc_gnuc_version >= __nolibc_version(4, 6, 0) || \
+	__nolibc_clang_version >= __nolibc_version(3, 0, 0)
+#  define __nolibc_static_assert(_t) _Static_assert(_t, "")
+#else
+#  define __nolibc_static_assert(_t)
+#endif
+
+/* Make the optimizer believe the variable can be manipulated arbitrarily. */
+#define _NOLIBC_OPTIMIZER_HIDE_VAR(var)	__asm__ ("" : "+r" (var))
+
+#if __nolibc_has_feature(undefined_behavior_sanitizer)
+#  if defined(__clang__)
+#    define __nolibc_no_sanitize_undefined __attribute__((no_sanitize("function")))
+#  else
+#    define __nolibc_no_sanitize_undefined __attribute__((no_sanitize_undefined))
+#  endif
+#else
+#  define __nolibc_no_sanitize_undefined
+#endif
+
+#endif /* _NOLIBC_COMPILER_H */