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, 83 insertions, 0 deletions

diff --git a/arch/csky/abiv2/memset.S b/arch/csky/abiv2/memset.S
new file mode 100644
index 000000000..a7e7d994b
--- /dev/null
+++ b/arch/csky/abiv2/memset.S
@@ -0,0 +1,83 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (C) 2018 Hangzhou C-SKY Microsystems co.,ltd.
+
+#include <linux/linkage.h>
+#include "sysdep.h"
+
+	.weak memset
+ENTRY(__memset)
+ENTRY(memset)
+	/* Test if len less than 4 bytes.  */
+	mov	r12, r0
+	cmplti	r2, 8
+	bt	.L_set_by_byte
+
+	andi	r13, r0, 3
+	movi	r19, 4
+	/* Test if dest is not 4 bytes aligned.  */
+	bnez	r13, .L_dest_not_aligned
+	/* Hardware can handle unaligned access directly.  */
+.L_dest_aligned:
+        zextb   r3, r1
+        lsli    r1, 8
+        or      r1, r3
+        lsli    r3, r1, 16
+        or      r3, r1
+
+	/* If dest is aligned, then copy.  */
+	zext	r18, r2, 31, 4
+	/* Test if len less than 16 bytes.  */
+	bez	r18, .L_len_less_16bytes
+
+	LABLE_ALIGN
+.L_len_larger_16bytes:
+	stw	r3, (r0, 0)
+	stw	r3, (r0, 4)
+	stw	r3, (r0, 8)
+	stw	r3, (r0, 12)
+	PRE_BNEZAD (r18)
+	addi	r0, 16
+	BNEZAD (r18, .L_len_larger_16bytes)
+
+.L_len_less_16bytes:
+	zext	r18, r2, 3, 2
+	andi	r2, 3
+	bez	r18, .L_set_by_byte
+.L_len_less_16bytes_loop:
+	stw	r3, (r0, 0)
+	PRE_BNEZAD (r18)
+	addi	r0, 4
+	BNEZAD (r18, .L_len_less_16bytes_loop)
+
+	/* Test if len less than 4 bytes.  */
+.L_set_by_byte:
+	zext	r18, r2, 2, 0
+	bez	r18, .L_return
+.L_set_by_byte_loop:
+	stb	r1, (r0, 0)
+	PRE_BNEZAD (r18)
+	addi	r0, 1
+	BNEZAD (r18, .L_set_by_byte_loop)
+
+.L_return:
+	mov	r0, r12
+	rts
+
+	/* If dest is not aligned, just set some bytes makes the dest
+	   align.  */
+
+.L_dest_not_aligned:
+	sub	r13, r19, r13
+	sub	r2, r13
+.L_dest_not_aligned_loop:
+	/* Makes the dest align.  */
+	stb	r1, (r0, 0)
+	PRE_BNEZAD (r13)
+	addi	r0, 1
+	BNEZAD (r13, .L_dest_not_aligned_loop)
+	cmplti	r2, 8
+	bt	.L_set_by_byte
+	/* Check whether the src is aligned.  */
+	jbr	.L_dest_aligned
+ENDPROC(memset)
+ENDPROC(__memset)