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, 94 insertions, 0 deletions

diff --git a/fs/ocfs2/blockcheck.h b/fs/ocfs2/blockcheck.h
new file mode 100644
index 000000000..d0578e98e
--- /dev/null
+++ b/fs/ocfs2/blockcheck.h
@@ -0,0 +1,94 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * blockcheck.h
+ *
+ * Checksum and ECC codes for the OCFS2 userspace library.
+ *
+ * Copyright (C) 2004, 2008 Oracle.  All rights reserved.
+ */
+
+#ifndef OCFS2_BLOCKCHECK_H
+#define OCFS2_BLOCKCHECK_H
+
+
+/* Count errors and error correction from blockcheck.c */
+struct ocfs2_blockcheck_stats {
+	spinlock_t b_lock;
+	u64 b_check_count;	/* Number of blocks we've checked */
+	u64 b_failure_count;	/* Number of failed checksums */
+	u64 b_recover_count;	/* Number of blocks fixed by ecc */
+
+	/*
+	 * debugfs entries, used if this is passed to
+	 * ocfs2_blockcheck_stats_debugfs_install()
+	 */
+	struct dentry *b_debug_dir;	/* Parent of the debugfs  files */
+};
+
+
+/* High level block API */
+void ocfs2_compute_meta_ecc(struct super_block *sb, void *data,
+			    struct ocfs2_block_check *bc);
+int ocfs2_validate_meta_ecc(struct super_block *sb, void *data,
+			    struct ocfs2_block_check *bc);
+void ocfs2_compute_meta_ecc_bhs(struct super_block *sb,
+				struct buffer_head **bhs, int nr,
+				struct ocfs2_block_check *bc);
+int ocfs2_validate_meta_ecc_bhs(struct super_block *sb,
+				struct buffer_head **bhs, int nr,
+				struct ocfs2_block_check *bc);
+
+/* Lower level API */
+void ocfs2_block_check_compute(void *data, size_t blocksize,
+			       struct ocfs2_block_check *bc);
+int ocfs2_block_check_validate(void *data, size_t blocksize,
+			       struct ocfs2_block_check *bc,
+			       struct ocfs2_blockcheck_stats *stats);
+void ocfs2_block_check_compute_bhs(struct buffer_head **bhs, int nr,
+				   struct ocfs2_block_check *bc);
+int ocfs2_block_check_validate_bhs(struct buffer_head **bhs, int nr,
+				   struct ocfs2_block_check *bc,
+				   struct ocfs2_blockcheck_stats *stats);
+
+/* Debug Initialization */
+void ocfs2_blockcheck_stats_debugfs_install(struct ocfs2_blockcheck_stats *stats,
+					    struct dentry *parent);
+void ocfs2_blockcheck_stats_debugfs_remove(struct ocfs2_blockcheck_stats *stats);
+
+/*
+ * Hamming code functions
+ */
+
+/*
+ * Encoding hamming code parity bits for a buffer.
+ *
+ * This is the low level encoder function.  It can be called across
+ * multiple hunks just like the crc32 code.  'd' is the number of bits
+ * _in_this_hunk_.  nr is the bit offset of this hunk.  So, if you had
+ * two 512B buffers, you would do it like so:
+ *
+ * parity = ocfs2_hamming_encode(0, buf1, 512 * 8, 0);
+ * parity = ocfs2_hamming_encode(parity, buf2, 512 * 8, 512 * 8);
+ *
+ * If you just have one buffer, use ocfs2_hamming_encode_block().
+ */
+u32 ocfs2_hamming_encode(u32 parity, void *data, unsigned int d,
+			 unsigned int nr);
+/*
+ * Fix a buffer with a bit error.  The 'fix' is the original parity
+ * xor'd with the parity calculated now.
+ *
+ * Like ocfs2_hamming_encode(), this can handle hunks.  nr is the bit
+ * offset of the current hunk.  If bit to be fixed is not part of the
+ * current hunk, this does nothing.
+ *
+ * If you only have one buffer, use ocfs2_hamming_fix_block().
+ */
+void ocfs2_hamming_fix(void *data, unsigned int d, unsigned int nr,
+		       unsigned int fix);
+
+/* Convenience wrappers for a single buffer of data */
+extern u32 ocfs2_hamming_encode_block(void *data, unsigned int blocksize);
+extern void ocfs2_hamming_fix_block(void *data, unsigned int blocksize,
+				    unsigned int fix);
+#endif