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, 109 insertions, 0 deletions

diff --git a/fs/jfs/jfs_debug.h b/fs/jfs/jfs_debug.h
new file mode 100644
index 000000000..48e2150c0
--- /dev/null
+++ b/fs/jfs/jfs_debug.h
@@ -0,0 +1,109 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ *   Copyright (C) International Business Machines Corp., 2000-2002
+ *   Portions Copyright (C) Christoph Hellwig, 2001-2002
+ */
+#ifndef _H_JFS_DEBUG
+#define _H_JFS_DEBUG
+
+/*
+ *	jfs_debug.h
+ *
+ * global debug message, data structure/macro definitions
+ * under control of CONFIG_JFS_DEBUG, CONFIG_JFS_STATISTICS;
+ */
+
+/*
+ * Create /proc/fs/jfs if procfs is enabled andeither
+ * CONFIG_JFS_DEBUG or CONFIG_JFS_STATISTICS is defined
+ */
+#if defined(CONFIG_PROC_FS) && (defined(CONFIG_JFS_DEBUG) || defined(CONFIG_JFS_STATISTICS))
+#define PROC_FS_JFS
+extern void jfs_proc_init(void);
+extern void jfs_proc_clean(void);
+#endif
+
+/*
+ *	assert with traditional printf/panic
+ */
+#define assert(p) do {	\
+	if (!(p)) {	\
+		printk(KERN_CRIT "BUG at %s:%d assert(%s)\n",	\
+		       __FILE__, __LINE__, #p);			\
+		BUG();	\
+	}		\
+} while (0)
+
+/*
+ *	debug ON
+ *	--------
+ */
+#ifdef CONFIG_JFS_DEBUG
+#define ASSERT(p) assert(p)
+
+/* printk verbosity */
+#define JFS_LOGLEVEL_ERR 1
+#define JFS_LOGLEVEL_WARN 2
+#define JFS_LOGLEVEL_DEBUG 3
+#define JFS_LOGLEVEL_INFO 4
+
+extern int jfsloglevel;
+
+int jfs_txanchor_proc_show(struct seq_file *m, void *v);
+
+/* information message: e.g., configuration, major event */
+#define jfs_info(fmt, arg...) do {			\
+	if (jfsloglevel >= JFS_LOGLEVEL_INFO)		\
+		printk(KERN_INFO fmt "\n", ## arg);	\
+} while (0)
+
+/* debug message: ad hoc */
+#define jfs_debug(fmt, arg...) do {			\
+	if (jfsloglevel >= JFS_LOGLEVEL_DEBUG)		\
+		printk(KERN_DEBUG fmt "\n", ## arg);	\
+} while (0)
+
+/* warn message: */
+#define jfs_warn(fmt, arg...) do {			\
+	if (jfsloglevel >= JFS_LOGLEVEL_WARN)		\
+		printk(KERN_WARNING fmt "\n", ## arg);	\
+} while (0)
+
+/* error event message: e.g., i/o error */
+#define jfs_err(fmt, arg...) do {			\
+	if (jfsloglevel >= JFS_LOGLEVEL_ERR)		\
+		printk(KERN_ERR fmt "\n", ## arg);	\
+} while (0)
+
+/*
+ *	debug OFF
+ *	---------
+ */
+#else				/* CONFIG_JFS_DEBUG */
+#define ASSERT(p) do {} while (0)
+#define jfs_info(fmt, arg...) do {} while (0)
+#define jfs_debug(fmt, arg...) do {} while (0)
+#define jfs_warn(fmt, arg...) do {} while (0)
+#define jfs_err(fmt, arg...) do {} while (0)
+#endif				/* CONFIG_JFS_DEBUG */
+
+/*
+ *	statistics
+ *	----------
+ */
+#ifdef	CONFIG_JFS_STATISTICS
+int jfs_lmstats_proc_show(struct seq_file *m, void *v);
+int jfs_txstats_proc_show(struct seq_file *m, void *v);
+int jfs_mpstat_proc_show(struct seq_file *m, void *v);
+int jfs_xtstat_proc_show(struct seq_file *m, void *v);
+
+#define	INCREMENT(x)		((x)++)
+#define	DECREMENT(x)		((x)--)
+#define	HIGHWATERMARK(x,y)	((x) = max((x), (y)))
+#else
+#define	INCREMENT(x)
+#define	DECREMENT(x)
+#define	HIGHWATERMARK(x,y)
+#endif				/* CONFIG_JFS_STATISTICS */
+
+#endif				/* _H_JFS_DEBUG */