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, 98 insertions, 0 deletions

diff --git a/arch/um/drivers/mconsole.h b/arch/um/drivers/mconsole.h
new file mode 100644
index 000000000..635637830
--- /dev/null
+++ b/arch/um/drivers/mconsole.h
@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org)
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#ifndef __MCONSOLE_H__
+#define __MCONSOLE_H__
+
+#ifdef __UM_HOST__
+#include <stdint.h>
+#define u32 uint32_t
+#endif
+
+#include <sysdep/ptrace.h>
+
+#define MCONSOLE_MAGIC (0xcafebabe)
+#define MCONSOLE_MAX_DATA (512)
+#define MCONSOLE_VERSION 2
+
+struct mconsole_request {
+	u32 magic;
+	u32 version;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mconsole_reply {
+	u32 err;
+	u32 more;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mconsole_notify {
+	u32 magic;
+	u32 version;
+	enum { MCONSOLE_SOCKET, MCONSOLE_PANIC, MCONSOLE_HANG,
+	       MCONSOLE_USER_NOTIFY } type;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mc_request;
+
+enum mc_context { MCONSOLE_INTR, MCONSOLE_PROC };
+
+struct mconsole_command
+{
+	char *command;
+	void (*handler)(struct mc_request *req);
+	enum mc_context context;
+};
+
+struct mc_request
+{
+	int len;
+	int as_interrupt;
+
+	int originating_fd;
+	unsigned int originlen;
+	unsigned char origin[128];			/* sockaddr_un */
+
+	struct mconsole_request request;
+	struct mconsole_command *cmd;
+	struct uml_pt_regs regs;
+};
+
+extern char mconsole_socket_name[];
+
+extern int mconsole_unlink_socket(void);
+extern int mconsole_reply_len(struct mc_request *req, const char *reply,
+			      int len, int err, int more);
+extern int mconsole_reply(struct mc_request *req, const char *str, int err,
+			  int more);
+
+extern void mconsole_version(struct mc_request *req);
+extern void mconsole_help(struct mc_request *req);
+extern void mconsole_halt(struct mc_request *req);
+extern void mconsole_reboot(struct mc_request *req);
+extern void mconsole_config(struct mc_request *req);
+extern void mconsole_remove(struct mc_request *req);
+extern void mconsole_sysrq(struct mc_request *req);
+extern void mconsole_cad(struct mc_request *req);
+extern void mconsole_stop(struct mc_request *req);
+extern void mconsole_go(struct mc_request *req);
+extern void mconsole_log(struct mc_request *req);
+extern void mconsole_proc(struct mc_request *req);
+extern void mconsole_stack(struct mc_request *req);
+
+extern int mconsole_get_request(int fd, struct mc_request *req);
+extern int mconsole_notify(char *sock_name, int type, const void *data,
+			   int len);
+extern char *mconsole_notify_socket(void);
+extern void lock_notify(void);
+extern void unlock_notify(void);
+
+#endif