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, 86 insertions, 0 deletions

diff --git a/drivers/comedi/comedi_internal.h b/drivers/comedi/comedi_internal.h
new file mode 100644
index 000000000..41a3b09f8
--- /dev/null
+++ b/drivers/comedi/comedi_internal.h
@@ -0,0 +1,86 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _COMEDI_INTERNAL_H
+#define _COMEDI_INTERNAL_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+
+/*
+ * various internal comedi stuff
+ */
+
+struct comedi_buf_map;
+struct comedi_devconfig;
+struct comedi_device;
+struct comedi_insn;
+struct comedi_rangeinfo;
+struct comedi_subdevice;
+struct device;
+
+int do_rangeinfo_ioctl(struct comedi_device *dev,
+		       struct comedi_rangeinfo *it);
+struct comedi_device *comedi_alloc_board_minor(struct device *hardware_device);
+void comedi_release_hardware_device(struct device *hardware_device);
+int comedi_alloc_subdevice_minor(struct comedi_subdevice *s);
+void comedi_free_subdevice_minor(struct comedi_subdevice *s);
+
+int comedi_buf_alloc(struct comedi_device *dev, struct comedi_subdevice *s,
+		     unsigned long new_size);
+void comedi_buf_reset(struct comedi_subdevice *s);
+bool comedi_buf_is_mmapped(struct comedi_subdevice *s);
+void comedi_buf_map_get(struct comedi_buf_map *bm);
+int comedi_buf_map_put(struct comedi_buf_map *bm);
+int comedi_buf_map_access(struct comedi_buf_map *bm, unsigned long offset,
+			  void *buf, int len, int write);
+struct comedi_buf_map *
+comedi_buf_map_from_subdev_get(struct comedi_subdevice *s);
+unsigned int comedi_buf_write_n_available(struct comedi_subdevice *s);
+unsigned int comedi_buf_write_n_allocated(struct comedi_subdevice *s);
+unsigned int _comedi_buf_write_alloc(struct comedi_subdevice *s,
+				     unsigned int nbytes);
+unsigned int _comedi_buf_write_free(struct comedi_subdevice *s,
+				    unsigned int nbytes);
+unsigned int _comedi_buf_read_n_available(struct comedi_subdevice *s);
+unsigned int _comedi_buf_read_alloc(struct comedi_subdevice *s,
+				    unsigned int nbytes);
+unsigned int _comedi_buf_read_free(struct comedi_subdevice *s,
+				   unsigned int nbytes);
+void _comedi_inc_scan_progress(struct comedi_subdevice *s,
+			       unsigned int num_bytes);
+void _comedi_event(struct comedi_device *dev, struct comedi_subdevice *s);
+void comedi_device_cancel_all(struct comedi_device *dev);
+bool comedi_can_auto_free_spriv(struct comedi_subdevice *s);
+
+extern unsigned int comedi_default_buf_size_kb;
+extern unsigned int comedi_default_buf_maxsize_kb;
+
+/* drivers.c */
+
+extern struct comedi_driver *comedi_drivers;
+extern struct mutex comedi_drivers_list_lock;
+
+int insn_inval(struct comedi_device *dev, struct comedi_subdevice *s,
+	       struct comedi_insn *insn, unsigned int *data);
+
+void comedi_device_detach_locked(struct comedi_device *dev);
+void comedi_device_detach(struct comedi_device *dev);
+int comedi_device_attach(struct comedi_device *dev,
+			 struct comedi_devconfig *it);
+
+#ifdef CONFIG_PROC_FS
+
+/* proc.c */
+
+void comedi_proc_init(void);
+void comedi_proc_cleanup(void);
+#else
+static inline void comedi_proc_init(void)
+{
+}
+
+static inline void comedi_proc_cleanup(void)
+{
+}
+#endif
+
+#endif /* _COMEDI_INTERNAL_H */