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, 269 insertions, 0 deletions

diff --git a/Documentation/ABI/testing/sysfs-fs-nilfs2 b/Documentation/ABI/testing/sysfs-fs-nilfs2
new file mode 100644
index 000000000..304ba84a9
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-fs-nilfs2
@@ -0,0 +1,269 @@
+
+What:		/sys/fs/nilfs2/features/revision
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show current revision of NILFS file system driver.
+		This value informs about file system revision that
+		driver is ready to support.
+
+What:		/sys/fs/nilfs2/features/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/features group.
+
+What:		/sys/fs/nilfs2/<device>/revision
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show NILFS file system revision on volume.
+		This value informs about metadata structures'
+		revision on mounted volume.
+
+What:		/sys/fs/nilfs2/<device>/blocksize
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show volume's block size in bytes.
+
+What:		/sys/fs/nilfs2/<device>/device_size
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show volume size in bytes.
+
+What:		/sys/fs/nilfs2/<device>/free_blocks
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show count of free blocks on volume.
+
+What:		/sys/fs/nilfs2/<device>/uuid
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show volume's UUID (Universally Unique Identifier).
+
+What:		/sys/fs/nilfs2/<device>/volume_name
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show volume's label.
+
+What:		/sys/fs/nilfs2/<device>/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device> group.
+
+What:		/sys/fs/nilfs2/<device>/superblock/sb_write_time
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show last write time of super block in human-readable
+		format.
+
+What:		/sys/fs/nilfs2/<device>/superblock/sb_write_time_secs
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show last write time of super block in seconds.
+
+What:		/sys/fs/nilfs2/<device>/superblock/sb_write_count
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show current write count of super block.
+
+What:		/sys/fs/nilfs2/<device>/superblock/sb_update_frequency
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show/Set interval of periodical update of superblock
+		(in seconds).
+
+What:		/sys/fs/nilfs2/<device>/superblock/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device>/superblock
+		group.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_pseg_block
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show start block number of the latest segment.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_seg_sequence
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show sequence value of the latest segment.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_seg_checkpoint
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show checkpoint number of the latest segment.
+
+What:		/sys/fs/nilfs2/<device>/segctor/current_seg_sequence
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show segment sequence counter.
+
+What:		/sys/fs/nilfs2/<device>/segctor/current_last_full_seg
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show index number of the latest full segment.
+
+What:		/sys/fs/nilfs2/<device>/segctor/next_full_seg
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show index number of the full segment index
+		to be used next.
+
+What:		/sys/fs/nilfs2/<device>/segctor/next_pseg_offset
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show offset of next partial segment in the current
+		full segment.
+
+What:		/sys/fs/nilfs2/<device>/segctor/next_checkpoint
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show next checkpoint number.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_seg_write_time
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show write time of the last segment in
+		human-readable format.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_seg_write_time_secs
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show write time of the last segment in seconds.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_nongc_write_time
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show write time of the last segment not for cleaner
+		operation in human-readable format.
+
+What:		/sys/fs/nilfs2/<device>/segctor/last_nongc_write_time_secs
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show write time of the last segment not for cleaner
+		operation in seconds.
+
+What:		/sys/fs/nilfs2/<device>/segctor/dirty_data_blocks_count
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of dirty data blocks.
+
+What:		/sys/fs/nilfs2/<device>/segctor/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device>/segctor
+		group.
+
+What:		/sys/fs/nilfs2/<device>/segments/segments_number
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of segments on a volume.
+
+What:		/sys/fs/nilfs2/<device>/segments/blocks_per_segment
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of blocks in segment.
+
+What:		/sys/fs/nilfs2/<device>/segments/clean_segments
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show count of clean segments.
+
+What:		/sys/fs/nilfs2/<device>/segments/dirty_segments
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show count of dirty segments.
+
+What:		/sys/fs/nilfs2/<device>/segments/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device>/segments
+		group.
+
+What:		/sys/fs/nilfs2/<device>/checkpoints/checkpoints_number
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of checkpoints on volume.
+
+What:		/sys/fs/nilfs2/<device>/checkpoints/snapshots_number
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of snapshots on volume.
+
+What:		/sys/fs/nilfs2/<device>/checkpoints/last_seg_checkpoint
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show checkpoint number of the latest segment.
+
+What:		/sys/fs/nilfs2/<device>/checkpoints/next_checkpoint
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show next checkpoint number.
+
+What:		/sys/fs/nilfs2/<device>/checkpoints/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device>/checkpoints
+		group.
+
+What:		/sys/fs/nilfs2/<device>/mounted_snapshots/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe content of /sys/fs/nilfs2/<device>/mounted_snapshots
+		group.
+
+What:		/sys/fs/nilfs2/<device>/mounted_snapshots/<id>/inodes_count
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of inodes for snapshot.
+
+What:		/sys/fs/nilfs2/<device>/mounted_snapshots/<id>/blocks_count
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Show number of blocks for snapshot.
+
+What:		/sys/fs/nilfs2/<device>/mounted_snapshots/<id>/README
+Date:		April 2014
+Contact:	"Vyacheslav Dubeyko" <slava@dubeyko.com>
+Description:
+		Describe attributes of /sys/fs/nilfs2/<device>/mounted_snapshots/<id>
+		group.