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, 83 insertions, 0 deletions

diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h
new file mode 100644
index 000000000..833e2fd98
--- /dev/null
+++ b/fs/btrfs/tree-checker.h
@@ -0,0 +1,83 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) Qu Wenruo 2017.  All rights reserved.
+ */
+
+#ifndef BTRFS_TREE_CHECKER_H
+#define BTRFS_TREE_CHECKER_H
+
+#include <linux/types.h>
+#include <uapi/linux/btrfs_tree.h>
+
+struct extent_buffer;
+struct btrfs_fs_info;
+struct btrfs_chunk;
+struct btrfs_key;
+
+/* All the extra info needed to verify the parentness of a tree block. */
+struct btrfs_tree_parent_check {
+	/*
+	 * The owner check against the tree block.
+	 *
+	 * Can be 0 to skip the owner check.
+	 */
+	u64 owner_root;
+
+	/*
+	 * Expected transid, can be 0 to skip the check, but such skip
+	 * should only be utilized for backref walk related code.
+	 */
+	u64 transid;
+
+	/*
+	 * The expected first key.
+	 *
+	 * This check can be skipped if @has_first_key is false, such skip
+	 * can happen for case where we don't have the parent node key,
+	 * e.g. reading the tree root, doing backref walk.
+	 */
+	struct btrfs_key first_key;
+	bool has_first_key;
+
+	/* The expected level. Should always be set. */
+	u8 level;
+};
+
+enum btrfs_tree_block_status {
+	BTRFS_TREE_BLOCK_CLEAN,
+	BTRFS_TREE_BLOCK_INVALID_NRITEMS,
+	BTRFS_TREE_BLOCK_INVALID_PARENT_KEY,
+	BTRFS_TREE_BLOCK_BAD_KEY_ORDER,
+	BTRFS_TREE_BLOCK_INVALID_LEVEL,
+	BTRFS_TREE_BLOCK_INVALID_FREE_SPACE,
+	BTRFS_TREE_BLOCK_INVALID_OFFSETS,
+	BTRFS_TREE_BLOCK_INVALID_BLOCKPTR,
+	BTRFS_TREE_BLOCK_INVALID_ITEM,
+	BTRFS_TREE_BLOCK_INVALID_OWNER,
+	BTRFS_TREE_BLOCK_WRITTEN_NOT_SET,
+};
+
+
+#define BTRFS_BLOCK_GROUP_VALID	(BTRFS_BLOCK_GROUP_TYPE_MASK | \
+				 BTRFS_BLOCK_GROUP_PROFILE_MASK | \
+				 BTRFS_BLOCK_GROUP_REMAPPED)
+
+/*
+ * Exported simply for btrfs-progs which wants to have the
+ * btrfs_tree_block_status return codes.
+ */
+enum btrfs_tree_block_status __btrfs_check_leaf(struct extent_buffer *leaf);
+enum btrfs_tree_block_status __btrfs_check_node(struct extent_buffer *node);
+
+int btrfs_check_leaf(struct extent_buffer *leaf);
+int btrfs_check_node(struct extent_buffer *node);
+
+int btrfs_check_chunk_valid(const struct btrfs_fs_info *fs_info,
+			    const struct extent_buffer *leaf,
+			    const struct btrfs_chunk *chunk, u64 logical,
+			    u32 sectorsize);
+int btrfs_check_eb_owner(const struct extent_buffer *eb, u64 root_owner);
+int btrfs_verify_level_key(struct extent_buffer *eb,
+			   const struct btrfs_tree_parent_check *check);
+
+#endif