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, 150 insertions, 0 deletions

diff --git a/fs/befs/befs.h b/fs/befs/befs.h
new file mode 100644
index 000000000..7cd472456
--- /dev/null
+++ b/fs/befs/befs.h
@@ -0,0 +1,150 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * befs.h
+ *
+ * Copyright (C) 2001-2002 Will Dyson <will_dyson@pobox.com>
+ * Copyright (C) 1999 Makoto Kato (m_kato@ga2.so-net.ne.jp)
+ */
+
+#ifndef _LINUX_BEFS_H
+#define _LINUX_BEFS_H
+
+#include "befs_fs_types.h"
+
+/* used in debug.c */
+#define BEFS_VERSION "0.9.3"
+
+
+typedef u64 befs_blocknr_t;
+/*
+ * BeFS in memory structures
+ */
+
+struct befs_mount_options {
+	kgid_t gid;
+	kuid_t uid;
+	int use_gid;
+	int use_uid;
+	int debug;
+	char *iocharset;
+};
+
+struct befs_sb_info {
+	u32 magic1;
+	u32 block_size;
+	u32 block_shift;
+	int byte_order;
+	befs_off_t num_blocks;
+	befs_off_t used_blocks;
+	u32 inode_size;
+	u32 magic2;
+
+	/* Allocation group information */
+	u32 blocks_per_ag;
+	u32 ag_shift;
+	u32 num_ags;
+
+	/* State of the superblock */
+	u32 flags;
+
+	/* Journal log entry */
+	befs_block_run log_blocks;
+	befs_off_t log_start;
+	befs_off_t log_end;
+
+	befs_inode_addr root_dir;
+	befs_inode_addr indices;
+	u32 magic3;
+
+	struct befs_mount_options mount_opts;
+	struct nls_table *nls;
+};
+
+struct befs_inode_info {
+	u32 i_flags;
+	u32 i_type;
+
+	befs_inode_addr i_inode_num;
+	befs_inode_addr i_parent;
+	befs_inode_addr i_attribute;
+
+	union {
+		befs_data_stream ds;
+		char symlink[BEFS_SYMLINK_LEN];
+	} i_data;
+
+	struct inode vfs_inode;
+};
+
+enum befs_err {
+	BEFS_OK,
+	BEFS_ERR,
+	BEFS_BAD_INODE,
+	BEFS_BT_END,
+	BEFS_BT_EMPTY,
+	BEFS_BT_MATCH,
+	BEFS_BT_OVERFLOW,
+	BEFS_BT_NOT_FOUND
+};
+
+
+/****************************/
+/* debug.c */
+__printf(2, 3)
+void befs_error(const struct super_block *sb, const char *fmt, ...);
+__printf(2, 3)
+void befs_warning(const struct super_block *sb, const char *fmt, ...);
+__printf(2, 3)
+void befs_debug(const struct super_block *sb, const char *fmt, ...);
+
+void befs_dump_super_block(const struct super_block *sb, befs_super_block *);
+void befs_dump_inode(const struct super_block *sb, befs_inode *);
+void befs_dump_index_entry(const struct super_block *sb, befs_disk_btree_super *);
+void befs_dump_index_node(const struct super_block *sb, befs_btree_nodehead *);
+/****************************/
+
+
+/* Gets a pointer to the private portion of the super_block
+ * structure from the public part
+ */
+static inline struct befs_sb_info *
+BEFS_SB(const struct super_block *super)
+{
+	return (struct befs_sb_info *) super->s_fs_info;
+}
+
+static inline struct befs_inode_info *
+BEFS_I(const struct inode *inode)
+{
+	return container_of(inode, struct befs_inode_info, vfs_inode);
+}
+
+static inline befs_blocknr_t
+iaddr2blockno(struct super_block *sb, const befs_inode_addr *iaddr)
+{
+	return ((iaddr->allocation_group << BEFS_SB(sb)->ag_shift) +
+		iaddr->start);
+}
+
+static inline befs_inode_addr
+blockno2iaddr(struct super_block *sb, befs_blocknr_t blockno)
+{
+	befs_inode_addr iaddr;
+
+	iaddr.allocation_group = blockno >> BEFS_SB(sb)->ag_shift;
+	iaddr.start =
+	    blockno - (iaddr.allocation_group << BEFS_SB(sb)->ag_shift);
+	iaddr.len = 1;
+
+	return iaddr;
+}
+
+static inline unsigned int
+befs_iaddrs_per_block(struct super_block *sb)
+{
+	return BEFS_SB(sb)->block_size / sizeof(befs_disk_inode_addr);
+}
+
+#include "endian.h"
+
+#endif				/* _LINUX_BEFS_H */