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, 119 insertions, 0 deletions

diff --git a/fs/hfs/string.c b/fs/hfs/string.c
new file mode 100644
index 000000000..0cfa35e82
--- /dev/null
+++ b/fs/hfs/string.c
@@ -0,0 +1,119 @@
+/*
+ *  linux/fs/hfs/string.c
+ *
+ * Copyright (C) 1995-1997  Paul H. Hargrove
+ * (C) 2003 Ardis Technologies <roman@ardistech.com>
+ * This file may be distributed under the terms of the GNU General Public License.
+ *
+ * This file contains the string comparison function for the
+ * Macintosh character set.
+ *
+ * The code in this file is derived from code which is copyright
+ * 1986, 1989, 1990 by Abacus Research and Development, Inc. (ARDI)
+ * It is used here by the permission of ARDI's president Cliff Matthews.
+ */
+
+#include "hfs_fs.h"
+#include <linux/dcache.h>
+
+#include <kunit/visibility.h>
+
+/*================ File-local variables ================*/
+
+/*
+ * unsigned char caseorder[]
+ *
+ * Defines the lexical ordering of characters on the Macintosh
+ *
+ * Composition of the 'casefold' and 'order' tables from ARDI's code
+ * with the entry for 0x20 changed to match that for 0xCA to remove
+ * special case for those two characters.
+ */
+static unsigned char caseorder[256] = {
+	0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F,
+	0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F,
+	0x20,0x22,0x23,0x28,0x29,0x2A,0x2B,0x2C,0x2F,0x30,0x31,0x32,0x33,0x34,0x35,0x36,
+	0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F,0x40,0x41,0x42,0x43,0x44,0x45,0x46,
+	0x47,0x48,0x57,0x59,0x5D,0x5F,0x66,0x68,0x6A,0x6C,0x72,0x74,0x76,0x78,0x7A,0x7E,
+	0x8C,0x8E,0x90,0x92,0x95,0x97,0x9E,0xA0,0xA2,0xA4,0xA7,0xA9,0xAA,0xAB,0xAC,0xAD,
+	0x4E,0x48,0x57,0x59,0x5D,0x5F,0x66,0x68,0x6A,0x6C,0x72,0x74,0x76,0x78,0x7A,0x7E,
+	0x8C,0x8E,0x90,0x92,0x95,0x97,0x9E,0xA0,0xA2,0xA4,0xA7,0xAF,0xB0,0xB1,0xB2,0xB3,
+	0x4A,0x4C,0x5A,0x60,0x7B,0x7F,0x98,0x4F,0x49,0x51,0x4A,0x4B,0x4C,0x5A,0x60,0x63,
+	0x64,0x65,0x6E,0x6F,0x70,0x71,0x7B,0x84,0x85,0x86,0x7F,0x80,0x9A,0x9B,0x9C,0x98,
+	0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0x94,0xBB,0xBC,0xBD,0xBE,0xBF,0xC0,0x4D,0x81,
+	0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0x55,0x8A,0xCC,0x4D,0x81,
+	0xCD,0xCE,0xCF,0xD0,0xD1,0xD2,0xD3,0x26,0x27,0xD4,0x20,0x49,0x4B,0x80,0x82,0x82,
+	0xD5,0xD6,0x24,0x25,0x2D,0x2E,0xD7,0xD8,0xA6,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xDF,
+	0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7,0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF,
+	0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF
+};
+
+/*================ Global functions ================*/
+
+/*
+ * Hash a string to an integer in a case-independent way
+ */
+int hfs_hash_dentry(const struct dentry *dentry, struct qstr *this)
+{
+	const unsigned char *name = this->name;
+	unsigned int hash, len = this->len;
+
+	if (len > HFS_NAMELEN)
+		len = HFS_NAMELEN;
+
+	hash = init_name_hash(dentry);
+	for (; len; len--)
+		hash = partial_name_hash(caseorder[*name++], hash);
+	this->hash = end_name_hash(hash);
+	return 0;
+}
+EXPORT_SYMBOL_IF_KUNIT(hfs_hash_dentry);
+
+/*
+ * Compare two strings in the HFS filename character ordering
+ * Returns positive, negative, or zero, not just 0 or (+/-)1
+ *
+ * Equivalent to ARDI's call:
+ *	ROMlib_RelString(s1+1, s2+1, true, false, (s1[0]<<16) | s2[0])
+ */
+int hfs_strcmp(const unsigned char *s1, unsigned int len1,
+	       const unsigned char *s2, unsigned int len2)
+{
+	int len, tmp;
+
+	len = (len1 > len2) ? len2 : len1;
+
+	while (len--) {
+		tmp = (int)caseorder[*(s1++)] - (int)caseorder[*(s2++)];
+		if (tmp)
+			return tmp;
+	}
+	return len1 - len2;
+}
+EXPORT_SYMBOL_IF_KUNIT(hfs_strcmp);
+
+/*
+ * Test for equality of two strings in the HFS filename character ordering.
+ * return 1 on failure and 0 on success
+ */
+int hfs_compare_dentry(const struct dentry *dentry,
+		unsigned int len, const char *str, const struct qstr *name)
+{
+	const unsigned char *n1, *n2;
+
+	if (len >= HFS_NAMELEN) {
+		if (name->len < HFS_NAMELEN)
+			return 1;
+		len = HFS_NAMELEN;
+	} else if (len != name->len)
+		return 1;
+
+	n1 = str;
+	n2 = name->name;
+	while (len--) {
+		if (caseorder[*n1++] != caseorder[*n2++])
+			return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_IF_KUNIT(hfs_compare_dentry);