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, 191 insertions, 0 deletions

diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c
new file mode 100644
index 000000000..1794d66b6
--- /dev/null
+++ b/drivers/mtd/ubi/misc.c
@@ -0,0 +1,191 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ *
+ * Author: Artem Bityutskiy (Битюцкий Артём)
+ */
+
+/* Here we keep miscellaneous functions which are used all over the UBI code */
+
+#include "ubi.h"
+
+/**
+ * ubi_calc_data_len - calculate how much real data is stored in a buffer.
+ * @ubi: UBI device description object
+ * @buf: a buffer with the contents of the physical eraseblock
+ * @length: the buffer length
+ *
+ * This function calculates how much "real data" is stored in @buf and returnes
+ * the length. Continuous 0xFF bytes at the end of the buffer are not
+ * considered as "real data".
+ */
+int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf,
+		      int length)
+{
+	int i;
+
+	ubi_assert(!(length & (ubi->min_io_size - 1)));
+
+	for (i = length - 1; i >= 0; i--)
+		if (((const uint8_t *)buf)[i] != 0xFF)
+			break;
+
+	/* The resulting length must be aligned to the minimum flash I/O size */
+	length = ALIGN(i + 1, ubi->min_io_size);
+	return length;
+}
+
+/**
+ * ubi_check_volume - check the contents of a static volume.
+ * @ubi: UBI device description object
+ * @vol_id: ID of the volume to check
+ *
+ * This function checks if static volume @vol_id is corrupted by fully reading
+ * it and checking data CRC. This function returns %0 if the volume is not
+ * corrupted, %1 if it is corrupted and a negative error code in case of
+ * failure. Dynamic volumes are not checked and zero is returned immediately.
+ */
+int ubi_check_volume(struct ubi_device *ubi, int vol_id)
+{
+	void *buf;
+	int err = 0, i;
+	struct ubi_volume *vol = ubi->volumes[vol_id];
+
+	if (vol->vol_type != UBI_STATIC_VOLUME)
+		return 0;
+
+	buf = vmalloc(vol->usable_leb_size);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; i < vol->used_ebs; i++) {
+		int size;
+
+		cond_resched();
+
+		if (i == vol->used_ebs - 1)
+			size = vol->last_eb_bytes;
+		else
+			size = vol->usable_leb_size;
+
+		err = ubi_eba_read_leb(ubi, vol, i, buf, 0, size, 1);
+		if (err) {
+			if (mtd_is_eccerr(err))
+				err = 1;
+			break;
+		}
+	}
+
+	vfree(buf);
+	return err;
+}
+
+/**
+ * ubi_update_reserved - update bad eraseblock handling accounting data.
+ * @ubi: UBI device description object
+ *
+ * This function calculates the gap between current number of PEBs reserved for
+ * bad eraseblock handling and the required level of PEBs that must be
+ * reserved, and if necessary, reserves more PEBs to fill that gap, according
+ * to availability. Should be called with ubi->volumes_lock held.
+ */
+void ubi_update_reserved(struct ubi_device *ubi)
+{
+	int need = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs;
+
+	if (need <= 0 || ubi->avail_pebs == 0)
+		return;
+
+	need = min_t(int, need, ubi->avail_pebs);
+	ubi->avail_pebs -= need;
+	ubi->rsvd_pebs += need;
+	ubi->beb_rsvd_pebs += need;
+	ubi_msg(ubi, "reserved more %d PEBs for bad PEB handling", need);
+}
+
+/**
+ * ubi_calculate_reserved - calculate how many PEBs must be reserved for bad
+ * eraseblock handling.
+ * @ubi: UBI device description object
+ */
+void ubi_calculate_reserved(struct ubi_device *ubi)
+{
+	/*
+	 * Calculate the actual number of PEBs currently needed to be reserved
+	 * for future bad eraseblock handling.
+	 */
+	ubi->beb_rsvd_level = ubi->bad_peb_limit - ubi->bad_peb_count;
+	if (ubi->beb_rsvd_level < 0) {
+		ubi->beb_rsvd_level = 0;
+		ubi_warn(ubi, "number of bad PEBs (%d) is above the expected limit (%d), not reserving any PEBs for bad PEB handling, will use available PEBs (if any)",
+			 ubi->bad_peb_count, ubi->bad_peb_limit);
+	}
+}
+
+/**
+ * ubi_check_pattern - check if buffer contains only a certain byte pattern.
+ * @buf: buffer to check
+ * @patt: the pattern to check
+ * @size: buffer size in bytes
+ *
+ * This function returns %1 in there are only @patt bytes in @buf, and %0 if
+ * something else was also found.
+ */
+int ubi_check_pattern(const void *buf, uint8_t patt, int size)
+{
+	int i;
+
+	for (i = 0; i < size; i++)
+		if (((const uint8_t *)buf)[i] != patt)
+			return 0;
+	return 1;
+}
+
+/* Normal UBI messages */
+void ubi_msg(const struct ubi_device *ubi, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	pr_notice(UBI_NAME_STR "%d: %pV\n", ubi->ubi_num, &vaf);
+
+	va_end(args);
+}
+
+/* UBI warning messages */
+void ubi_warn(const struct ubi_device *ubi, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	pr_warn(UBI_NAME_STR "%d warning: %ps: %pV\n",
+		ubi->ubi_num, __builtin_return_address(0), &vaf);
+
+	va_end(args);
+}
+
+/* UBI error messages */
+void ubi_err(const struct ubi_device *ubi, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list args;
+
+	va_start(args, fmt);
+
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	pr_err(UBI_NAME_STR "%d error: %ps: %pV\n",
+	       ubi->ubi_num, __builtin_return_address(0), &vaf);
+	va_end(args);
+}