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, 144 insertions, 0 deletions

diff --git a/drivers/base/regmap/regmap-raw-ram.c b/drivers/base/regmap/regmap-raw-ram.c
new file mode 100644
index 000000000..60d6e95cd
--- /dev/null
+++ b/drivers/base/regmap/regmap-raw-ram.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Register map access API - Memory region with raw access
+//
+// This is intended for testing only
+//
+// Copyright (c) 2023, Arm Ltd
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+#include <linux/swab.h>
+
+#include "internal.h"
+
+static unsigned int decode_reg(enum regmap_endian endian, const void *reg)
+{
+	const u16 *r = reg;
+
+	if (endian == REGMAP_ENDIAN_BIG)
+		return be16_to_cpu(*r);
+	else
+		return le16_to_cpu(*r);
+}
+
+static int regmap_raw_ram_gather_write(void *context,
+				       const void *reg, size_t reg_len,
+				       const void *val, size_t val_len)
+{
+	struct regmap_ram_data *data = context;
+	unsigned int r;
+	u16 *our_buf = (u16 *)data->vals;
+	int i;
+
+	if (reg_len != 2)
+		return -EINVAL;
+	if (val_len % 2)
+		return -EINVAL;
+
+	r = decode_reg(data->reg_endian, reg);
+	if (data->noinc_reg && data->noinc_reg(data, r)) {
+		memcpy(&our_buf[r], val + val_len - 2, 2);
+		data->written[r] = true;
+	} else {
+		memcpy(&our_buf[r], val, val_len);
+
+		for (i = 0; i < val_len / 2; i++)
+			data->written[r + i] = true;
+	}
+	
+	return 0;
+}
+
+static int regmap_raw_ram_write(void *context, const void *data, size_t count)
+{
+	return regmap_raw_ram_gather_write(context, data, 2,
+					   data + 2, count - 2);
+}
+
+static int regmap_raw_ram_read(void *context,
+			       const void *reg, size_t reg_len,
+			       void *val, size_t val_len)
+{
+	struct regmap_ram_data *data = context;
+	unsigned int r;
+	u16 *our_buf = (u16 *)data->vals;
+	int i;
+
+	if (reg_len != 2)
+		return -EINVAL;
+	if (val_len % 2)
+		return -EINVAL;
+
+	r = decode_reg(data->reg_endian, reg);
+	if (data->noinc_reg && data->noinc_reg(data, r)) {
+		for (i = 0; i < val_len; i += 2)
+			memcpy(val + i, &our_buf[r], 2);
+		data->read[r] = true;
+	} else {
+		memcpy(val, &our_buf[r], val_len);
+
+		for (i = 0; i < val_len / 2; i++)
+			data->read[r + i] = true;
+	}
+
+	return 0;
+}
+
+static void regmap_raw_ram_free_context(void *context)
+{
+	struct regmap_ram_data *data = context;
+
+	kfree(data->vals);
+	kfree(data->read);
+	kfree(data->written);
+	kfree(data);
+}
+
+static const struct regmap_bus regmap_raw_ram = {
+	.fast_io = true,
+	.write = regmap_raw_ram_write,
+	.gather_write = regmap_raw_ram_gather_write,
+	.read = regmap_raw_ram_read,
+	.free_context = regmap_raw_ram_free_context,
+};
+
+struct regmap *__regmap_init_raw_ram(struct device *dev,
+				     const struct regmap_config *config,
+				     struct regmap_ram_data *data,
+				     struct lock_class_key *lock_key,
+				     const char *lock_name)
+{
+	struct regmap *map;
+
+	if (config->reg_bits != 16)
+		return ERR_PTR(-EINVAL);
+
+	if (!config->max_register) {
+		pr_crit("No max_register specified for RAM regmap\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	data->read = kzalloc_objs(bool, config->max_register + 1);
+	if (!data->read)
+		return ERR_PTR(-ENOMEM);
+
+	data->written = kzalloc_objs(bool, config->max_register + 1);
+	if (!data->written)
+		return ERR_PTR(-ENOMEM);
+
+	data->reg_endian = config->reg_format_endian;
+
+	map = __regmap_init(dev, &regmap_raw_ram, data, config,
+			    lock_key, lock_name);
+
+	return map;
+}
+EXPORT_SYMBOL_GPL(__regmap_init_raw_ram);
+
+MODULE_DESCRIPTION("Register map access API - Memory region with raw access");
+MODULE_LICENSE("GPL v2");