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, 184 insertions, 0 deletions

diff --git a/drivers/hwmon/ltc4260.c b/drivers/hwmon/ltc4260.c
new file mode 100644
index 000000000..37a85125d
--- /dev/null
+++ b/drivers/hwmon/ltc4260.c
@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Linear Technology LTC4260 I2C Positive Voltage Hot Swap Controller
+ *
+ * Copyright (c) 2014 Guenter Roeck
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/i2c.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/jiffies.h>
+#include <linux/regmap.h>
+
+/* chip registers */
+#define LTC4260_CONTROL	0x00
+#define LTC4260_ALERT	0x01
+#define LTC4260_STATUS	0x02
+#define LTC4260_FAULT	0x03
+#define LTC4260_SENSE	0x04
+#define LTC4260_SOURCE	0x05
+#define LTC4260_ADIN	0x06
+
+/*
+ * Fault register bits
+ */
+#define FAULT_OV	(1 << 0)
+#define FAULT_UV	(1 << 1)
+#define FAULT_OC	(1 << 2)
+#define FAULT_POWER_BAD	(1 << 3)
+#define FAULT_FET_SHORT	(1 << 5)
+
+/* Return the voltage from the given register in mV or mA */
+static int ltc4260_get_value(struct device *dev, u8 reg)
+{
+	struct regmap *regmap = dev_get_drvdata(dev);
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(regmap, reg, &val);
+	if (ret < 0)
+		return ret;
+
+	switch (reg) {
+	case LTC4260_ADIN:
+		/* 10 mV resolution. Convert to mV. */
+		val = val * 10;
+		break;
+	case LTC4260_SOURCE:
+		/* 400 mV resolution. Convert to mV. */
+		val = val * 400;
+		break;
+	case LTC4260_SENSE:
+		/*
+		 * 300 uV resolution. Convert to current as measured with
+		 * an 1 mOhm sense resistor, in mA. If a different sense
+		 * resistor is installed, calculate the actual current by
+		 * dividing the reported current by the sense resistor value
+		 * in mOhm.
+		 */
+		val = val * 300;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return val;
+}
+
+static ssize_t ltc4260_value_show(struct device *dev,
+				  struct device_attribute *da, char *buf)
+{
+	struct sensor_device_attribute *attr = to_sensor_dev_attr(da);
+	int value;
+
+	value = ltc4260_get_value(dev, attr->index);
+	if (value < 0)
+		return value;
+	return sysfs_emit(buf, "%d\n", value);
+}
+
+static ssize_t ltc4260_bool_show(struct device *dev,
+				 struct device_attribute *da, char *buf)
+{
+	struct sensor_device_attribute *attr = to_sensor_dev_attr(da);
+	struct regmap *regmap = dev_get_drvdata(dev);
+	unsigned int fault;
+	int ret;
+
+	ret = regmap_read(regmap, LTC4260_FAULT, &fault);
+	if (ret < 0)
+		return ret;
+
+	fault &= attr->index;
+	if (fault)		/* Clear reported faults in chip register */
+		regmap_update_bits(regmap, LTC4260_FAULT, attr->index, 0);
+
+	return sysfs_emit(buf, "%d\n", !!fault);
+}
+
+/* Voltages */
+static SENSOR_DEVICE_ATTR_RO(in1_input, ltc4260_value, LTC4260_SOURCE);
+static SENSOR_DEVICE_ATTR_RO(in2_input, ltc4260_value, LTC4260_ADIN);
+
+/*
+ * Voltage alarms
+ * UV/OV faults are associated with the input voltage, and the POWER BAD and
+ * FET SHORT faults are associated with the output voltage.
+ */
+static SENSOR_DEVICE_ATTR_RO(in1_min_alarm, ltc4260_bool, FAULT_UV);
+static SENSOR_DEVICE_ATTR_RO(in1_max_alarm, ltc4260_bool, FAULT_OV);
+static SENSOR_DEVICE_ATTR_RO(in2_alarm, ltc4260_bool,
+			     FAULT_POWER_BAD | FAULT_FET_SHORT);
+
+/* Current (via sense resistor) */
+static SENSOR_DEVICE_ATTR_RO(curr1_input, ltc4260_value, LTC4260_SENSE);
+
+/* Overcurrent alarm */
+static SENSOR_DEVICE_ATTR_RO(curr1_max_alarm, ltc4260_bool, FAULT_OC);
+
+static struct attribute *ltc4260_attrs[] = {
+	&sensor_dev_attr_in1_input.dev_attr.attr,
+	&sensor_dev_attr_in1_min_alarm.dev_attr.attr,
+	&sensor_dev_attr_in1_max_alarm.dev_attr.attr,
+	&sensor_dev_attr_in2_input.dev_attr.attr,
+	&sensor_dev_attr_in2_alarm.dev_attr.attr,
+
+	&sensor_dev_attr_curr1_input.dev_attr.attr,
+	&sensor_dev_attr_curr1_max_alarm.dev_attr.attr,
+
+	NULL,
+};
+ATTRIBUTE_GROUPS(ltc4260);
+
+static const struct regmap_config ltc4260_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = LTC4260_ADIN,
+};
+
+static int ltc4260_probe(struct i2c_client *client)
+{
+	struct device *dev = &client->dev;
+	struct device *hwmon_dev;
+	struct regmap *regmap;
+
+	regmap = devm_regmap_init_i2c(client, &ltc4260_regmap_config);
+	if (IS_ERR(regmap)) {
+		dev_err(dev, "failed to allocate register map\n");
+		return PTR_ERR(regmap);
+	}
+
+	/* Clear faults */
+	regmap_write(regmap, LTC4260_FAULT, 0x00);
+
+	hwmon_dev = devm_hwmon_device_register_with_groups(dev, client->name,
+							   regmap,
+							   ltc4260_groups);
+	return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+static const struct i2c_device_id ltc4260_id[] = {
+	{ .name = "ltc4260" },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(i2c, ltc4260_id);
+
+static struct i2c_driver ltc4260_driver = {
+	.driver = {
+		   .name = "ltc4260",
+		   },
+	.probe = ltc4260_probe,
+	.id_table = ltc4260_id,
+};
+
+module_i2c_driver(ltc4260_driver);
+
+MODULE_AUTHOR("Guenter Roeck <linux@roeck-us.net>");
+MODULE_DESCRIPTION("LTC4260 driver");
+MODULE_LICENSE("GPL");