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, 217 insertions, 0 deletions

diff --git a/drivers/hwmon/lm70.c b/drivers/hwmon/lm70.c
new file mode 100644
index 000000000..8b525725f
--- /dev/null
+++ b/drivers/hwmon/lm70.c
@@ -0,0 +1,217 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * lm70.c
+ *
+ * The LM70 is a temperature sensor chip from National Semiconductor (NS).
+ * Copyright (C) 2006 Kaiwan N Billimoria <kaiwan@designergraphix.com>
+ *
+ * The LM70 communicates with a host processor via an SPI/Microwire Bus
+ * interface. The complete datasheet is available at TI's website
+ * here:
+ * https://www.ti.com/product/LM70
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/sysfs.h>
+#include <linux/hwmon.h>
+#include <linux/mutex.h>
+#include <linux/of.h>
+#include <linux/property.h>
+#include <linux/spi/spi.h>
+#include <linux/slab.h>
+
+#define DRVNAME		"lm70"
+
+#define LM70_CHIP_LM70		0	/* original NS LM70 */
+#define LM70_CHIP_TMP121	1	/* TI TMP121/TMP123 */
+#define LM70_CHIP_LM71		2	/* NS LM71 */
+#define LM70_CHIP_LM74		3	/* NS LM74 */
+#define LM70_CHIP_TMP122	4	/* TI TMP122/TMP124 */
+#define LM70_CHIP_TMP125	5	/* TI TMP125 */
+
+struct lm70 {
+	struct spi_device *spi;
+	struct mutex lock;
+	unsigned int chip;
+};
+
+/* sysfs hook function */
+static ssize_t temp1_input_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct lm70 *p_lm70 = dev_get_drvdata(dev);
+	struct spi_device *spi = p_lm70->spi;
+	int status, val = 0;
+	u8 rxbuf[2];
+	s16 raw = 0;
+
+	if (mutex_lock_interruptible(&p_lm70->lock))
+		return -ERESTARTSYS;
+
+	/*
+	 * spi_read() requires a DMA-safe buffer; so we use
+	 * spi_write_then_read(), transmitting 0 bytes.
+	 */
+	status = spi_write_then_read(spi, NULL, 0, &rxbuf[0], 2);
+	if (status < 0) {
+		dev_warn(dev, "spi_write_then_read failed with status %d\n",
+			 status);
+		goto out;
+	}
+	raw = (rxbuf[0] << 8) + rxbuf[1];
+	dev_dbg(dev, "rxbuf[0] : 0x%02x rxbuf[1] : 0x%02x raw=0x%04x\n",
+		rxbuf[0], rxbuf[1], raw);
+
+	/*
+	 * LM70:
+	 * The "raw" temperature read into rxbuf[] is a 16-bit signed 2's
+	 * complement value. Only the MSB 11 bits (1 sign + 10 temperature
+	 * bits) are meaningful; the LSB 5 bits are to be discarded.
+	 * See the datasheet.
+	 *
+	 * Further, each bit represents 0.25 degrees Celsius; so, multiply
+	 * by 0.25. Also multiply by 1000 to represent in millidegrees
+	 * Celsius.
+	 * So it's equivalent to multiplying by 0.25 * 1000 = 250.
+	 *
+	 * LM74 and TMP121/TMP122/TMP123/TMP124:
+	 * 13 bits of 2's complement data, discard LSB 3 bits,
+	 * resolution 0.0625 degrees celsius.
+	 *
+	 * LM71:
+	 * 14 bits of 2's complement data, discard LSB 2 bits,
+	 * resolution 0.0312 degrees celsius.
+	 *
+	 * TMP125:
+	 * MSB/D15 is a leading zero. D14 is the sign-bit. This is
+	 * followed by 9 temperature bits (D13..D5) in 2's complement
+	 * data format with a resolution of 0.25 degrees celsius per unit.
+	 * LSB 5 bits (D4..D0) share the same value as D5 and get discarded.
+	 */
+	switch (p_lm70->chip) {
+	case LM70_CHIP_LM70:
+		val = ((int)raw / 32) * 250;
+		break;
+
+	case LM70_CHIP_TMP121:
+	case LM70_CHIP_TMP122:
+	case LM70_CHIP_LM74:
+		val = ((int)raw / 8) * 625 / 10;
+		break;
+
+	case LM70_CHIP_LM71:
+		val = ((int)raw / 4) * 3125 / 100;
+		break;
+
+	case LM70_CHIP_TMP125:
+		val = (sign_extend32(raw, 14) / 32) * 250;
+		break;
+	}
+
+	status = sprintf(buf, "%d\n", val); /* millidegrees Celsius */
+out:
+	mutex_unlock(&p_lm70->lock);
+	return status;
+}
+
+static DEVICE_ATTR_RO(temp1_input);
+
+static struct attribute *lm70_attrs[] = {
+	&dev_attr_temp1_input.attr,
+	NULL
+};
+
+ATTRIBUTE_GROUPS(lm70);
+
+/*----------------------------------------------------------------------*/
+
+#ifdef CONFIG_OF
+static const struct of_device_id lm70_of_ids[] = {
+	{
+		.compatible = "ti,lm70",
+		.data = (void *) LM70_CHIP_LM70,
+	},
+	{
+		.compatible = "ti,tmp121",
+		.data = (void *) LM70_CHIP_TMP121,
+	},
+	{
+		.compatible = "ti,tmp122",
+		.data = (void *) LM70_CHIP_TMP122,
+	},
+	{
+		.compatible = "ti,tmp125",
+		.data = (void *) LM70_CHIP_TMP125,
+	},
+	{
+		.compatible = "ti,lm71",
+		.data = (void *) LM70_CHIP_LM71,
+	},
+	{
+		.compatible = "ti,lm74",
+		.data = (void *) LM70_CHIP_LM74,
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, lm70_of_ids);
+#endif
+
+static int lm70_probe(struct spi_device *spi)
+{
+	struct device *hwmon_dev;
+	struct lm70 *p_lm70;
+	int chip;
+
+	chip = (kernel_ulong_t)spi_get_device_match_data(spi);
+
+	/* signaling is SPI_MODE_0 */
+	if ((spi->mode & SPI_MODE_X_MASK) != SPI_MODE_0)
+		return -EINVAL;
+
+	/* NOTE:  we assume 8-bit words, and convert to 16 bits manually */
+
+	p_lm70 = devm_kzalloc(&spi->dev, sizeof(*p_lm70), GFP_KERNEL);
+	if (!p_lm70)
+		return -ENOMEM;
+
+	mutex_init(&p_lm70->lock);
+	p_lm70->chip = chip;
+	p_lm70->spi = spi;
+
+	hwmon_dev = devm_hwmon_device_register_with_groups(&spi->dev,
+							   spi->modalias,
+							   p_lm70, lm70_groups);
+	return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+static const struct spi_device_id lm70_ids[] = {
+	{ "lm70",   LM70_CHIP_LM70 },
+	{ "tmp121", LM70_CHIP_TMP121 },
+	{ "tmp122", LM70_CHIP_TMP122 },
+	{ "tmp125", LM70_CHIP_TMP125 },
+	{ "lm71",   LM70_CHIP_LM71 },
+	{ "lm74",   LM70_CHIP_LM74 },
+	{ },
+};
+MODULE_DEVICE_TABLE(spi, lm70_ids);
+
+static struct spi_driver lm70_driver = {
+	.driver = {
+		.name	= "lm70",
+		.of_match_table	= of_match_ptr(lm70_of_ids),
+	},
+	.id_table = lm70_ids,
+	.probe	= lm70_probe,
+};
+
+module_spi_driver(lm70_driver);
+
+MODULE_AUTHOR("Kaiwan N Billimoria");
+MODULE_DESCRIPTION("NS LM70 and compatibles Linux driver");
+MODULE_LICENSE("GPL");