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, 200 insertions, 0 deletions

diff --git a/drivers/iio/trigger/iio-trig-hrtimer.c b/drivers/iio/trigger/iio-trig-hrtimer.c
new file mode 100644
index 000000000..57550b581
--- /dev/null
+++ b/drivers/iio/trigger/iio-trig-hrtimer.c
@@ -0,0 +1,200 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * The industrial I/O periodic hrtimer trigger driver
+ *
+ * Copyright (C) Intuitive Aerial AB
+ * Written by Marten Svanfeldt, marten@intuitiveaerial.com
+ * Copyright (C) 2012, Analog Devices Inc.
+ *	Author: Lars-Peter Clausen <lars@metafoo.de>
+ * Copyright (C) 2015, Intel Corporation
+ */
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/hrtimer.h>
+
+#include <linux/iio/iio.h>
+#include <linux/iio/trigger.h>
+#include <linux/iio/sw_trigger.h>
+
+/* Defined locally, not in time64.h yet. */
+#define PSEC_PER_SEC   1000000000000LL
+
+/* default sampling frequency - 100Hz */
+#define HRTIMER_DEFAULT_SAMPLING_FREQUENCY 100
+
+struct iio_hrtimer_info {
+	struct iio_sw_trigger swt;
+	struct hrtimer timer;
+	int sampling_frequency[2];
+	ktime_t period;
+};
+
+static const struct config_item_type iio_hrtimer_type = {
+	.ct_owner = THIS_MODULE,
+};
+
+static
+ssize_t iio_hrtimer_show_sampling_frequency(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	struct iio_trigger *trig = to_iio_trigger(dev);
+	struct iio_hrtimer_info *info = iio_trigger_get_drvdata(trig);
+
+	return iio_format_value(buf, IIO_VAL_INT_PLUS_MICRO,
+			ARRAY_SIZE(info->sampling_frequency),
+			info->sampling_frequency);
+}
+
+static
+ssize_t iio_hrtimer_store_sampling_frequency(struct device *dev,
+					     struct device_attribute *attr,
+					     const char *buf, size_t len)
+{
+	struct iio_trigger *trig = to_iio_trigger(dev);
+	struct iio_hrtimer_info *info = iio_trigger_get_drvdata(trig);
+	unsigned long long val;
+	u64 period;
+	int integer, fract, ret;
+
+	ret = iio_str_to_fixpoint(buf, 100, &integer, &fract);
+	if (ret)
+		return ret;
+	if (integer < 0 || fract < 0)
+		return -ERANGE;
+
+	val = fract + 1000ULL * integer;  /* mHz */
+
+	if (!val || val > UINT_MAX)
+		return -EINVAL;
+
+	info->sampling_frequency[0] = integer;  /* Hz */
+	info->sampling_frequency[1] = fract * 1000;  /* uHz */
+	period = PSEC_PER_SEC;
+	do_div(period, val);
+	info->period = period;  /* nS */
+
+	return len;
+}
+
+static DEVICE_ATTR(sampling_frequency, S_IRUGO | S_IWUSR,
+		   iio_hrtimer_show_sampling_frequency,
+		   iio_hrtimer_store_sampling_frequency);
+
+static struct attribute *iio_hrtimer_attrs[] = {
+	&dev_attr_sampling_frequency.attr,
+	NULL
+};
+
+static const struct attribute_group iio_hrtimer_attr_group = {
+	.attrs = iio_hrtimer_attrs,
+};
+
+static const struct attribute_group *iio_hrtimer_attr_groups[] = {
+	&iio_hrtimer_attr_group,
+	NULL
+};
+
+static enum hrtimer_restart iio_hrtimer_trig_handler(struct hrtimer *timer)
+{
+	struct iio_hrtimer_info *info;
+
+	info = container_of(timer, struct iio_hrtimer_info, timer);
+
+	hrtimer_forward_now(timer, info->period);
+	iio_trigger_poll(info->swt.trigger);
+
+	return HRTIMER_RESTART;
+}
+
+static int iio_trig_hrtimer_set_state(struct iio_trigger *trig, bool state)
+{
+	struct iio_hrtimer_info *trig_info;
+
+	trig_info = iio_trigger_get_drvdata(trig);
+
+	if (state)
+		hrtimer_start(&trig_info->timer, trig_info->period,
+			      HRTIMER_MODE_REL_HARD);
+	else
+		hrtimer_cancel(&trig_info->timer);
+
+	return 0;
+}
+
+static const struct iio_trigger_ops iio_hrtimer_trigger_ops = {
+	.set_trigger_state = iio_trig_hrtimer_set_state,
+};
+
+static struct iio_sw_trigger *iio_trig_hrtimer_probe(const char *name)
+{
+	struct iio_hrtimer_info *trig_info;
+	int ret;
+
+	trig_info = kzalloc_obj(*trig_info);
+	if (!trig_info)
+		return ERR_PTR(-ENOMEM);
+
+	trig_info->swt.trigger = iio_trigger_alloc(NULL, "%s", name);
+	if (!trig_info->swt.trigger) {
+		ret = -ENOMEM;
+		goto err_free_trig_info;
+	}
+
+	iio_trigger_set_drvdata(trig_info->swt.trigger, trig_info);
+	trig_info->swt.trigger->ops = &iio_hrtimer_trigger_ops;
+	trig_info->swt.trigger->dev.groups = iio_hrtimer_attr_groups;
+
+	hrtimer_setup(&trig_info->timer, iio_hrtimer_trig_handler, CLOCK_MONOTONIC,
+		      HRTIMER_MODE_REL_HARD);
+
+	trig_info->sampling_frequency[0] = HRTIMER_DEFAULT_SAMPLING_FREQUENCY;
+	trig_info->period = NSEC_PER_SEC / trig_info->sampling_frequency[0];
+
+	ret = iio_trigger_register(trig_info->swt.trigger);
+	if (ret)
+		goto err_free_trigger;
+
+	iio_swt_group_init_type_name(&trig_info->swt, name, &iio_hrtimer_type);
+	return &trig_info->swt;
+err_free_trigger:
+	iio_trigger_free(trig_info->swt.trigger);
+err_free_trig_info:
+	kfree(trig_info);
+
+	return ERR_PTR(ret);
+}
+
+static int iio_trig_hrtimer_remove(struct iio_sw_trigger *swt)
+{
+	struct iio_hrtimer_info *trig_info;
+
+	trig_info = iio_trigger_get_drvdata(swt->trigger);
+
+	iio_trigger_unregister(swt->trigger);
+
+	/* cancel the timer after unreg to make sure no one rearms it */
+	hrtimer_cancel(&trig_info->timer);
+	iio_trigger_free(swt->trigger);
+	kfree(trig_info);
+
+	return 0;
+}
+
+static const struct iio_sw_trigger_ops iio_trig_hrtimer_ops = {
+	.probe		= iio_trig_hrtimer_probe,
+	.remove		= iio_trig_hrtimer_remove,
+};
+
+static struct iio_sw_trigger_type iio_trig_hrtimer = {
+	.name = "hrtimer",
+	.owner = THIS_MODULE,
+	.ops = &iio_trig_hrtimer_ops,
+};
+
+module_iio_sw_trigger_driver(iio_trig_hrtimer);
+
+MODULE_AUTHOR("Marten Svanfeldt <marten@intuitiveaerial.com>");
+MODULE_AUTHOR("Daniel Baluta <daniel.baluta@intel.com>");
+MODULE_DESCRIPTION("Periodic hrtimer trigger for the IIO subsystem");
+MODULE_LICENSE("GPL v2");