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, 158 insertions, 0 deletions

diff --git a/net/atm/atm_sysfs.c b/net/atm/atm_sysfs.c
new file mode 100644
index 000000000..0676a9c33
--- /dev/null
+++ b/net/atm/atm_sysfs.c
@@ -0,0 +1,158 @@
+// SPDX-License-Identifier: GPL-2.0
+/* ATM driver model support. */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/kobject.h>
+#include <linux/atmdev.h>
+#include "common.h"
+#include "resources.h"
+
+#define to_atm_dev(cldev) container_of(cldev, struct atm_dev, class_dev)
+
+static ssize_t type_show(struct device *cdev,
+			 struct device_attribute *attr, char *buf)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+
+	return scnprintf(buf, PAGE_SIZE, "%s\n", adev->type);
+}
+
+static ssize_t address_show(struct device *cdev,
+			    struct device_attribute *attr, char *buf)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+
+	return scnprintf(buf, PAGE_SIZE, "%pM\n", adev->esi);
+}
+
+static ssize_t atmindex_show(struct device *cdev,
+			     struct device_attribute *attr, char *buf)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n", adev->number);
+}
+
+static ssize_t carrier_show(struct device *cdev,
+			    struct device_attribute *attr, char *buf)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+
+	return scnprintf(buf, PAGE_SIZE, "%d\n",
+			 adev->signal == ATM_PHY_SIG_LOST ? 0 : 1);
+}
+
+static ssize_t link_rate_show(struct device *cdev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+	int link_rate;
+
+	/* show the link rate, not the data rate */
+	switch (adev->link_rate) {
+	case ATM_OC3_PCR:
+		link_rate = 155520000;
+		break;
+	case ATM_OC12_PCR:
+		link_rate = 622080000;
+		break;
+	case ATM_25_PCR:
+		link_rate = 25600000;
+		break;
+	default:
+		link_rate = adev->link_rate * 8 * 53;
+	}
+	return scnprintf(buf, PAGE_SIZE, "%d\n", link_rate);
+}
+
+static DEVICE_ATTR_RO(address);
+static DEVICE_ATTR_RO(atmindex);
+static DEVICE_ATTR_RO(carrier);
+static DEVICE_ATTR_RO(type);
+static DEVICE_ATTR_RO(link_rate);
+
+static struct device_attribute *atm_attrs[] = {
+	&dev_attr_address,
+	&dev_attr_atmindex,
+	&dev_attr_carrier,
+	&dev_attr_type,
+	&dev_attr_link_rate,
+	NULL
+};
+
+
+static int atm_uevent(const struct device *cdev, struct kobj_uevent_env *env)
+{
+	const struct atm_dev *adev;
+
+	if (!cdev)
+		return -ENODEV;
+
+	adev = to_atm_dev(cdev);
+
+	if (add_uevent_var(env, "NAME=%s%d", adev->type, adev->number))
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void atm_release(struct device *cdev)
+{
+	struct atm_dev *adev = to_atm_dev(cdev);
+
+	kfree(adev);
+}
+
+static struct class atm_class = {
+	.name		= "atm",
+	.dev_release	= atm_release,
+	.dev_uevent		= atm_uevent,
+};
+
+int atm_register_sysfs(struct atm_dev *adev, struct device *parent)
+{
+	struct device *cdev = &adev->class_dev;
+	int i, j, err;
+
+	cdev->class = &atm_class;
+	cdev->parent = parent;
+	dev_set_drvdata(cdev, adev);
+
+	dev_set_name(cdev, "%s%d", adev->type, adev->number);
+	err = device_register(cdev);
+	if (err < 0)
+		return err;
+
+	for (i = 0; atm_attrs[i]; i++) {
+		err = device_create_file(cdev, atm_attrs[i]);
+		if (err)
+			goto err_out;
+	}
+
+	return 0;
+
+err_out:
+	for (j = 0; j < i; j++)
+		device_remove_file(cdev, atm_attrs[j]);
+	device_del(cdev);
+	return err;
+}
+
+void atm_unregister_sysfs(struct atm_dev *adev)
+{
+	struct device *cdev = &adev->class_dev;
+
+	device_del(cdev);
+}
+
+int __init atm_sysfs_init(void)
+{
+	return class_register(&atm_class);
+}
+
+void __exit atm_sysfs_exit(void)
+{
+	class_unregister(&atm_class);
+}