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, 121 insertions, 0 deletions

diff --git a/drivers/power/reset/qemu-virt-ctrl.c b/drivers/power/reset/qemu-virt-ctrl.c
new file mode 100644
index 000000000..97875007f
--- /dev/null
+++ b/drivers/power/reset/qemu-virt-ctrl.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * QEMU Virt Machine System Controller Driver
+ *
+ * Copyright (C) 2026 Kuan-Wei Chiu <visitorckw@gmail.com>
+ */
+
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+
+/* Registers */
+#define VIRT_CTRL_REG_FEATURES	0x00
+#define VIRT_CTRL_REG_CMD	0x04
+
+/* Commands */
+#define CMD_NOOP	0
+#define CMD_RESET	1
+#define CMD_HALT	2
+#define CMD_PANIC	3
+
+struct qemu_virt_ctrl {
+	void __iomem *base;
+	struct notifier_block reboot_nb;
+};
+
+static inline void virt_ctrl_write32(u32 val, void __iomem *addr)
+{
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		iowrite32be(val, addr);
+	else
+		iowrite32(val, addr);
+}
+
+static int qemu_virt_ctrl_power_off(struct sys_off_data *data)
+{
+	struct qemu_virt_ctrl *ctrl = data->cb_data;
+
+	virt_ctrl_write32(CMD_HALT, ctrl->base + VIRT_CTRL_REG_CMD);
+
+	return NOTIFY_DONE;
+}
+
+static int qemu_virt_ctrl_restart(struct sys_off_data *data)
+{
+	struct qemu_virt_ctrl *ctrl = data->cb_data;
+
+	virt_ctrl_write32(CMD_RESET, ctrl->base + VIRT_CTRL_REG_CMD);
+
+	return NOTIFY_DONE;
+}
+
+static int qemu_virt_ctrl_reboot_notify(struct notifier_block *nb,
+					unsigned long action, void *data)
+{
+	struct qemu_virt_ctrl *ctrl = container_of(nb, struct qemu_virt_ctrl, reboot_nb);
+
+	if (action == SYS_HALT)
+		virt_ctrl_write32(CMD_HALT, ctrl->base + VIRT_CTRL_REG_CMD);
+
+	return NOTIFY_DONE;
+}
+
+static int qemu_virt_ctrl_probe(struct platform_device *pdev)
+{
+	struct qemu_virt_ctrl *ctrl;
+	int ret;
+
+	ctrl = devm_kzalloc(&pdev->dev, sizeof(*ctrl), GFP_KERNEL);
+	if (!ctrl)
+		return -ENOMEM;
+
+	ctrl->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(ctrl->base))
+		return PTR_ERR(ctrl->base);
+
+	ret = devm_register_sys_off_handler(&pdev->dev,
+					    SYS_OFF_MODE_RESTART,
+					    SYS_OFF_PRIO_DEFAULT,
+					    qemu_virt_ctrl_restart,
+					    ctrl);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret,
+				     "cannot register restart handler\n");
+
+	ret = devm_register_sys_off_handler(&pdev->dev,
+					    SYS_OFF_MODE_POWER_OFF,
+					    SYS_OFF_PRIO_DEFAULT,
+					    qemu_virt_ctrl_power_off,
+					    ctrl);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret,
+				     "cannot register power-off handler\n");
+
+	ctrl->reboot_nb.notifier_call = qemu_virt_ctrl_reboot_notify;
+	ret = devm_register_reboot_notifier(&pdev->dev, &ctrl->reboot_nb);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret, "cannot register reboot notifier\n");
+
+	return 0;
+}
+
+static const struct platform_device_id qemu_virt_ctrl_id[] = {
+	{ .name = "qemu-virt-ctrl" },
+	{ }
+};
+MODULE_DEVICE_TABLE(platform, qemu_virt_ctrl_id);
+
+static struct platform_driver qemu_virt_ctrl_driver = {
+	.probe = qemu_virt_ctrl_probe,
+	.driver = {
+		.name = "qemu-virt-ctrl",
+	},
+	.id_table = qemu_virt_ctrl_id,
+};
+module_platform_driver(qemu_virt_ctrl_driver);
+
+MODULE_AUTHOR("Kuan-Wei Chiu <visitorckw@gmail.com>");
+MODULE_DESCRIPTION("QEMU Virt Machine System Controller Driver");
+MODULE_LICENSE("GPL");