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, 145 insertions, 0 deletions

diff --git a/drivers/reset/reset-zynqmp.c b/drivers/reset/reset-zynqmp.c
new file mode 100644
index 000000000..c770ea3a1
--- /dev/null
+++ b/drivers/reset/reset-zynqmp.c
@@ -0,0 +1,145 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright (C) 2018 Xilinx, Inc.
+ *
+ */
+
+#include <linux/err.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/reset-controller.h>
+#include <linux/firmware/xlnx-zynqmp.h>
+
+#define ZYNQMP_NR_RESETS (ZYNQMP_PM_RESET_END - ZYNQMP_PM_RESET_START)
+#define ZYNQMP_RESET_ID ZYNQMP_PM_RESET_START
+#define VERSAL_NR_RESETS	95
+#define VERSAL_NET_NR_RESETS	176
+
+struct zynqmp_reset_soc_data {
+	u32 reset_id;
+	u32 num_resets;
+};
+
+struct zynqmp_reset_data {
+	struct reset_controller_dev rcdev;
+	const struct zynqmp_reset_soc_data *data;
+};
+
+static inline struct zynqmp_reset_data *
+to_zynqmp_reset_data(struct reset_controller_dev *rcdev)
+{
+	return container_of(rcdev, struct zynqmp_reset_data, rcdev);
+}
+
+static int zynqmp_reset_assert(struct reset_controller_dev *rcdev,
+			       unsigned long id)
+{
+	struct zynqmp_reset_data *priv = to_zynqmp_reset_data(rcdev);
+
+	return zynqmp_pm_reset_assert(priv->data->reset_id + id,
+				      PM_RESET_ACTION_ASSERT);
+}
+
+static int zynqmp_reset_deassert(struct reset_controller_dev *rcdev,
+				 unsigned long id)
+{
+	struct zynqmp_reset_data *priv = to_zynqmp_reset_data(rcdev);
+
+	return zynqmp_pm_reset_assert(priv->data->reset_id + id,
+				      PM_RESET_ACTION_RELEASE);
+}
+
+static int zynqmp_reset_status(struct reset_controller_dev *rcdev,
+			       unsigned long id)
+{
+	struct zynqmp_reset_data *priv = to_zynqmp_reset_data(rcdev);
+	int err;
+	u32 val;
+
+	err = zynqmp_pm_reset_get_status(priv->data->reset_id + id, &val);
+	if (err)
+		return err;
+
+	return val;
+}
+
+static int zynqmp_reset_reset(struct reset_controller_dev *rcdev,
+			      unsigned long id)
+{
+	struct zynqmp_reset_data *priv = to_zynqmp_reset_data(rcdev);
+
+	return zynqmp_pm_reset_assert(priv->data->reset_id + id,
+				      PM_RESET_ACTION_PULSE);
+}
+
+static int zynqmp_reset_of_xlate(struct reset_controller_dev *rcdev,
+				 const struct of_phandle_args *reset_spec)
+{
+	return reset_spec->args[0];
+}
+
+static const struct zynqmp_reset_soc_data zynqmp_reset_data = {
+	.reset_id = ZYNQMP_RESET_ID,
+	.num_resets = ZYNQMP_NR_RESETS,
+};
+
+static const struct zynqmp_reset_soc_data versal_reset_data = {
+	.reset_id = 0,
+	.num_resets = VERSAL_NR_RESETS,
+};
+
+static const struct zynqmp_reset_soc_data versal_net_reset_data = {
+	.reset_id = 0,
+	.num_resets = VERSAL_NET_NR_RESETS,
+};
+
+static const struct reset_control_ops zynqmp_reset_ops = {
+	.reset = zynqmp_reset_reset,
+	.assert = zynqmp_reset_assert,
+	.deassert = zynqmp_reset_deassert,
+	.status = zynqmp_reset_status,
+};
+
+static int zynqmp_reset_probe(struct platform_device *pdev)
+{
+	struct zynqmp_reset_data *priv;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->data = of_device_get_match_data(&pdev->dev);
+	if (!priv->data)
+		return -EINVAL;
+
+	priv->rcdev.ops = &zynqmp_reset_ops;
+	priv->rcdev.owner = THIS_MODULE;
+	priv->rcdev.of_node = pdev->dev.of_node;
+	priv->rcdev.nr_resets = priv->data->num_resets;
+	priv->rcdev.of_reset_n_cells = 1;
+	priv->rcdev.of_xlate = zynqmp_reset_of_xlate;
+
+	return devm_reset_controller_register(&pdev->dev, &priv->rcdev);
+}
+
+static const struct of_device_id zynqmp_reset_dt_ids[] = {
+	{ .compatible = "xlnx,zynqmp-reset", .data = &zynqmp_reset_data, },
+	{ .compatible = "xlnx,versal-reset", .data = &versal_reset_data, },
+	{ .compatible = "xlnx,versal-net-reset", .data = &versal_net_reset_data, },
+	{ /* sentinel */ },
+};
+
+static struct platform_driver zynqmp_reset_driver = {
+	.probe	= zynqmp_reset_probe,
+	.driver = {
+		.name		= KBUILD_MODNAME,
+		.of_match_table	= zynqmp_reset_dt_ids,
+	},
+};
+
+static int __init zynqmp_reset_init(void)
+{
+	return platform_driver_register(&zynqmp_reset_driver);
+}
+
+arch_initcall(zynqmp_reset_init);