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, 128 insertions, 0 deletions

diff --git a/drivers/reset/reset-a10sr.c b/drivers/reset/reset-a10sr.c
new file mode 100644
index 000000000..99b3bc838
--- /dev/null
+++ b/drivers/reset/reset-a10sr.c
@@ -0,0 +1,128 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Copyright Intel Corporation (C) 2017. All Rights Reserved
+ *
+ * Reset driver for Altera Arria10 MAX5 System Resource Chip
+ *
+ * Adapted from reset-socfpga.c
+ */
+
+#include <linux/err.h>
+#include <linux/mfd/altera-a10sr.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/reset-controller.h>
+
+#include <dt-bindings/reset/altr,rst-mgr-a10sr.h>
+
+struct a10sr_reset {
+	struct reset_controller_dev     rcdev;
+	struct regmap *regmap;
+};
+
+static inline struct a10sr_reset *to_a10sr_rst(struct reset_controller_dev *rc)
+{
+	return container_of(rc, struct a10sr_reset, rcdev);
+}
+
+static inline int a10sr_reset_shift(unsigned long id)
+{
+	switch (id) {
+	case A10SR_RESET_ENET_HPS:
+		return 1;
+	case A10SR_RESET_PCIE:
+	case A10SR_RESET_FILE:
+	case A10SR_RESET_BQSPI:
+	case A10SR_RESET_USB:
+		return id + 11;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int a10sr_reset_update(struct reset_controller_dev *rcdev,
+			      unsigned long id, bool assert)
+{
+	struct a10sr_reset *a10r = to_a10sr_rst(rcdev);
+	int offset = a10sr_reset_shift(id);
+	u8 mask = ALTR_A10SR_REG_BIT_MASK(offset);
+	int index = ALTR_A10SR_HPS_RST_REG + ALTR_A10SR_REG_OFFSET(offset);
+
+	return regmap_update_bits(a10r->regmap, index, mask, assert ? 0 : mask);
+}
+
+static int a10sr_reset_assert(struct reset_controller_dev *rcdev,
+			      unsigned long id)
+{
+	return a10sr_reset_update(rcdev, id, true);
+}
+
+static int a10sr_reset_deassert(struct reset_controller_dev *rcdev,
+				unsigned long id)
+{
+	return a10sr_reset_update(rcdev, id, false);
+}
+
+static int a10sr_reset_status(struct reset_controller_dev *rcdev,
+			      unsigned long id)
+{
+	int ret;
+	struct a10sr_reset *a10r = to_a10sr_rst(rcdev);
+	int offset = a10sr_reset_shift(id);
+	u8 mask = ALTR_A10SR_REG_BIT_MASK(offset);
+	int index = ALTR_A10SR_HPS_RST_REG + ALTR_A10SR_REG_OFFSET(offset);
+	unsigned int value;
+
+	ret = regmap_read(a10r->regmap, index, &value);
+	if (ret < 0)
+		return ret;
+
+	return !!(value & mask);
+}
+
+static const struct reset_control_ops a10sr_reset_ops = {
+	.assert		= a10sr_reset_assert,
+	.deassert	= a10sr_reset_deassert,
+	.status		= a10sr_reset_status,
+};
+
+static int a10sr_reset_probe(struct platform_device *pdev)
+{
+	struct altr_a10sr *a10sr = dev_get_drvdata(pdev->dev.parent);
+	struct a10sr_reset *a10r;
+
+	a10r = devm_kzalloc(&pdev->dev, sizeof(struct a10sr_reset),
+			    GFP_KERNEL);
+	if (!a10r)
+		return -ENOMEM;
+
+	a10r->rcdev.owner = THIS_MODULE;
+	a10r->rcdev.nr_resets = A10SR_RESET_NUM;
+	a10r->rcdev.ops = &a10sr_reset_ops;
+	a10r->rcdev.of_node = pdev->dev.of_node;
+	a10r->regmap = a10sr->regmap;
+
+	platform_set_drvdata(pdev, a10r);
+
+	return devm_reset_controller_register(&pdev->dev, &a10r->rcdev);
+}
+
+static const struct of_device_id a10sr_reset_of_match[] = {
+	{ .compatible = "altr,a10sr-reset" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, a10sr_reset_of_match);
+
+static struct platform_driver a10sr_reset_driver = {
+	.probe	= a10sr_reset_probe,
+	.driver = {
+		.name		= "altr_a10sr_reset",
+		.of_match_table	= a10sr_reset_of_match,
+	},
+};
+module_platform_driver(a10sr_reset_driver);
+
+MODULE_AUTHOR("Thor Thayer <thor.thayer@linux.intel.com>");
+MODULE_DESCRIPTION("Altera Arria10 System Resource Reset Controller Driver");
+MODULE_LICENSE("GPL v2");