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, 134 insertions, 0 deletions

diff --git a/drivers/memory/pl353-smc.c b/drivers/memory/pl353-smc.c
new file mode 100644
index 000000000..28a8cc560
--- /dev/null
+++ b/drivers/memory/pl353-smc.c
@@ -0,0 +1,134 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ARM PL353 SMC driver
+ *
+ * Copyright (C) 2012 - 2018 Xilinx, Inc
+ * Author: Punnaiah Choudary Kalluri <punnaiah@xilinx.com>
+ * Author: Naga Sureshkumar Relli <nagasure@xilinx.com>
+ */
+
+#include <linux/clk.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/amba/bus.h>
+
+/**
+ * struct pl353_smc_data - Private smc driver structure
+ * @memclk:		Pointer to the peripheral clock
+ * @aclk:		Pointer to the AXI peripheral clock
+ */
+struct pl353_smc_data {
+	struct clk		*memclk;
+	struct clk		*aclk;
+};
+
+static int __maybe_unused pl353_smc_suspend(struct device *dev)
+{
+	struct pl353_smc_data *pl353_smc = dev_get_drvdata(dev);
+
+	clk_disable(pl353_smc->memclk);
+	clk_disable(pl353_smc->aclk);
+
+	return 0;
+}
+
+static int __maybe_unused pl353_smc_resume(struct device *dev)
+{
+	struct pl353_smc_data *pl353_smc = dev_get_drvdata(dev);
+	int ret;
+
+	ret = clk_enable(pl353_smc->aclk);
+	if (ret) {
+		dev_err(dev, "Cannot enable axi domain clock.\n");
+		return ret;
+	}
+
+	ret = clk_enable(pl353_smc->memclk);
+	if (ret) {
+		dev_err(dev, "Cannot enable memory clock.\n");
+		clk_disable(pl353_smc->aclk);
+		return ret;
+	}
+
+	return ret;
+}
+
+static SIMPLE_DEV_PM_OPS(pl353_smc_dev_pm_ops, pl353_smc_suspend,
+			 pl353_smc_resume);
+
+static const struct of_device_id pl353_smc_supported_children[] = {
+	{
+		.compatible = "cfi-flash"
+	},
+	{
+		.compatible = "arm,pl353-nand-r2p1",
+	},
+	{}
+};
+
+static int pl353_smc_probe(struct amba_device *adev, const struct amba_id *id)
+{
+	struct device_node *of_node = adev->dev.of_node;
+	const struct of_device_id *match = NULL;
+	struct pl353_smc_data *pl353_smc;
+
+	pl353_smc = devm_kzalloc(&adev->dev, sizeof(*pl353_smc), GFP_KERNEL);
+	if (!pl353_smc)
+		return -ENOMEM;
+
+	pl353_smc->aclk = devm_clk_get_enabled(&adev->dev, "apb_pclk");
+	if (IS_ERR(pl353_smc->aclk))
+		return dev_err_probe(&adev->dev, PTR_ERR(pl353_smc->aclk),
+				     "aclk clock not found.\n");
+
+	pl353_smc->memclk = devm_clk_get_enabled(&adev->dev, "memclk");
+	if (IS_ERR(pl353_smc->memclk))
+		return dev_err_probe(&adev->dev, PTR_ERR(pl353_smc->memclk),
+				     "memclk clock not found.\n");
+
+	amba_set_drvdata(adev, pl353_smc);
+
+	/* Find compatible children. Only a single child is supported */
+	for_each_available_child_of_node_scoped(of_node, child) {
+		match = of_match_node(pl353_smc_supported_children, child);
+		if (!match) {
+			dev_warn(&adev->dev, "unsupported child node\n");
+			continue;
+		}
+		of_platform_device_create(child, NULL, &adev->dev);
+		break;
+	}
+	if (!match) {
+		dev_err(&adev->dev, "no matching children\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static const struct amba_id pl353_ids[] = {
+	{
+		.id = 0x00041353,
+		.mask = 0x000fffff,
+	},
+	{ 0, 0 },
+};
+MODULE_DEVICE_TABLE(amba, pl353_ids);
+
+static struct amba_driver pl353_smc_driver = {
+	.drv = {
+		.name = "pl353-smc",
+		.pm = &pl353_smc_dev_pm_ops,
+	},
+	.id_table = pl353_ids,
+	.probe = pl353_smc_probe,
+};
+
+module_amba_driver(pl353_smc_driver);
+
+MODULE_AUTHOR("Xilinx, Inc.");
+MODULE_DESCRIPTION("ARM PL353 SMC Driver");
+MODULE_LICENSE("GPL");