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, 187 insertions, 0 deletions

diff --git a/drivers/mcb/mcb-lpc.c b/drivers/mcb/mcb-lpc.c
new file mode 100644
index 000000000..070aa787a
--- /dev/null
+++ b/drivers/mcb/mcb-lpc.c
@@ -0,0 +1,187 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * MEN Chameleon Bus.
+ *
+ * Copyright (C) 2014 MEN Mikroelektronik GmbH (www.men.de)
+ * Author: Andreas Werner <andreas.werner@men.de>
+ */
+
+#include <linux/platform_device.h>
+#include <linux/module.h>
+#include <linux/dmi.h>
+#include <linux/mcb.h>
+#include <linux/io.h>
+#include "mcb-internal.h"
+
+struct priv {
+	struct mcb_bus *bus;
+	struct resource *mem;
+	void __iomem *base;
+};
+
+static int mcb_lpc_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct priv *priv;
+	int ret = 0, table_size;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	priv->mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!priv->mem) {
+		dev_err(&pdev->dev, "No Memory resource\n");
+		return -ENODEV;
+	}
+
+	res = devm_request_mem_region(&pdev->dev, priv->mem->start,
+				      resource_size(priv->mem),
+				      KBUILD_MODNAME);
+	if (!res) {
+		dev_err(&pdev->dev, "Failed to request IO memory\n");
+		return -EBUSY;
+	}
+
+	priv->base = devm_ioremap(&pdev->dev, priv->mem->start,
+				  resource_size(priv->mem));
+	if (!priv->base) {
+		dev_err(&pdev->dev, "Cannot ioremap\n");
+		return -ENOMEM;
+	}
+
+	platform_set_drvdata(pdev, priv);
+
+	priv->bus = mcb_alloc_bus(&pdev->dev);
+	if (IS_ERR(priv->bus))
+		return PTR_ERR(priv->bus);
+
+	ret = chameleon_parse_cells(priv->bus, priv->mem->start, priv->base);
+	if (ret < 0) {
+		goto out_mcb_bus;
+	}
+
+	table_size = ret;
+
+	if (table_size < CHAM_HEADER_SIZE) {
+		/* Release the previous resources */
+		devm_iounmap(&pdev->dev, priv->base);
+		devm_release_mem_region(&pdev->dev, priv->mem->start, resource_size(priv->mem));
+
+		/* Then, allocate it again with the actual chameleon table size */
+		res = devm_request_mem_region(&pdev->dev, priv->mem->start,
+					      table_size,
+					      KBUILD_MODNAME);
+		if (!res) {
+			dev_err(&pdev->dev, "Failed to request PCI memory\n");
+			ret = -EBUSY;
+			goto out_mcb_bus;
+		}
+
+		priv->base = devm_ioremap(&pdev->dev, priv->mem->start, table_size);
+		if (!priv->base) {
+			dev_err(&pdev->dev, "Cannot ioremap\n");
+			ret = -ENOMEM;
+			goto out_mcb_bus;
+		}
+
+		platform_set_drvdata(pdev, priv);
+	}
+
+	mcb_bus_add_devices(priv->bus);
+
+	return 0;
+
+out_mcb_bus:
+	mcb_release_bus(priv->bus);
+	return ret;
+}
+
+static void mcb_lpc_remove(struct platform_device *pdev)
+{
+	struct priv *priv = platform_get_drvdata(pdev);
+
+	mcb_release_bus(priv->bus);
+}
+
+static struct platform_device *mcb_lpc_pdev;
+
+static int mcb_lpc_create_platform_device(const struct dmi_system_id *id)
+{
+	struct resource *res = id->driver_data;
+	int ret;
+
+	mcb_lpc_pdev = platform_device_alloc("mcb-lpc", -1);
+	if (!mcb_lpc_pdev)
+		return -ENOMEM;
+
+	ret = platform_device_add_resources(mcb_lpc_pdev, res, 1);
+	if (ret)
+		goto out_put;
+
+	ret = platform_device_add(mcb_lpc_pdev);
+	if (ret)
+		goto out_put;
+
+	return 0;
+
+out_put:
+	platform_device_put(mcb_lpc_pdev);
+	return ret;
+}
+
+static struct resource sc24_fpga_resource = DEFINE_RES_MEM(0xe000e000, CHAM_HEADER_SIZE);
+static struct resource sc31_fpga_resource = DEFINE_RES_MEM(0xf000e000, CHAM_HEADER_SIZE);
+
+static struct platform_driver mcb_lpc_driver = {
+	.driver		= {
+		.name = "mcb-lpc",
+	},
+	.probe		= mcb_lpc_probe,
+	.remove		= mcb_lpc_remove,
+};
+
+static const struct dmi_system_id mcb_lpc_dmi_table[] = {
+	{
+		.ident = "SC24",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "MEN"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "14SC24"),
+		},
+		.driver_data = (void *)&sc24_fpga_resource,
+		.callback = mcb_lpc_create_platform_device,
+	},
+	{
+		.ident = "SC31",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "MEN"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "14SC31"),
+		},
+		.driver_data = (void *)&sc31_fpga_resource,
+		.callback = mcb_lpc_create_platform_device,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(dmi, mcb_lpc_dmi_table);
+
+static int __init mcb_lpc_init(void)
+{
+	if (!dmi_check_system(mcb_lpc_dmi_table))
+		return -ENODEV;
+
+	return platform_driver_register(&mcb_lpc_driver);
+}
+
+static void __exit mcb_lpc_exit(void)
+{
+	platform_device_unregister(mcb_lpc_pdev);
+	platform_driver_unregister(&mcb_lpc_driver);
+}
+
+module_init(mcb_lpc_init);
+module_exit(mcb_lpc_exit);
+
+MODULE_AUTHOR("Andreas Werner <andreas.werner@men.de>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("MCB over LPC support");
+MODULE_IMPORT_NS("MCB");