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, 156 insertions, 0 deletions

diff --git a/drivers/power/reset/keystone-reset.c b/drivers/power/reset/keystone-reset.c
new file mode 100644
index 000000000..3c44cd6ce
--- /dev/null
+++ b/drivers/power/reset/keystone-reset.c
@@ -0,0 +1,156 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * TI keystone reboot driver
+ *
+ * Copyright (C) 2014 Texas Instruments Incorporated. https://www.ti.com/
+ *
+ * Author: Ivan Khoronzhuk <ivan.khoronzhuk@ti.com>
+ */
+
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+#include <linux/regmap.h>
+#include <linux/mfd/syscon.h>
+#include <linux/of.h>
+
+#define RSCTRL_RG			0x4
+#define RSCFG_RG			0x8
+#define RSISO_RG			0xc
+
+#define RSCTRL_KEY_MASK			0x0000ffff
+#define RSCTRL_RESET_MASK		BIT(16)
+#define RSCTRL_KEY			0x5a69
+
+#define RSMUX_OMODE_MASK		0xe
+#define RSMUX_OMODE_RESET_ON		0xa
+#define RSMUX_OMODE_RESET_OFF		0x0
+#define RSMUX_LOCK_SET			0x1
+
+#define RSCFG_RSTYPE_SOFT		0x300f
+#define RSCFG_RSTYPE_HARD		0x0
+
+#define WDT_MUX_NUMBER			0x4
+
+static int rspll_offset;
+static struct regmap *pllctrl_regs;
+
+/**
+ * rsctrl_enable_rspll_write - enable access to RSCTRL, RSCFG
+ * To be able to access to RSCTRL, RSCFG registers
+ * we have to write a key before
+ */
+static inline int rsctrl_enable_rspll_write(void)
+{
+	return regmap_update_bits(pllctrl_regs, rspll_offset + RSCTRL_RG,
+				  RSCTRL_KEY_MASK, RSCTRL_KEY);
+}
+
+static int rsctrl_restart_handler(struct sys_off_data *data)
+{
+	/* enable write access to RSTCTRL */
+	rsctrl_enable_rspll_write();
+
+	/* reset the SOC */
+	regmap_update_bits(pllctrl_regs, rspll_offset + RSCTRL_RG,
+			   RSCTRL_RESET_MASK, 0);
+
+	return NOTIFY_DONE;
+}
+
+static const struct of_device_id rsctrl_of_match[] = {
+	{.compatible = "ti,keystone-reset", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, rsctrl_of_match);
+
+static int rsctrl_probe(struct platform_device *pdev)
+{
+	int i;
+	int ret;
+	u32 val;
+	unsigned int rg;
+	u32 rsmux_offset;
+	struct regmap *devctrl_regs;
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+
+	if (!np)
+		return -ENODEV;
+
+	/* get regmaps */
+	pllctrl_regs = syscon_regmap_lookup_by_phandle_args(np, "ti,syscon-pll",
+							    1, &rspll_offset);
+	if (IS_ERR(pllctrl_regs))
+		return PTR_ERR(pllctrl_regs);
+
+	devctrl_regs = syscon_regmap_lookup_by_phandle_args(np, "ti,syscon-dev",
+							    1, &rsmux_offset);
+	if (IS_ERR(devctrl_regs))
+		return PTR_ERR(devctrl_regs);
+
+	/* set soft/hard reset */
+	val = of_property_read_bool(np, "ti,soft-reset");
+	val = val ? RSCFG_RSTYPE_SOFT : RSCFG_RSTYPE_HARD;
+
+	ret = rsctrl_enable_rspll_write();
+	if (ret)
+		return ret;
+
+	ret = regmap_write(pllctrl_regs, rspll_offset + RSCFG_RG, val);
+	if (ret)
+		return ret;
+
+	/* disable a reset isolation for all module clocks */
+	ret = regmap_write(pllctrl_regs, rspll_offset + RSISO_RG, 0);
+	if (ret)
+		return ret;
+
+	/* enable a reset for watchdogs from wdt-list */
+	for (i = 0; i < WDT_MUX_NUMBER; i++) {
+		ret = of_property_read_u32_index(np, "ti,wdt-list", i, &val);
+		if (ret == -EOVERFLOW && !i) {
+			dev_err(dev, "ti,wdt-list property has to contain at"
+				"least one entry\n");
+			return -EINVAL;
+		} else if (ret) {
+			break;
+		}
+
+		if (val >= WDT_MUX_NUMBER) {
+			dev_err(dev, "ti,wdt-list property can contain "
+				"only numbers < 4\n");
+			return -EINVAL;
+		}
+
+		rg = rsmux_offset + val * 4;
+
+		ret = regmap_update_bits(devctrl_regs, rg, RSMUX_OMODE_MASK,
+					 RSMUX_OMODE_RESET_ON |
+					 RSMUX_LOCK_SET);
+		if (ret)
+			return ret;
+	}
+
+	ret = devm_register_sys_off_handler(dev, SYS_OFF_MODE_RESTART, 128,
+					    rsctrl_restart_handler, NULL);
+	if (ret)
+		dev_err(dev, "cannot register restart handler (err=%d)\n", ret);
+
+	return ret;
+}
+
+static struct platform_driver rsctrl_driver = {
+	.probe = rsctrl_probe,
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = rsctrl_of_match,
+	},
+};
+module_platform_driver(rsctrl_driver);
+
+MODULE_AUTHOR("Ivan Khoronzhuk <ivan.khoronzhuk@ti.com>");
+MODULE_DESCRIPTION("Texas Instruments keystone reset driver");
+MODULE_ALIAS("platform:" KBUILD_MODNAME);