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, 348 insertions, 0 deletions

diff --git a/drivers/usb/phy/phy-generic.c b/drivers/usb/phy/phy-generic.c
new file mode 100644
index 000000000..de26b3023
--- /dev/null
+++ b/drivers/usb/phy/phy-generic.c
@@ -0,0 +1,348 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * NOP USB transceiver for all USB transceiver which are either built-in
+ * into USB IP or which are mostly autonomous.
+ *
+ * Copyright (C) 2009 Texas Instruments Inc
+ * Author: Ajay Kumar Gupta <ajay.gupta@ti.com>
+ *
+ * Current status:
+ *	This provides a "nop" transceiver for PHYs which are
+ *	autonomous such as isp1504, isp1707, etc.
+ */
+
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/usb/gadget.h>
+#include <linux/usb/otg.h>
+#include <linux/usb/usb_phy_generic.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/regulator/consumer.h>
+#include <linux/property.h>
+#include <linux/gpio/consumer.h>
+#include <linux/delay.h>
+
+#include "phy-generic.h"
+
+#define VBUS_IRQ_FLAGS \
+	(IRQF_SHARED | IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | \
+		IRQF_ONESHOT)
+
+struct platform_device *usb_phy_generic_register(void)
+{
+	return platform_device_register_simple("usb_phy_generic",
+			PLATFORM_DEVID_AUTO, NULL, 0);
+}
+EXPORT_SYMBOL_GPL(usb_phy_generic_register);
+
+void usb_phy_generic_unregister(struct platform_device *pdev)
+{
+	platform_device_unregister(pdev);
+}
+EXPORT_SYMBOL_GPL(usb_phy_generic_unregister);
+
+static int nop_set_suspend(struct usb_phy *x, int suspend)
+{
+	struct usb_phy_generic *nop = dev_get_drvdata(x->dev);
+	int ret = 0;
+
+	if (suspend) {
+		clk_disable_unprepare(nop->clk);
+		if (!IS_ERR(nop->vcc) && !device_may_wakeup(x->dev))
+			ret = regulator_disable(nop->vcc);
+	} else {
+		if (!IS_ERR(nop->vcc) && !device_may_wakeup(x->dev))
+			ret = regulator_enable(nop->vcc);
+		clk_prepare_enable(nop->clk);
+	}
+
+	return ret;
+}
+
+static void nop_reset(struct usb_phy_generic *nop)
+{
+	if (!nop->gpiod_reset)
+		return;
+
+	gpiod_set_value_cansleep(nop->gpiod_reset, 1);
+	usleep_range(10000, 20000);
+	gpiod_set_value_cansleep(nop->gpiod_reset, 0);
+	usleep_range(10000, 30000);
+}
+
+/* interface to regulator framework */
+static int nop_set_vbus(struct usb_otg *otg, bool enable)
+{
+	int ret = 0;
+	struct usb_phy_generic *nop = dev_get_drvdata(otg->usb_phy->dev);
+
+	if (!nop->vbus_draw)
+		return 0;
+
+	if (enable && !nop->vbus_draw_enabled) {
+		ret = regulator_enable(nop->vbus_draw);
+		if (ret)
+			nop->vbus_draw_enabled = false;
+		else
+			nop->vbus_draw_enabled = true;
+
+	} else if (!enable && nop->vbus_draw_enabled) {
+		ret = regulator_disable(nop->vbus_draw);
+		nop->vbus_draw_enabled = false;
+	}
+	return ret;
+}
+
+
+static irqreturn_t nop_gpio_vbus_thread(int irq, void *data)
+{
+	struct usb_phy_generic *nop = data;
+	struct usb_otg *otg = nop->phy.otg;
+	int vbus, status;
+
+	vbus = gpiod_get_value(nop->gpiod_vbus);
+	if ((vbus ^ nop->vbus) == 0)
+		return IRQ_HANDLED;
+	nop->vbus = vbus;
+
+	if (vbus) {
+		status = USB_EVENT_VBUS;
+		otg->state = OTG_STATE_B_PERIPHERAL;
+		nop->phy.last_event = status;
+
+		atomic_notifier_call_chain(&nop->phy.notifier, status,
+					   otg->gadget);
+	} else {
+		status = USB_EVENT_NONE;
+		otg->state = OTG_STATE_B_IDLE;
+		nop->phy.last_event = status;
+
+		atomic_notifier_call_chain(&nop->phy.notifier, status,
+					   otg->gadget);
+	}
+	return IRQ_HANDLED;
+}
+
+int usb_gen_phy_init(struct usb_phy *phy)
+{
+	struct usb_phy_generic *nop = dev_get_drvdata(phy->dev);
+	int ret;
+
+	if (!IS_ERR(nop->vcc)) {
+		if (regulator_enable(nop->vcc))
+			dev_err(phy->dev, "Failed to enable power\n");
+	}
+
+	ret = clk_prepare_enable(nop->clk);
+	if (ret)
+		return ret;
+
+	nop_reset(nop);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(usb_gen_phy_init);
+
+void usb_gen_phy_shutdown(struct usb_phy *phy)
+{
+	struct usb_phy_generic *nop = dev_get_drvdata(phy->dev);
+
+	gpiod_set_value_cansleep(nop->gpiod_reset, 1);
+
+	clk_disable_unprepare(nop->clk);
+
+	if (!IS_ERR(nop->vcc)) {
+		if (regulator_disable(nop->vcc))
+			dev_err(phy->dev, "Failed to disable power\n");
+	}
+}
+EXPORT_SYMBOL_GPL(usb_gen_phy_shutdown);
+
+static int nop_set_peripheral(struct usb_otg *otg, struct usb_gadget *gadget)
+{
+	if (!otg)
+		return -ENODEV;
+
+	if (!gadget) {
+		otg->gadget = NULL;
+		return -ENODEV;
+	}
+
+	otg->gadget = gadget;
+	if (otg->state == OTG_STATE_B_PERIPHERAL)
+		atomic_notifier_call_chain(&otg->usb_phy->notifier,
+					   USB_EVENT_VBUS, otg->gadget);
+	else
+		otg->state = OTG_STATE_B_IDLE;
+	return 0;
+}
+
+static int nop_set_host(struct usb_otg *otg, struct usb_bus *host)
+{
+	if (!otg)
+		return -ENODEV;
+
+	if (!host) {
+		otg->host = NULL;
+		return -ENODEV;
+	}
+
+	otg->host = host;
+	return 0;
+}
+
+int usb_phy_gen_create_phy(struct device *dev, struct usb_phy_generic *nop)
+{
+	enum usb_phy_type type = USB_PHY_TYPE_USB2;
+	int err = 0;
+	u32 clk_rate = 0;
+
+	device_property_read_u32(dev, "clock-frequency", &clk_rate);
+	nop->gpiod_reset = devm_gpiod_get_optional(dev, "reset",
+						   GPIOD_ASIS);
+	err = PTR_ERR_OR_ZERO(nop->gpiod_reset);
+	if (!err) {
+		nop->gpiod_vbus = devm_gpiod_get_optional(dev,
+						 "vbus-detect",
+						 GPIOD_ASIS);
+		err = PTR_ERR_OR_ZERO(nop->gpiod_vbus);
+	}
+
+	if (err)
+		return dev_err_probe(dev, err,
+				     "Error requesting RESET or VBUS GPIO\n");
+	if (nop->gpiod_reset)
+		gpiod_direction_output(nop->gpiod_reset, 1);
+
+	nop->phy.otg = devm_kzalloc(dev, sizeof(*nop->phy.otg),
+			GFP_KERNEL);
+	if (!nop->phy.otg)
+		return -ENOMEM;
+
+	nop->clk = devm_clk_get_optional(dev, "main_clk");
+	if (IS_ERR(nop->clk))
+		return dev_err_probe(dev, PTR_ERR(nop->clk),
+				     "Can't get phy clock\n");
+
+	if (clk_rate) {
+		err = clk_set_rate(nop->clk, clk_rate);
+		if (err)
+			return dev_err_probe(dev, err,
+					     "Error setting clock rate\n");
+	}
+
+	nop->vcc = devm_regulator_get_optional(dev, "vcc");
+	if (IS_ERR(nop->vcc) && PTR_ERR(nop->vcc) != -ENODEV)
+		return dev_err_probe(dev, PTR_ERR(nop->vcc),
+				     "could not get vcc regulator\n");
+
+	nop->vbus_draw = devm_regulator_get_exclusive(dev, "vbus");
+	if (PTR_ERR(nop->vbus_draw) == -ENODEV)
+		nop->vbus_draw = NULL;
+	if (IS_ERR(nop->vbus_draw))
+		return dev_err_probe(dev, PTR_ERR(nop->vbus_draw),
+				     "could not get vbus regulator\n");
+
+	nop->dev		= dev;
+	nop->phy.dev		= nop->dev;
+	nop->phy.label		= "nop-xceiv";
+	nop->phy.set_suspend	= nop_set_suspend;
+	nop->phy.type		= type;
+
+	nop->phy.otg->state		= OTG_STATE_UNDEFINED;
+	nop->phy.otg->usb_phy		= &nop->phy;
+	nop->phy.otg->set_host		= nop_set_host;
+	nop->phy.otg->set_peripheral	= nop_set_peripheral;
+	nop->phy.otg->set_vbus          = nop_set_vbus;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(usb_phy_gen_create_phy);
+
+static int usb_phy_generic_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct usb_phy_generic	*nop;
+	int err;
+
+	nop = devm_kzalloc(dev, sizeof(*nop), GFP_KERNEL);
+	if (!nop)
+		return -ENOMEM;
+
+	err = usb_phy_gen_create_phy(dev, nop);
+	if (err)
+		return err;
+
+	if (nop->gpiod_vbus) {
+		err = devm_request_threaded_irq(dev,
+						gpiod_to_irq(nop->gpiod_vbus),
+						NULL, nop_gpio_vbus_thread,
+						VBUS_IRQ_FLAGS, "vbus_detect",
+						nop);
+		if (err)
+			return dev_err_probe(dev, err, "can't request irq %i\n",
+					     gpiod_to_irq(nop->gpiod_vbus));
+
+		nop->phy.otg->state = gpiod_get_value(nop->gpiod_vbus) ?
+			OTG_STATE_B_PERIPHERAL : OTG_STATE_B_IDLE;
+	}
+
+	nop->phy.init		= usb_gen_phy_init;
+	nop->phy.shutdown	= usb_gen_phy_shutdown;
+
+	err = usb_add_phy_dev(&nop->phy);
+	if (err)
+		return dev_err_probe(dev, err, "can't register transceiver\n");
+
+	platform_set_drvdata(pdev, nop);
+
+	device_set_wakeup_capable(dev,
+				  device_property_read_bool(dev, "wakeup-source"));
+
+	return 0;
+}
+
+static void usb_phy_generic_remove(struct platform_device *pdev)
+{
+	struct usb_phy_generic *nop = platform_get_drvdata(pdev);
+
+	usb_remove_phy(&nop->phy);
+
+	if (nop->vbus_draw && nop->vbus_draw_enabled)
+		regulator_disable(nop->vbus_draw);
+}
+
+static const struct of_device_id nop_xceiv_dt_ids[] = {
+	{ .compatible = "usb-nop-xceiv" },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, nop_xceiv_dt_ids);
+
+static struct platform_driver usb_phy_generic_driver = {
+	.probe		= usb_phy_generic_probe,
+	.remove		= usb_phy_generic_remove,
+	.driver		= {
+		.name	= "usb_phy_generic",
+		.of_match_table = nop_xceiv_dt_ids,
+	},
+};
+
+static int __init usb_phy_generic_init(void)
+{
+	return platform_driver_register(&usb_phy_generic_driver);
+}
+subsys_initcall(usb_phy_generic_init);
+
+static void __exit usb_phy_generic_exit(void)
+{
+	platform_driver_unregister(&usb_phy_generic_driver);
+}
+module_exit(usb_phy_generic_exit);
+
+MODULE_ALIAS("platform:usb_phy_generic");
+MODULE_AUTHOR("Texas Instruments Inc");
+MODULE_DESCRIPTION("NOP USB Transceiver driver");
+MODULE_LICENSE("GPL");