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, 324 insertions, 0 deletions

diff --git a/drivers/rtc/rtc-cpcap.c b/drivers/rtc/rtc-cpcap.c
new file mode 100644
index 000000000..a7db3173d
--- /dev/null
+++ b/drivers/rtc/rtc-cpcap.c
@@ -0,0 +1,324 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Motorola CPCAP PMIC RTC driver
+ *
+ * Based on cpcap-regulator.c from Motorola Linux kernel tree
+ * Copyright (C) 2009 Motorola, Inc.
+ *
+ * Rewritten for mainline kernel
+ *  - use DT
+ *  - use regmap
+ *  - use standard interrupt framework
+ *  - use managed device resources
+ *  - remove custom "secure clock daemon" helpers
+ *
+ * Copyright (C) 2017 Sebastian Reichel <sre@kernel.org>
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/rtc.h>
+#include <linux/err.h>
+#include <linux/regmap.h>
+#include <linux/mfd/motorola-cpcap.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+
+#define SECS_PER_DAY 86400
+#define DAY_MASK  0x7FFF
+#define TOD1_MASK 0x00FF
+#define TOD2_MASK 0x01FF
+
+struct cpcap_time {
+	int day;
+	int tod1;
+	int tod2;
+};
+
+struct cpcap_rtc {
+	struct regmap *regmap;
+	struct rtc_device *rtc_dev;
+	u16 vendor;
+	int alarm_irq;
+	bool alarm_enabled;
+	int update_irq;
+	bool update_enabled;
+};
+
+static void cpcap2rtc_time(struct rtc_time *rtc, struct cpcap_time *cpcap)
+{
+	unsigned long int tod;
+	unsigned long int time;
+
+	tod = (cpcap->tod1 & TOD1_MASK) | ((cpcap->tod2 & TOD2_MASK) << 8);
+	time = tod + ((cpcap->day & DAY_MASK) * SECS_PER_DAY);
+
+	rtc_time64_to_tm(time, rtc);
+}
+
+static void rtc2cpcap_time(struct cpcap_time *cpcap, struct rtc_time *rtc)
+{
+	unsigned long time;
+
+	time = rtc_tm_to_time64(rtc);
+
+	cpcap->day = time / SECS_PER_DAY;
+	time %= SECS_PER_DAY;
+	cpcap->tod2 = (time >> 8) & TOD2_MASK;
+	cpcap->tod1 = time & TOD1_MASK;
+}
+
+static int cpcap_rtc_alarm_irq_enable(struct device *dev, unsigned int enabled)
+{
+	struct cpcap_rtc *rtc = dev_get_drvdata(dev);
+
+	if (rtc->alarm_enabled == enabled)
+		return 0;
+
+	if (enabled)
+		enable_irq(rtc->alarm_irq);
+	else
+		disable_irq(rtc->alarm_irq);
+
+	rtc->alarm_enabled = !!enabled;
+
+	return 0;
+}
+
+static int cpcap_rtc_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct cpcap_rtc *rtc;
+	struct cpcap_time cpcap_tm;
+	int temp_tod2;
+	int ret;
+
+	rtc = dev_get_drvdata(dev);
+
+	ret = regmap_read(rtc->regmap, CPCAP_REG_TOD2, &temp_tod2);
+	ret |= regmap_read(rtc->regmap, CPCAP_REG_DAY, &cpcap_tm.day);
+	ret |= regmap_read(rtc->regmap, CPCAP_REG_TOD1, &cpcap_tm.tod1);
+	ret |= regmap_read(rtc->regmap, CPCAP_REG_TOD2, &cpcap_tm.tod2);
+
+	if (temp_tod2 > cpcap_tm.tod2)
+		ret |= regmap_read(rtc->regmap, CPCAP_REG_DAY, &cpcap_tm.day);
+
+	if (ret) {
+		dev_err(dev, "Failed to read time\n");
+		return -EIO;
+	}
+
+	cpcap2rtc_time(tm, &cpcap_tm);
+
+	return 0;
+}
+
+static int cpcap_rtc_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct cpcap_rtc *rtc;
+	struct cpcap_time cpcap_tm;
+	int ret = 0;
+
+	rtc = dev_get_drvdata(dev);
+
+	rtc2cpcap_time(&cpcap_tm, tm);
+
+	if (rtc->alarm_enabled)
+		disable_irq(rtc->alarm_irq);
+	if (rtc->update_enabled)
+		disable_irq(rtc->update_irq);
+
+	if (rtc->vendor == CPCAP_VENDOR_ST) {
+		/* The TOD1 and TOD2 registers MUST be written in this order
+		 * for the change to properly set.
+		 */
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TOD1,
+					  TOD1_MASK, cpcap_tm.tod1);
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TOD2,
+					  TOD2_MASK, cpcap_tm.tod2);
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_DAY,
+					  DAY_MASK, cpcap_tm.day);
+	} else {
+		/* Clearing the upper lower 8 bits of the TOD guarantees that
+		 * the upper half of TOD (TOD2) will not increment for 0xFF RTC
+		 * ticks (255 seconds).  During this time we can safely write
+		 * to DAY, TOD2, then TOD1 (in that order) and expect RTC to be
+		 * synchronized to the exact time requested upon the final write
+		 * to TOD1.
+		 */
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TOD1,
+					  TOD1_MASK, 0);
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_DAY,
+					  DAY_MASK, cpcap_tm.day);
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TOD2,
+					  TOD2_MASK, cpcap_tm.tod2);
+		ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TOD1,
+					  TOD1_MASK, cpcap_tm.tod1);
+	}
+
+	if (rtc->update_enabled)
+		enable_irq(rtc->update_irq);
+	if (rtc->alarm_enabled)
+		enable_irq(rtc->alarm_irq);
+
+	return ret;
+}
+
+static int cpcap_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	struct cpcap_rtc *rtc;
+	struct cpcap_time cpcap_tm;
+	int ret;
+
+	rtc = dev_get_drvdata(dev);
+
+	alrm->enabled = rtc->alarm_enabled;
+
+	ret = regmap_read(rtc->regmap, CPCAP_REG_DAYA, &cpcap_tm.day);
+	ret |= regmap_read(rtc->regmap, CPCAP_REG_TODA2, &cpcap_tm.tod2);
+	ret |= regmap_read(rtc->regmap, CPCAP_REG_TODA1, &cpcap_tm.tod1);
+
+	if (ret) {
+		dev_err(dev, "Failed to read time\n");
+		return -EIO;
+	}
+
+	cpcap2rtc_time(&alrm->time, &cpcap_tm);
+	return rtc_valid_tm(&alrm->time);
+}
+
+static int cpcap_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	struct cpcap_rtc *rtc;
+	struct cpcap_time cpcap_tm;
+	int ret;
+
+	rtc = dev_get_drvdata(dev);
+
+	rtc2cpcap_time(&cpcap_tm, &alrm->time);
+
+	if (rtc->alarm_enabled)
+		disable_irq(rtc->alarm_irq);
+
+	ret = regmap_update_bits(rtc->regmap, CPCAP_REG_DAYA, DAY_MASK,
+				 cpcap_tm.day);
+	ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TODA2, TOD2_MASK,
+				  cpcap_tm.tod2);
+	ret |= regmap_update_bits(rtc->regmap, CPCAP_REG_TODA1, TOD1_MASK,
+				  cpcap_tm.tod1);
+
+	if (!ret) {
+		enable_irq(rtc->alarm_irq);
+		rtc->alarm_enabled = true;
+	}
+
+	return ret;
+}
+
+static const struct rtc_class_ops cpcap_rtc_ops = {
+	.read_time		= cpcap_rtc_read_time,
+	.set_time		= cpcap_rtc_set_time,
+	.read_alarm		= cpcap_rtc_read_alarm,
+	.set_alarm		= cpcap_rtc_set_alarm,
+	.alarm_irq_enable	= cpcap_rtc_alarm_irq_enable,
+};
+
+static irqreturn_t cpcap_rtc_alarm_irq(int irq, void *data)
+{
+	struct cpcap_rtc *rtc = data;
+
+	rtc_update_irq(rtc->rtc_dev, 1, RTC_AF | RTC_IRQF);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t cpcap_rtc_update_irq(int irq, void *data)
+{
+	struct cpcap_rtc *rtc = data;
+
+	rtc_update_irq(rtc->rtc_dev, 1, RTC_UF | RTC_IRQF);
+	return IRQ_HANDLED;
+}
+
+static int cpcap_rtc_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct cpcap_rtc *rtc;
+	int err;
+
+	rtc = devm_kzalloc(dev, sizeof(*rtc), GFP_KERNEL);
+	if (!rtc)
+		return -ENOMEM;
+
+	rtc->regmap = dev_get_regmap(dev->parent, NULL);
+	if (!rtc->regmap)
+		return -ENODEV;
+
+	platform_set_drvdata(pdev, rtc);
+	rtc->rtc_dev = devm_rtc_allocate_device(dev);
+	if (IS_ERR(rtc->rtc_dev))
+		return PTR_ERR(rtc->rtc_dev);
+
+	rtc->rtc_dev->ops = &cpcap_rtc_ops;
+	rtc->rtc_dev->range_max = (timeu64_t) (DAY_MASK + 1) * SECS_PER_DAY - 1;
+
+	err = cpcap_get_vendor(dev, rtc->regmap, &rtc->vendor);
+	if (err)
+		return err;
+
+	rtc->alarm_irq = platform_get_irq(pdev, 0);
+	err = devm_request_threaded_irq(dev, rtc->alarm_irq, NULL,
+					cpcap_rtc_alarm_irq,
+					IRQF_TRIGGER_NONE | IRQF_ONESHOT,
+					"rtc_alarm", rtc);
+	if (err) {
+		dev_err(dev, "Could not request alarm irq: %d\n", err);
+		return err;
+	}
+	disable_irq(rtc->alarm_irq);
+
+	/* Stock Android uses the 1 Hz interrupt for "secure clock daemon",
+	 * which is not supported by the mainline kernel. The mainline kernel
+	 * does not use the irq at the moment, but we explicitly request and
+	 * disable it, so that its masked and does not wake up the processor
+	 * every second.
+	 */
+	rtc->update_irq = platform_get_irq(pdev, 1);
+	err = devm_request_threaded_irq(dev, rtc->update_irq, NULL,
+					cpcap_rtc_update_irq,
+					IRQF_TRIGGER_NONE | IRQF_ONESHOT,
+					"rtc_1hz", rtc);
+	if (err) {
+		dev_err(dev, "Could not request update irq: %d\n", err);
+		return err;
+	}
+	disable_irq(rtc->update_irq);
+
+	err = device_init_wakeup(dev, true);
+	if (err) {
+		dev_err(dev, "wakeup initialization failed (%d)\n", err);
+		/* ignore error and continue without wakeup support */
+	}
+
+	return devm_rtc_register_device(rtc->rtc_dev);
+}
+
+static const struct of_device_id cpcap_rtc_of_match[] = {
+	{ .compatible = "motorola,cpcap-rtc", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, cpcap_rtc_of_match);
+
+static struct platform_driver cpcap_rtc_driver = {
+	.probe		= cpcap_rtc_probe,
+	.driver		= {
+		.name	= "cpcap-rtc",
+		.of_match_table = cpcap_rtc_of_match,
+	},
+};
+
+module_platform_driver(cpcap_rtc_driver);
+
+MODULE_DESCRIPTION("CPCAP RTC driver");
+MODULE_AUTHOR("Sebastian Reichel <sre@kernel.org>");
+MODULE_LICENSE("GPL");