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, 223 insertions, 0 deletions

diff --git a/drivers/clocksource/timer-npcm7xx.c b/drivers/clocksource/timer-npcm7xx.c
new file mode 100644
index 000000000..9af30af5f
--- /dev/null
+++ b/drivers/clocksource/timer-npcm7xx.c
@@ -0,0 +1,223 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2014-2018 Nuvoton Technologies tomer.maimon@nuvoton.com
+ * All rights reserved.
+ *
+ * Copyright 2017 Google, Inc.
+ */
+
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/err.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/clockchips.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include "timer-of.h"
+
+/* Timers registers */
+#define NPCM7XX_REG_TCSR0	0x0 /* Timer 0 Control and Status Register */
+#define NPCM7XX_REG_TICR0	0x8 /* Timer 0 Initial Count Register */
+#define NPCM7XX_REG_TCSR1	0x4 /* Timer 1 Control and Status Register */
+#define NPCM7XX_REG_TICR1	0xc /* Timer 1 Initial Count Register */
+#define NPCM7XX_REG_TDR1	0x14 /* Timer 1 Data Register */
+#define NPCM7XX_REG_TISR	0x18 /* Timer Interrupt Status Register */
+
+/* Timers control */
+#define NPCM7XX_Tx_RESETINT		0x1f
+#define NPCM7XX_Tx_PERIOD		BIT(27)
+#define NPCM7XX_Tx_INTEN		BIT(29)
+#define NPCM7XX_Tx_COUNTEN		BIT(30)
+#define NPCM7XX_Tx_ONESHOT		0x0
+#define NPCM7XX_Tx_OPER			GENMASK(28, 27)
+#define NPCM7XX_Tx_MIN_PRESCALE		0x1
+#define NPCM7XX_Tx_TDR_MASK_BITS	24
+#define NPCM7XX_Tx_MAX_CNT		0xFFFFFF
+#define NPCM7XX_T0_CLR_INT		0x1
+#define NPCM7XX_Tx_CLR_CSR		0x0
+
+/* Timers operating mode */
+#define NPCM7XX_START_PERIODIC_Tx (NPCM7XX_Tx_PERIOD | NPCM7XX_Tx_COUNTEN | \
+					NPCM7XX_Tx_INTEN | \
+					NPCM7XX_Tx_MIN_PRESCALE)
+
+#define NPCM7XX_START_ONESHOT_Tx (NPCM7XX_Tx_ONESHOT | NPCM7XX_Tx_COUNTEN | \
+					NPCM7XX_Tx_INTEN | \
+					NPCM7XX_Tx_MIN_PRESCALE)
+
+#define NPCM7XX_START_Tx (NPCM7XX_Tx_COUNTEN | NPCM7XX_Tx_PERIOD | \
+				NPCM7XX_Tx_MIN_PRESCALE)
+
+#define NPCM7XX_DEFAULT_CSR (NPCM7XX_Tx_CLR_CSR | NPCM7XX_Tx_MIN_PRESCALE)
+
+static int npcm7xx_timer_resume(struct clock_event_device *evt)
+{
+	struct timer_of *to = to_timer_of(evt);
+	u32 val;
+
+	val = readl(timer_of_base(to) + NPCM7XX_REG_TCSR0);
+	val |= NPCM7XX_Tx_COUNTEN;
+	writel(val, timer_of_base(to) + NPCM7XX_REG_TCSR0);
+
+	return 0;
+}
+
+static int npcm7xx_timer_shutdown(struct clock_event_device *evt)
+{
+	struct timer_of *to = to_timer_of(evt);
+	u32 val;
+
+	val = readl(timer_of_base(to) + NPCM7XX_REG_TCSR0);
+	val &= ~NPCM7XX_Tx_COUNTEN;
+	writel(val, timer_of_base(to) + NPCM7XX_REG_TCSR0);
+
+	return 0;
+}
+
+static int npcm7xx_timer_oneshot(struct clock_event_device *evt)
+{
+	struct timer_of *to = to_timer_of(evt);
+	u32 val;
+
+	val = readl(timer_of_base(to) + NPCM7XX_REG_TCSR0);
+	val &= ~NPCM7XX_Tx_OPER;
+	val |= NPCM7XX_START_ONESHOT_Tx;
+	writel(val, timer_of_base(to) + NPCM7XX_REG_TCSR0);
+
+	return 0;
+}
+
+static int npcm7xx_timer_periodic(struct clock_event_device *evt)
+{
+	struct timer_of *to = to_timer_of(evt);
+	u32 val;
+
+	writel(timer_of_period(to), timer_of_base(to) + NPCM7XX_REG_TICR0);
+
+	val = readl(timer_of_base(to) + NPCM7XX_REG_TCSR0);
+	val &= ~NPCM7XX_Tx_OPER;
+	val |= NPCM7XX_START_PERIODIC_Tx;
+	writel(val, timer_of_base(to) + NPCM7XX_REG_TCSR0);
+
+	return 0;
+}
+
+static int npcm7xx_clockevent_set_next_event(unsigned long evt,
+		struct clock_event_device *clk)
+{
+	struct timer_of *to = to_timer_of(clk);
+	u32 val;
+
+	writel(evt, timer_of_base(to) + NPCM7XX_REG_TICR0);
+	val = readl(timer_of_base(to) + NPCM7XX_REG_TCSR0);
+	val |= NPCM7XX_START_Tx;
+	writel(val, timer_of_base(to) + NPCM7XX_REG_TCSR0);
+
+	return 0;
+}
+
+static irqreturn_t npcm7xx_timer0_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *evt = (struct clock_event_device *)dev_id;
+	struct timer_of *to = to_timer_of(evt);
+
+	writel(NPCM7XX_T0_CLR_INT, timer_of_base(to) + NPCM7XX_REG_TISR);
+
+	evt->event_handler(evt);
+
+	return IRQ_HANDLED;
+}
+
+static struct timer_of npcm7xx_to = {
+	.flags = TIMER_OF_IRQ | TIMER_OF_BASE | TIMER_OF_CLOCK,
+
+	.clkevt = {
+		.name		    = "npcm7xx-timer0",
+		.features	    = CLOCK_EVT_FEAT_PERIODIC |
+				      CLOCK_EVT_FEAT_ONESHOT,
+		.set_next_event	    = npcm7xx_clockevent_set_next_event,
+		.set_state_shutdown = npcm7xx_timer_shutdown,
+		.set_state_periodic = npcm7xx_timer_periodic,
+		.set_state_oneshot  = npcm7xx_timer_oneshot,
+		.tick_resume	    = npcm7xx_timer_resume,
+		.rating		    = 300,
+	},
+
+	.of_irq = {
+		.handler = npcm7xx_timer0_interrupt,
+		.flags = IRQF_TIMER | IRQF_IRQPOLL,
+	},
+};
+
+static void __init npcm7xx_clockevents_init(void)
+{
+	writel(NPCM7XX_DEFAULT_CSR,
+		timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TCSR0);
+
+	writel(NPCM7XX_Tx_RESETINT,
+		timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TISR);
+
+	npcm7xx_to.clkevt.cpumask = cpumask_of(0);
+	clockevents_config_and_register(&npcm7xx_to.clkevt,
+					timer_of_rate(&npcm7xx_to),
+					0x1, NPCM7XX_Tx_MAX_CNT);
+}
+
+static void __init npcm7xx_clocksource_init(void)
+{
+	u32 val;
+
+	writel(NPCM7XX_DEFAULT_CSR,
+		timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TCSR1);
+	writel(NPCM7XX_Tx_MAX_CNT,
+		timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TICR1);
+
+	val = readl(timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TCSR1);
+	val |= NPCM7XX_START_Tx;
+	writel(val, timer_of_base(&npcm7xx_to) + NPCM7XX_REG_TCSR1);
+
+	clocksource_mmio_init(timer_of_base(&npcm7xx_to) +
+				NPCM7XX_REG_TDR1,
+				"npcm7xx-timer1", timer_of_rate(&npcm7xx_to),
+				200, (unsigned int)NPCM7XX_Tx_TDR_MASK_BITS,
+				clocksource_mmio_readl_down);
+}
+
+static int __init npcm7xx_timer_init(struct device_node *np)
+{
+	struct clk *clk;
+	int ret;
+
+	ret = timer_of_init(np, &npcm7xx_to);
+	if (ret)
+		return ret;
+
+	/* Clock input is divided by PRESCALE + 1 before it is fed */
+	/* to the counter */
+	npcm7xx_to.of_clk.rate = npcm7xx_to.of_clk.rate /
+		(NPCM7XX_Tx_MIN_PRESCALE + 1);
+
+	/* Enable the clock for timer1, if it exists */
+	clk = of_clk_get(np, 1);
+	if (clk) {
+		if (!IS_ERR(clk))
+			clk_prepare_enable(clk);
+		else
+			pr_warn("%pOF: Failed to get clock for timer1: %pe", np, clk);
+	}
+
+	npcm7xx_clocksource_init();
+	npcm7xx_clockevents_init();
+
+	pr_info("Enabling NPCM7xx clocksource timer base: %px, IRQ: %d ",
+		timer_of_base(&npcm7xx_to), timer_of_irq(&npcm7xx_to));
+
+	return 0;
+}
+
+TIMER_OF_DECLARE(wpcm450, "nuvoton,wpcm450-timer", npcm7xx_timer_init);
+TIMER_OF_DECLARE(npcm7xx, "nuvoton,npcm750-timer", npcm7xx_timer_init);
+