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, 206 insertions, 0 deletions

diff --git a/arch/mips/bcm63xx/timer.c b/arch/mips/bcm63xx/timer.c
new file mode 100644
index 000000000..74b83807d
--- /dev/null
+++ b/arch/mips/bcm63xx/timer.c
@@ -0,0 +1,206 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2008 Maxime Bizon <mbizon@freebox.fr>
+ */
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/export.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+#include <linux/clk.h>
+#include <bcm63xx_cpu.h>
+#include <bcm63xx_io.h>
+#include <bcm63xx_timer.h>
+#include <bcm63xx_regs.h>
+
+static DEFINE_RAW_SPINLOCK(timer_reg_lock);
+static DEFINE_RAW_SPINLOCK(timer_data_lock);
+static struct clk *periph_clk;
+
+static struct timer_data {
+	void	(*cb)(void *);
+	void	*data;
+} timer_data[BCM63XX_TIMER_COUNT];
+
+static irqreturn_t timer_interrupt(int irq, void *dev_id)
+{
+	u32 stat;
+	int i;
+
+	raw_spin_lock(&timer_reg_lock);
+	stat = bcm_timer_readl(TIMER_IRQSTAT_REG);
+	bcm_timer_writel(stat, TIMER_IRQSTAT_REG);
+	raw_spin_unlock(&timer_reg_lock);
+
+	for (i = 0; i < BCM63XX_TIMER_COUNT; i++) {
+		if (!(stat & TIMER_IRQSTAT_TIMER_CAUSE(i)))
+			continue;
+
+		raw_spin_lock(&timer_data_lock);
+		if (!timer_data[i].cb) {
+			raw_spin_unlock(&timer_data_lock);
+			continue;
+		}
+
+		timer_data[i].cb(timer_data[i].data);
+		raw_spin_unlock(&timer_data_lock);
+	}
+
+	return IRQ_HANDLED;
+}
+
+int bcm63xx_timer_enable(int id)
+{
+	u32 reg;
+	unsigned long flags;
+
+	if (id >= BCM63XX_TIMER_COUNT)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&timer_reg_lock, flags);
+
+	reg = bcm_timer_readl(TIMER_CTLx_REG(id));
+	reg |= TIMER_CTL_ENABLE_MASK;
+	bcm_timer_writel(reg, TIMER_CTLx_REG(id));
+
+	reg = bcm_timer_readl(TIMER_IRQSTAT_REG);
+	reg |= TIMER_IRQSTAT_TIMER_IR_EN(id);
+	bcm_timer_writel(reg, TIMER_IRQSTAT_REG);
+
+	raw_spin_unlock_irqrestore(&timer_reg_lock, flags);
+	return 0;
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_enable);
+
+int bcm63xx_timer_disable(int id)
+{
+	u32 reg;
+	unsigned long flags;
+
+	if (id >= BCM63XX_TIMER_COUNT)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&timer_reg_lock, flags);
+
+	reg = bcm_timer_readl(TIMER_CTLx_REG(id));
+	reg &= ~TIMER_CTL_ENABLE_MASK;
+	bcm_timer_writel(reg, TIMER_CTLx_REG(id));
+
+	reg = bcm_timer_readl(TIMER_IRQSTAT_REG);
+	reg &= ~TIMER_IRQSTAT_TIMER_IR_EN(id);
+	bcm_timer_writel(reg, TIMER_IRQSTAT_REG);
+
+	raw_spin_unlock_irqrestore(&timer_reg_lock, flags);
+	return 0;
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_disable);
+
+int bcm63xx_timer_register(int id, void (*callback)(void *data), void *data)
+{
+	unsigned long flags;
+	int ret;
+
+	if (id >= BCM63XX_TIMER_COUNT || !callback)
+		return -EINVAL;
+
+	ret = 0;
+	raw_spin_lock_irqsave(&timer_data_lock, flags);
+	if (timer_data[id].cb) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	timer_data[id].cb = callback;
+	timer_data[id].data = data;
+
+out:
+	raw_spin_unlock_irqrestore(&timer_data_lock, flags);
+	return ret;
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_register);
+
+void bcm63xx_timer_unregister(int id)
+{
+	unsigned long flags;
+
+	if (id >= BCM63XX_TIMER_COUNT)
+		return;
+
+	raw_spin_lock_irqsave(&timer_data_lock, flags);
+	timer_data[id].cb = NULL;
+	raw_spin_unlock_irqrestore(&timer_data_lock, flags);
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_unregister);
+
+unsigned int bcm63xx_timer_countdown(unsigned int countdown_us)
+{
+	return (clk_get_rate(periph_clk) / (1000 * 1000)) * countdown_us;
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_countdown);
+
+int bcm63xx_timer_set(int id, int monotonic, unsigned int countdown_us)
+{
+	u32 reg, countdown;
+	unsigned long flags;
+
+	if (id >= BCM63XX_TIMER_COUNT)
+		return -EINVAL;
+
+	countdown = bcm63xx_timer_countdown(countdown_us);
+	if (countdown & ~TIMER_CTL_COUNTDOWN_MASK)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&timer_reg_lock, flags);
+	reg = bcm_timer_readl(TIMER_CTLx_REG(id));
+
+	if (monotonic)
+		reg &= ~TIMER_CTL_MONOTONIC_MASK;
+	else
+		reg |= TIMER_CTL_MONOTONIC_MASK;
+
+	reg &= ~TIMER_CTL_COUNTDOWN_MASK;
+	reg |= countdown;
+	bcm_timer_writel(reg, TIMER_CTLx_REG(id));
+
+	raw_spin_unlock_irqrestore(&timer_reg_lock, flags);
+	return 0;
+}
+
+EXPORT_SYMBOL(bcm63xx_timer_set);
+
+static int bcm63xx_timer_init(void)
+{
+	int ret, irq;
+	u32 reg;
+
+	reg = bcm_timer_readl(TIMER_IRQSTAT_REG);
+	reg &= ~TIMER_IRQSTAT_TIMER0_IR_EN;
+	reg &= ~TIMER_IRQSTAT_TIMER1_IR_EN;
+	reg &= ~TIMER_IRQSTAT_TIMER2_IR_EN;
+	bcm_timer_writel(reg, TIMER_IRQSTAT_REG);
+
+	periph_clk = clk_get(NULL, "periph");
+	if (IS_ERR(periph_clk))
+		return -ENODEV;
+
+	irq = bcm63xx_get_irq_number(IRQ_TIMER);
+	ret = request_irq(irq, timer_interrupt, 0, "bcm63xx_timer", NULL);
+	if (ret) {
+		pr_err("%s: failed to register irq\n", __func__);
+		return ret;
+	}
+
+	return 0;
+}
+
+arch_initcall(bcm63xx_timer_init);