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, 103 insertions, 0 deletions

diff --git a/arch/arm/lib/delay.c b/arch/arm/lib/delay.c
new file mode 100644
index 000000000..7e9304012
--- /dev/null
+++ b/arch/arm/lib/delay.c
@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Delay loops based on the OpenRISC implementation.
+ *
+ * Copyright (C) 2012 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#include <linux/clocksource.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+/*
+ * Default to the loop-based delay implementation.
+ */
+struct arm_delay_ops arm_delay_ops __ro_after_init = {
+	.delay		= __loop_delay,
+	.const_udelay	= __loop_const_udelay,
+	.udelay		= __loop_udelay,
+};
+
+static const struct delay_timer *delay_timer;
+static bool delay_calibrated;
+static u64 delay_res;
+
+bool delay_read_timer(unsigned long *timer_val)
+{
+	if (!delay_timer)
+		return false;
+	*timer_val = delay_timer->read_current_timer();
+	return true;
+}
+EXPORT_SYMBOL_GPL(delay_read_timer);
+
+static inline u64 cyc_to_ns(u64 cyc, u32 mult, u32 shift)
+{
+	return (cyc * mult) >> shift;
+}
+
+static void __timer_delay(unsigned long cycles)
+{
+	cycles_t start = get_cycles();
+
+	while ((get_cycles() - start) < cycles)
+		cpu_relax();
+}
+
+static void __timer_const_udelay(unsigned long xloops)
+{
+	unsigned long long loops = xloops;
+	loops *= arm_delay_ops.ticks_per_jiffy;
+	__timer_delay(loops >> UDELAY_SHIFT);
+}
+
+static void __timer_udelay(unsigned long usecs)
+{
+	__timer_const_udelay(usecs * UDELAY_MULT);
+}
+
+void __init register_current_timer_delay(const struct delay_timer *timer)
+{
+	u32 new_mult, new_shift;
+	u64 res;
+
+	clocks_calc_mult_shift(&new_mult, &new_shift, timer->freq,
+			       NSEC_PER_SEC, 3600);
+	res = cyc_to_ns(1ULL, new_mult, new_shift);
+
+	if (res > 1000) {
+		pr_err("Ignoring delay timer %ps, which has insufficient resolution of %lluns\n",
+			timer, res);
+		return;
+	}
+
+	if (!delay_calibrated && (!delay_res || (res < delay_res))) {
+		pr_info("Switching to timer-based delay loop, resolution %lluns\n", res);
+		delay_timer			= timer;
+		lpj_fine			= timer->freq / HZ;
+		delay_res			= res;
+
+		/* cpufreq may scale loops_per_jiffy, so keep a private copy */
+		arm_delay_ops.ticks_per_jiffy	= lpj_fine;
+		arm_delay_ops.delay		= __timer_delay;
+		arm_delay_ops.const_udelay	= __timer_const_udelay;
+		arm_delay_ops.udelay		= __timer_udelay;
+	} else {
+		pr_info("Ignoring duplicate/late registration of read_current_timer delay\n");
+	}
+}
+
+unsigned long calibrate_delay_is_known(void)
+{
+	delay_calibrated = true;
+	return lpj_fine;
+}
+
+void calibration_delay_done(void)
+{
+	delay_calibrated = true;
+}