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, 90 insertions, 0 deletions

diff --git a/arch/mips/generic/board-ranchu.c b/arch/mips/generic/board-ranchu.c
new file mode 100644
index 000000000..930c45041
--- /dev/null
+++ b/arch/mips/generic/board-ranchu.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Support code for virtual Ranchu board for MIPS.
+ *
+ * Author: Miodrag Dinic <miodrag.dinic@mips.com>
+ */
+
+#include <linux/of_address.h>
+#include <linux/types.h>
+
+#include <asm/machine.h>
+#include <asm/mipsregs.h>
+#include <asm/time.h>
+
+#define GOLDFISH_TIMER_LOW		0x00
+#define GOLDFISH_TIMER_HIGH		0x04
+
+static __init u64 read_rtc_time(void __iomem *base)
+{
+	u32 time_low;
+	u32 time_high;
+
+	/*
+	 * Reading the low address latches the high value
+	 * as well so there is no fear that we may read
+	 * inaccurate high value.
+	 */
+	time_low = readl(base + GOLDFISH_TIMER_LOW);
+	time_high = readl(base + GOLDFISH_TIMER_HIGH);
+
+	return ((u64)time_high << 32) | time_low;
+}
+
+static __init unsigned int ranchu_measure_hpt_freq(void)
+{
+	u64 rtc_start, rtc_current, rtc_delta;
+	unsigned int start, count;
+	struct device_node *np;
+	void __iomem *rtc_base;
+
+	np = of_find_compatible_node(NULL, NULL, "google,goldfish-rtc");
+	if (!np)
+		panic("%s(): Failed to find 'google,goldfish-rtc' dt node!",
+		      __func__);
+
+	rtc_base = of_iomap(np, 0);
+	of_node_put(np);
+	if (!rtc_base)
+		panic("%s(): Failed to ioremap Goldfish RTC base!", __func__);
+
+	/*
+	 * Poll the nanosecond resolution RTC for one
+	 * second to calibrate the CPU frequency.
+	 */
+	rtc_start = read_rtc_time(rtc_base);
+	start = read_c0_count();
+
+	do {
+		rtc_current = read_rtc_time(rtc_base);
+		rtc_delta = rtc_current - rtc_start;
+	} while (rtc_delta < NSEC_PER_SEC);
+
+	count = read_c0_count() - start;
+
+	/*
+	 * Make sure the frequency will be a round number.
+	 * Without this correction, the returned value may vary
+	 * between subsequent emulation executions.
+	 *
+	 * TODO: Set this value using device tree.
+	 */
+	count += 5000;
+	count -= count % 10000;
+
+	iounmap(rtc_base);
+
+	return count;
+}
+
+static const struct of_device_id ranchu_of_match[] __initconst = {
+	{
+		.compatible = "mti,ranchu",
+	},
+	{}
+};
+
+MIPS_MACHINE(ranchu) = {
+	.matches = ranchu_of_match,
+	.measure_hpt_freq = ranchu_measure_hpt_freq,
+};