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, 127 insertions, 0 deletions

diff --git a/drivers/pinctrl/mediatek/mtk-eint.h b/drivers/pinctrl/mediatek/mtk-eint.h
new file mode 100644
index 000000000..3cdd6f631
--- /dev/null
+++ b/drivers/pinctrl/mediatek/mtk-eint.h
@@ -0,0 +1,127 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2014-2025 MediaTek Inc.
+ *
+ * Author: Maoguang Meng <maoguang.meng@mediatek.com>
+ *	   Sean Wang <sean.wang@mediatek.com>
+ *	   Hao Chang <ot_chhao.chang@mediatek.com>
+ *	   Qingliang Li <qingliang.li@mediatek.com>
+ */
+#ifndef __MTK_EINT_H
+#define __MTK_EINT_H
+
+#include <linux/irqdomain.h>
+
+struct mtk_eint_regs {
+	unsigned int	stat;
+	unsigned int	ack;
+	unsigned int	mask;
+	unsigned int	mask_set;
+	unsigned int	mask_clr;
+	unsigned int	sens;
+	unsigned int	sens_set;
+	unsigned int	sens_clr;
+	unsigned int	soft;
+	unsigned int	soft_set;
+	unsigned int	soft_clr;
+	unsigned int	pol;
+	unsigned int	pol_set;
+	unsigned int	pol_clr;
+	unsigned int	dom_en;
+	unsigned int	dbnc_ctrl;
+	unsigned int	dbnc_set;
+	unsigned int	dbnc_clr;
+};
+
+struct mtk_eint_hw {
+	u8		port_mask;
+	u8		ports;
+	unsigned int	ap_num;
+	unsigned int	db_cnt;
+	const unsigned int *db_time;
+};
+
+struct mtk_eint_pin {
+	u16 number;
+	u8 instance;
+	u8 index;
+	bool debounce;
+	bool dual_edge;
+};
+
+extern const unsigned int debounce_time_mt2701[];
+extern const unsigned int debounce_time_mt6765[];
+extern const unsigned int debounce_time_mt6795[];
+extern const unsigned int debounce_time_mt6878[];
+
+struct mtk_eint;
+
+struct mtk_eint_xt {
+	int (*get_gpio_n)(void *data, unsigned long eint_n,
+			  unsigned int *gpio_n,
+			  struct gpio_chip **gpio_chip);
+	int (*get_gpio_state)(void *data, unsigned long eint_n);
+	int (*set_gpio_as_eint)(void *data, unsigned long eint_n);
+};
+
+struct mtk_eint {
+	struct device *dev;
+	void __iomem **base;
+	int nbase;
+	u16 *base_pin_num;
+	struct irq_domain *domain;
+	int irq;
+
+	int *dual_edge;
+	u16 **pin_list;
+	u32 **wake_mask;
+	u32 **cur_mask;
+
+	/* Used to fit into various EINT device */
+	const struct mtk_eint_hw *hw;
+	const struct mtk_eint_regs *regs;
+	struct mtk_eint_pin *pins;
+	u16 num_db_time;
+
+	/* Used to fit into various pinctrl device */
+	void *pctl;
+	const struct mtk_eint_xt *gpio_xlate;
+};
+
+#if IS_ENABLED(CONFIG_EINT_MTK)
+int mtk_eint_do_init(struct mtk_eint *eint, struct mtk_eint_pin *eint_pin);
+int mtk_eint_do_suspend(struct mtk_eint *eint);
+int mtk_eint_do_resume(struct mtk_eint *eint);
+int mtk_eint_set_debounce(struct mtk_eint *eint, unsigned long eint_n,
+			  unsigned int debounce);
+int mtk_eint_find_irq(struct mtk_eint *eint, unsigned long eint_n);
+
+#else
+static inline int mtk_eint_do_init(struct mtk_eint *eint,
+				   struct mtk_eint_pin *eint_pin)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int mtk_eint_do_suspend(struct mtk_eint *eint)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int mtk_eint_do_resume(struct mtk_eint *eint)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int mtk_eint_set_debounce(struct mtk_eint *eint, unsigned long eint_n,
+			  unsigned int debounce)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int mtk_eint_find_irq(struct mtk_eint *eint, unsigned long eint_n)
+{
+	return -EOPNOTSUPP;
+}
+#endif
+#endif /* __MTK_EINT_H */