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, 161 insertions, 0 deletions

diff --git a/drivers/ipack/carriers/tpci200.h b/drivers/ipack/carriers/tpci200.h
new file mode 100644
index 000000000..a2bf31257
--- /dev/null
+++ b/drivers/ipack/carriers/tpci200.h
@@ -0,0 +1,161 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * driver for the carrier TEWS TPCI-200
+ *
+ * Copyright (C) 2009-2012 CERN (www.cern.ch)
+ * Author: Nicolas Serafini, EIC2 SA
+ * Author: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
+ */
+
+#ifndef _TPCI200_H_
+#define _TPCI200_H_
+
+#include <linux/limits.h>
+#include <linux/pci.h>
+#include <linux/spinlock.h>
+#include <linux/swab.h>
+#include <linux/io.h>
+#include <linux/ipack.h>
+
+#define TPCI200_NB_SLOT               0x4
+#define TPCI200_NB_BAR                0x6
+
+#define TPCI200_VENDOR_ID             0x1498
+#define TPCI200_DEVICE_ID             0x30C8
+#define TPCI200_SUBVENDOR_ID          0x1498
+#define TPCI200_SUBDEVICE_ID          0x300A
+
+#define TPCI200_CFG_MEM_BAR           0
+#define TPCI200_IP_INTERFACE_BAR      2
+#define TPCI200_IO_ID_INT_SPACES_BAR  3
+#define TPCI200_MEM16_SPACE_BAR       4
+#define TPCI200_MEM8_SPACE_BAR        5
+
+struct tpci200_regs {
+	__le16	revision;
+	/* writes to control should occur with the mutex held to protect
+	 * read-modify-write operations */
+	__le16  control[4];
+	__le16	reset;
+	__le16	status;
+	u8	reserved[242];
+} __packed;
+
+#define TPCI200_IFACE_SIZE            0x100
+
+#define TPCI200_IO_SPACE_OFF          0x0000
+#define TPCI200_IO_SPACE_INTERVAL     0x0100
+#define TPCI200_IO_SPACE_SIZE         0x0080
+#define TPCI200_ID_SPACE_OFF          0x0080
+#define TPCI200_ID_SPACE_INTERVAL     0x0100
+#define TPCI200_ID_SPACE_SIZE         0x0040
+#define TPCI200_INT_SPACE_OFF         0x00C0
+#define TPCI200_INT_SPACE_INTERVAL    0x0100
+#define TPCI200_INT_SPACE_SIZE        0x0040
+#define TPCI200_IOIDINT_SIZE          0x0400
+
+#define TPCI200_MEM8_SPACE_INTERVAL   0x00400000
+#define TPCI200_MEM8_SPACE_SIZE       0x00400000
+#define TPCI200_MEM16_SPACE_INTERVAL  0x00800000
+#define TPCI200_MEM16_SPACE_SIZE      0x00800000
+
+/* control field in tpci200_regs */
+#define TPCI200_INT0_EN               0x0040
+#define TPCI200_INT1_EN               0x0080
+#define TPCI200_INT0_EDGE             0x0010
+#define TPCI200_INT1_EDGE             0x0020
+#define TPCI200_ERR_INT_EN            0x0008
+#define TPCI200_TIME_INT_EN           0x0004
+#define TPCI200_RECOVER_EN            0x0002
+#define TPCI200_CLK32                 0x0001
+
+/* reset field in tpci200_regs */
+#define TPCI200_A_RESET               0x0001
+#define TPCI200_B_RESET               0x0002
+#define TPCI200_C_RESET               0x0004
+#define TPCI200_D_RESET               0x0008
+
+/* status field in tpci200_regs */
+#define TPCI200_A_TIMEOUT             0x1000
+#define TPCI200_B_TIMEOUT             0x2000
+#define TPCI200_C_TIMEOUT             0x4000
+#define TPCI200_D_TIMEOUT             0x8000
+
+#define TPCI200_A_ERROR               0x0100
+#define TPCI200_B_ERROR               0x0200
+#define TPCI200_C_ERROR               0x0400
+#define TPCI200_D_ERROR               0x0800
+
+#define TPCI200_A_INT0                0x0001
+#define TPCI200_A_INT1                0x0002
+#define TPCI200_B_INT0                0x0004
+#define TPCI200_B_INT1                0x0008
+#define TPCI200_C_INT0                0x0010
+#define TPCI200_C_INT1                0x0020
+#define TPCI200_D_INT0                0x0040
+#define TPCI200_D_INT1                0x0080
+
+#define TPCI200_SLOT_INT_MASK         0x00FF
+
+/* PCI Configuration registers. The PCI bridge is a PLX Technology PCI9030. */
+#define LAS1_DESC		      0x2C
+#define LAS2_DESC		      0x30
+
+/* Bits in the LAS?_DESC registers */
+#define LAS_BIT_BIGENDIAN	      24
+
+#define VME_IOID_SPACE  "IOID"
+#define VME_MEM_SPACE  "MEM"
+
+/**
+ * struct slot_irq - slot IRQ definition.
+ * @vector	Vector number
+ * @handler	Handler called when IRQ arrives
+ * @arg		Handler argument
+ *
+ */
+struct slot_irq {
+	struct ipack_device *holder;
+	int		vector;
+	irqreturn_t	(*handler)(void *);
+	void		*arg;
+};
+
+/**
+ * struct tpci200_slot - data specific to the tpci200 slot.
+ * @slot_id	Slot identification gived to external interface
+ * @irq		Slot IRQ infos
+ * @io_phys	IO physical base address register of the slot
+ * @id_phys	ID physical base address register of the slot
+ * @int_phys	INT physical base address register of the slot
+ * @mem_phys	MEM physical base address register of the slot
+ *
+ */
+struct tpci200_slot {
+	struct slot_irq	    *irq;
+};
+
+/**
+ * struct tpci200_infos - informations specific of the TPCI200 tpci200.
+ * @pci_dev		PCI device
+ * @interface_regs	Pointer to IP interface space (Bar 2)
+ * @ioidint_space	Pointer to IP ID, IO and INT space (Bar 3)
+ * @mem8_space		Pointer to MEM space (Bar 4)
+ *
+ */
+struct tpci200_infos {
+	struct pci_dev			*pdev;
+	struct tpci200_regs __iomem	*interface_regs;
+	void __iomem			*cfg_regs;
+	struct ipack_bus_device		*ipack_bus;
+};
+struct tpci200_board {
+	unsigned int		number;
+	struct mutex		mutex;
+	spinlock_t		regs_lock;
+	struct tpci200_slot	*slots;
+	struct tpci200_infos	*info;
+	phys_addr_t             mod_mem[IPACK_SPACE_COUNT];
+};
+
+#endif /* _TPCI200_H_ */