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, 220 insertions, 0 deletions

diff --git a/arch/m68k/atari/stdma.c b/arch/m68k/atari/stdma.c
new file mode 100644
index 000000000..155fefff1
--- /dev/null
+++ b/arch/m68k/atari/stdma.c
@@ -0,0 +1,220 @@
+/*
+ *  linux/arch/m68k/atari/stmda.c
+ *
+ *  Copyright (C) 1994 Roman Hodek
+ *
+ *
+ * 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.
+ */
+
+
+/* This file contains some function for controlling the access to the  */
+/* ST-DMA chip that may be shared between devices. Currently we have:  */
+/*   TT:     Floppy and ACSI bus                                       */
+/*   Falcon: Floppy and SCSI                                           */
+/*                                                                     */
+/* The controlling functions set up a wait queue for access to the     */
+/* ST-DMA chip. Callers to stdma_lock() that cannot granted access are */
+/* put onto a queue and waked up later if the owner calls              */
+/* stdma_release(). Additionally, the caller gives his interrupt       */
+/* service routine to stdma_lock().                                    */
+/*                                                                     */
+/* On the Falcon, the IDE bus uses just the ACSI/Floppy interrupt, but */
+/* not the ST-DMA chip itself. So falhd.c needs not to lock the        */
+/* chip. The interrupt is routed to falhd.c if IDE is configured, the  */
+/* model is a Falcon and the interrupt was caused by the HD controller */
+/* (can be determined by looking at its status register).              */
+
+
+#include <linux/types.h>
+#include <linux/kdev_t.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/wait.h>
+#include <linux/module.h>
+
+#include <asm/atari_stdma.h>
+#include <asm/atariints.h>
+#include <asm/atarihw.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+
+static int stdma_locked;			/* the semaphore */
+						/* int func to be called */
+static irq_handler_t stdma_isr;
+static void *stdma_isr_data;			/* data passed to isr */
+static DECLARE_WAIT_QUEUE_HEAD(stdma_wait);	/* wait queue for ST-DMA */
+
+
+
+
+/***************************** Prototypes *****************************/
+
+static irqreturn_t stdma_int (int irq, void *dummy);
+
+/************************* End of Prototypes **************************/
+
+
+/**
+ * stdma_try_lock - attempt to acquire ST DMA interrupt "lock"
+ * @handler: interrupt handler to use after acquisition
+ * @data: cookie passed to the interrupt handler function
+ *
+ * Returns !0 if lock was acquired; otherwise 0.
+ */
+
+int stdma_try_lock(irq_handler_t handler, void *data)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	if (stdma_locked) {
+		local_irq_restore(flags);
+		return 0;
+	}
+
+	stdma_locked   = 1;
+	stdma_isr      = handler;
+	stdma_isr_data = data;
+	local_irq_restore(flags);
+	return 1;
+}
+EXPORT_SYMBOL(stdma_try_lock);
+
+
+/*
+ * Function: void stdma_lock( isrfunc isr, void *data )
+ *
+ * Purpose: Tries to get a lock on the ST-DMA chip that is used by more
+ *   then one device driver. Waits on stdma_wait until lock is free.
+ *   stdma_lock() may not be called from an interrupt! You have to
+ *   get the lock in your main routine and release it when your
+ *   request is finished.
+ *
+ * Inputs: A interrupt function that is called until the lock is
+ *   released.
+ *
+ * Returns: nothing
+ *
+ */
+
+void stdma_lock(irq_handler_t handler, void *data)
+{
+	/* Since the DMA is used for file system purposes, we
+	 have to sleep uninterruptible (there may be locked
+	 buffers) */
+	wait_event(stdma_wait, stdma_try_lock(handler, data));
+}
+EXPORT_SYMBOL(stdma_lock);
+
+
+/*
+ * Function: void stdma_release( void )
+ *
+ * Purpose: Releases the lock on the ST-DMA chip.
+ *
+ * Inputs: none
+ *
+ * Returns: nothing
+ *
+ */
+
+void stdma_release(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+
+	stdma_locked   = 0;
+	stdma_isr      = NULL;
+	stdma_isr_data = NULL;
+	wake_up(&stdma_wait);
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(stdma_release);
+
+
+/**
+ * stdma_is_locked_by - allow lock holder to check whether it needs to release.
+ * @handler: interrupt handler previously used to acquire lock.
+ *
+ * Returns !0 if locked for the given handler; 0 otherwise.
+ */
+
+int stdma_is_locked_by(irq_handler_t handler)
+{
+	unsigned long flags;
+	int result;
+
+	local_irq_save(flags);
+	result = stdma_locked && (stdma_isr == handler);
+	local_irq_restore(flags);
+
+	return result;
+}
+EXPORT_SYMBOL(stdma_is_locked_by);
+
+
+/*
+ * Function: int stdma_islocked( void )
+ *
+ * Purpose: Check if the ST-DMA is currently locked.
+ * Note: Returned status is only valid if ints are disabled while calling and
+ *       as long as they remain disabled.
+ *       If called with ints enabled, status can change only from locked to
+ *       unlocked, because ints may not lock the ST-DMA.
+ *
+ * Inputs: none
+ *
+ * Returns: != 0 if locked, 0 otherwise
+ *
+ */
+
+int stdma_islocked(void)
+{
+	return stdma_locked;
+}
+EXPORT_SYMBOL(stdma_islocked);
+
+
+/*
+ * Function: void stdma_init( void )
+ *
+ * Purpose: Initialize the ST-DMA chip access controlling.
+ *   It sets up the interrupt and its service routine. The int is registered
+ *   as slow int, client devices have to live with that (no problem
+ *   currently).
+ *
+ * Inputs: none
+ *
+ * Return: nothing
+ *
+ */
+
+void __init stdma_init(void)
+{
+	stdma_isr = NULL;
+	if (request_irq(IRQ_MFP_FDC, stdma_int, IRQF_SHARED,
+			"ST-DMA floppy,ACSI,IDE,Falcon-SCSI", stdma_int))
+		pr_err("Couldn't register ST-DMA interrupt\n");
+}
+
+
+/*
+ * Function: void stdma_int()
+ *
+ * Purpose: The interrupt routine for the ST-DMA. It calls the isr
+ *   registered by stdma_lock().
+ *
+ */
+
+static irqreturn_t stdma_int(int irq, void *dummy)
+{
+  if (stdma_isr)
+      (*stdma_isr)(irq, stdma_isr_data);
+  return IRQ_HANDLED;
+}