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, 124 insertions, 0 deletions

diff --git a/drivers/dma-buf/sync_debug.c b/drivers/dma-buf/sync_debug.c
new file mode 100644
index 000000000..9e5d662cd
--- /dev/null
+++ b/drivers/dma-buf/sync_debug.c
@@ -0,0 +1,124 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Sync File validation framework and debug information
+ *
+ * Copyright (C) 2012 Google, Inc.
+ */
+
+#include <linux/debugfs.h>
+#include "sync_debug.h"
+
+static struct dentry *dbgfs;
+
+static LIST_HEAD(sync_timeline_list_head);
+static DEFINE_SPINLOCK(sync_timeline_list_lock);
+
+void sync_timeline_debug_add(struct sync_timeline *obj)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sync_timeline_list_lock, flags);
+	list_add_tail(&obj->sync_timeline_list, &sync_timeline_list_head);
+	spin_unlock_irqrestore(&sync_timeline_list_lock, flags);
+}
+
+void sync_timeline_debug_remove(struct sync_timeline *obj)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sync_timeline_list_lock, flags);
+	list_del(&obj->sync_timeline_list);
+	spin_unlock_irqrestore(&sync_timeline_list_lock, flags);
+}
+
+static const char *sync_status_str(int status)
+{
+	if (status < 0)
+		return "error";
+
+	if (status > 0)
+		return "signaled";
+
+	return "active";
+}
+
+static void sync_print_fence(struct seq_file *s,
+			     struct dma_fence *fence, bool show)
+{
+	struct sync_timeline *parent = dma_fence_parent(fence);
+	int status;
+
+	status = dma_fence_get_status_locked(fence);
+
+	seq_printf(s, "  %s%sfence %s",
+		   show ? parent->name : "",
+		   show ? "_" : "",
+		   sync_status_str(status));
+
+	if (test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT, &fence->flags)) {
+		struct timespec64 ts64 =
+			ktime_to_timespec64(fence->timestamp);
+
+		seq_printf(s, "@%ptSp", &ts64);
+	}
+
+	seq_printf(s, ": %lld", fence->seqno);
+	seq_printf(s, " / %d", parent->value);
+	seq_putc(s, '\n');
+}
+
+static void sync_print_obj(struct seq_file *s, struct sync_timeline *obj)
+{
+	struct list_head *pos;
+
+	seq_printf(s, "%s: %d\n", obj->name, obj->value);
+
+	spin_lock(&obj->lock); /* Caller already disabled IRQ. */
+	list_for_each(pos, &obj->pt_list) {
+		struct sync_pt *pt = container_of(pos, struct sync_pt, link);
+		sync_print_fence(s, &pt->base, false);
+	}
+	spin_unlock(&obj->lock);
+}
+
+static int sync_info_debugfs_show(struct seq_file *s, void *unused)
+{
+	struct list_head *pos;
+
+	seq_puts(s, "objs:\n--------------\n");
+
+	spin_lock_irq(&sync_timeline_list_lock);
+	list_for_each(pos, &sync_timeline_list_head) {
+		struct sync_timeline *obj =
+			container_of(pos, struct sync_timeline,
+				     sync_timeline_list);
+
+		sync_print_obj(s, obj);
+		seq_putc(s, '\n');
+	}
+	spin_unlock_irq(&sync_timeline_list_lock);
+
+	seq_puts(s, "fences:\n--------------\n");
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(sync_info_debugfs);
+
+static __init int sync_debugfs_init(void)
+{
+	dbgfs = debugfs_create_dir("sync", NULL);
+
+	/*
+	 * The debugfs files won't ever get removed and thus, there is
+	 * no need to protect it against removal races. The use of
+	 * debugfs_create_file_unsafe() is actually safe here.
+	 */
+	debugfs_create_file_unsafe("info", 0444, dbgfs, NULL,
+				   &sync_info_debugfs_fops);
+	debugfs_create_file_unsafe("sw_sync", 0644, dbgfs, NULL,
+				   &sw_sync_debugfs_fops);
+
+	return 0;
+}
+late_initcall(sync_debugfs_init);