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, 485 insertions, 0 deletions

diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
new file mode 100644
index 000000000..8df20b021
--- /dev/null
+++ b/drivers/dma-buf/sw_sync.c
@@ -0,0 +1,485 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Sync File validation framework
+ *
+ * Copyright (C) 2012 Google, Inc.
+ */
+
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/uaccess.h>
+#include <linux/panic.h>
+#include <linux/slab.h>
+#include <linux/sync_file.h>
+
+#include "sync_debug.h"
+
+#define CREATE_TRACE_POINTS
+#include "sync_trace.h"
+
+/*
+ * SW SYNC validation framework
+ *
+ * A sync object driver that uses a 32bit counter to coordinate
+ * synchronization.  Useful when there is no hardware primitive backing
+ * the synchronization.
+ *
+ * To start the framework just open:
+ *
+ * <debugfs>/sync/sw_sync
+ *
+ * That will create a sync timeline, all fences created under this timeline
+ * file descriptor will belong to the this timeline.
+ *
+ * The 'sw_sync' file can be opened many times as to create different
+ * timelines.
+ *
+ * Fences can be created with SW_SYNC_IOC_CREATE_FENCE ioctl with struct
+ * sw_sync_create_fence_data as parameter.
+ *
+ * To increment the timeline counter, SW_SYNC_IOC_INC ioctl should be used
+ * with the increment as u32. This will update the last signaled value
+ * from the timeline and signal any fence that has a seqno smaller or equal
+ * to it.
+ *
+ * struct sw_sync_create_fence_data
+ * @value:	the seqno to initialise the fence with
+ * @name:	the name of the new sync point
+ * @fence:	return the fd of the new sync_file with the created fence
+ */
+struct sw_sync_create_fence_data {
+	__u32	value;
+	char	name[32];
+	__s32	fence; /* fd of new fence */
+};
+
+/**
+ * struct sw_sync_get_deadline - get the deadline hint of a sw_sync fence
+ * @deadline_ns: absolute time of the deadline
+ * @pad:	must be zero
+ * @fence_fd:	the sw_sync fence fd (in)
+ *
+ * Return the earliest deadline set on the fence.  The timebase for the
+ * deadline is CLOCK_MONOTONIC (same as vblank).  If there is no deadline
+ * set on the fence, this ioctl will return -ENOENT.
+ */
+struct sw_sync_get_deadline {
+	__u64	deadline_ns;
+	__u32	pad;
+	__s32	fence_fd;
+};
+
+#define SW_SYNC_IOC_MAGIC	'W'
+
+#define SW_SYNC_IOC_CREATE_FENCE	_IOWR(SW_SYNC_IOC_MAGIC, 0,\
+		struct sw_sync_create_fence_data)
+
+#define SW_SYNC_IOC_INC			_IOW(SW_SYNC_IOC_MAGIC, 1, __u32)
+#define SW_SYNC_GET_DEADLINE		_IOWR(SW_SYNC_IOC_MAGIC, 2, \
+		struct sw_sync_get_deadline)
+
+
+#define SW_SYNC_HAS_DEADLINE_BIT	DMA_FENCE_FLAG_USER_BITS
+
+static const struct dma_fence_ops timeline_fence_ops;
+
+static inline struct sync_pt *dma_fence_to_sync_pt(struct dma_fence *fence)
+{
+	if (fence->ops != &timeline_fence_ops)
+		return NULL;
+	return container_of(fence, struct sync_pt, base);
+}
+
+/**
+ * sync_timeline_create() - creates a sync object
+ * @name:	sync_timeline name
+ *
+ * Creates a new sync_timeline. Returns the sync_timeline object or NULL in
+ * case of error.
+ */
+static struct sync_timeline *sync_timeline_create(const char *name)
+{
+	struct sync_timeline *obj;
+
+	obj = kzalloc_obj(*obj);
+	if (!obj)
+		return NULL;
+
+	kref_init(&obj->kref);
+	obj->context = dma_fence_context_alloc(1);
+	strscpy(obj->name, name, sizeof(obj->name));
+
+	obj->pt_tree = RB_ROOT;
+	INIT_LIST_HEAD(&obj->pt_list);
+	spin_lock_init(&obj->lock);
+
+	sync_timeline_debug_add(obj);
+
+	return obj;
+}
+
+static void sync_timeline_free(struct kref *kref)
+{
+	struct sync_timeline *obj =
+		container_of(kref, struct sync_timeline, kref);
+
+	sync_timeline_debug_remove(obj);
+
+	kfree(obj);
+}
+
+static void sync_timeline_get(struct sync_timeline *obj)
+{
+	kref_get(&obj->kref);
+}
+
+static void sync_timeline_put(struct sync_timeline *obj)
+{
+	kref_put(&obj->kref, sync_timeline_free);
+}
+
+static const char *timeline_fence_get_driver_name(struct dma_fence *fence)
+{
+	return "sw_sync";
+}
+
+static const char *timeline_fence_get_timeline_name(struct dma_fence *fence)
+{
+	struct sync_timeline *parent = dma_fence_parent(fence);
+
+	return parent->name;
+}
+
+static void timeline_fence_release(struct dma_fence *fence)
+{
+	struct sync_pt *pt = dma_fence_to_sync_pt(fence);
+	struct sync_timeline *parent = dma_fence_parent(fence);
+	unsigned long flags;
+
+	dma_fence_lock_irqsave(fence, flags);
+	if (!list_empty(&pt->link)) {
+		list_del(&pt->link);
+		rb_erase(&pt->node, &parent->pt_tree);
+	}
+	dma_fence_unlock_irqrestore(fence, flags);
+
+	sync_timeline_put(parent);
+	dma_fence_free(fence);
+}
+
+static bool timeline_fence_signaled(struct dma_fence *fence)
+{
+	struct sync_timeline *parent = dma_fence_parent(fence);
+
+	return !__dma_fence_is_later(fence, fence->seqno, parent->value);
+}
+
+static void timeline_fence_set_deadline(struct dma_fence *fence, ktime_t deadline)
+{
+	struct sync_pt *pt = dma_fence_to_sync_pt(fence);
+	unsigned long flags;
+
+	dma_fence_lock_irqsave(fence, flags);
+	if (test_bit(SW_SYNC_HAS_DEADLINE_BIT, &fence->flags)) {
+		if (ktime_before(deadline, pt->deadline))
+			pt->deadline = deadline;
+	} else {
+		pt->deadline = deadline;
+		__set_bit(SW_SYNC_HAS_DEADLINE_BIT, &fence->flags);
+	}
+	dma_fence_unlock_irqrestore(fence, flags);
+}
+
+static const struct dma_fence_ops timeline_fence_ops = {
+	.get_driver_name = timeline_fence_get_driver_name,
+	.get_timeline_name = timeline_fence_get_timeline_name,
+	.signaled = timeline_fence_signaled,
+	.release = timeline_fence_release,
+	.set_deadline = timeline_fence_set_deadline,
+};
+
+/**
+ * sync_timeline_signal() - signal a status change on a sync_timeline
+ * @obj:	sync_timeline to signal
+ * @inc:	num to increment on timeline->value
+ *
+ * A sync implementation should call this any time one of it's fences
+ * has signaled or has an error condition.
+ */
+static void sync_timeline_signal(struct sync_timeline *obj, unsigned int inc)
+{
+	LIST_HEAD(signalled);
+	struct sync_pt *pt, *next;
+
+	trace_sync_timeline(obj);
+
+	spin_lock_irq(&obj->lock);
+
+	obj->value += inc;
+
+	list_for_each_entry_safe(pt, next, &obj->pt_list, link) {
+		if (!timeline_fence_signaled(&pt->base))
+			break;
+
+		dma_fence_get(&pt->base);
+
+		list_move_tail(&pt->link, &signalled);
+		rb_erase(&pt->node, &obj->pt_tree);
+
+		dma_fence_signal_locked(&pt->base);
+	}
+
+	spin_unlock_irq(&obj->lock);
+
+	list_for_each_entry_safe(pt, next, &signalled, link) {
+		list_del_init(&pt->link);
+		dma_fence_put(&pt->base);
+	}
+}
+
+/**
+ * sync_pt_create() - creates a sync pt
+ * @obj:	parent sync_timeline
+ * @value:	value of the fence
+ *
+ * Creates a new sync_pt (fence) as a child of @parent.  @size bytes will be
+ * allocated allowing for implementation specific data to be kept after
+ * the generic sync_timeline struct. Returns the sync_pt object or
+ * NULL in case of error.
+ */
+static struct sync_pt *sync_pt_create(struct sync_timeline *obj,
+				      unsigned int value)
+{
+	struct sync_pt *pt;
+
+	pt = kzalloc_obj(*pt);
+	if (!pt)
+		return NULL;
+
+	sync_timeline_get(obj);
+	dma_fence_init(&pt->base, &timeline_fence_ops, &obj->lock,
+		       obj->context, value);
+	INIT_LIST_HEAD(&pt->link);
+
+	spin_lock_irq(&obj->lock);
+	if (!dma_fence_is_signaled_locked(&pt->base)) {
+		struct rb_node **p = &obj->pt_tree.rb_node;
+		struct rb_node *parent = NULL;
+
+		while (*p) {
+			struct sync_pt *other;
+			int cmp;
+
+			parent = *p;
+			other = rb_entry(parent, typeof(*pt), node);
+			cmp = value - other->base.seqno;
+			if (cmp > 0) {
+				p = &parent->rb_right;
+			} else if (cmp < 0) {
+				p = &parent->rb_left;
+			} else {
+				if (dma_fence_get_rcu(&other->base)) {
+					sync_timeline_put(obj);
+					kfree(pt);
+					pt = other;
+					goto unlock;
+				}
+				p = &parent->rb_left;
+			}
+		}
+		rb_link_node(&pt->node, parent, p);
+		rb_insert_color(&pt->node, &obj->pt_tree);
+
+		parent = rb_next(&pt->node);
+		list_add_tail(&pt->link,
+			      parent ? &rb_entry(parent, typeof(*pt), node)->link : &obj->pt_list);
+	}
+unlock:
+	spin_unlock_irq(&obj->lock);
+
+	return pt;
+}
+
+/*
+ * *WARNING*
+ *
+ * improper use of this can result in deadlocking kernel drivers from userspace.
+ */
+
+/* opening sw_sync create a new sync obj */
+static int sw_sync_debugfs_open(struct inode *inode, struct file *file)
+{
+	struct sync_timeline *obj;
+	char task_comm[TASK_COMM_LEN];
+
+	get_task_comm(task_comm, current);
+
+	obj = sync_timeline_create(task_comm);
+	if (!obj)
+		return -ENOMEM;
+
+	file->private_data = obj;
+
+	return 0;
+}
+
+static int sw_sync_debugfs_release(struct inode *inode, struct file *file)
+{
+	struct sync_timeline *obj = file->private_data;
+	struct sync_pt *pt, *next;
+
+	spin_lock_irq(&obj->lock);
+
+	list_for_each_entry_safe(pt, next, &obj->pt_list, link) {
+		dma_fence_set_error(&pt->base, -ENOENT);
+		dma_fence_signal_locked(&pt->base);
+	}
+
+	spin_unlock_irq(&obj->lock);
+
+	sync_timeline_put(obj);
+	return 0;
+}
+
+static long sw_sync_ioctl_create_fence(struct sync_timeline *obj,
+				       unsigned long arg)
+{
+	int fd = get_unused_fd_flags(O_CLOEXEC);
+	int err;
+	struct sync_pt *pt;
+	struct sync_file *sync_file;
+	struct sw_sync_create_fence_data data;
+
+	/* SW sync fence are inherently unsafe and can deadlock the kernel */
+	add_taint(TAINT_SOFTLOCKUP, LOCKDEP_STILL_OK);
+
+	if (fd < 0)
+		return fd;
+
+	if (copy_from_user(&data, (void __user *)arg, sizeof(data))) {
+		err = -EFAULT;
+		goto err;
+	}
+
+	pt = sync_pt_create(obj, data.value);
+	if (!pt) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	sync_file = sync_file_create(&pt->base);
+	dma_fence_put(&pt->base);
+	if (!sync_file) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	data.fence = fd;
+	if (copy_to_user((void __user *)arg, &data, sizeof(data))) {
+		fput(sync_file->file);
+		err = -EFAULT;
+		goto err;
+	}
+
+	fd_install(fd, sync_file->file);
+
+	return 0;
+
+err:
+	put_unused_fd(fd);
+	return err;
+}
+
+static long sw_sync_ioctl_inc(struct sync_timeline *obj, unsigned long arg)
+{
+	u32 value;
+
+	if (copy_from_user(&value, (void __user *)arg, sizeof(value)))
+		return -EFAULT;
+
+	while (value > INT_MAX)  {
+		sync_timeline_signal(obj, INT_MAX);
+		value -= INT_MAX;
+	}
+
+	sync_timeline_signal(obj, value);
+
+	return 0;
+}
+
+static int sw_sync_ioctl_get_deadline(struct sync_timeline *obj, unsigned long arg)
+{
+	struct sw_sync_get_deadline data;
+	struct dma_fence *fence;
+	unsigned long flags;
+	struct sync_pt *pt;
+	int ret = 0;
+
+	if (copy_from_user(&data, (void __user *)arg, sizeof(data)))
+		return -EFAULT;
+
+	if (data.deadline_ns || data.pad)
+		return -EINVAL;
+
+	fence = sync_file_get_fence(data.fence_fd);
+	if (!fence)
+		return -EINVAL;
+
+	pt = dma_fence_to_sync_pt(fence);
+	if (!pt) {
+		ret = -EINVAL;
+		goto put_fence;
+	}
+
+	dma_fence_lock_irqsave(fence, flags);
+	if (!test_bit(SW_SYNC_HAS_DEADLINE_BIT, &fence->flags)) {
+		ret = -ENOENT;
+		goto unlock;
+	}
+	data.deadline_ns = ktime_to_ns(pt->deadline);
+	dma_fence_unlock_irqrestore(fence, flags);
+
+	dma_fence_put(fence);
+
+	if (ret)
+		return ret;
+
+	if (copy_to_user((void __user *)arg, &data, sizeof(data)))
+		return -EFAULT;
+
+	return 0;
+
+unlock:
+	dma_fence_unlock_irqrestore(fence, flags);
+put_fence:
+	dma_fence_put(fence);
+
+	return ret;
+}
+
+static long sw_sync_ioctl(struct file *file, unsigned int cmd,
+			  unsigned long arg)
+{
+	struct sync_timeline *obj = file->private_data;
+
+	switch (cmd) {
+	case SW_SYNC_IOC_CREATE_FENCE:
+		return sw_sync_ioctl_create_fence(obj, arg);
+
+	case SW_SYNC_IOC_INC:
+		return sw_sync_ioctl_inc(obj, arg);
+
+	case SW_SYNC_GET_DEADLINE:
+		return sw_sync_ioctl_get_deadline(obj, arg);
+
+	default:
+		return -ENOTTY;
+	}
+}
+
+const struct file_operations sw_sync_debugfs_fops = {
+	.open           = sw_sync_debugfs_open,
+	.release        = sw_sync_debugfs_release,
+	.unlocked_ioctl = sw_sync_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
+};