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, 176 insertions, 0 deletions

diff --git a/net/qrtr/tun.c b/net/qrtr/tun.c
new file mode 100644
index 000000000..9726db31f
--- /dev/null
+++ b/net/qrtr/tun.c
@@ -0,0 +1,176 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2018, Linaro Ltd */
+
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/skbuff.h>
+#include <linux/uaccess.h>
+
+#include "qrtr.h"
+
+struct qrtr_tun {
+	struct qrtr_endpoint ep;
+
+	struct sk_buff_head queue;
+	wait_queue_head_t readq;
+};
+
+static int qrtr_tun_send(struct qrtr_endpoint *ep, struct sk_buff *skb)
+{
+	struct qrtr_tun *tun = container_of(ep, struct qrtr_tun, ep);
+
+	skb_queue_tail(&tun->queue, skb);
+
+	/* wake up any blocking processes, waiting for new data */
+	wake_up_interruptible(&tun->readq);
+
+	return 0;
+}
+
+static int qrtr_tun_open(struct inode *inode, struct file *filp)
+{
+	struct qrtr_tun *tun;
+	int ret;
+
+	tun = kzalloc_obj(*tun);
+	if (!tun)
+		return -ENOMEM;
+
+	skb_queue_head_init(&tun->queue);
+	init_waitqueue_head(&tun->readq);
+
+	tun->ep.xmit = qrtr_tun_send;
+
+	filp->private_data = tun;
+
+	ret = qrtr_endpoint_register(&tun->ep, QRTR_EP_NID_AUTO);
+	if (ret)
+		goto out;
+
+	return 0;
+
+out:
+	filp->private_data = NULL;
+	kfree(tun);
+	return ret;
+}
+
+static ssize_t qrtr_tun_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct file *filp = iocb->ki_filp;
+	struct qrtr_tun *tun = filp->private_data;
+	struct sk_buff *skb;
+	int count;
+
+	while (!(skb = skb_dequeue(&tun->queue))) {
+		if (filp->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		/* Wait until we get data or the endpoint goes away */
+		if (wait_event_interruptible(tun->readq,
+					     !skb_queue_empty(&tun->queue)))
+			return -ERESTARTSYS;
+	}
+
+	count = min_t(size_t, iov_iter_count(to), skb->len);
+	if (copy_to_iter(skb->data, count, to) != count)
+		count = -EFAULT;
+
+	kfree_skb(skb);
+
+	return count;
+}
+
+static ssize_t qrtr_tun_write_iter(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct file *filp = iocb->ki_filp;
+	struct qrtr_tun *tun = filp->private_data;
+	size_t len = iov_iter_count(from);
+	ssize_t ret;
+	void *kbuf;
+
+	if (!len)
+		return -EINVAL;
+
+	if (len > KMALLOC_MAX_SIZE)
+		return -ENOMEM;
+
+	kbuf = kzalloc(len, GFP_KERNEL);
+	if (!kbuf)
+		return -ENOMEM;
+
+	if (!copy_from_iter_full(kbuf, len, from)) {
+		kfree(kbuf);
+		return -EFAULT;
+	}
+
+	ret = qrtr_endpoint_post(&tun->ep, kbuf, len);
+
+	kfree(kbuf);
+	return ret < 0 ? ret : len;
+}
+
+static __poll_t qrtr_tun_poll(struct file *filp, poll_table *wait)
+{
+	struct qrtr_tun *tun = filp->private_data;
+	__poll_t mask = 0;
+
+	poll_wait(filp, &tun->readq, wait);
+
+	if (!skb_queue_empty(&tun->queue))
+		mask |= EPOLLIN | EPOLLRDNORM;
+
+	return mask;
+}
+
+static int qrtr_tun_release(struct inode *inode, struct file *filp)
+{
+	struct qrtr_tun *tun = filp->private_data;
+
+	qrtr_endpoint_unregister(&tun->ep);
+
+	/* Discard all SKBs */
+	skb_queue_purge(&tun->queue);
+
+	kfree(tun);
+
+	return 0;
+}
+
+static const struct file_operations qrtr_tun_ops = {
+	.owner = THIS_MODULE,
+	.open = qrtr_tun_open,
+	.poll = qrtr_tun_poll,
+	.read_iter = qrtr_tun_read_iter,
+	.write_iter = qrtr_tun_write_iter,
+	.release = qrtr_tun_release,
+};
+
+static struct miscdevice qrtr_tun_miscdev = {
+	MISC_DYNAMIC_MINOR,
+	"qrtr-tun",
+	&qrtr_tun_ops,
+};
+
+static int __init qrtr_tun_init(void)
+{
+	int ret;
+
+	ret = misc_register(&qrtr_tun_miscdev);
+	if (ret)
+		pr_err("failed to register Qualcomm IPC Router tun device\n");
+
+	return ret;
+}
+
+static void __exit qrtr_tun_exit(void)
+{
+	misc_deregister(&qrtr_tun_miscdev);
+}
+
+module_init(qrtr_tun_init);
+module_exit(qrtr_tun_exit);
+
+MODULE_DESCRIPTION("Qualcomm IPC Router TUN device");
+MODULE_LICENSE("GPL v2");