On 5/8/26 16:34, Stefano Garzarella wrote:
On Tue, Aug 04, 2026 at 09:52:36PM +0800, [email protected] wrote:
From: Nguyen Dinh Phi <[email protected]>
Syzbot reported an issue which can be reproduced with these steps:
r0 = socket(AF_VSOCK, SOCK_STREAM, 0)
bind(r0, {VMADDR_CID_ANY, PORT})
connect(r0, {VMADDR_CID_LOCAL, PORT}) -> -1, EPROTO (self-connect)
listen(r0, backlog) -> 0
r1 = socket(AF_VSOCK, SOCK_STREAM, 0)
connect(r1, {VMADDR_CID_LOCAL, PORT}) -> 0
accept(r0) -> -1, EPROTO (stale sk_err)
Basically, it creates a socket (r0) and triggers a self-connect after
binding it. This self-connect fails with EPROTO because it loops back to
r0 while the socket is still in the TCP_SYN_SENT state, causing it to be
incorrectly dispatched to the connecting-client path. The unexpected
packet type encountered there sets sk_err to EPROTO.
After that, it invokes a listen() call on the same socket. This listen()
call succeeds because the kernel's listening path never inspects or
clears sk_err. Then, a new socket (r1) is created as a normal client and
connects to r0. However, vsock_accept() rejects this incoming connection
because the listener's sk_err still holds the EPROTO error from the
earlier failed self-connect.
This rejection causes the child socket created for r1's connection to
never be freed on virtio or hyperv transports; only the VMCI transport
implements pending_work to revisit and clean up a rejected socket.
Fix the issue in blocking connect() by using sock_error() to read the
sk_err to prevent the rejection branch from occurring in this scenario.
sock_error() atomically reads and clears sk_err, ensuring the error is
consumed when vsock_connect() returns and cannot affect subsequent
operations on the same socket. This matches the established pattern
used by other protocol connect() implementations in the network
stack like __inet_stream_connect(), tipc_wait_for_connect()...
For non-blocking connection, vsock_connect_timeout() may set
sk->sk_err after vsock_connect() has returned. To handle it, we also
remove the sk_err checks from vsock_accept(). Nothing in vsock sets
sk_err on a listening socket, so accept() has no reason to inspect it
at all.
Reported-by: [email protected]
Closes: https://syzkaller.appspot.com/bug?extid=1b2c9c4a0f8708082678
Fixes: d021c344051af ("VSOCK: Introduce VM Sockets")
Suggested-by: Michal Luczaj <[email protected]>
Signed-off-by: Nguyen Dinh Phi <[email protected]>
Tested-by: Wupeng Ma <[email protected]>
---
V2: Add reproducer steps to commit message.
V3: Fix truncated title and add annotations to reproducer steps.
V4: Remove sk_err checks from vsock_accept()
net/vmw_vsock/af_vsock.c | 13 ++++---------
1 file changed, 4 insertions(+), 9 deletions(-)
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 622dbd046799..594fe27d2ebe 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -1847,12 +1847,10 @@ static int vsock_connect(struct socket *sock,
struct sockaddr_unsized *addr,
prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
}
- if (sk->sk_err) {
- err = -sk->sk_err;
+ err = sock_error(sk);
+ if (err) {
sk->sk_state = TCP_CLOSE;
sock->state = SS_UNCONNECTED;
- } else {
- err = 0;
}
out_wait:
@@ -1893,7 +1891,7 @@ static int vsock_accept(struct socket *sock,
struct socket *newsock,
timeout = sock_rcvtimeo(listener, arg->flags & O_NONBLOCK);
while ((connected = vsock_dequeue_accept(listener)) == NULL &&
- listener->sk_err == 0 && timeout != 0) {
+ timeout != 0) {
prepare_to_wait(sk_sleep(listener), &wait, TASK_INTERRUPTIBLE);
release_sock(listener);
timeout = schedule_timeout(timeout);
@@ -1906,11 +1904,8 @@ static int vsock_accept(struct socket *sock,
struct socket *newsock,
}
}
- if (listener->sk_err) {
- err = -listener->sk_err;
- } else if (!connected) {
+ if (!connected)
err = -EAGAIN;
- }
if (connected) {
Can this become an `} else {` ?
Or just add a `goto out` when setting `err = -EAGAIN`.
I will update it.
sk_acceptq_removed(listener);
lock_sock_nested(connected, SINGLE_DEPTH_NESTING);
vconnected = vsock_sk(connected);
/* If the listener socket has received an error, then we should
* reject this socket and return. Note that we simply mark the
* socket rejected, drop our reference, and let the cleanup
* function handle the cleanup; the fact that we found it in
* the listener's accept queue guarantees that the cleanup
* function hasn't run yet.
*/
if (err) {
vconnected->rejected = true;
} else {
Should we update this comment too and maybe remove the `if (err)` at
all. With that change I guess `rejected` is never set at the end and
maybe we can remove it at all from `struct vsock_sock`.
Looking at commit d021c344051a ("VSOCK: Introduce VM Sockets") where
`rejected` was introduced, I can't see any path where sk_err is set on
a listener socket, so I guess that path was dead since the beginning.
That seems true, let me verify it.
So now I'm thinking if it's better to split in 2 patches (both with the
same Fixes tag):
- Patch 1: "vsock: remove stale sk_err checks from vsock_accept()"
Where we can also remove `rejected` since it's never set to true since
the beginning
- Patch 2: "vsock: use sock_error() to consume sk_err after a failed
connect"
WDYT?
Yes, I felt the same when I was writing the commit message, but honestly I
didn't know that I could split it into a series when sending the new
version.
So, it may contain 3 patches, if the rejected flag can be removed from
vsock_sock.
Thanks,
Phi
