On Wed, Dec 10, 2025 at 04:00:19PM +0100, Melbin K Mathew wrote:

The virtio vsock transport currently derives its TX credit directly
from peer_buf_alloc, which is set from the remote endpoint's
SO_VM_SOCKETS_BUFFER_SIZE value.

Why removing the target tree [net] from the tags?

Also this is a v2, so the tags should have been [PATCH net v2], please 
check it in next versions, more info:

https://www.kernel.org/doc/html/latest/process/submitting-patches.html#subject-line

On the host side this means that the amount of data we are willing to
queue for a connection is scaled by a guest-chosen buffer size,
rather than the host's own vsock configuration. A malicious guest can
advertise a large buffer and read slowly, causing the host to allocate
a correspondingly large amount of sk_buff memory.

Introduce a small helper, virtio_transport_peer_buf_alloc(), that
returns min(peer_buf_alloc, buf_alloc), and use it wherever we consume
peer_buf_alloc:

 - virtio_transport_get_credit()
 - virtio_transport_has_space()
 - virtio_transport_seqpacket_enqueue()

This ensures the effective TX window is bounded by both the peer's
advertised buffer and our own buf_alloc (already clamped to
buffer_max_size via SO_VM_SOCKETS_BUFFER_MAX_SIZE), so a remote guest
cannot force the host to queue more data than allowed by the host's
own vsock settings.

On an unpatched Ubuntu 22.04 host (~64 GiB RAM), running a PoC with
32 guest vsock connections advertising 2 GiB each and reading slowly
drove Slab/SUnreclaim from ~0.5 GiB to ~57 GiB and the system only
recovered after killing the QEMU process.

With this patch applied, rerunning the same PoC yields:

 Before:
   MemFree:        ~61.6 GiB
   MemAvailable:   ~62.3 GiB
   Slab:           ~142 MiB
   SUnreclaim:     ~117 MiB

 After 32 high-credit connections:
   MemFree:        ~61.5 GiB
   MemAvailable:   ~62.3 GiB
   Slab:           ~178 MiB
   SUnreclaim:     ~152 MiB

i.e. only ~35 MiB increase in Slab/SUnreclaim, no host OOM, and the
guest remains responsive.

I think we should include here a summary of what you replied to Michael 
about other transports.

I can't find your reply in the archive, but I mean the reply to
https://lore.kernel.org/netdev/20251210084318-mutt-send-email-mst@kernel.org/ (local)

nit: block comment in this file doesn't leave empty line, so I'd follow
it:

@@ -491,8 +491,7 @@ void virtio_transport_consume_skb_sent(struct sk_buff *skb, bool consume)
  }
  EXPORT_SYMBOL_GPL(virtio_transport_consume_skb_sent);

-/*
- * Return the effective peer buffer size for TX credit computation.
+/* Return the effective peer buffer size for TX credit computation.
   *
   * The peer advertises its receive buffer via peer_buf_alloc, but we
   * cap that to our local buf_alloc (derived from

+ *
+ * The peer advertises its receive buffer via peer_buf_alloc, but we
+ * cap that to our local buf_alloc (derived from
+ * SO_VM_SOCKETS_BUFFER_SIZE and already clamped to buffer_max_size)
+ * so that a remote endpoint cannot force us to queue more data than
+ * our own configuration allows.
+ */
+static u32 virtio_transport_tx_buf_alloc(struct virtio_vsock_sock *vvs)
+{
+	u32 peer  = vvs->peer_buf_alloc;
+	u32 local = vvs->buf_alloc;
+
+	if (peer > local)
+		return local;
+	return peer;
+}
+

I think here Michael was suggesting this:

@@ -502,12 +502,7 @@ EXPORT_SYMBOL_GPL(virtio_transport_consume_skb_sent);
   */
  static u32 virtio_transport_tx_buf_alloc(struct virtio_vsock_sock *vvs)
  {
-       u32 peer  = vvs->peer_buf_alloc;
-       u32 local = vvs->buf_alloc;
-
-       if (peer > local)
-               return local;
-       return peer;
+       return min(vvs->peer_buf_alloc, vvs->buf_alloc);
  }

nit: please align this:

@@ -903,7 +898,7 @@ static s64 virtio_transport_has_space(struct vsock_sock *vsk)
         s64 bytes;

         bytes = (s64)virtio_transport_tx_buf_alloc(vvs) -
-             (vvs->tx_cnt - vvs->peer_fwd_cnt);
+               (vvs->tx_cnt - vvs->peer_fwd_cnt);
         if (bytes < 0)
                 bytes = 0;

Just minor things, but the patch LGTM, thanks!
Stefano

	if (bytes < 0)
		bytes = 0;

-- 
2.34.1