Received-SPF: pass (google.com: domain of linux-kernel+bounces-202429-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org,
	stable@vger.kernel.org
Cc: Mike Christie <michael.christie@oracle.com>,
	"Michael S . Tsirkin" <mst@redhat.com>,
	Sasha Levin <sashal@kernel.org>,
	jasowang@redhat.com,
	virtualization@lists.linux.dev,
	kvm@vger.kernel.org,
	netdev@vger.kernel.org
Subject: [PATCH AUTOSEL 6.8 6/6] vhost-scsi: Handle vhost_vq_work_queue failures for events
Date: Wed,  5 Jun 2024 07:55:58 -0400
Message-ID: <20240605115602.2964993-6-sashal@kernel.org>
In-Reply-To: <20240605115602.2964993-1-sashal@kernel.org>
References: <20240605115602.2964993-1-sashal@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Mike Christie <michael.christie@oracle.com>

[ Upstream commit b1b2ce58ed23c5d56e0ab299a5271ac01f95b75c ]

Currently, we can try to queue an event's work before the vhost_task is
created. When this happens we just drop it in vhost_scsi_do_plug before
even calling vhost_vq_work_queue. During a device shutdown we do the
same thing after vhost_scsi_clear_endpoint has cleared the backends.

In the next patches we will be able to kill the vhost_task before we
have cleared the endpoint. In that case, vhost_vq_work_queue can fail
and we will leak the event's memory. This has handle the failure by
just freeing the event. This is safe to do, because
vhost_vq_work_queue will only return failure for us when the vhost_task
is killed and so userspace will not be able to handle events if we
sent them.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
Message-Id: <20240316004707.45557-2-michael.christie@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/vhost/scsi.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 282aac45c6909..f34f9895b8984 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -497,10 +497,8 @@ vhost_scsi_do_evt_work(struct vhost_scsi *vs, struct vhost_scsi_evt *evt)
 		vq_err(vq, "Faulted on vhost_scsi_send_event\n");
 }
 
-static void vhost_scsi_evt_work(struct vhost_work *work)
+static void vhost_scsi_complete_events(struct vhost_scsi *vs, bool drop)
 {
-	struct vhost_scsi *vs = container_of(work, struct vhost_scsi,
-					vs_event_work);
 	struct vhost_virtqueue *vq = &vs->vqs[VHOST_SCSI_VQ_EVT].vq;
 	struct vhost_scsi_evt *evt, *t;
 	struct llist_node *llnode;
@@ -508,12 +506,20 @@ static void vhost_scsi_evt_work(struct vhost_work *work)
 	mutex_lock(&vq->mutex);
 	llnode = llist_del_all(&vs->vs_event_list);
 	llist_for_each_entry_safe(evt, t, llnode, list) {
-		vhost_scsi_do_evt_work(vs, evt);
+		if (!drop)
+			vhost_scsi_do_evt_work(vs, evt);
 		vhost_scsi_free_evt(vs, evt);
 	}
 	mutex_unlock(&vq->mutex);
 }
 
+static void vhost_scsi_evt_work(struct vhost_work *work)
+{
+	struct vhost_scsi *vs = container_of(work, struct vhost_scsi,
+					     vs_event_work);
+	vhost_scsi_complete_events(vs, false);
+}
+
 static int vhost_scsi_copy_sgl_to_iov(struct vhost_scsi_cmd *cmd)
 {
 	struct iov_iter *iter = &cmd->saved_iter;
@@ -1509,7 +1515,8 @@ vhost_scsi_send_evt(struct vhost_scsi *vs, struct vhost_virtqueue *vq,
 	}
 
 	llist_add(&evt->list, &vs->vs_event_list);
-	vhost_vq_work_queue(vq, &vs->vs_event_work);
+	if (!vhost_vq_work_queue(vq, &vs->vs_event_work))
+		vhost_scsi_complete_events(vs, true);
 }
 
 static void vhost_scsi_evt_handle_kick(struct vhost_work *work)
-- 
2.43.0