vhost_scsi: flush IO vqs then send TMF rsp

With one worker we will always send the scsi cmd responses then send the
TMF rsp, because LIO will always complete the scsi cmds first then call
into us to send the TMF response.

With multiple workers, the IO vq workers could be running while the
TMF/ctl vq worker is running so this has us do a flush before completing
the TMF to make sure cmds are completed when it's work is later queued
and run.
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Message-Id: <20230626232307.97930-12-michael.christie@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

drivers/vhost/scsi.c

@@ -1133,12 +1133,27 @@ static void vhost_scsi_tmf_resp_work(struct vhost_work *work)
 {
 	struct vhost_scsi_tmf *tmf = container_of(work, struct vhost_scsi_tmf,
 						  vwork);
-	int resp_code;
+	struct vhost_virtqueue *ctl_vq, *vq;
+	int resp_code, i;
+
+	if (tmf->scsi_resp == TMR_FUNCTION_COMPLETE) {
+		/*
+		 * Flush IO vqs that don't share a worker with the ctl to make
+		 * sure they have sent their responses before us.
+		 */
+		ctl_vq = &tmf->vhost->vqs[VHOST_SCSI_VQ_CTL].vq;
+		for (i = VHOST_SCSI_VQ_IO; i < tmf->vhost->dev.nvqs; i++) {
+			vq = &tmf->vhost->vqs[i].vq;
+
+			if (vhost_vq_is_setup(vq) &&
+			    vq->worker != ctl_vq->worker)
+				vhost_vq_flush(vq);
+		}
 
-	if (tmf->scsi_resp == TMR_FUNCTION_COMPLETE)
 		resp_code = VIRTIO_SCSI_S_FUNCTION_SUCCEEDED;
-	else
+	} else {
 		resp_code = VIRTIO_SCSI_S_FUNCTION_REJECTED;
+	}
 
 	vhost_scsi_send_tmf_resp(tmf->vhost, &tmf->svq->vq, tmf->in_iovs,
 				 tmf->vq_desc, &tmf->resp_iov, resp_code);