Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

Pull virtio updates from Michael Tsirkin:

 - resume support in vdpa/solidrun

 - structure size optimizations in virtio_pci

 - new pds_vdpa driver

 - immediate initialization mechanism for vdpa/ifcvf

 - interrupt bypass for vdpa/mlx5

 - multiple worker support for vhost

 - viirtio net in Intel F2000X-PL support for vdpa/ifcvf

 - fixes, cleanups all over the place

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: (48 commits)
  vhost: Make parameter name match of vhost_get_vq_desc()
  vduse: fix NULL pointer dereference
  vhost: Allow worker switching while work is queueing
  vhost_scsi: add support for worker ioctls
  vhost: allow userspace to create workers
  vhost: replace single worker pointer with xarray
  vhost: add helper to parse userspace vring state/file
  vhost: remove vhost_work_queue
  vhost_scsi: flush IO vqs then send TMF rsp
  vhost_scsi: convert to vhost_vq_work_queue
  vhost_scsi: make SCSI cmd completion per vq
  vhost_sock: convert to vhost_vq_work_queue
  vhost: convert poll work to be vq based
  vhost: take worker or vq for flushing
  vhost: take worker or vq instead of dev for queueing
  vhost, vhost_net: add helper to check if vq has work
  vhost: add vhost_worker pointer to vhost_virtqueue
  vhost: dynamically allocate vhost_worker
  vhost: create worker at end of vhost_dev_set_owner
  virtio_bt: call scheduler when we free unused buffs
  ...

1 files changed, 61 insertions, 42 deletions

diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index bb10fa4bb4f6..c83f7f043470 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -167,6 +167,7 @@ MODULE_PARM_DESC(max_io_vqs, "Set the max number of IO virtqueues a vhost scsi d
 
 struct vhost_scsi_virtqueue {
 	struct vhost_virtqueue vq;
+	struct vhost_scsi *vs;
 	/*
 	 * Reference counting for inflight reqs, used for flush operation. At
 	 * each time, one reference tracks new commands submitted, while we
@@ -181,6 +182,9 @@ struct vhost_scsi_virtqueue {
 	struct vhost_scsi_cmd *scsi_cmds;
 	struct sbitmap scsi_tags;
 	int max_cmds;
+
+	struct vhost_work completion_work;
+	struct llist_head completion_list;
 };
 
 struct vhost_scsi {
@@ -190,12 +194,8 @@ struct vhost_scsi {
 
 	struct vhost_dev dev;
 	struct vhost_scsi_virtqueue *vqs;
-	unsigned long *compl_bitmap;
 	struct vhost_scsi_inflight **old_inflight;
 
-	struct vhost_work vs_completion_work; /* cmd completion work item */
-	struct llist_head vs_completion_list; /* cmd completion queue */
-
 	struct vhost_work vs_event_work; /* evt injection work item */
 	struct llist_head vs_event_list; /* evt injection queue */
 
@@ -353,15 +353,17 @@ static void vhost_scsi_release_cmd(struct se_cmd *se_cmd)
 	if (se_cmd->se_cmd_flags & SCF_SCSI_TMR_CDB) {
 		struct vhost_scsi_tmf *tmf = container_of(se_cmd,
 					struct vhost_scsi_tmf, se_cmd);
+		struct vhost_virtqueue *vq = &tmf->svq->vq;
 
-		vhost_work_queue(&tmf->vhost->dev, &tmf->vwork);
+		vhost_vq_work_queue(vq, &tmf->vwork);
 	} else {
 		struct vhost_scsi_cmd *cmd = container_of(se_cmd,
 					struct vhost_scsi_cmd, tvc_se_cmd);
-		struct vhost_scsi *vs = cmd->tvc_vhost;
+		struct vhost_scsi_virtqueue *svq =  container_of(cmd->tvc_vq,
+					struct vhost_scsi_virtqueue, vq);
 
-		llist_add(&cmd->tvc_completion_list, &vs->vs_completion_list);
-		vhost_work_queue(&vs->dev, &vs->vs_completion_work);
+		llist_add(&cmd->tvc_completion_list, &svq->completion_list);
+		vhost_vq_work_queue(&svq->vq, &svq->completion_work);
 	}
 }
 
@@ -509,17 +511,17 @@ static void vhost_scsi_evt_work(struct vhost_work *work)
  */
 static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 {
-	struct vhost_scsi *vs = container_of(work, struct vhost_scsi,
-					vs_completion_work);
+	struct vhost_scsi_virtqueue *svq = container_of(work,
+				struct vhost_scsi_virtqueue, completion_work);
 	struct virtio_scsi_cmd_resp v_rsp;
 	struct vhost_scsi_cmd *cmd, *t;
 	struct llist_node *llnode;
 	struct se_cmd *se_cmd;
 	struct iov_iter iov_iter;
-	int ret, vq;
+	bool signal = false;
+	int ret;
 
-	bitmap_zero(vs->compl_bitmap, vs->dev.nvqs);
-	llnode = llist_del_all(&vs->vs_completion_list);
+	llnode = llist_del_all(&svq->completion_list);
 	llist_for_each_entry_safe(cmd, t, llnode, tvc_completion_list) {
 		se_cmd = &cmd->tvc_se_cmd;
 
@@ -539,21 +541,17 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 			      cmd->tvc_in_iovs, sizeof(v_rsp));
 		ret = copy_to_iter(&v_rsp, sizeof(v_rsp), &iov_iter);
 		if (likely(ret == sizeof(v_rsp))) {
-			struct vhost_scsi_virtqueue *q;
+			signal = true;
+
 			vhost_add_used(cmd->tvc_vq, cmd->tvc_vq_desc, 0);
-			q = container_of(cmd->tvc_vq, struct vhost_scsi_virtqueue, vq);
-			vq = q - vs->vqs;
-			__set_bit(vq, vs->compl_bitmap);
 		} else
 			pr_err("Faulted on virtio_scsi_cmd_resp\n");
 
 		vhost_scsi_release_cmd_res(se_cmd);
 	}
 
-	vq = -1;
-	while ((vq = find_next_bit(vs->compl_bitmap, vs->dev.nvqs, vq + 1))
-		< vs->dev.nvqs)
-		vhost_signal(&vs->dev, &vs->vqs[vq].vq);
+	if (signal)
+		vhost_signal(&svq->vs->dev, &svq->vq);
 }
 
 static struct vhost_scsi_cmd *
@@ -1135,12 +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);
@@ -1335,11 +1348,9 @@ static void vhost_scsi_ctl_handle_kick(struct vhost_work *work)
 }
 
 static void
-vhost_scsi_send_evt(struct vhost_scsi *vs,
-		   struct vhost_scsi_tpg *tpg,
-		   struct se_lun *lun,
-		   u32 event,
-		   u32 reason)
+vhost_scsi_send_evt(struct vhost_scsi *vs, struct vhost_virtqueue *vq,
+		    struct vhost_scsi_tpg *tpg, struct se_lun *lun,
+		    u32 event, u32 reason)
 {
 	struct vhost_scsi_evt *evt;
 
@@ -1361,7 +1372,7 @@ vhost_scsi_send_evt(struct vhost_scsi *vs,
 	}
 
 	llist_add(&evt->list, &vs->vs_event_list);
-	vhost_work_queue(&vs->dev, &vs->vs_event_work);
+	vhost_vq_work_queue(vq, &vs->vs_event_work);
 }
 
 static void vhost_scsi_evt_handle_kick(struct vhost_work *work)
@@ -1375,7 +1386,8 @@ static void vhost_scsi_evt_handle_kick(struct vhost_work *work)
 		goto out;
 
 	if (vs->vs_events_missed)
-		vhost_scsi_send_evt(vs, NULL, NULL, VIRTIO_SCSI_T_NO_EVENT, 0);
+		vhost_scsi_send_evt(vs, vq, NULL, NULL, VIRTIO_SCSI_T_NO_EVENT,
+				    0);
 out:
 	mutex_unlock(&vq->mutex);
 }
@@ -1770,6 +1782,7 @@ static int vhost_scsi_set_features(struct vhost_scsi *vs, u64 features)
 
 static int vhost_scsi_open(struct inode *inode, struct file *f)
 {
+	struct vhost_scsi_virtqueue *svq;
 	struct vhost_scsi *vs;
 	struct vhost_virtqueue **vqs;
 	int r = -ENOMEM, i, nvqs = vhost_scsi_max_io_vqs;
@@ -1788,10 +1801,6 @@ static int vhost_scsi_open(struct inode *inode, struct file *f)
 	}
 	nvqs += VHOST_SCSI_VQ_IO;
 
-	vs->compl_bitmap = bitmap_alloc(nvqs, GFP_KERNEL);
-	if (!vs->compl_bitmap)
-		goto err_compl_bitmap;
-
 	vs->old_inflight = kmalloc_array(nvqs, sizeof(*vs->old_inflight),
 					 GFP_KERNEL | __GFP_ZERO);
 	if (!vs->old_inflight)
@@ -1806,7 +1815,6 @@ static int vhost_scsi_open(struct inode *inode, struct file *f)
 	if (!vqs)
 		goto err_local_vqs;
 
-	vhost_work_init(&vs->vs_completion_work, vhost_scsi_complete_cmd_work);
 	vhost_work_init(&vs->vs_event_work, vhost_scsi_evt_work);
 
 	vs->vs_events_nr = 0;
@@ -1817,8 +1825,14 @@ static int vhost_scsi_open(struct inode *inode, struct file *f)
 	vs->vqs[VHOST_SCSI_VQ_CTL].vq.handle_kick = vhost_scsi_ctl_handle_kick;
 	vs->vqs[VHOST_SCSI_VQ_EVT].vq.handle_kick = vhost_scsi_evt_handle_kick;
 	for (i = VHOST_SCSI_VQ_IO; i < nvqs; i++) {
-		vqs[i] = &vs->vqs[i].vq;
-		vs->vqs[i].vq.handle_kick = vhost_scsi_handle_kick;
+		svq = &vs->vqs[i];
+
+		vqs[i] = &svq->vq;
+		svq->vs = vs;
+		init_llist_head(&svq->completion_list);
+		vhost_work_init(&svq->completion_work,
+				vhost_scsi_complete_cmd_work);
+		svq->vq.handle_kick = vhost_scsi_handle_kick;
 	}
 	vhost_dev_init(&vs->dev, vqs, nvqs, UIO_MAXIOV,
 		       VHOST_SCSI_WEIGHT, 0, true, NULL);
@@ -1833,8 +1847,6 @@ err_local_vqs:
 err_vqs:
 	kfree(vs->old_inflight);
 err_inflight:
-	bitmap_free(vs->compl_bitmap);
-err_compl_bitmap:
 	kvfree(vs);
 err_vs:
 	return r;
@@ -1854,7 +1866,6 @@ static int vhost_scsi_release(struct inode *inode, struct file *f)
 	kfree(vs->dev.vqs);
 	kfree(vs->vqs);
 	kfree(vs->old_inflight);
-	bitmap_free(vs->compl_bitmap);
 	kvfree(vs);
 	return 0;
 }
@@ -1916,6 +1927,14 @@ vhost_scsi_ioctl(struct file *f,
 		if (copy_from_user(&features, featurep, sizeof features))
 			return -EFAULT;
 		return vhost_scsi_set_features(vs, features);
+	case VHOST_NEW_WORKER:
+	case VHOST_FREE_WORKER:
+	case VHOST_ATTACH_VRING_WORKER:
+	case VHOST_GET_VRING_WORKER:
+		mutex_lock(&vs->dev.mutex);
+		r = vhost_worker_ioctl(&vs->dev, ioctl, argp);
+		mutex_unlock(&vs->dev.mutex);
+		return r;
 	default:
 		mutex_lock(&vs->dev.mutex);
 		r = vhost_dev_ioctl(&vs->dev, ioctl, argp);
@@ -1995,7 +2014,7 @@ vhost_scsi_do_plug(struct vhost_scsi_tpg *tpg,
 		goto unlock;
 
 	if (vhost_has_feature(vq, VIRTIO_SCSI_F_HOTPLUG))
-		vhost_scsi_send_evt(vs, tpg, lun,
+		vhost_scsi_send_evt(vs, vq, tpg, lun,
 				   VIRTIO_SCSI_T_TRANSPORT_RESET, reason);
 unlock:
 	mutex_unlock(&vq->mutex);