4 files changed, 247 insertions, 116 deletions

diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index f7d083bb3bd5..d826bf3e62c8 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -106,7 +106,7 @@ struct nvme_queue {
 	dma_addr_t cq_dma_addr;
 	u32 __iomem *q_db;
 	u16 q_depth;
-	u16 cq_vector;
+	s16 cq_vector;
 	u16 sq_head;
 	u16 sq_tail;
 	u16 cq_head;
@@ -215,6 +215,7 @@ static void nvme_set_info(struct nvme_cmd_info *cmd, void *ctx,
 	cmd->fn = handler;
 	cmd->ctx = ctx;
 	cmd->aborted = 0;
+	blk_mq_start_request(blk_mq_rq_from_pdu(cmd));
 }
 
 /* Special values must be less than 0x1000 */
@@ -431,8 +432,13 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx,
 	if (unlikely(status)) {
 		if (!(status & NVME_SC_DNR || blk_noretry_request(req))
 		    && (jiffies - req->start_time) < req->timeout) {
+			unsigned long flags;
+
 			blk_mq_requeue_request(req);
-			blk_mq_kick_requeue_list(req->q);
+			spin_lock_irqsave(req->q->queue_lock, flags);
+			if (!blk_queue_stopped(req->q))
+				blk_mq_kick_requeue_list(req->q);
+			spin_unlock_irqrestore(req->q->queue_lock, flags);
 			return;
 		}
 		req->errors = nvme_error_status(status);
@@ -664,8 +670,6 @@ static int nvme_queue_rq(struct blk_mq_hw_ctx *hctx,
 		}
 	}
 
-	blk_mq_start_request(req);
-
 	nvme_set_info(cmd, iod, req_completion);
 	spin_lock_irq(&nvmeq->q_lock);
 	if (req->cmd_flags & REQ_DISCARD)
@@ -835,6 +839,7 @@ static int nvme_submit_async_admin_req(struct nvme_dev *dev)
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
+	req->cmd_flags |= REQ_NO_TIMEOUT;
 	cmd_info = blk_mq_rq_to_pdu(req);
 	nvme_set_info(cmd_info, req, async_req_completion);
 
@@ -1016,14 +1021,19 @@ static void nvme_abort_req(struct request *req)
 	struct nvme_command cmd;
 
 	if (!nvmeq->qid || cmd_rq->aborted) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&dev_list_lock, flags);
 		if (work_busy(&dev->reset_work))
-			return;
+			goto out;
 		list_del_init(&dev->node);
 		dev_warn(&dev->pci_dev->dev,
 			"I/O %d QID %d timeout, reset controller\n",
 							req->tag, nvmeq->qid);
 		dev->reset_workfn = nvme_reset_failed_dev;
 		queue_work(nvme_workq, &dev->reset_work);
+ out:
+		spin_unlock_irqrestore(&dev_list_lock, flags);
 		return;
 	}
 
@@ -1064,15 +1074,22 @@ static void nvme_cancel_queue_ios(struct blk_mq_hw_ctx *hctx,
 	void *ctx;
 	nvme_completion_fn fn;
 	struct nvme_cmd_info *cmd;
-	static struct nvme_completion cqe = {
-		.status = cpu_to_le16(NVME_SC_ABORT_REQ << 1),
-	};
+	struct nvme_completion cqe;
+
+	if (!blk_mq_request_started(req))
+		return;
 
 	cmd = blk_mq_rq_to_pdu(req);
 
 	if (cmd->ctx == CMD_CTX_CANCELLED)
 		return;
 
+	if (blk_queue_dying(req->q))
+		cqe.status = cpu_to_le16((NVME_SC_ABORT_REQ | NVME_SC_DNR) << 1);
+	else
+		cqe.status = cpu_to_le16(NVME_SC_ABORT_REQ << 1);
+
+
 	dev_warn(nvmeq->q_dmadev, "Cancelling I/O %d QID %d\n",
 						req->tag, nvmeq->qid);
 	ctx = cancel_cmd_info(cmd, &fn);
@@ -1084,17 +1101,29 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req, bool reserved)
 	struct nvme_cmd_info *cmd = blk_mq_rq_to_pdu(req);
 	struct nvme_queue *nvmeq = cmd->nvmeq;
 
-	dev_warn(nvmeq->q_dmadev, "Timeout I/O %d QID %d\n", req->tag,
-							nvmeq->qid);
-	if (nvmeq->dev->initialized)
-		nvme_abort_req(req);
-
 	/*
 	 * The aborted req will be completed on receiving the abort req.
 	 * We enable the timer again. If hit twice, it'll cause a device reset,
 	 * as the device then is in a faulty state.
 	 */
-	return BLK_EH_RESET_TIMER;
+	int ret = BLK_EH_RESET_TIMER;
+
+	dev_warn(nvmeq->q_dmadev, "Timeout I/O %d QID %d\n", req->tag,
+							nvmeq->qid);
+
+	spin_lock_irq(&nvmeq->q_lock);
+	if (!nvmeq->dev->initialized) {
+		/*
+		 * Force cancelled command frees the request, which requires we
+		 * return BLK_EH_NOT_HANDLED.
+		 */
+		nvme_cancel_queue_ios(nvmeq->hctx, req, nvmeq, reserved);
+		ret = BLK_EH_NOT_HANDLED;
+	} else
+		nvme_abort_req(req);
+	spin_unlock_irq(&nvmeq->q_lock);
+
+	return ret;
 }
 
 static void nvme_free_queue(struct nvme_queue *nvmeq)
@@ -1175,6 +1204,8 @@ static void nvme_disable_queue(struct nvme_dev *dev, int qid)
 		adapter_delete_sq(dev, qid);
 		adapter_delete_cq(dev, qid);
 	}
+	if (!qid && dev->admin_q)
+		blk_mq_freeze_queue_start(dev->admin_q);
 	nvme_clear_queue(nvmeq);
 }
 
@@ -1361,6 +1392,14 @@ static struct blk_mq_ops nvme_mq_ops = {
 	.timeout	= nvme_timeout,
 };
 
+static void nvme_dev_remove_admin(struct nvme_dev *dev)
+{
+	if (dev->admin_q && !blk_queue_dying(dev->admin_q)) {
+		blk_cleanup_queue(dev->admin_q);
+		blk_mq_free_tag_set(&dev->admin_tagset);
+	}
+}
+
 static int nvme_alloc_admin_tags(struct nvme_dev *dev)
 {
 	if (!dev->admin_q) {
@@ -1380,17 +1419,16 @@ static int nvme_alloc_admin_tags(struct nvme_dev *dev)
 			blk_mq_free_tag_set(&dev->admin_tagset);
 			return -ENOMEM;
 		}
-	}
+		if (!blk_get_queue(dev->admin_q)) {
+			nvme_dev_remove_admin(dev);
+			return -ENODEV;
+		}
+	} else
+		blk_mq_unfreeze_queue(dev->admin_q);
 
 	return 0;
 }
 
-static void nvme_free_admin_tags(struct nvme_dev *dev)
-{
-	if (dev->admin_q)
-		blk_mq_free_tag_set(&dev->admin_tagset);
-}
-
 static int nvme_configure_admin_queue(struct nvme_dev *dev)
 {
 	int result;
@@ -1445,19 +1483,13 @@ static int nvme_configure_admin_queue(struct nvme_dev *dev)
 	if (result)
 		goto free_nvmeq;
 
-	result = nvme_alloc_admin_tags(dev);
-	if (result)
-		goto free_nvmeq;
-
 	nvmeq->cq_vector = 0;
 	result = queue_request_irq(dev, nvmeq, nvmeq->irqname);
 	if (result)
-		goto free_tags;
+		goto free_nvmeq;
 
 	return result;
 
- free_tags:
-	nvme_free_admin_tags(dev);
  free_nvmeq:
 	nvme_free_queues(dev, 0);
 	return result;
@@ -2242,13 +2274,18 @@ static void nvme_wait_dq(struct nvme_delq_ctx *dq, struct nvme_dev *dev)
 			break;
 		if (!schedule_timeout(ADMIN_TIMEOUT) ||
 					fatal_signal_pending(current)) {
+			/*
+			 * Disable the controller first since we can't trust it
+			 * at this point, but leave the admin queue enabled
+			 * until all queue deletion requests are flushed.
+			 * FIXME: This may take a while if there are more h/w
+			 * queues than admin tags.
+			 */
 			set_current_state(TASK_RUNNING);
-
 			nvme_disable_ctrl(dev, readq(&dev->bar->cap));
-			nvme_disable_queue(dev, 0);
-
-			send_sig(SIGKILL, dq->worker->task, 1);
+			nvme_clear_queue(dev->queues[0]);
 			flush_kthread_worker(dq->worker);
+			nvme_disable_queue(dev, 0);
 			return;
 		}
 	}
@@ -2325,7 +2362,6 @@ static void nvme_del_queue_start(struct kthread_work *work)
 {
 	struct nvme_queue *nvmeq = container_of(work, struct nvme_queue,
 							cmdinfo.work);
-	allow_signal(SIGKILL);
 	if (nvme_delete_sq(nvmeq))
 		nvme_del_queue_end(nvmeq);
 }
@@ -2383,6 +2419,34 @@ static void nvme_dev_list_remove(struct nvme_dev *dev)
 		kthread_stop(tmp);
 }
 
+static void nvme_freeze_queues(struct nvme_dev *dev)
+{
+	struct nvme_ns *ns;
+
+	list_for_each_entry(ns, &dev->namespaces, list) {
+		blk_mq_freeze_queue_start(ns->queue);
+
+		spin_lock(ns->queue->queue_lock);
+		queue_flag_set(QUEUE_FLAG_STOPPED, ns->queue);
+		spin_unlock(ns->queue->queue_lock);
+
+		blk_mq_cancel_requeue_work(ns->queue);
+		blk_mq_stop_hw_queues(ns->queue);
+	}
+}
+
+static void nvme_unfreeze_queues(struct nvme_dev *dev)
+{
+	struct nvme_ns *ns;
+
+	list_for_each_entry(ns, &dev->namespaces, list) {
+		queue_flag_clear_unlocked(QUEUE_FLAG_STOPPED, ns->queue);
+		blk_mq_unfreeze_queue(ns->queue);
+		blk_mq_start_stopped_hw_queues(ns->queue, true);
+		blk_mq_kick_requeue_list(ns->queue);
+	}
+}
+
 static void nvme_dev_shutdown(struct nvme_dev *dev)
 {
 	int i;
@@ -2391,8 +2455,10 @@ static void nvme_dev_shutdown(struct nvme_dev *dev)
 	dev->initialized = 0;
 	nvme_dev_list_remove(dev);
 
-	if (dev->bar)
+	if (dev->bar) {
+		nvme_freeze_queues(dev);
 		csts = readl(&dev->bar->csts);
+	}
 	if (csts & NVME_CSTS_CFS || !(csts & NVME_CSTS_RDY)) {
 		for (i = dev->queue_count - 1; i >= 0; i--) {
 			struct nvme_queue *nvmeq = dev->queues[i];
@@ -2407,12 +2473,6 @@ static void nvme_dev_shutdown(struct nvme_dev *dev)
 	nvme_dev_unmap(dev);
 }
 
-static void nvme_dev_remove_admin(struct nvme_dev *dev)
-{
-	if (dev->admin_q && !blk_queue_dying(dev->admin_q))
-		blk_cleanup_queue(dev->admin_q);
-}
-
 static void nvme_dev_remove(struct nvme_dev *dev)
 {
 	struct nvme_ns *ns;
@@ -2420,8 +2480,10 @@ static void nvme_dev_remove(struct nvme_dev *dev)
 	list_for_each_entry(ns, &dev->namespaces, list) {
 		if (ns->disk->flags & GENHD_FL_UP)
 			del_gendisk(ns->disk);
-		if (!blk_queue_dying(ns->queue))
+		if (!blk_queue_dying(ns->queue)) {
+			blk_mq_abort_requeue_list(ns->queue);
 			blk_cleanup_queue(ns->queue);
+		}
 	}
 }
 
@@ -2502,6 +2564,7 @@ static void nvme_free_dev(struct kref *kref)
 	nvme_free_namespaces(dev);
 	nvme_release_instance(dev);
 	blk_mq_free_tag_set(&dev->tagset);
+	blk_put_queue(dev->admin_q);
 	kfree(dev->queues);
 	kfree(dev->entry);
 	kfree(dev);
@@ -2598,15 +2661,20 @@ static int nvme_dev_start(struct nvme_dev *dev)
 	}
 
 	nvme_init_queue(dev->queues[0], 0);
+	result = nvme_alloc_admin_tags(dev);
+	if (result)
+		goto disable;
 
 	result = nvme_setup_io_queues(dev);
 	if (result)
-		goto disable;
+		goto free_tags;
 
 	nvme_set_irq_hints(dev);
 
 	return result;
 
+ free_tags:
+	nvme_dev_remove_admin(dev);
  disable:
 	nvme_disable_queue(dev, 0);
 	nvme_dev_list_remove(dev);
@@ -2646,6 +2714,9 @@ static int nvme_dev_resume(struct nvme_dev *dev)
 		dev->reset_workfn = nvme_remove_disks;
 		queue_work(nvme_workq, &dev->reset_work);
 		spin_unlock(&dev_list_lock);
+	} else {
+		nvme_unfreeze_queues(dev);
+		nvme_set_irq_hints(dev);
 	}
 	dev->initialized = 1;
 	return 0;
@@ -2783,11 +2854,10 @@ static void nvme_remove(struct pci_dev *pdev)
 	pci_set_drvdata(pdev, NULL);
 	flush_work(&dev->reset_work);
 	misc_deregister(&dev->miscdev);
-	nvme_dev_remove(dev);
 	nvme_dev_shutdown(dev);
+	nvme_dev_remove(dev);
 	nvme_dev_remove_admin(dev);
 	nvme_free_queues(dev, 0);
-	nvme_free_admin_tags(dev);
 	nvme_release_prp_pools(dev);
 	kref_put(&dev->kref, nvme_free_dev);
 }
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 3ec85dfce124..8a86b62466f7 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -2098,32 +2098,26 @@ static void rbd_dev_parent_put(struct rbd_device *rbd_dev)
  * If an image has a non-zero parent overlap, get a reference to its
  * parent.
  *
- * We must get the reference before checking for the overlap to
- * coordinate properly with zeroing the parent overlap in
- * rbd_dev_v2_parent_info() when an image gets flattened.  We
- * drop it again if there is no overlap.
- *
  * Returns true if the rbd device has a parent with a non-zero
  * overlap and a reference for it was successfully taken, or
  * false otherwise.
  */
 static bool rbd_dev_parent_get(struct rbd_device *rbd_dev)
 {
-	int counter;
+	int counter = 0;
 
 	if (!rbd_dev->parent_spec)
 		return false;
 
-	counter = atomic_inc_return_safe(&rbd_dev->parent_ref);
-	if (counter > 0 && rbd_dev->parent_overlap)
-		return true;
-
-	/* Image was flattened, but parent is not yet torn down */
+	down_read(&rbd_dev->header_rwsem);
+	if (rbd_dev->parent_overlap)
+		counter = atomic_inc_return_safe(&rbd_dev->parent_ref);
+	up_read(&rbd_dev->header_rwsem);
 
 	if (counter < 0)
 		rbd_warn(rbd_dev, "parent reference overflow");
 
-	return false;
+	return counter > 0;
 }
 
 /*
@@ -4239,7 +4233,6 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev)
 		 */
 		if (rbd_dev->parent_overlap) {
 			rbd_dev->parent_overlap = 0;
-			smp_mb();
 			rbd_dev_parent_put(rbd_dev);
 			pr_info("%s: clone image has been flattened\n",
 				rbd_dev->disk->disk_name);
@@ -4285,7 +4278,6 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev)
 	 * treat it specially.
 	 */
 	rbd_dev->parent_overlap = overlap;
-	smp_mb();
 	if (!overlap) {
 
 		/* A null parent_spec indicates it's the initial probe */
@@ -5114,10 +5106,7 @@ static void rbd_dev_unprobe(struct rbd_device *rbd_dev)
 {
 	struct rbd_image_header	*header;
 
-	/* Drop parent reference unless it's already been done (or none) */
-
-	if (rbd_dev->parent_overlap)
-		rbd_dev_parent_put(rbd_dev);
+	rbd_dev_parent_put(rbd_dev);
 
 	/* Free dynamic fields from the header, then zero it out */
 
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index 63fc7f06a014..2a04d341e598 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -47,6 +47,7 @@
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
 #include <xen/balloon.h>
+#include <xen/grant_table.h>
 #include "common.h"
 
 /*
@@ -100,7 +101,7 @@ module_param(log_stats, int, 0644);
 
 #define BLKBACK_INVALID_HANDLE (~0)
 
-/* Number of free pages to remove on each call to free_xenballooned_pages */
+/* Number of free pages to remove on each call to gnttab_free_pages */
 #define NUM_BATCH_FREE_PAGES 10
 
 static inline int get_free_page(struct xen_blkif *blkif, struct page **page)
@@ -111,7 +112,7 @@ static inline int get_free_page(struct xen_blkif *blkif, struct page **page)
 	if (list_empty(&blkif->free_pages)) {
 		BUG_ON(blkif->free_pages_num != 0);
 		spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
-		return alloc_xenballooned_pages(1, page, false);
+		return gnttab_alloc_pages(1, page);
 	}
 	BUG_ON(blkif->free_pages_num == 0);
 	page[0] = list_first_entry(&blkif->free_pages, struct page, lru);
@@ -151,14 +152,14 @@ static inline void shrink_free_pagepool(struct xen_blkif *blkif, int num)
 		blkif->free_pages_num--;
 		if (++num_pages == NUM_BATCH_FREE_PAGES) {
 			spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
-			free_xenballooned_pages(num_pages, page);
+			gnttab_free_pages(num_pages, page);
 			spin_lock_irqsave(&blkif->free_pages_lock, flags);
 			num_pages = 0;
 		}
 	}
 	spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
 	if (num_pages != 0)
-		free_xenballooned_pages(num_pages, page);
+		gnttab_free_pages(num_pages, page);
 }
 
 #define vaddr(page) ((unsigned long)pfn_to_kaddr(page_to_pfn(page)))
@@ -262,6 +263,17 @@ static void put_persistent_gnt(struct xen_blkif *blkif,
 	atomic_dec(&blkif->persistent_gnt_in_use);
 }
 
+static void free_persistent_gnts_unmap_callback(int result,
+						struct gntab_unmap_queue_data *data)
+{
+	struct completion *c = data->data;
+
+	/* BUG_ON used to reproduce existing behaviour,
+	   but is this the best way to deal with this? */
+	BUG_ON(result);
+	complete(c);
+}
+
 static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
                                  unsigned int num)
 {
@@ -269,8 +281,17 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
 	struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	struct persistent_gnt *persistent_gnt;
 	struct rb_node *n;
-	int ret = 0;
 	int segs_to_unmap = 0;
+	struct gntab_unmap_queue_data unmap_data;
+	struct completion unmap_completion;
+
+	init_completion(&unmap_completion);
+
+	unmap_data.data = &unmap_completion;
+	unmap_data.done = &free_persistent_gnts_unmap_callback;
+	unmap_data.pages = pages;
+	unmap_data.unmap_ops = unmap;
+	unmap_data.kunmap_ops = NULL;
 
 	foreach_grant_safe(persistent_gnt, n, root, node) {
 		BUG_ON(persistent_gnt->handle ==
@@ -285,9 +306,11 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
 
 		if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST ||
 			!rb_next(&persistent_gnt->node)) {
-			ret = gnttab_unmap_refs(unmap, NULL, pages,
-				segs_to_unmap);
-			BUG_ON(ret);
+
+			unmap_data.count = segs_to_unmap;
+			gnttab_unmap_refs_async(&unmap_data);
+			wait_for_completion(&unmap_completion);
+
 			put_free_pages(blkif, pages, segs_to_unmap);
 			segs_to_unmap = 0;
 		}
@@ -653,18 +676,14 @@ void xen_blkbk_free_caches(struct xen_blkif *blkif)
 	shrink_free_pagepool(blkif, 0 /* All */);
 }
 
-/*
- * Unmap the grant references, and also remove the M2P over-rides
- * used in the 'pending_req'.
- */
-static void xen_blkbk_unmap(struct xen_blkif *blkif,
-                            struct grant_page *pages[],
-                            int num)
+static unsigned int xen_blkbk_unmap_prepare(
+	struct xen_blkif *blkif,
+	struct grant_page **pages,
+	unsigned int num,
+	struct gnttab_unmap_grant_ref *unmap_ops,
+	struct page **unmap_pages)
 {
-	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
-	struct page *unmap_pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	unsigned int i, invcount = 0;
-	int ret;
 
 	for (i = 0; i < num; i++) {
 		if (pages[i]->persistent_gnt != NULL) {
@@ -674,21 +693,95 @@ static void xen_blkbk_unmap(struct xen_blkif *blkif,
 		if (pages[i]->handle == BLKBACK_INVALID_HANDLE)
 			continue;
 		unmap_pages[invcount] = pages[i]->page;
-		gnttab_set_unmap_op(&unmap[invcount], vaddr(pages[i]->page),
+		gnttab_set_unmap_op(&unmap_ops[invcount], vaddr(pages[i]->page),
 				    GNTMAP_host_map, pages[i]->handle);
 		pages[i]->handle = BLKBACK_INVALID_HANDLE;
-		if (++invcount == BLKIF_MAX_SEGMENTS_PER_REQUEST) {
-			ret = gnttab_unmap_refs(unmap, NULL, unmap_pages,
-			                        invcount);
+		invcount++;
+       }
+
+       return invcount;
+}
+
+static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_queue_data *data)
+{
+	struct pending_req* pending_req = (struct pending_req*) (data->data);
+	struct xen_blkif *blkif = pending_req->blkif;
+
+	/* BUG_ON used to reproduce existing behaviour,
+	   but is this the best way to deal with this? */
+	BUG_ON(result);
+
+	put_free_pages(blkif, data->pages, data->count);
+	make_response(blkif, pending_req->id,
+		      pending_req->operation, pending_req->status);
+	free_req(blkif, pending_req);
+	/*
+	 * Make sure the request is freed before releasing blkif,
+	 * or there could be a race between free_req and the
+	 * cleanup done in xen_blkif_free during shutdown.
+	 *
+	 * NB: The fact that we might try to wake up pending_free_wq
+	 * before drain_complete (in case there's a drain going on)
+	 * it's not a problem with our current implementation
+	 * because we can assure there's no thread waiting on
+	 * pending_free_wq if there's a drain going on, but it has
+	 * to be taken into account if the current model is changed.
+	 */
+	if (atomic_dec_and_test(&blkif->inflight) && atomic_read(&blkif->drain)) {
+		complete(&blkif->drain_complete);
+	}
+	xen_blkif_put(blkif);
+}
+
+static void xen_blkbk_unmap_and_respond(struct pending_req *req)
+{
+	struct gntab_unmap_queue_data* work = &req->gnttab_unmap_data;
+	struct xen_blkif *blkif = req->blkif;
+	struct grant_page **pages = req->segments;
+	unsigned int invcount;
+
+	invcount = xen_blkbk_unmap_prepare(blkif, pages, req->nr_pages,
+					   req->unmap, req->unmap_pages);
+
+	work->data = req;
+	work->done = xen_blkbk_unmap_and_respond_callback;
+	work->unmap_ops = req->unmap;
+	work->kunmap_ops = NULL;
+	work->pages = req->unmap_pages;
+	work->count = invcount;
+
+	gnttab_unmap_refs_async(&req->gnttab_unmap_data);
+}
+
+
+/*
+ * Unmap the grant references.
+ *
+ * This could accumulate ops up to the batch size to reduce the number
+ * of hypercalls, but since this is only used in error paths there's
+ * no real need.
+ */
+static void xen_blkbk_unmap(struct xen_blkif *blkif,
+                            struct grant_page *pages[],
+                            int num)
+{
+	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	struct page *unmap_pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	unsigned int invcount = 0;
+	int ret;
+
+	while (num) {
+		unsigned int batch = min(num, BLKIF_MAX_SEGMENTS_PER_REQUEST);
+		
+		invcount = xen_blkbk_unmap_prepare(blkif, pages, batch,
+						   unmap, unmap_pages);
+		if (invcount) {
+			ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
 			BUG_ON(ret);
 			put_free_pages(blkif, unmap_pages, invcount);
-			invcount = 0;
 		}
-	}
-	if (invcount) {
-		ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
-		BUG_ON(ret);
-		put_free_pages(blkif, unmap_pages, invcount);
+		pages += batch;
+		num -= batch;
 	}
 }
 
@@ -982,32 +1075,8 @@ static void __end_block_io_op(struct pending_req *pending_req, int error)
 	 * the grant references associated with 'request' and provide
 	 * the proper response on the ring.
 	 */
-	if (atomic_dec_and_test(&pending_req->pendcnt)) {
-		struct xen_blkif *blkif = pending_req->blkif;
-
-		xen_blkbk_unmap(blkif,
-		                pending_req->segments,
-		                pending_req->nr_pages);
-		make_response(blkif, pending_req->id,
-			      pending_req->operation, pending_req->status);
-		free_req(blkif, pending_req);
-		/*
-		 * Make sure the request is freed before releasing blkif,
-		 * or there could be a race between free_req and the
-		 * cleanup done in xen_blkif_free during shutdown.
-		 *
-		 * NB: The fact that we might try to wake up pending_free_wq
-		 * before drain_complete (in case there's a drain going on)
-		 * it's not a problem with our current implementation
-		 * because we can assure there's no thread waiting on
-		 * pending_free_wq if there's a drain going on, but it has
-		 * to be taken into account if the current model is changed.
-		 */
-		if (atomic_dec_and_test(&blkif->inflight) && atomic_read(&blkif->drain)) {
-			complete(&blkif->drain_complete);
-		}
-		xen_blkif_put(blkif);
-	}
+	if (atomic_dec_and_test(&pending_req->pendcnt))
+		xen_blkbk_unmap_and_respond(pending_req);
 }
 
 /*
diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h
index f65b807e3236..cc90a840e616 100644
--- a/drivers/block/xen-blkback/common.h
+++ b/drivers/block/xen-blkback/common.h
@@ -350,6 +350,9 @@ struct pending_req {
 	struct grant_page	*indirect_pages[MAX_INDIRECT_PAGES];
 	struct seg_buf		seg[MAX_INDIRECT_SEGMENTS];
 	struct bio		*biolist[MAX_INDIRECT_SEGMENTS];
+	struct gnttab_unmap_grant_ref unmap[MAX_INDIRECT_SEGMENTS];
+	struct page                   *unmap_pages[MAX_INDIRECT_SEGMENTS];
+	struct gntab_unmap_queue_data gnttab_unmap_data;
 };