Merge branch 'stable/for-jens-4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen into for-4.5/drivers

Konrad writes:

The pull is based on converting the backend driver into an multiqueue
driver and exposing more than one queue to the frontend. As such we had
to modify the frontend and also fix a bunch of bugs around this.

The original work is based on Arianna Avanzini's work as an OPW intern.
Bob took over the work and had been massaging it for quite some time.

Also included are are features to 64KB page support for ARM and various
bug-fixes.

drivers/block/xen-blkback/blkback.c

@@ -83,6 +83,16 @@ module_param_named(max_persistent_grants, xen_blkif_max_pgrants, int, 0644);
 MODULE_PARM_DESC(max_persistent_grants,
                  "Maximum number of grants to map persistently");
 
+/*
+ * Maximum number of rings/queues blkback supports, allow as many queues as there
+ * are CPUs if user has not specified a value.
+ */
+unsigned int xenblk_max_queues;
+module_param_named(max_queues, xenblk_max_queues, uint, 0644);
+MODULE_PARM_DESC(max_queues,
+		 "Maximum number of hardware queues per virtual disk." \
+		 "By default it is the number of online CPUs.");
+
 /*
  * Maximum order of pages to be used for the shared ring between front and
  * backend, 4KB page granularity is used.
@@ -113,71 +123,71 @@ module_param(log_stats, int, 0644);
 /* 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)
+static inline int get_free_page(struct xen_blkif_ring *ring, struct page **page)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&blkif->free_pages_lock, flags);
-	if (list_empty(&blkif->free_pages)) {
-		BUG_ON(blkif->free_pages_num != 0);
-		spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
+	spin_lock_irqsave(&ring->free_pages_lock, flags);
+	if (list_empty(&ring->free_pages)) {
+		BUG_ON(ring->free_pages_num != 0);
+		spin_unlock_irqrestore(&ring->free_pages_lock, flags);
 		return gnttab_alloc_pages(1, page);
 	}
-	BUG_ON(blkif->free_pages_num == 0);
-	page[0] = list_first_entry(&blkif->free_pages, struct page, lru);
+	BUG_ON(ring->free_pages_num == 0);
+	page[0] = list_first_entry(&ring->free_pages, struct page, lru);
 	list_del(&page[0]->lru);
-	blkif->free_pages_num--;
-	spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
+	ring->free_pages_num--;
+	spin_unlock_irqrestore(&ring->free_pages_lock, flags);
 
 	return 0;
 }
 
-static inline void put_free_pages(struct xen_blkif *blkif, struct page **page,
+static inline void put_free_pages(struct xen_blkif_ring *ring, struct page **page,
                                   int num)
 {
 	unsigned long flags;
 	int i;
 
-	spin_lock_irqsave(&blkif->free_pages_lock, flags);
+	spin_lock_irqsave(&ring->free_pages_lock, flags);
 	for (i = 0; i < num; i++)
-		list_add(&page[i]->lru, &blkif->free_pages);
-	blkif->free_pages_num += num;
-	spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
+		list_add(&page[i]->lru, &ring->free_pages);
+	ring->free_pages_num += num;
+	spin_unlock_irqrestore(&ring->free_pages_lock, flags);
 }
 
-static inline void shrink_free_pagepool(struct xen_blkif *blkif, int num)
+static inline void shrink_free_pagepool(struct xen_blkif_ring *ring, int num)
 {
 	/* Remove requested pages in batches of NUM_BATCH_FREE_PAGES */
 	struct page *page[NUM_BATCH_FREE_PAGES];
 	unsigned int num_pages = 0;
 	unsigned long flags;
 
-	spin_lock_irqsave(&blkif->free_pages_lock, flags);
-	while (blkif->free_pages_num > num) {
-		BUG_ON(list_empty(&blkif->free_pages));
-		page[num_pages] = list_first_entry(&blkif->free_pages,
+	spin_lock_irqsave(&ring->free_pages_lock, flags);
+	while (ring->free_pages_num > num) {
+		BUG_ON(list_empty(&ring->free_pages));
+		page[num_pages] = list_first_entry(&ring->free_pages,
 		                                   struct page, lru);
 		list_del(&page[num_pages]->lru);
-		blkif->free_pages_num--;
+		ring->free_pages_num--;
 		if (++num_pages == NUM_BATCH_FREE_PAGES) {
-			spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
+			spin_unlock_irqrestore(&ring->free_pages_lock, flags);
 			gnttab_free_pages(num_pages, page);
-			spin_lock_irqsave(&blkif->free_pages_lock, flags);
+			spin_lock_irqsave(&ring->free_pages_lock, flags);
 			num_pages = 0;
 		}
 	}
-	spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
+	spin_unlock_irqrestore(&ring->free_pages_lock, flags);
 	if (num_pages != 0)
 		gnttab_free_pages(num_pages, page);
 }
 
 #define vaddr(page) ((unsigned long)pfn_to_kaddr(page_to_pfn(page)))
 
-static int do_block_io_op(struct xen_blkif *blkif);
-static int dispatch_rw_block_io(struct xen_blkif *blkif,
+static int do_block_io_op(struct xen_blkif_ring *ring);
+static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 				struct blkif_request *req,
 				struct pending_req *pending_req);
-static void make_response(struct xen_blkif *blkif, u64 id,
+static void make_response(struct xen_blkif_ring *ring, u64 id,
 			  unsigned short op, int st);
 
 #define foreach_grant_safe(pos, n, rbtree, node) \
@@ -190,7 +200,7 @@ static void make_response(struct xen_blkif *blkif, u64 id,
 
 /*
  * We don't need locking around the persistent grant helpers
- * because blkback uses a single-thread for each backed, so we
+ * because blkback uses a single-thread for each backend, so we
  * can be sure that this functions will never be called recursively.
  *
  * The only exception to that is put_persistent_grant, that can be called
@@ -198,19 +208,20 @@ static void make_response(struct xen_blkif *blkif, u64 id,
  * bit operations to modify the flags of a persistent grant and to count
  * the number of used grants.
  */
-static int add_persistent_gnt(struct xen_blkif *blkif,
+static int add_persistent_gnt(struct xen_blkif_ring *ring,
 			       struct persistent_gnt *persistent_gnt)
 {
 	struct rb_node **new = NULL, *parent = NULL;
 	struct persistent_gnt *this;
+	struct xen_blkif *blkif = ring->blkif;
 
-	if (blkif->persistent_gnt_c >= xen_blkif_max_pgrants) {
+	if (ring->persistent_gnt_c >= xen_blkif_max_pgrants) {
 		if (!blkif->vbd.overflow_max_grants)
 			blkif->vbd.overflow_max_grants = 1;
 		return -EBUSY;
 	}
 	/* Figure out where to put new node */
-	new = &blkif->persistent_gnts.rb_node;
+	new = &ring->persistent_gnts.rb_node;
 	while (*new) {
 		this = container_of(*new, struct persistent_gnt, node);
 
@@ -229,19 +240,19 @@ static int add_persistent_gnt(struct xen_blkif *blkif,
 	set_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags);
 	/* Add new node and rebalance tree. */
 	rb_link_node(&(persistent_gnt->node), parent, new);
-	rb_insert_color(&(persistent_gnt->node), &blkif->persistent_gnts);
-	blkif->persistent_gnt_c++;
-	atomic_inc(&blkif->persistent_gnt_in_use);
+	rb_insert_color(&(persistent_gnt->node), &ring->persistent_gnts);
+	ring->persistent_gnt_c++;
+	atomic_inc(&ring->persistent_gnt_in_use);
 	return 0;
 }
 
-static struct persistent_gnt *get_persistent_gnt(struct xen_blkif *blkif,
+static struct persistent_gnt *get_persistent_gnt(struct xen_blkif_ring *ring,
 						 grant_ref_t gref)
 {
 	struct persistent_gnt *data;
 	struct rb_node *node = NULL;
 
-	node = blkif->persistent_gnts.rb_node;
+	node = ring->persistent_gnts.rb_node;
 	while (node) {
 		data = container_of(node, struct persistent_gnt, node);
 
@@ -255,24 +266,24 @@ static struct persistent_gnt *get_persistent_gnt(struct xen_blkif *blkif,
 				return NULL;
 			}
 			set_bit(PERSISTENT_GNT_ACTIVE, data->flags);
-			atomic_inc(&blkif->persistent_gnt_in_use);
+			atomic_inc(&ring->persistent_gnt_in_use);
 			return data;
 		}
 	}
 	return NULL;
 }
 
-static void put_persistent_gnt(struct xen_blkif *blkif,
+static void put_persistent_gnt(struct xen_blkif_ring *ring,
                                struct persistent_gnt *persistent_gnt)
 {
 	if(!test_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags))
 		pr_alert_ratelimited("freeing a grant already unused\n");
 	set_bit(PERSISTENT_GNT_WAS_ACTIVE, persistent_gnt->flags);
 	clear_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags);
-	atomic_dec(&blkif->persistent_gnt_in_use);
+	atomic_dec(&ring->persistent_gnt_in_use);
 }
 
-static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
+static void free_persistent_gnts(struct xen_blkif_ring *ring, struct rb_root *root,
                                  unsigned int num)
 {
 	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
@@ -303,7 +314,7 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
 			unmap_data.count = segs_to_unmap;
 			BUG_ON(gnttab_unmap_refs_sync(&unmap_data));
 
-			put_free_pages(blkif, pages, segs_to_unmap);
+			put_free_pages(ring, pages, segs_to_unmap);
 			segs_to_unmap = 0;
 		}
 
@@ -320,15 +331,15 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work)
 	struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	struct persistent_gnt *persistent_gnt;
 	int segs_to_unmap = 0;
-	struct xen_blkif *blkif = container_of(work, typeof(*blkif), persistent_purge_work);
+	struct xen_blkif_ring *ring = container_of(work, typeof(*ring), persistent_purge_work);
 	struct gntab_unmap_queue_data unmap_data;
 
 	unmap_data.pages = pages;
 	unmap_data.unmap_ops = unmap;
 	unmap_data.kunmap_ops = NULL;
 
-	while(!list_empty(&blkif->persistent_purge_list)) {
-		persistent_gnt = list_first_entry(&blkif->persistent_purge_list,
+	while(!list_empty(&ring->persistent_purge_list)) {
+		persistent_gnt = list_first_entry(&ring->persistent_purge_list,
 		                                  struct persistent_gnt,
 		                                  remove_node);
 		list_del(&persistent_gnt->remove_node);
@@ -343,7 +354,7 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work)
 		if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST) {
 			unmap_data.count = segs_to_unmap;
 			BUG_ON(gnttab_unmap_refs_sync(&unmap_data));
-			put_free_pages(blkif, pages, segs_to_unmap);
+			put_free_pages(ring, pages, segs_to_unmap);
 			segs_to_unmap = 0;
 		}
 		kfree(persistent_gnt);
@@ -351,11 +362,11 @@ void xen_blkbk_unmap_purged_grants(struct work_struct *work)
 	if (segs_to_unmap > 0) {
 		unmap_data.count = segs_to_unmap;
 		BUG_ON(gnttab_unmap_refs_sync(&unmap_data));
-		put_free_pages(blkif, pages, segs_to_unmap);
+		put_free_pages(ring, pages, segs_to_unmap);
 	}
 }
 
-static void purge_persistent_gnt(struct xen_blkif *blkif)
+static void purge_persistent_gnt(struct xen_blkif_ring *ring)
 {
 	struct persistent_gnt *persistent_gnt;
 	struct rb_node *n;
@@ -363,23 +374,23 @@ static void purge_persistent_gnt(struct xen_blkif *blkif)
 	bool scan_used = false, clean_used = false;
 	struct rb_root *root;
 
-	if (blkif->persistent_gnt_c < xen_blkif_max_pgrants ||
-	    (blkif->persistent_gnt_c == xen_blkif_max_pgrants &&
-	    !blkif->vbd.overflow_max_grants)) {
-		return;
+	if (ring->persistent_gnt_c < xen_blkif_max_pgrants ||
+	    (ring->persistent_gnt_c == xen_blkif_max_pgrants &&
+	    !ring->blkif->vbd.overflow_max_grants)) {
+		goto out;
 	}
 
-	if (work_busy(&blkif->persistent_purge_work)) {
+	if (work_busy(&ring->persistent_purge_work)) {
 		pr_alert_ratelimited("Scheduled work from previous purge is still busy, cannot purge list\n");
-		return;
+		goto out;
 	}
 
 	num_clean = (xen_blkif_max_pgrants / 100) * LRU_PERCENT_CLEAN;
-	num_clean = blkif->persistent_gnt_c - xen_blkif_max_pgrants + num_clean;
-	num_clean = min(blkif->persistent_gnt_c, num_clean);
+	num_clean = ring->persistent_gnt_c - xen_blkif_max_pgrants + num_clean;
+	num_clean = min(ring->persistent_gnt_c, num_clean);
 	if ((num_clean == 0) ||
-	    (num_clean > (blkif->persistent_gnt_c - atomic_read(&blkif->persistent_gnt_in_use))))
-		return;
+	    (num_clean > (ring->persistent_gnt_c - atomic_read(&ring->persistent_gnt_in_use))))
+		goto out;
 
 	/*
 	 * At this point, we can assure that there will be no calls
@@ -394,8 +405,8 @@ static void purge_persistent_gnt(struct xen_blkif *blkif)
 
 	pr_debug("Going to purge %u persistent grants\n", num_clean);
 
-	BUG_ON(!list_empty(&blkif->persistent_purge_list));
-	root = &blkif->persistent_gnts;
+	BUG_ON(!list_empty(&ring->persistent_purge_list));
+	root = &ring->persistent_gnts;
 purge_list:
 	foreach_grant_safe(persistent_gnt, n, root, node) {
 		BUG_ON(persistent_gnt->handle ==
@@ -414,7 +425,7 @@ purge_list:
 
 		rb_erase(&persistent_gnt->node, root);
 		list_add(&persistent_gnt->remove_node,
-		         &blkif->persistent_purge_list);
+			 &ring->persistent_purge_list);
 		if (--num_clean == 0)
 			goto finished;
 	}
@@ -435,30 +446,32 @@ finished:
 		goto purge_list;
 	}
 
-	blkif->persistent_gnt_c -= (total - num_clean);
-	blkif->vbd.overflow_max_grants = 0;
+	ring->persistent_gnt_c -= (total - num_clean);
+	ring->blkif->vbd.overflow_max_grants = 0;
 
 	/* We can defer this work */
-	schedule_work(&blkif->persistent_purge_work);
+	schedule_work(&ring->persistent_purge_work);
 	pr_debug("Purged %u/%u\n", (total - num_clean), total);
+
+out:
 	return;
 }
 
 /*
  * Retrieve from the 'pending_reqs' a free pending_req structure to be used.
  */
-static struct pending_req *alloc_req(struct xen_blkif *blkif)
+static struct pending_req *alloc_req(struct xen_blkif_ring *ring)
 {
 	struct pending_req *req = NULL;
 	unsigned long flags;
 
-	spin_lock_irqsave(&blkif->pending_free_lock, flags);
-	if (!list_empty(&blkif->pending_free)) {
-		req = list_entry(blkif->pending_free.next, struct pending_req,
+	spin_lock_irqsave(&ring->pending_free_lock, flags);
+	if (!list_empty(&ring->pending_free)) {
+		req = list_entry(ring->pending_free.next, struct pending_req,
 				 free_list);
 		list_del(&req->free_list);
 	}
-	spin_unlock_irqrestore(&blkif->pending_free_lock, flags);
+	spin_unlock_irqrestore(&ring->pending_free_lock, flags);
 	return req;
 }
 
@@ -466,17 +479,17 @@ static struct pending_req *alloc_req(struct xen_blkif *blkif)
  * Return the 'pending_req' structure back to the freepool. We also
  * wake up the thread if it was waiting for a free page.
  */
-static void free_req(struct xen_blkif *blkif, struct pending_req *req)
+static void free_req(struct xen_blkif_ring *ring, struct pending_req *req)
 {
 	unsigned long flags;
 	int was_empty;
 
-	spin_lock_irqsave(&blkif->pending_free_lock, flags);
-	was_empty = list_empty(&blkif->pending_free);
-	list_add(&req->free_list, &blkif->pending_free);
-	spin_unlock_irqrestore(&blkif->pending_free_lock, flags);
+	spin_lock_irqsave(&ring->pending_free_lock, flags);
+	was_empty = list_empty(&ring->pending_free);
+	list_add(&req->free_list, &ring->pending_free);
+	spin_unlock_irqrestore(&ring->pending_free_lock, flags);
 	if (was_empty)
-		wake_up(&blkif->pending_free_wq);
+		wake_up(&ring->pending_free_wq);
 }
 
 /*
@@ -556,10 +569,10 @@ abort:
 /*
  * Notification from the guest OS.
  */
-static void blkif_notify_work(struct xen_blkif *blkif)
+static void blkif_notify_work(struct xen_blkif_ring *ring)
 {
-	blkif->waiting_reqs = 1;
-	wake_up(&blkif->wq);
+	ring->waiting_reqs = 1;
+	wake_up(&ring->wq);
 }
 
 irqreturn_t xen_blkif_be_int(int irq, void *dev_id)
@@ -572,31 +585,33 @@ irqreturn_t xen_blkif_be_int(int irq, void *dev_id)
  * SCHEDULER FUNCTIONS
  */
 
-static void print_stats(struct xen_blkif *blkif)
+static void print_stats(struct xen_blkif_ring *ring)
 {
 	pr_info("(%s): oo %3llu  |  rd %4llu  |  wr %4llu  |  f %4llu"
 		 "  |  ds %4llu | pg: %4u/%4d\n",
-		 current->comm, blkif->st_oo_req,
-		 blkif->st_rd_req, blkif->st_wr_req,
-		 blkif->st_f_req, blkif->st_ds_req,
-		 blkif->persistent_gnt_c,
+		 current->comm, ring->st_oo_req,
+		 ring->st_rd_req, ring->st_wr_req,
+		 ring->st_f_req, ring->st_ds_req,
+		 ring->persistent_gnt_c,
 		 xen_blkif_max_pgrants);
-	blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000);
-	blkif->st_rd_req = 0;
-	blkif->st_wr_req = 0;
-	blkif->st_oo_req = 0;
-	blkif->st_ds_req = 0;
+	ring->st_print = jiffies + msecs_to_jiffies(10 * 1000);
+	ring->st_rd_req = 0;
+	ring->st_wr_req = 0;
+	ring->st_oo_req = 0;
+	ring->st_ds_req = 0;
 }
 
 int xen_blkif_schedule(void *arg)
 {
-	struct xen_blkif *blkif = arg;
+	struct xen_blkif_ring *ring = arg;
+	struct xen_blkif *blkif = ring->blkif;
 	struct xen_vbd *vbd = &blkif->vbd;
 	unsigned long timeout;
 	int ret;
 
 	xen_blkif_get(blkif);
 
+	set_freezable();
 	while (!kthread_should_stop()) {
 		if (try_to_freeze())
 			continue;
@@ -606,50 +621,50 @@ int xen_blkif_schedule(void *arg)
 		timeout = msecs_to_jiffies(LRU_INTERVAL);
 
 		timeout = wait_event_interruptible_timeout(
-			blkif->wq,
-			blkif->waiting_reqs || kthread_should_stop(),
+			ring->wq,
+			ring->waiting_reqs || kthread_should_stop(),
 			timeout);
 		if (timeout == 0)
 			goto purge_gnt_list;
 		timeout = wait_event_interruptible_timeout(
-			blkif->pending_free_wq,
-			!list_empty(&blkif->pending_free) ||
+			ring->pending_free_wq,
+			!list_empty(&ring->pending_free) ||
 			kthread_should_stop(),
 			timeout);
 		if (timeout == 0)
 			goto purge_gnt_list;
 
-		blkif->waiting_reqs = 0;
+		ring->waiting_reqs = 0;
 		smp_mb(); /* clear flag *before* checking for work */
 
-		ret = do_block_io_op(blkif);
+		ret = do_block_io_op(ring);
 		if (ret > 0)
-			blkif->waiting_reqs = 1;
+			ring->waiting_reqs = 1;
 		if (ret == -EACCES)
-			wait_event_interruptible(blkif->shutdown_wq,
+			wait_event_interruptible(ring->shutdown_wq,
 						 kthread_should_stop());
 
 purge_gnt_list:
 		if (blkif->vbd.feature_gnt_persistent &&
-		    time_after(jiffies, blkif->next_lru)) {
-			purge_persistent_gnt(blkif);
-			blkif->next_lru = jiffies + msecs_to_jiffies(LRU_INTERVAL);
+		    time_after(jiffies, ring->next_lru)) {
+			purge_persistent_gnt(ring);
+			ring->next_lru = jiffies + msecs_to_jiffies(LRU_INTERVAL);
 		}
 
 		/* Shrink if we have more than xen_blkif_max_buffer_pages */
-		shrink_free_pagepool(blkif, xen_blkif_max_buffer_pages);
+		shrink_free_pagepool(ring, xen_blkif_max_buffer_pages);
 
-		if (log_stats && time_after(jiffies, blkif->st_print))
-			print_stats(blkif);
+		if (log_stats && time_after(jiffies, ring->st_print))
+			print_stats(ring);
 	}
 
 	/* Drain pending purge work */
-	flush_work(&blkif->persistent_purge_work);
+	flush_work(&ring->persistent_purge_work);
 
 	if (log_stats)
-		print_stats(blkif);
+		print_stats(ring);
 
-	blkif->xenblkd = NULL;
+	ring->xenblkd = NULL;
 	xen_blkif_put(blkif);
 
 	return 0;
@@ -658,22 +673,22 @@ purge_gnt_list:
 /*
  * Remove persistent grants and empty the pool of free pages
  */
-void xen_blkbk_free_caches(struct xen_blkif *blkif)
+void xen_blkbk_free_caches(struct xen_blkif_ring *ring)
 {
 	/* Free all persistent grant pages */
-	if (!RB_EMPTY_ROOT(&blkif->persistent_gnts))
-		free_persistent_gnts(blkif, &blkif->persistent_gnts,
-			blkif->persistent_gnt_c);
+	if (!RB_EMPTY_ROOT(&ring->persistent_gnts))
+		free_persistent_gnts(ring, &ring->persistent_gnts,
+			ring->persistent_gnt_c);
 
-	BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts));
-	blkif->persistent_gnt_c = 0;
+	BUG_ON(!RB_EMPTY_ROOT(&ring->persistent_gnts));
+	ring->persistent_gnt_c = 0;
 
 	/* Since we are shutting down remove all pages from the buffer */
-	shrink_free_pagepool(blkif, 0 /* All */);
+	shrink_free_pagepool(ring, 0 /* All */);
 }
 
 static unsigned int xen_blkbk_unmap_prepare(
-	struct xen_blkif *blkif,
+	struct xen_blkif_ring *ring,
 	struct grant_page **pages,
 	unsigned int num,
 	struct gnttab_unmap_grant_ref *unmap_ops,
@@ -683,7 +698,7 @@ static unsigned int xen_blkbk_unmap_prepare(
 
 	for (i = 0; i < num; i++) {
 		if (pages[i]->persistent_gnt != NULL) {
-			put_persistent_gnt(blkif, pages[i]->persistent_gnt);
+			put_persistent_gnt(ring, pages[i]->persistent_gnt);
 			continue;
 		}
 		if (pages[i]->handle == BLKBACK_INVALID_HANDLE)
@@ -700,17 +715,18 @@ static unsigned int xen_blkbk_unmap_prepare(
 
 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;
+	struct pending_req *pending_req = (struct pending_req *)(data->data);
+	struct xen_blkif_ring *ring = pending_req->ring;
+	struct xen_blkif *blkif = ring->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,
+	put_free_pages(ring, data->pages, data->count);
+	make_response(ring, pending_req->id,
 		      pending_req->operation, pending_req->status);
-	free_req(blkif, pending_req);
+	free_req(ring, pending_req);
 	/*
 	 * Make sure the request is freed before releasing blkif,
 	 * or there could be a race between free_req and the
@@ -723,7 +739,7 @@ static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_
 	 * 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)) {
+	if (atomic_dec_and_test(&ring->inflight) && atomic_read(&blkif->drain)) {
 		complete(&blkif->drain_complete);
 	}
 	xen_blkif_put(blkif);
@@ -732,11 +748,11 @@ static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_
 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 xen_blkif_ring *ring = req->ring;
 	struct grant_page **pages = req->segments;
 	unsigned int invcount;
 
-	invcount = xen_blkbk_unmap_prepare(blkif, pages, req->nr_segs,
+	invcount = xen_blkbk_unmap_prepare(ring, pages, req->nr_segs,
 					   req->unmap, req->unmap_pages);
 
 	work->data = req;
@@ -757,7 +773,7 @@ static void xen_blkbk_unmap_and_respond(struct pending_req *req)
  * 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,
+static void xen_blkbk_unmap(struct xen_blkif_ring *ring,
                             struct grant_page *pages[],
                             int num)
 {
@@ -768,20 +784,20 @@ static void xen_blkbk_unmap(struct xen_blkif *blkif,
 
 	while (num) {
 		unsigned int batch = min(num, BLKIF_MAX_SEGMENTS_PER_REQUEST);
-		
-		invcount = xen_blkbk_unmap_prepare(blkif, pages, batch,
+
+		invcount = xen_blkbk_unmap_prepare(ring, 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);
+			put_free_pages(ring, unmap_pages, invcount);
 		}
 		pages += batch;
 		num -= batch;
 	}
 }
 
-static int xen_blkbk_map(struct xen_blkif *blkif,
+static int xen_blkbk_map(struct xen_blkif_ring *ring,
 			 struct grant_page *pages[],
 			 int num, bool ro)
 {
@@ -794,6 +810,7 @@ static int xen_blkbk_map(struct xen_blkif *blkif,
 	int ret = 0;
 	int last_map = 0, map_until = 0;
 	int use_persistent_gnts;
+	struct xen_blkif *blkif = ring->blkif;
 
 	use_persistent_gnts = (blkif->vbd.feature_gnt_persistent);
 
@@ -806,10 +823,11 @@ again:
 	for (i = map_until; i < num; i++) {
 		uint32_t flags;
 
-		if (use_persistent_gnts)
+		if (use_persistent_gnts) {
 			persistent_gnt = get_persistent_gnt(
-				blkif,
+				ring,
 				pages[i]->gref);
+		}
 
 		if (persistent_gnt) {
 			/*
@@ -819,7 +837,7 @@ again:
 			pages[i]->page = persistent_gnt->page;
 			pages[i]->persistent_gnt = persistent_gnt;
 		} else {
-			if (get_free_page(blkif, &pages[i]->page))
+			if (get_free_page(ring, &pages[i]->page))
 				goto out_of_memory;
 			addr = vaddr(pages[i]->page);
 			pages_to_gnt[segs_to_map] = pages[i]->page;
@@ -852,7 +870,7 @@ again:
 			BUG_ON(new_map_idx >= segs_to_map);
 			if (unlikely(map[new_map_idx].status != 0)) {
 				pr_debug("invalid buffer -- could not remap it\n");
-				put_free_pages(blkif, &pages[seg_idx]->page, 1);
+				put_free_pages(ring, &pages[seg_idx]->page, 1);
 				pages[seg_idx]->handle = BLKBACK_INVALID_HANDLE;
 				ret |= 1;
 				goto next;
@@ -862,7 +880,7 @@ again:
 			continue;
 		}
 		if (use_persistent_gnts &&
-		    blkif->persistent_gnt_c < xen_blkif_max_pgrants) {
+		    ring->persistent_gnt_c < xen_blkif_max_pgrants) {
 			/*
 			 * We are using persistent grants, the grant is
 			 * not mapped but we might have room for it.
@@ -880,7 +898,7 @@ again:
 			persistent_gnt->gnt = map[new_map_idx].ref;
 			persistent_gnt->handle = map[new_map_idx].handle;
 			persistent_gnt->page = pages[seg_idx]->page;
-			if (add_persistent_gnt(blkif,
+			if (add_persistent_gnt(ring,
 			                       persistent_gnt)) {
 				kfree(persistent_gnt);
 				persistent_gnt = NULL;
@@ -888,7 +906,7 @@ again:
 			}
 			pages[seg_idx]->persistent_gnt = persistent_gnt;
 			pr_debug("grant %u added to the tree of persistent grants, using %u/%u\n",
-				 persistent_gnt->gnt, blkif->persistent_gnt_c,
+				 persistent_gnt->gnt, ring->persistent_gnt_c,
 				 xen_blkif_max_pgrants);
 			goto next;
 		}
@@ -913,7 +931,7 @@ next:
 
 out_of_memory:
 	pr_alert("%s: out of memory\n", __func__);
-	put_free_pages(blkif, pages_to_gnt, segs_to_map);
+	put_free_pages(ring, pages_to_gnt, segs_to_map);
 	return -ENOMEM;
 }
 
@@ -921,7 +939,7 @@ static int xen_blkbk_map_seg(struct pending_req *pending_req)
 {
 	int rc;
 
-	rc = xen_blkbk_map(pending_req->blkif, pending_req->segments,
+	rc = xen_blkbk_map(pending_req->ring, pending_req->segments,
 			   pending_req->nr_segs,
 	                   (pending_req->operation != BLKIF_OP_READ));
 
@@ -934,7 +952,7 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req,
 				    struct phys_req *preq)
 {
 	struct grant_page **pages = pending_req->indirect_pages;
-	struct xen_blkif *blkif = pending_req->blkif;
+	struct xen_blkif_ring *ring = pending_req->ring;
 	int indirect_grefs, rc, n, nseg, i;
 	struct blkif_request_segment *segments = NULL;
 
@@ -945,7 +963,7 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req,
 	for (i = 0; i < indirect_grefs; i++)
 		pages[i]->gref = req->u.indirect.indirect_grefs[i];
 
-	rc = xen_blkbk_map(blkif, pages, indirect_grefs, true);
+	rc = xen_blkbk_map(ring, pages, indirect_grefs, true);
 	if (rc)
 		goto unmap;
 
@@ -972,15 +990,16 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req,
 unmap:
 	if (segments)
 		kunmap_atomic(segments);
-	xen_blkbk_unmap(blkif, pages, indirect_grefs);
+	xen_blkbk_unmap(ring, pages, indirect_grefs);
 	return rc;
 }
 
-static int dispatch_discard_io(struct xen_blkif *blkif,
+static int dispatch_discard_io(struct xen_blkif_ring *ring,
 				struct blkif_request *req)
 {
 	int err = 0;
 	int status = BLKIF_RSP_OKAY;
+	struct xen_blkif *blkif = ring->blkif;
 	struct block_device *bdev = blkif->vbd.bdev;
 	unsigned long secure;
 	struct phys_req preq;
@@ -997,7 +1016,7 @@ static int dispatch_discard_io(struct xen_blkif *blkif,
 			preq.sector_number + preq.nr_sects, blkif->vbd.pdevice);
 		goto fail_response;
 	}
-	blkif->st_ds_req++;
+	ring->st_ds_req++;
 
 	secure = (blkif->vbd.discard_secure &&
 		 (req->u.discard.flag & BLKIF_DISCARD_SECURE)) ?
@@ -1013,26 +1032,28 @@ fail_response:
 	} else if (err)
 		status = BLKIF_RSP_ERROR;
 
-	make_response(blkif, req->u.discard.id, req->operation, status);
+	make_response(ring, req->u.discard.id, req->operation, status);
 	xen_blkif_put(blkif);
 	return err;
 }
 
-static int dispatch_other_io(struct xen_blkif *blkif,
+static int dispatch_other_io(struct xen_blkif_ring *ring,
 			     struct blkif_request *req,
 			     struct pending_req *pending_req)
 {
-	free_req(blkif, pending_req);
-	make_response(blkif, req->u.other.id, req->operation,
+	free_req(ring, pending_req);
+	make_response(ring, req->u.other.id, req->operation,
 		      BLKIF_RSP_EOPNOTSUPP);
 	return -EIO;
 }
 
-static void xen_blk_drain_io(struct xen_blkif *blkif)
+static void xen_blk_drain_io(struct xen_blkif_ring *ring)
 {
+	struct xen_blkif *blkif = ring->blkif;
+
 	atomic_set(&blkif->drain, 1);
 	do {
-		if (atomic_read(&blkif->inflight) == 0)
+		if (atomic_read(&ring->inflight) == 0)
 			break;
 		wait_for_completion_interruptible_timeout(
 				&blkif->drain_complete, HZ);
@@ -1053,12 +1074,12 @@ static void __end_block_io_op(struct pending_req *pending_req, int error)
 	if ((pending_req->operation == BLKIF_OP_FLUSH_DISKCACHE) &&
 	    (error == -EOPNOTSUPP)) {
 		pr_debug("flush diskcache op failed, not supported\n");
-		xen_blkbk_flush_diskcache(XBT_NIL, pending_req->blkif->be, 0);
+		xen_blkbk_flush_diskcache(XBT_NIL, pending_req->ring->blkif->be, 0);
 		pending_req->status = BLKIF_RSP_EOPNOTSUPP;
 	} else if ((pending_req->operation == BLKIF_OP_WRITE_BARRIER) &&
 		    (error == -EOPNOTSUPP)) {
 		pr_debug("write barrier op failed, not supported\n");
-		xen_blkbk_barrier(XBT_NIL, pending_req->blkif->be, 0);
+		xen_blkbk_barrier(XBT_NIL, pending_req->ring->blkif->be, 0);
 		pending_req->status = BLKIF_RSP_EOPNOTSUPP;
 	} else if (error) {
 		pr_debug("Buffer not up-to-date at end of operation,"
@@ -1092,9 +1113,9 @@ static void end_block_io_op(struct bio *bio)
  * and transmute  it to the block API to hand it over to the proper block disk.
  */
 static int
-__do_block_io_op(struct xen_blkif *blkif)
+__do_block_io_op(struct xen_blkif_ring *ring)
 {
-	union blkif_back_rings *blk_rings = &blkif->blk_rings;
+	union blkif_back_rings *blk_rings = &ring->blk_rings;
 	struct blkif_request req;
 	struct pending_req *pending_req;
 	RING_IDX rc, rp;
@@ -1107,7 +1128,7 @@ __do_block_io_op(struct xen_blkif *blkif)
 	if (RING_REQUEST_PROD_OVERFLOW(&blk_rings->common, rp)) {
 		rc = blk_rings->common.rsp_prod_pvt;
 		pr_warn("Frontend provided bogus ring requests (%d - %d = %d). Halting ring processing on dev=%04x\n",
-			rp, rc, rp - rc, blkif->vbd.pdevice);
+			rp, rc, rp - rc, ring->blkif->vbd.pdevice);
 		return -EACCES;
 	}
 	while (rc != rp) {
@@ -1120,14 +1141,14 @@ __do_block_io_op(struct xen_blkif *blkif)
 			break;
 		}
 
-		pending_req = alloc_req(blkif);
+		pending_req = alloc_req(ring);
 		if (NULL == pending_req) {
-			blkif->st_oo_req++;
+			ring->st_oo_req++;
 			more_to_do = 1;
 			break;
 		}
 
-		switch (blkif->blk_protocol) {
+		switch (ring->blkif->blk_protocol) {
 		case BLKIF_PROTOCOL_NATIVE:
 			memcpy(&req, RING_GET_REQUEST(&blk_rings->native, rc), sizeof(req));
 			break;
@@ -1151,16 +1172,16 @@ __do_block_io_op(struct xen_blkif *blkif)
 		case BLKIF_OP_WRITE_BARRIER:
 		case BLKIF_OP_FLUSH_DISKCACHE:
 		case BLKIF_OP_INDIRECT:
-			if (dispatch_rw_block_io(blkif, &req, pending_req))
+			if (dispatch_rw_block_io(ring, &req, pending_req))
 				goto done;
 			break;
 		case BLKIF_OP_DISCARD:
-			free_req(blkif, pending_req);
-			if (dispatch_discard_io(blkif, &req))
+			free_req(ring, pending_req);
+			if (dispatch_discard_io(ring, &req))
 				goto done;
 			break;
 		default:
-			if (dispatch_other_io(blkif, &req, pending_req))
+			if (dispatch_other_io(ring, &req, pending_req))
 				goto done;
 			break;
 		}
@@ -1173,13 +1194,13 @@ done:
 }
 
 static int
-do_block_io_op(struct xen_blkif *blkif)
+do_block_io_op(struct xen_blkif_ring *ring)
 {
-	union blkif_back_rings *blk_rings = &blkif->blk_rings;
+	union blkif_back_rings *blk_rings = &ring->blk_rings;
 	int more_to_do;
 
 	do {
-		more_to_do = __do_block_io_op(blkif);
+		more_to_do = __do_block_io_op(ring);
 		if (more_to_do)
 			break;
 
@@ -1192,7 +1213,7 @@ do_block_io_op(struct xen_blkif *blkif)
  * Transmutation of the 'struct blkif_request' to a proper 'struct bio'
  * and call the 'submit_bio' to pass it to the underlying storage.
  */
-static int dispatch_rw_block_io(struct xen_blkif *blkif,
+static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 				struct blkif_request *req,
 				struct pending_req *pending_req)
 {
@@ -1220,17 +1241,17 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 
 	switch (req_operation) {
 	case BLKIF_OP_READ:
-		blkif->st_rd_req++;
+		ring->st_rd_req++;
 		operation = READ;
 		break;
 	case BLKIF_OP_WRITE:
-		blkif->st_wr_req++;
+		ring->st_wr_req++;
 		operation = WRITE_ODIRECT;
 		break;
 	case BLKIF_OP_WRITE_BARRIER:
 		drain = true;
 	case BLKIF_OP_FLUSH_DISKCACHE:
-		blkif->st_f_req++;
+		ring->st_f_req++;
 		operation = WRITE_FLUSH;
 		break;
 	default:
@@ -1255,7 +1276,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 
 	preq.nr_sects      = 0;
 
-	pending_req->blkif     = blkif;
+	pending_req->ring      = ring;
 	pending_req->id        = req->u.rw.id;
 	pending_req->operation = req_operation;
 	pending_req->status    = BLKIF_RSP_OKAY;
@@ -1282,12 +1303,12 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 			goto fail_response;
 	}
 
-	if (xen_vbd_translate(&preq, blkif, operation) != 0) {
+	if (xen_vbd_translate(&preq, ring->blkif, operation) != 0) {
 		pr_debug("access denied: %s of [%llu,%llu] on dev=%04x\n",
 			 operation == READ ? "read" : "write",
 			 preq.sector_number,
 			 preq.sector_number + preq.nr_sects,
-			 blkif->vbd.pdevice);
+			 ring->blkif->vbd.pdevice);
 		goto fail_response;
 	}
 
@@ -1299,7 +1320,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 		if (((int)preq.sector_number|(int)seg[i].nsec) &
 		    ((bdev_logical_block_size(preq.bdev) >> 9) - 1)) {
 			pr_debug("Misaligned I/O request from domain %d\n",
-				 blkif->domid);
+				 ring->blkif->domid);
 			goto fail_response;
 		}
 	}
@@ -1308,7 +1329,7 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 	 * issue the WRITE_FLUSH.
 	 */
 	if (drain)
-		xen_blk_drain_io(pending_req->blkif);
+		xen_blk_drain_io(pending_req->ring);
 
 	/*
 	 * If we have failed at this point, we need to undo the M2P override,
@@ -1323,8 +1344,8 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 	 * This corresponding xen_blkif_put is done in __end_block_io_op, or
 	 * below (in "!bio") if we are handling a BLKIF_OP_DISCARD.
 	 */
-	xen_blkif_get(blkif);
-	atomic_inc(&blkif->inflight);
+	xen_blkif_get(ring->blkif);
+	atomic_inc(&ring->inflight);
 
 	for (i = 0; i < nseg; i++) {
 		while ((bio == NULL) ||
@@ -1372,19 +1393,19 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 	blk_finish_plug(&plug);
 
 	if (operation == READ)
-		blkif->st_rd_sect += preq.nr_sects;
+		ring->st_rd_sect += preq.nr_sects;
 	else if (operation & WRITE)
-		blkif->st_wr_sect += preq.nr_sects;
+		ring->st_wr_sect += preq.nr_sects;
 
 	return 0;
 
  fail_flush:
-	xen_blkbk_unmap(blkif, pending_req->segments,
+	xen_blkbk_unmap(ring, pending_req->segments,
 	                pending_req->nr_segs);
  fail_response:
 	/* Haven't submitted any bio's yet. */
-	make_response(blkif, req->u.rw.id, req_operation, BLKIF_RSP_ERROR);
-	free_req(blkif, pending_req);
+	make_response(ring, req->u.rw.id, req_operation, BLKIF_RSP_ERROR);
+	free_req(ring, pending_req);
 	msleep(1); /* back off a bit */
 	return -EIO;
 
@@ -1402,21 +1423,22 @@ static int dispatch_rw_block_io(struct xen_blkif *blkif,
 /*
  * Put a response on the ring on how the operation fared.
  */
-static void make_response(struct xen_blkif *blkif, u64 id,
+static void make_response(struct xen_blkif_ring *ring, u64 id,
 			  unsigned short op, int st)
 {
 	struct blkif_response  resp;
 	unsigned long     flags;
-	union blkif_back_rings *blk_rings = &blkif->blk_rings;
+	union blkif_back_rings *blk_rings;
 	int notify;
 
 	resp.id        = id;
 	resp.operation = op;
 	resp.status    = st;
 
-	spin_lock_irqsave(&blkif->blk_ring_lock, flags);
+	spin_lock_irqsave(&ring->blk_ring_lock, flags);
+	blk_rings = &ring->blk_rings;
 	/* Place on the response ring for the relevant domain. */
-	switch (blkif->blk_protocol) {
+	switch (ring->blkif->blk_protocol) {
 	case BLKIF_PROTOCOL_NATIVE:
 		memcpy(RING_GET_RESPONSE(&blk_rings->native, blk_rings->native.rsp_prod_pvt),
 		       &resp, sizeof(resp));
@@ -1434,9 +1456,9 @@ static void make_response(struct xen_blkif *blkif, u64 id,
 	}
 	blk_rings->common.rsp_prod_pvt++;
 	RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&blk_rings->common, notify);
-	spin_unlock_irqrestore(&blkif->blk_ring_lock, flags);
+	spin_unlock_irqrestore(&ring->blk_ring_lock, flags);
 	if (notify)
-		notify_remote_via_irq(blkif->irq);
+		notify_remote_via_irq(ring->irq);
 }
 
 static int __init xen_blkif_init(void)
@@ -1452,6 +1474,9 @@ static int __init xen_blkif_init(void)
 		xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
 	}
 
+	if (xenblk_max_queues == 0)
+		xenblk_max_queues = num_online_cpus();
+
 	rc = xen_blkif_interface_init();
 	if (rc)
 		goto failed_init;

drivers/block/xen-blkback/common.h

@@ -46,6 +46,7 @@
 #include <xen/interface/io/protocols.h>
 
 extern unsigned int xen_blkif_max_ring_order;
+extern unsigned int xenblk_max_queues;
 /*
  * This is the maximum number of segments that would be allowed in indirect
  * requests. This value will also be passed to the frontend.
@@ -269,68 +270,79 @@ struct persistent_gnt {
 	struct list_head remove_node;
 };
 
-struct xen_blkif {
-	/* Unique identifier for this interface. */
-	domid_t			domid;
-	unsigned int		handle;
+/* Per-ring information. */
+struct xen_blkif_ring {
 	/* Physical parameters of the comms window. */
 	unsigned int		irq;
-	/* Comms information. */
-	enum blkif_protocol	blk_protocol;
 	union blkif_back_rings	blk_rings;
 	void			*blk_ring;
-	/* The VBD attached to this interface. */
-	struct xen_vbd		vbd;
-	/* Back pointer to the backend_info. */
-	struct backend_info	*be;
 	/* Private fields. */
 	spinlock_t		blk_ring_lock;
-	atomic_t		refcnt;
 
 	wait_queue_head_t	wq;
-	/* for barrier (drain) requests */
-	struct completion	drain_complete;
-	atomic_t		drain;
 	atomic_t		inflight;
-	/* One thread per one blkif. */
+	/* One thread per blkif ring. */
 	struct task_struct	*xenblkd;
 	unsigned int		waiting_reqs;
 
-	/* tree to store persistent grants */
+	/* List of all 'pending_req' available */
+	struct list_head	pending_free;
+	/* And its spinlock. */
+	spinlock_t		pending_free_lock;
+	wait_queue_head_t	pending_free_wq;
+
+	/* Tree to store persistent grants. */
+	spinlock_t		pers_gnts_lock;
 	struct rb_root		persistent_gnts;
 	unsigned int		persistent_gnt_c;
 	atomic_t		persistent_gnt_in_use;
 	unsigned long           next_lru;
 
-	/* used by the kworker that offload work from the persistent purge */
+	/* Statistics. */
+	unsigned long		st_print;
+	unsigned long long	st_rd_req;
+	unsigned long long	st_wr_req;
+	unsigned long long	st_oo_req;
+	unsigned long long	st_f_req;
+	unsigned long long	st_ds_req;
+	unsigned long long	st_rd_sect;
+	unsigned long long	st_wr_sect;
+
+	/* Used by the kworker that offload work from the persistent purge. */
 	struct list_head	persistent_purge_list;
 	struct work_struct	persistent_purge_work;
 
-	/* buffer of free pages to map grant refs */
+	/* Buffer of free pages to map grant refs. */
 	spinlock_t		free_pages_lock;
 	int			free_pages_num;
 	struct list_head	free_pages;
 
-	/* List of all 'pending_req' available */
-	struct list_head	pending_free;
-	/* And its spinlock. */
-	spinlock_t		pending_free_lock;
-	wait_queue_head_t	pending_free_wq;
-
-	/* statistics */
-	unsigned long		st_print;
-	unsigned long long			st_rd_req;
-	unsigned long long			st_wr_req;
-	unsigned long long			st_oo_req;
-	unsigned long long			st_f_req;
-	unsigned long long			st_ds_req;
-	unsigned long long			st_rd_sect;
-	unsigned long long			st_wr_sect;
-
 	struct work_struct	free_work;
 	/* Thread shutdown wait queue. */
 	wait_queue_head_t	shutdown_wq;
-	unsigned int nr_ring_pages;
+	struct xen_blkif 	*blkif;
+};
+
+struct xen_blkif {
+	/* Unique identifier for this interface. */
+	domid_t			domid;
+	unsigned int		handle;
+	/* Comms information. */
+	enum blkif_protocol	blk_protocol;
+	/* The VBD attached to this interface. */
+	struct xen_vbd		vbd;
+	/* Back pointer to the backend_info. */
+	struct backend_info	*be;
+	atomic_t		refcnt;
+	/* for barrier (drain) requests */
+	struct completion	drain_complete;
+	atomic_t		drain;
+
+	struct work_struct	free_work;
+	unsigned int 		nr_ring_pages;
+	/* All rings for this device. */
+	struct xen_blkif_ring	*rings;
+	unsigned int		nr_rings;
 };
 
 struct seg_buf {
@@ -352,7 +364,7 @@ struct grant_page {
  * response queued for it, with the saved 'id' passed back.
  */
 struct pending_req {
-	struct xen_blkif	*blkif;
+	struct xen_blkif_ring   *ring;
 	u64			id;
 	int			nr_segs;
 	atomic_t		pendcnt;
@@ -394,7 +406,7 @@ int xen_blkif_xenbus_init(void);
 irqreturn_t xen_blkif_be_int(int irq, void *dev_id);
 int xen_blkif_schedule(void *arg);
 int xen_blkif_purge_persistent(void *arg);
-void xen_blkbk_free_caches(struct xen_blkif *blkif);
+void xen_blkbk_free_caches(struct xen_blkif_ring *ring);
 
 int xen_blkbk_flush_diskcache(struct xenbus_transaction xbt,
 			      struct backend_info *be, int state);

drivers/block/xen-blkback/xenbus.c

@@ -86,9 +86,11 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 {
 	int err;
 	char name[BLKBACK_NAME_LEN];
+	struct xen_blkif_ring *ring;
+	int i;
 
 	/* Not ready to connect? */
-	if (!blkif->irq || !blkif->vbd.bdev)
+	if (!blkif->rings || !blkif->rings[0].irq || !blkif->vbd.bdev)
 		return;
 
 	/* Already connected? */
@@ -113,13 +115,55 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 	}
 	invalidate_inode_pages2(blkif->vbd.bdev->bd_inode->i_mapping);
 
-	blkif->xenblkd = kthread_run(xen_blkif_schedule, blkif, "%s", name);
-	if (IS_ERR(blkif->xenblkd)) {
-		err = PTR_ERR(blkif->xenblkd);
-		blkif->xenblkd = NULL;
-		xenbus_dev_error(blkif->be->dev, err, "start xenblkd");
-		return;
+	for (i = 0; i < blkif->nr_rings; i++) {
+		ring = &blkif->rings[i];
+		ring->xenblkd = kthread_run(xen_blkif_schedule, ring, "%s-%d", name, i);
+		if (IS_ERR(ring->xenblkd)) {
+			err = PTR_ERR(ring->xenblkd);
+			ring->xenblkd = NULL;
+			xenbus_dev_fatal(blkif->be->dev, err,
+					"start %s-%d xenblkd", name, i);
+			goto out;
+		}
+	}
+	return;
+
+out:
+	while (--i >= 0) {
+		ring = &blkif->rings[i];
+		kthread_stop(ring->xenblkd);
+	}
+	return;
+}
+
+static int xen_blkif_alloc_rings(struct xen_blkif *blkif)
+{
+	unsigned int r;
+
+	blkif->rings = kzalloc(blkif->nr_rings * sizeof(struct xen_blkif_ring), GFP_KERNEL);
+	if (!blkif->rings)
+		return -ENOMEM;
+
+	for (r = 0; r < blkif->nr_rings; r++) {
+		struct xen_blkif_ring *ring = &blkif->rings[r];
+
+		spin_lock_init(&ring->blk_ring_lock);
+		init_waitqueue_head(&ring->wq);
+		INIT_LIST_HEAD(&ring->pending_free);
+		INIT_LIST_HEAD(&ring->persistent_purge_list);
+		INIT_WORK(&ring->persistent_purge_work, xen_blkbk_unmap_purged_grants);
+		spin_lock_init(&ring->free_pages_lock);
+		INIT_LIST_HEAD(&ring->free_pages);
+
+		spin_lock_init(&ring->pending_free_lock);
+		init_waitqueue_head(&ring->pending_free_wq);
+		init_waitqueue_head(&ring->shutdown_wq);
+		ring->blkif = blkif;
+		ring->st_print = jiffies;
+		xen_blkif_get(blkif);
 	}
+
+	return 0;
 }
 
 static struct xen_blkif *xen_blkif_alloc(domid_t domid)
@@ -133,41 +177,25 @@ static struct xen_blkif *xen_blkif_alloc(domid_t domid)
 		return ERR_PTR(-ENOMEM);
 
 	blkif->domid = domid;
-	spin_lock_init(&blkif->blk_ring_lock);
 	atomic_set(&blkif->refcnt, 1);
-	init_waitqueue_head(&blkif->wq);
 	init_completion(&blkif->drain_complete);
-	atomic_set(&blkif->drain, 0);
-	blkif->st_print = jiffies;
-	blkif->persistent_gnts.rb_node = NULL;
-	spin_lock_init(&blkif->free_pages_lock);
-	INIT_LIST_HEAD(&blkif->free_pages);
-	INIT_LIST_HEAD(&blkif->persistent_purge_list);
-	blkif->free_pages_num = 0;
-	atomic_set(&blkif->persistent_gnt_in_use, 0);
-	atomic_set(&blkif->inflight, 0);
-	INIT_WORK(&blkif->persistent_purge_work, xen_blkbk_unmap_purged_grants);
-
-	INIT_LIST_HEAD(&blkif->pending_free);
 	INIT_WORK(&blkif->free_work, xen_blkif_deferred_free);
-	spin_lock_init(&blkif->pending_free_lock);
-	init_waitqueue_head(&blkif->pending_free_wq);
-	init_waitqueue_head(&blkif->shutdown_wq);
 
 	return blkif;
 }
 
-static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref,
+static int xen_blkif_map(struct xen_blkif_ring *ring, grant_ref_t *gref,
 			 unsigned int nr_grefs, unsigned int evtchn)
 {
 	int err;
+	struct xen_blkif *blkif = ring->blkif;
 
 	/* Already connected through? */
-	if (blkif->irq)
+	if (ring->irq)
 		return 0;
 
 	err = xenbus_map_ring_valloc(blkif->be->dev, gref, nr_grefs,
-				     &blkif->blk_ring);
+				     &ring->blk_ring);
 	if (err < 0)
 		return err;
 
@@ -175,24 +203,24 @@ static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref,
 	case BLKIF_PROTOCOL_NATIVE:
 	{
 		struct blkif_sring *sring;
-		sring = (struct blkif_sring *)blkif->blk_ring;
-		BACK_RING_INIT(&blkif->blk_rings.native, sring,
+		sring = (struct blkif_sring *)ring->blk_ring;
+		BACK_RING_INIT(&ring->blk_rings.native, sring,
 			       XEN_PAGE_SIZE * nr_grefs);
 		break;
 	}
 	case BLKIF_PROTOCOL_X86_32:
 	{
 		struct blkif_x86_32_sring *sring_x86_32;
-		sring_x86_32 = (struct blkif_x86_32_sring *)blkif->blk_ring;
-		BACK_RING_INIT(&blkif->blk_rings.x86_32, sring_x86_32,
+		sring_x86_32 = (struct blkif_x86_32_sring *)ring->blk_ring;
+		BACK_RING_INIT(&ring->blk_rings.x86_32, sring_x86_32,
 			       XEN_PAGE_SIZE * nr_grefs);
 		break;
 	}
 	case BLKIF_PROTOCOL_X86_64:
 	{
 		struct blkif_x86_64_sring *sring_x86_64;
-		sring_x86_64 = (struct blkif_x86_64_sring *)blkif->blk_ring;
-		BACK_RING_INIT(&blkif->blk_rings.x86_64, sring_x86_64,
+		sring_x86_64 = (struct blkif_x86_64_sring *)ring->blk_ring;
+		BACK_RING_INIT(&ring->blk_rings.x86_64, sring_x86_64,
 			       XEN_PAGE_SIZE * nr_grefs);
 		break;
 	}
@@ -202,13 +230,13 @@ static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref,
 
 	err = bind_interdomain_evtchn_to_irqhandler(blkif->domid, evtchn,
 						    xen_blkif_be_int, 0,
-						    "blkif-backend", blkif);
+						    "blkif-backend", ring);
 	if (err < 0) {
-		xenbus_unmap_ring_vfree(blkif->be->dev, blkif->blk_ring);
-		blkif->blk_rings.common.sring = NULL;
+		xenbus_unmap_ring_vfree(blkif->be->dev, ring->blk_ring);
+		ring->blk_rings.common.sring = NULL;
 		return err;
 	}
-	blkif->irq = err;
+	ring->irq = err;
 
 	return 0;
 }
@@ -216,50 +244,69 @@ static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref,
 static int xen_blkif_disconnect(struct xen_blkif *blkif)
 {
 	struct pending_req *req, *n;
-	int i = 0, j;
+	unsigned int j, r;
 
-	if (blkif->xenblkd) {
-		kthread_stop(blkif->xenblkd);
-		wake_up(&blkif->shutdown_wq);
-		blkif->xenblkd = NULL;
-	}
+	for (r = 0; r < blkif->nr_rings; r++) {
+		struct xen_blkif_ring *ring = &blkif->rings[r];
+		unsigned int i = 0;
 
-	/* The above kthread_stop() guarantees that at this point we
-	 * don't have any discard_io or other_io requests. So, checking
-	 * for inflight IO is enough.
-	 */
-	if (atomic_read(&blkif->inflight) > 0)
-		return -EBUSY;
+		if (ring->xenblkd) {
+			kthread_stop(ring->xenblkd);
+			wake_up(&ring->shutdown_wq);
+			ring->xenblkd = NULL;
+		}
 
-	if (blkif->irq) {
-		unbind_from_irqhandler(blkif->irq, blkif);
-		blkif->irq = 0;
-	}
+		/* The above kthread_stop() guarantees that at this point we
+		 * don't have any discard_io or other_io requests. So, checking
+		 * for inflight IO is enough.
+		 */
+		if (atomic_read(&ring->inflight) > 0)
+			return -EBUSY;
 
-	if (blkif->blk_rings.common.sring) {
-		xenbus_unmap_ring_vfree(blkif->be->dev, blkif->blk_ring);
-		blkif->blk_rings.common.sring = NULL;
-	}
+		if (ring->irq) {
+			unbind_from_irqhandler(ring->irq, ring);
+			ring->irq = 0;
+		}
 
-	/* Remove all persistent grants and the cache of ballooned pages. */
-	xen_blkbk_free_caches(blkif);
+		if (ring->blk_rings.common.sring) {
+			xenbus_unmap_ring_vfree(blkif->be->dev, ring->blk_ring);
+			ring->blk_rings.common.sring = NULL;
+		}
 
-	/* Check that there is no request in use */
-	list_for_each_entry_safe(req, n, &blkif->pending_free, free_list) {
-		list_del(&req->free_list);
+		/* Remove all persistent grants and the cache of ballooned pages. */
+		xen_blkbk_free_caches(ring);
 
-		for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++)
-			kfree(req->segments[j]);
+		/* Check that there is no request in use */
+		list_for_each_entry_safe(req, n, &ring->pending_free, free_list) {
+			list_del(&req->free_list);
 
-		for (j = 0; j < MAX_INDIRECT_PAGES; j++)
-			kfree(req->indirect_pages[j]);
+			for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++)
+				kfree(req->segments[j]);
 
-		kfree(req);
-		i++;
-	}
+			for (j = 0; j < MAX_INDIRECT_PAGES; j++)
+				kfree(req->indirect_pages[j]);
+
+			kfree(req);
+			i++;
+		}
 
-	WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages));
+		BUG_ON(atomic_read(&ring->persistent_gnt_in_use) != 0);
+		BUG_ON(!list_empty(&ring->persistent_purge_list));
+		BUG_ON(!RB_EMPTY_ROOT(&ring->persistent_gnts));
+		BUG_ON(!list_empty(&ring->free_pages));
+		BUG_ON(ring->free_pages_num != 0);
+		BUG_ON(ring->persistent_gnt_c != 0);
+		WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages));
+		xen_blkif_put(blkif);
+	}
 	blkif->nr_ring_pages = 0;
+	/*
+	 * blkif->rings was allocated in connect_ring, so we should free it in
+	 * here.
+	 */
+	kfree(blkif->rings);
+	blkif->rings = NULL;
+	blkif->nr_rings = 0;
 
 	return 0;
 }
@@ -271,13 +318,6 @@ static void xen_blkif_free(struct xen_blkif *blkif)
 	xen_vbd_free(&blkif->vbd);
 
 	/* Make sure everything is drained before shutting down */
-	BUG_ON(blkif->persistent_gnt_c != 0);
-	BUG_ON(atomic_read(&blkif->persistent_gnt_in_use) != 0);
-	BUG_ON(blkif->free_pages_num != 0);
-	BUG_ON(!list_empty(&blkif->persistent_purge_list));
-	BUG_ON(!list_empty(&blkif->free_pages));
-	BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts));
-
 	kmem_cache_free(xen_blkif_cachep, blkif);
 }
 
@@ -296,25 +336,38 @@ int __init xen_blkif_interface_init(void)
  *  sysfs interface for VBD I/O requests
  */
 
-#define VBD_SHOW(name, format, args...)					\
+#define VBD_SHOW_ALLRING(name, format)					\
 	static ssize_t show_##name(struct device *_dev,			\
 				   struct device_attribute *attr,	\
 				   char *buf)				\
 	{								\
 		struct xenbus_device *dev = to_xenbus_device(_dev);	\
 		struct backend_info *be = dev_get_drvdata(&dev->dev);	\
+		struct xen_blkif *blkif = be->blkif;			\
+		unsigned int i;						\
+		unsigned long long result = 0;				\
 									\
-		return sprintf(buf, format, ##args);			\
+		if (!blkif->rings)				\
+			goto out;					\
+									\
+		for (i = 0; i < blkif->nr_rings; i++) {		\
+			struct xen_blkif_ring *ring = &blkif->rings[i];	\
+									\
+			result += ring->st_##name;			\
+		}							\
+									\
+out:									\
+		return sprintf(buf, format, result);			\
 	}								\
 	static DEVICE_ATTR(name, S_IRUGO, show_##name, NULL)
 
-VBD_SHOW(oo_req,  "%llu\n", be->blkif->st_oo_req);
-VBD_SHOW(rd_req,  "%llu\n", be->blkif->st_rd_req);
-VBD_SHOW(wr_req,  "%llu\n", be->blkif->st_wr_req);
-VBD_SHOW(f_req,  "%llu\n", be->blkif->st_f_req);
-VBD_SHOW(ds_req,  "%llu\n", be->blkif->st_ds_req);
-VBD_SHOW(rd_sect, "%llu\n", be->blkif->st_rd_sect);
-VBD_SHOW(wr_sect, "%llu\n", be->blkif->st_wr_sect);
+VBD_SHOW_ALLRING(oo_req,  "%llu\n");
+VBD_SHOW_ALLRING(rd_req,  "%llu\n");
+VBD_SHOW_ALLRING(wr_req,  "%llu\n");
+VBD_SHOW_ALLRING(f_req,  "%llu\n");
+VBD_SHOW_ALLRING(ds_req,  "%llu\n");
+VBD_SHOW_ALLRING(rd_sect, "%llu\n");
+VBD_SHOW_ALLRING(wr_sect, "%llu\n");
 
 static struct attribute *xen_vbdstat_attrs[] = {
 	&dev_attr_oo_req.attr,
@@ -332,6 +385,18 @@ static struct attribute_group xen_vbdstat_group = {
 	.attrs = xen_vbdstat_attrs,
 };
 
+#define VBD_SHOW(name, format, args...)					\
+	static ssize_t show_##name(struct device *_dev,			\
+				   struct device_attribute *attr,	\
+				   char *buf)				\
+	{								\
+		struct xenbus_device *dev = to_xenbus_device(_dev);	\
+		struct backend_info *be = dev_get_drvdata(&dev->dev);	\
+									\
+		return sprintf(buf, format, ##args);			\
+	}								\
+	static DEVICE_ATTR(name, S_IRUGO, show_##name, NULL)
+
 VBD_SHOW(physical_device, "%x:%x\n", be->major, be->minor);
 VBD_SHOW(mode, "%s\n", be->mode);
 
@@ -440,11 +505,11 @@ static int xen_blkbk_remove(struct xenbus_device *dev)
 
 	dev_set_drvdata(&dev->dev, NULL);
 
-	if (be->blkif) {
+	if (be->blkif)
 		xen_blkif_disconnect(be->blkif);
-		xen_blkif_put(be->blkif);
-	}
 
+	/* Put the reference we set in xen_blkif_alloc(). */
+	xen_blkif_put(be->blkif);
 	kfree(be->mode);
 	kfree(be);
 	return 0;
@@ -553,6 +618,12 @@ static int xen_blkbk_probe(struct xenbus_device *dev,
 		goto fail;
 	}
 
+	/* Multi-queue: advertise how many queues are supported by us.*/
+	err = xenbus_printf(XBT_NIL, dev->nodename,
+			    "multi-queue-max-queues", "%u", xenblk_max_queues);
+	if (err)
+		pr_warn("Error writing multi-queue-max-queues\n");
+
 	/* setup back pointer */
 	be->blkif->be = be;
 
@@ -708,8 +779,14 @@ static void frontend_changed(struct xenbus_device *dev,
 		}
 
 		err = connect_ring(be);
-		if (err)
+		if (err) {
+			/*
+			 * Clean up so that memory resources can be used by
+			 * other devices. connect_ring reported already error.
+			 */
+			xen_blkif_disconnect(be->blkif);
 			break;
+		}
 		xen_update_blkif_status(be->blkif);
 		break;
 
@@ -825,50 +902,43 @@ again:
 	xenbus_transaction_end(xbt, 1);
 }
 
-
-static int connect_ring(struct backend_info *be)
+/*
+ * Each ring may have multi pages, depends on "ring-page-order".
+ */
+static int read_per_ring_refs(struct xen_blkif_ring *ring, const char *dir)
 {
-	struct xenbus_device *dev = be->dev;
 	unsigned int ring_ref[XENBUS_MAX_RING_GRANTS];
-	unsigned int evtchn, nr_grefs, ring_page_order;
-	unsigned int pers_grants;
-	char protocol[64] = "";
 	struct pending_req *req, *n;
 	int err, i, j;
+	struct xen_blkif *blkif = ring->blkif;
+	struct xenbus_device *dev = blkif->be->dev;
+	unsigned int ring_page_order, nr_grefs, evtchn;
 
-	pr_debug("%s %s\n", __func__, dev->otherend);
-
-	err = xenbus_scanf(XBT_NIL, dev->otherend, "event-channel", "%u",
+	err = xenbus_scanf(XBT_NIL, dir, "event-channel", "%u",
 			  &evtchn);
 	if (err != 1) {
 		err = -EINVAL;
-		xenbus_dev_fatal(dev, err, "reading %s/event-channel",
-				 dev->otherend);
+		xenbus_dev_fatal(dev, err, "reading %s/event-channel", dir);
 		return err;
 	}
-	pr_info("event-channel %u\n", evtchn);
 
 	err = xenbus_scanf(XBT_NIL, dev->otherend, "ring-page-order", "%u",
 			  &ring_page_order);
 	if (err != 1) {
-		err = xenbus_scanf(XBT_NIL, dev->otherend, "ring-ref",
-				  "%u", &ring_ref[0]);
+		err = xenbus_scanf(XBT_NIL, dir, "ring-ref", "%u", &ring_ref[0]);
 		if (err != 1) {
 			err = -EINVAL;
-			xenbus_dev_fatal(dev, err, "reading %s/ring-ref",
-					 dev->otherend);
+			xenbus_dev_fatal(dev, err, "reading %s/ring-ref", dir);
 			return err;
 		}
 		nr_grefs = 1;
-		pr_info("%s:using single page: ring-ref %d\n", dev->otherend,
-			ring_ref[0]);
 	} else {
 		unsigned int i;
 
 		if (ring_page_order > xen_blkif_max_ring_order) {
 			err = -EINVAL;
 			xenbus_dev_fatal(dev, err, "%s/request %d ring page order exceed max:%d",
-					 dev->otherend, ring_page_order,
+					 dir, ring_page_order,
 					 xen_blkif_max_ring_order);
 			return err;
 		}
@@ -878,52 +948,23 @@ static int connect_ring(struct backend_info *be)
 			char ring_ref_name[RINGREF_NAME_LEN];
 
 			snprintf(ring_ref_name, RINGREF_NAME_LEN, "ring-ref%u", i);
-			err = xenbus_scanf(XBT_NIL, dev->otherend, ring_ref_name,
+			err = xenbus_scanf(XBT_NIL, dir, ring_ref_name,
 					   "%u", &ring_ref[i]);
 			if (err != 1) {
 				err = -EINVAL;
 				xenbus_dev_fatal(dev, err, "reading %s/%s",
-						 dev->otherend, ring_ref_name);
+						 dir, ring_ref_name);
 				return err;
 			}
-			pr_info("ring-ref%u: %u\n", i, ring_ref[i]);
 		}
 	}
-
-	be->blkif->blk_protocol = BLKIF_PROTOCOL_DEFAULT;
-	err = xenbus_gather(XBT_NIL, dev->otherend, "protocol",
-			    "%63s", protocol, NULL);
-	if (err)
-		strcpy(protocol, "unspecified, assuming default");
-	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_NATIVE))
-		be->blkif->blk_protocol = BLKIF_PROTOCOL_NATIVE;
-	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_X86_32))
-		be->blkif->blk_protocol = BLKIF_PROTOCOL_X86_32;
-	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_X86_64))
-		be->blkif->blk_protocol = BLKIF_PROTOCOL_X86_64;
-	else {
-		xenbus_dev_fatal(dev, err, "unknown fe protocol %s", protocol);
-		return -1;
-	}
-	err = xenbus_gather(XBT_NIL, dev->otherend,
-			    "feature-persistent", "%u",
-			    &pers_grants, NULL);
-	if (err)
-		pers_grants = 0;
-
-	be->blkif->vbd.feature_gnt_persistent = pers_grants;
-	be->blkif->vbd.overflow_max_grants = 0;
-	be->blkif->nr_ring_pages = nr_grefs;
-
-	pr_info("ring-pages:%d, event-channel %d, protocol %d (%s) %s\n",
-		nr_grefs, evtchn, be->blkif->blk_protocol, protocol,
-		pers_grants ? "persistent grants" : "");
+	blkif->nr_ring_pages = nr_grefs;
 
 	for (i = 0; i < nr_grefs * XEN_BLKIF_REQS_PER_PAGE; i++) {
 		req = kzalloc(sizeof(*req), GFP_KERNEL);
 		if (!req)
 			goto fail;
-		list_add_tail(&req->free_list, &be->blkif->pending_free);
+		list_add_tail(&req->free_list, &ring->pending_free);
 		for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++) {
 			req->segments[j] = kzalloc(sizeof(*req->segments[0]), GFP_KERNEL);
 			if (!req->segments[j])
@@ -938,7 +979,7 @@ static int connect_ring(struct backend_info *be)
 	}
 
 	/* Map the shared frame, irq etc. */
-	err = xen_blkif_map(be->blkif, ring_ref, nr_grefs, evtchn);
+	err = xen_blkif_map(ring, ring_ref, nr_grefs, evtchn);
 	if (err) {
 		xenbus_dev_fatal(dev, err, "mapping ring-ref port %u", evtchn);
 		return err;
@@ -947,7 +988,7 @@ static int connect_ring(struct backend_info *be)
 	return 0;
 
 fail:
-	list_for_each_entry_safe(req, n, &be->blkif->pending_free, free_list) {
+	list_for_each_entry_safe(req, n, &ring->pending_free, free_list) {
 		list_del(&req->free_list);
 		for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++) {
 			if (!req->segments[j])
@@ -962,6 +1003,93 @@ fail:
 		kfree(req);
 	}
 	return -ENOMEM;
+
+}
+
+static int connect_ring(struct backend_info *be)
+{
+	struct xenbus_device *dev = be->dev;
+	unsigned int pers_grants;
+	char protocol[64] = "";
+	int err, i;
+	char *xspath;
+	size_t xspathsize;
+	const size_t xenstore_path_ext_size = 11; /* sufficient for "/queue-NNN" */
+	unsigned int requested_num_queues = 0;
+
+	pr_debug("%s %s\n", __func__, dev->otherend);
+
+	be->blkif->blk_protocol = BLKIF_PROTOCOL_DEFAULT;
+	err = xenbus_gather(XBT_NIL, dev->otherend, "protocol",
+			    "%63s", protocol, NULL);
+	if (err)
+		strcpy(protocol, "unspecified, assuming default");
+	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_NATIVE))
+		be->blkif->blk_protocol = BLKIF_PROTOCOL_NATIVE;
+	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_X86_32))
+		be->blkif->blk_protocol = BLKIF_PROTOCOL_X86_32;
+	else if (0 == strcmp(protocol, XEN_IO_PROTO_ABI_X86_64))
+		be->blkif->blk_protocol = BLKIF_PROTOCOL_X86_64;
+	else {
+		xenbus_dev_fatal(dev, err, "unknown fe protocol %s", protocol);
+		return -ENOSYS;
+	}
+	err = xenbus_gather(XBT_NIL, dev->otherend,
+			    "feature-persistent", "%u",
+			    &pers_grants, NULL);
+	if (err)
+		pers_grants = 0;
+
+	be->blkif->vbd.feature_gnt_persistent = pers_grants;
+	be->blkif->vbd.overflow_max_grants = 0;
+
+	/*
+	 * Read the number of hardware queues from frontend.
+	 */
+	err = xenbus_scanf(XBT_NIL, dev->otherend, "multi-queue-num-queues",
+			   "%u", &requested_num_queues);
+	if (err < 0) {
+		requested_num_queues = 1;
+	} else {
+		if (requested_num_queues > xenblk_max_queues
+		    || requested_num_queues == 0) {
+			/* Buggy or malicious guest. */
+			xenbus_dev_fatal(dev, err,
+					"guest requested %u queues, exceeding the maximum of %u.",
+					requested_num_queues, xenblk_max_queues);
+			return -ENOSYS;
+		}
+	}
+	be->blkif->nr_rings = requested_num_queues;
+	if (xen_blkif_alloc_rings(be->blkif))
+		return -ENOMEM;
+
+	pr_info("%s: using %d queues, protocol %d (%s) %s\n", dev->nodename,
+		 be->blkif->nr_rings, be->blkif->blk_protocol, protocol,
+		 pers_grants ? "persistent grants" : "");
+
+	if (be->blkif->nr_rings == 1)
+		return read_per_ring_refs(&be->blkif->rings[0], dev->otherend);
+	else {
+		xspathsize = strlen(dev->otherend) + xenstore_path_ext_size;
+		xspath = kmalloc(xspathsize, GFP_KERNEL);
+		if (!xspath) {
+			xenbus_dev_fatal(dev, -ENOMEM, "reading ring references");
+			return -ENOMEM;
+		}
+
+		for (i = 0; i < be->blkif->nr_rings; i++) {
+			memset(xspath, 0, xspathsize);
+			snprintf(xspath, xspathsize, "%s/queue-%u", dev->otherend, i);
+			err = read_per_ring_refs(&be->blkif->rings[i], xspath);
+			if (err) {
+				kfree(xspath);
+				return err;
+			}
+		}
+		kfree(xspath);
+	}
+	return 0;
 }
 
 static const struct xenbus_device_id xen_blkbk_ids[] = {

drivers/block/xen-blkfront.c

@@ -60,6 +60,20 @@
 
 #include <asm/xen/hypervisor.h>
 
+/*
+ * The minimal size of segment supported by the block framework is PAGE_SIZE.
+ * When Linux is using a different page size than Xen, it may not be possible
+ * to put all the data in a single segment.
+ * This can happen when the backend doesn't support indirect descriptor and
+ * therefore the maximum amount of data that a request can carry is
+ * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE = 44KB
+ *
+ * Note that we only support one extra request. So the Linux page size
+ * should be <= ( 2 * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) =
+ * 88KB.
+ */
+#define HAS_EXTRA_REQ (BLKIF_MAX_SEGMENTS_PER_REQUEST < XEN_PFN_PER_PAGE)
+
 enum blkif_state {
 	BLKIF_STATE_DISCONNECTED,
 	BLKIF_STATE_CONNECTED,
@@ -72,6 +86,13 @@ struct grant {
 	struct list_head node;
 };
 
+enum blk_req_status {
+	REQ_WAITING,
+	REQ_DONE,
+	REQ_ERROR,
+	REQ_EOPNOTSUPP,
+};
+
 struct blk_shadow {
 	struct blkif_request req;
 	struct request *request;
@@ -79,6 +100,14 @@ struct blk_shadow {
 	struct grant **indirect_grants;
 	struct scatterlist *sg;
 	unsigned int num_sg;
+	enum blk_req_status status;
+
+	#define NO_ASSOCIATED_ID ~0UL
+	/*
+	 * Id of the sibling if we ever need 2 requests when handling a
+	 * block I/O request
+	 */
+	unsigned long associated_id;
 };
 
 struct split_bio {
@@ -99,6 +128,10 @@ static unsigned int xen_blkif_max_segments = 32;
 module_param_named(max, xen_blkif_max_segments, int, S_IRUGO);
 MODULE_PARM_DESC(max, "Maximum amount of segments in indirect requests (default is 32)");
 
+static unsigned int xen_blkif_max_queues = 4;
+module_param_named(max_queues, xen_blkif_max_queues, uint, S_IRUGO);
+MODULE_PARM_DESC(max_queues, "Maximum number of hardware queues/rings used per virtual disk");
+
 /*
  * Maximum order of pages to be used for the shared ring between front and
  * backend, 4KB page granularity is used.
@@ -114,10 +147,35 @@ MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the
 	__CONST_RING_SIZE(blkif, XEN_PAGE_SIZE * XENBUS_MAX_RING_GRANTS)
 
 /*
- * ring-ref%i i=(-1UL) would take 11 characters + 'ring-ref' is 8, so 19
- * characters are enough. Define to 20 to keep consist with backend.
+ * ring-ref%u i=(-1UL) would take 11 characters + 'ring-ref' is 8, so 19
+ * characters are enough. Define to 20 to keep consistent with backend.
  */
 #define RINGREF_NAME_LEN (20)
+/*
+ * queue-%u would take 7 + 10(UINT_MAX) = 17 characters.
+ */
+#define QUEUE_NAME_LEN (17)
+
+/*
+ *  Per-ring info.
+ *  Every blkfront device can associate with one or more blkfront_ring_info,
+ *  depending on how many hardware queues/rings to be used.
+ */
+struct blkfront_ring_info {
+	/* Lock to protect data in every ring buffer. */
+	spinlock_t ring_lock;
+	struct blkif_front_ring ring;
+	unsigned int ring_ref[XENBUS_MAX_RING_GRANTS];
+	unsigned int evtchn, irq;
+	struct work_struct work;
+	struct gnttab_free_callback callback;
+	struct blk_shadow shadow[BLK_MAX_RING_SIZE];
+	struct list_head indirect_pages;
+	struct list_head grants;
+	unsigned int persistent_gnts_c;
+	unsigned long shadow_free;
+	struct blkfront_info *dev_info;
+};
 
 /*
  * We have one of these per vbd, whether ide, scsi or 'other'.  They
@@ -126,25 +184,15 @@ MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the
  */
 struct blkfront_info
 {
-	spinlock_t io_lock;
 	struct mutex mutex;
 	struct xenbus_device *xbdev;
 	struct gendisk *gd;
 	int vdevice;
 	blkif_vdev_t handle;
 	enum blkif_state connected;
-	int ring_ref[XENBUS_MAX_RING_GRANTS];
+	/* Number of pages per ring buffer. */
 	unsigned int nr_ring_pages;
-	struct blkif_front_ring ring;
-	unsigned int evtchn, irq;
 	struct request_queue *rq;
-	struct work_struct work;
-	struct gnttab_free_callback callback;
-	struct blk_shadow shadow[BLK_MAX_RING_SIZE];
-	struct list_head grants;
-	struct list_head indirect_pages;
-	unsigned int persistent_gnts_c;
-	unsigned long shadow_free;
 	unsigned int feature_flush;
 	unsigned int feature_discard:1;
 	unsigned int feature_secdiscard:1;
@@ -155,6 +203,8 @@ struct blkfront_info
 	unsigned int max_indirect_segments;
 	int is_ready;
 	struct blk_mq_tag_set tag_set;
+	struct blkfront_ring_info *rinfo;
+	unsigned int nr_rings;
 };
 
 static unsigned int nr_minors;
@@ -198,38 +248,40 @@ static DEFINE_SPINLOCK(minor_lock);
 
 #define GREFS(_psegs)	((_psegs) * GRANTS_PER_PSEG)
 
-static int blkfront_setup_indirect(struct blkfront_info *info);
-static int blkfront_gather_backend_features(struct blkfront_info *info);
+static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo);
+static void blkfront_gather_backend_features(struct blkfront_info *info);
 
-static int get_id_from_freelist(struct blkfront_info *info)
+static int get_id_from_freelist(struct blkfront_ring_info *rinfo)
 {
-	unsigned long free = info->shadow_free;
-	BUG_ON(free >= BLK_RING_SIZE(info));
-	info->shadow_free = info->shadow[free].req.u.rw.id;
-	info->shadow[free].req.u.rw.id = 0x0fffffee; /* debug */
+	unsigned long free = rinfo->shadow_free;
+
+	BUG_ON(free >= BLK_RING_SIZE(rinfo->dev_info));
+	rinfo->shadow_free = rinfo->shadow[free].req.u.rw.id;
+	rinfo->shadow[free].req.u.rw.id = 0x0fffffee; /* debug */
 	return free;
 }
 
-static int add_id_to_freelist(struct blkfront_info *info,
-			       unsigned long id)
+static int add_id_to_freelist(struct blkfront_ring_info *rinfo,
+			      unsigned long id)
 {
-	if (info->shadow[id].req.u.rw.id != id)
+	if (rinfo->shadow[id].req.u.rw.id != id)
 		return -EINVAL;
-	if (info->shadow[id].request == NULL)
+	if (rinfo->shadow[id].request == NULL)
 		return -EINVAL;
-	info->shadow[id].req.u.rw.id  = info->shadow_free;
-	info->shadow[id].request = NULL;
-	info->shadow_free = id;
+	rinfo->shadow[id].req.u.rw.id  = rinfo->shadow_free;
+	rinfo->shadow[id].request = NULL;
+	rinfo->shadow_free = id;
 	return 0;
 }
 
-static int fill_grant_buffer(struct blkfront_info *info, int num)
+static int fill_grant_buffer(struct blkfront_ring_info *rinfo, int num)
 {
+	struct blkfront_info *info = rinfo->dev_info;
 	struct page *granted_page;
 	struct grant *gnt_list_entry, *n;
 	int i = 0;
 
-	while(i < num) {
+	while (i < num) {
 		gnt_list_entry = kzalloc(sizeof(struct grant), GFP_NOIO);
 		if (!gnt_list_entry)
 			goto out_of_memory;
@@ -244,7 +296,7 @@ static int fill_grant_buffer(struct blkfront_info *info, int num)
 		}
 
 		gnt_list_entry->gref = GRANT_INVALID_REF;
-		list_add(&gnt_list_entry->node, &info->grants);
+		list_add(&gnt_list_entry->node, &rinfo->grants);
 		i++;
 	}
 
@@ -252,7 +304,7 @@ static int fill_grant_buffer(struct blkfront_info *info, int num)
 
 out_of_memory:
 	list_for_each_entry_safe(gnt_list_entry, n,
-	                         &info->grants, node) {
+	                         &rinfo->grants, node) {
 		list_del(&gnt_list_entry->node);
 		if (info->feature_persistent)
 			__free_page(gnt_list_entry->page);
@@ -263,17 +315,17 @@ out_of_memory:
 	return -ENOMEM;
 }
 
-static struct grant *get_free_grant(struct blkfront_info *info)
+static struct grant *get_free_grant(struct blkfront_ring_info *rinfo)
 {
 	struct grant *gnt_list_entry;
 
-	BUG_ON(list_empty(&info->grants));
-	gnt_list_entry = list_first_entry(&info->grants, struct grant,
+	BUG_ON(list_empty(&rinfo->grants));
+	gnt_list_entry = list_first_entry(&rinfo->grants, struct grant,
 					  node);
 	list_del(&gnt_list_entry->node);
 
 	if (gnt_list_entry->gref != GRANT_INVALID_REF)
-		info->persistent_gnts_c--;
+		rinfo->persistent_gnts_c--;
 
 	return gnt_list_entry;
 }
@@ -289,9 +341,10 @@ static inline void grant_foreign_access(const struct grant *gnt_list_entry,
 
 static struct grant *get_grant(grant_ref_t *gref_head,
 			       unsigned long gfn,
-			       struct blkfront_info *info)
+			       struct blkfront_ring_info *rinfo)
 {
-	struct grant *gnt_list_entry = get_free_grant(info);
+	struct grant *gnt_list_entry = get_free_grant(rinfo);
+	struct blkfront_info *info = rinfo->dev_info;
 
 	if (gnt_list_entry->gref != GRANT_INVALID_REF)
 		return gnt_list_entry;
@@ -312,9 +365,10 @@ static struct grant *get_grant(grant_ref_t *gref_head,
 }
 
 static struct grant *get_indirect_grant(grant_ref_t *gref_head,
-					struct blkfront_info *info)
+					struct blkfront_ring_info *rinfo)
 {
-	struct grant *gnt_list_entry = get_free_grant(info);
+	struct grant *gnt_list_entry = get_free_grant(rinfo);
+	struct blkfront_info *info = rinfo->dev_info;
 
 	if (gnt_list_entry->gref != GRANT_INVALID_REF)
 		return gnt_list_entry;
@@ -326,8 +380,8 @@ static struct grant *get_indirect_grant(grant_ref_t *gref_head,
 		struct page *indirect_page;
 
 		/* Fetch a pre-allocated page to use for indirect grefs */
-		BUG_ON(list_empty(&info->indirect_pages));
-		indirect_page = list_first_entry(&info->indirect_pages,
+		BUG_ON(list_empty(&rinfo->indirect_pages));
+		indirect_page = list_first_entry(&rinfo->indirect_pages,
 						 struct page, lru);
 		list_del(&indirect_page->lru);
 		gnt_list_entry->page = indirect_page;
@@ -403,8 +457,8 @@ static void xlbd_release_minors(unsigned int minor, unsigned int nr)
 
 static void blkif_restart_queue_callback(void *arg)
 {
-	struct blkfront_info *info = (struct blkfront_info *)arg;
-	schedule_work(&info->work);
+	struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)arg;
+	schedule_work(&rinfo->work);
 }
 
 static int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg)
@@ -456,16 +510,33 @@ static int blkif_ioctl(struct block_device *bdev, fmode_t mode,
 	return 0;
 }
 
-static int blkif_queue_discard_req(struct request *req)
+static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo,
+					    struct request *req,
+					    struct blkif_request **ring_req)
 {
-	struct blkfront_info *info = req->rq_disk->private_data;
+	unsigned long id;
+
+	*ring_req = RING_GET_REQUEST(&rinfo->ring, rinfo->ring.req_prod_pvt);
+	rinfo->ring.req_prod_pvt++;
+
+	id = get_id_from_freelist(rinfo);
+	rinfo->shadow[id].request = req;
+	rinfo->shadow[id].status = REQ_WAITING;
+	rinfo->shadow[id].associated_id = NO_ASSOCIATED_ID;
+
+	(*ring_req)->u.rw.id = id;
+
+	return id;
+}
+
+static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_info *rinfo)
+{
+	struct blkfront_info *info = rinfo->dev_info;
 	struct blkif_request *ring_req;
 	unsigned long id;
 
 	/* Fill out a communications ring structure. */
-	ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
-	id = get_id_from_freelist(info);
-	info->shadow[id].request = req;
+	id = blkif_ring_get_request(rinfo, req, &ring_req);
 
 	ring_req->operation = BLKIF_OP_DISCARD;
 	ring_req->u.discard.nr_sectors = blk_rq_sectors(req);
@@ -476,10 +547,8 @@ static int blkif_queue_discard_req(struct request *req)
 	else
 		ring_req->u.discard.flag = 0;
 
-	info->ring.req_prod_pvt++;
-
 	/* Keep a private copy so we can reissue requests when recovering. */
-	info->shadow[id].req = *ring_req;
+	rinfo->shadow[id].req = *ring_req;
 
 	return 0;
 }
@@ -487,7 +556,7 @@ static int blkif_queue_discard_req(struct request *req)
 struct setup_rw_req {
 	unsigned int grant_idx;
 	struct blkif_request_segment *segments;
-	struct blkfront_info *info;
+	struct blkfront_ring_info *rinfo;
 	struct blkif_request *ring_req;
 	grant_ref_t gref_head;
 	unsigned int id;
@@ -495,6 +564,9 @@ struct setup_rw_req {
 	bool need_copy;
 	unsigned int bvec_off;
 	char *bvec_data;
+
+	bool require_extra_req;
+	struct blkif_request *extra_ring_req;
 };
 
 static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
@@ -507,8 +579,24 @@ static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
 	/* Convenient aliases */
 	unsigned int grant_idx = setup->grant_idx;
 	struct blkif_request *ring_req = setup->ring_req;
-	struct blkfront_info *info = setup->info;
-	struct blk_shadow *shadow = &info->shadow[setup->id];
+	struct blkfront_ring_info *rinfo = setup->rinfo;
+	/*
+	 * We always use the shadow of the first request to store the list
+	 * of grant associated to the block I/O request. This made the
+	 * completion more easy to handle even if the block I/O request is
+	 * split.
+	 */
+	struct blk_shadow *shadow = &rinfo->shadow[setup->id];
+
+	if (unlikely(setup->require_extra_req &&
+		     grant_idx >= BLKIF_MAX_SEGMENTS_PER_REQUEST)) {
+		/*
+		 * We are using the second request, setup grant_idx
+		 * to be the index of the segment array.
+		 */
+		grant_idx -= BLKIF_MAX_SEGMENTS_PER_REQUEST;
+		ring_req = setup->extra_ring_req;
+	}
 
 	if ((ring_req->operation == BLKIF_OP_INDIRECT) &&
 	    (grant_idx % GRANTS_PER_INDIRECT_FRAME == 0)) {
@@ -516,15 +604,19 @@ static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
 			kunmap_atomic(setup->segments);
 
 		n = grant_idx / GRANTS_PER_INDIRECT_FRAME;
-		gnt_list_entry = get_indirect_grant(&setup->gref_head, info);
+		gnt_list_entry = get_indirect_grant(&setup->gref_head, rinfo);
 		shadow->indirect_grants[n] = gnt_list_entry;
 		setup->segments = kmap_atomic(gnt_list_entry->page);
 		ring_req->u.indirect.indirect_grefs[n] = gnt_list_entry->gref;
 	}
 
-	gnt_list_entry = get_grant(&setup->gref_head, gfn, info);
+	gnt_list_entry = get_grant(&setup->gref_head, gfn, rinfo);
 	ref = gnt_list_entry->gref;
-	shadow->grants_used[grant_idx] = gnt_list_entry;
+	/*
+	 * All the grants are stored in the shadow of the first
+	 * request. Therefore we have to use the global index.
+	 */
+	shadow->grants_used[setup->grant_idx] = gnt_list_entry;
 
 	if (setup->need_copy) {
 		void *shared_data;
@@ -566,16 +658,36 @@ static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
 	(setup->grant_idx)++;
 }
 
-static int blkif_queue_rw_req(struct request *req)
+static void blkif_setup_extra_req(struct blkif_request *first,
+				  struct blkif_request *second)
 {
-	struct blkfront_info *info = req->rq_disk->private_data;
-	struct blkif_request *ring_req;
-	unsigned long id;
+	uint16_t nr_segments = first->u.rw.nr_segments;
+
+	/*
+	 * The second request is only present when the first request uses
+	 * all its segments. It's always the continuity of the first one.
+	 */
+	first->u.rw.nr_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
+
+	second->u.rw.nr_segments = nr_segments - BLKIF_MAX_SEGMENTS_PER_REQUEST;
+	second->u.rw.sector_number = first->u.rw.sector_number +
+		(BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) / 512;
+
+	second->u.rw.handle = first->u.rw.handle;
+	second->operation = first->operation;
+}
+
+static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *rinfo)
+{
+	struct blkfront_info *info = rinfo->dev_info;
+	struct blkif_request *ring_req, *extra_ring_req = NULL;
+	unsigned long id, extra_id = NO_ASSOCIATED_ID;
+	bool require_extra_req = false;
 	int i;
 	struct setup_rw_req setup = {
 		.grant_idx = 0,
 		.segments = NULL,
-		.info = info,
+		.rinfo = rinfo,
 		.need_copy = rq_data_dir(req) && info->feature_persistent,
 	};
 
@@ -584,7 +696,6 @@ static int blkif_queue_rw_req(struct request *req)
 	 * existing persistent grants, or if we have to get new grants,
 	 * as there are not sufficiently many free.
 	 */
-	bool new_persistent_gnts;
 	struct scatterlist *sg;
 	int num_sg, max_grefs, num_grant;
 
@@ -596,41 +707,36 @@ static int blkif_queue_rw_req(struct request *req)
 		 */
 		max_grefs += INDIRECT_GREFS(max_grefs);
 
-	/* Check if we have enough grants to allocate a requests */
-	if (info->persistent_gnts_c < max_grefs) {
-		new_persistent_gnts = 1;
-		if (gnttab_alloc_grant_references(
-		    max_grefs - info->persistent_gnts_c,
-		    &setup.gref_head) < 0) {
+	/*
+	 * We have to reserve 'max_grefs' grants because persistent
+	 * grants are shared by all rings.
+	 */
+	if (max_grefs > 0)
+		if (gnttab_alloc_grant_references(max_grefs, &setup.gref_head) < 0) {
 			gnttab_request_free_callback(
-				&info->callback,
+				&rinfo->callback,
 				blkif_restart_queue_callback,
-				info,
+				rinfo,
 				max_grefs);
 			return 1;
 		}
-	} else
-		new_persistent_gnts = 0;
 
 	/* Fill out a communications ring structure. */
-	ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
-	id = get_id_from_freelist(info);
-	info->shadow[id].request = req;
-
-	BUG_ON(info->max_indirect_segments == 0 &&
-	       GREFS(req->nr_phys_segments) > BLKIF_MAX_SEGMENTS_PER_REQUEST);
-	BUG_ON(info->max_indirect_segments &&
-	       GREFS(req->nr_phys_segments) > info->max_indirect_segments);
+	id = blkif_ring_get_request(rinfo, req, &ring_req);
 
-	num_sg = blk_rq_map_sg(req->q, req, info->shadow[id].sg);
+	num_sg = blk_rq_map_sg(req->q, req, rinfo->shadow[id].sg);
 	num_grant = 0;
 	/* Calculate the number of grant used */
-	for_each_sg(info->shadow[id].sg, sg, num_sg, i)
+	for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i)
 	       num_grant += gnttab_count_grant(sg->offset, sg->length);
 
-	ring_req->u.rw.id = id;
-	info->shadow[id].num_sg = num_sg;
-	if (num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST) {
+	require_extra_req = info->max_indirect_segments == 0 &&
+		num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST;
+	BUG_ON(!HAS_EXTRA_REQ && require_extra_req);
+
+	rinfo->shadow[id].num_sg = num_sg;
+	if (num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST &&
+	    likely(!require_extra_req)) {
 		/*
 		 * The indirect operation can only be a BLKIF_OP_READ or
 		 * BLKIF_OP_WRITE
@@ -670,11 +776,31 @@ static int blkif_queue_rw_req(struct request *req)
 			}
 		}
 		ring_req->u.rw.nr_segments = num_grant;
+		if (unlikely(require_extra_req)) {
+			extra_id = blkif_ring_get_request(rinfo, req,
+							  &extra_ring_req);
+			/*
+			 * Only the first request contains the scatter-gather
+			 * list.
+			 */
+			rinfo->shadow[extra_id].num_sg = 0;
+
+			blkif_setup_extra_req(ring_req, extra_ring_req);
+
+			/* Link the 2 requests together */
+			rinfo->shadow[extra_id].associated_id = id;
+			rinfo->shadow[id].associated_id = extra_id;
+		}
 	}
 
 	setup.ring_req = ring_req;
 	setup.id = id;
-	for_each_sg(info->shadow[id].sg, sg, num_sg, i) {
+
+	setup.require_extra_req = require_extra_req;
+	if (unlikely(require_extra_req))
+		setup.extra_ring_req = extra_ring_req;
+
+	for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i) {
 		BUG_ON(sg->offset + sg->length > PAGE_SIZE);
 
 		if (setup.need_copy) {
@@ -694,12 +820,12 @@ static int blkif_queue_rw_req(struct request *req)
 	if (setup.segments)
 		kunmap_atomic(setup.segments);
 
-	info->ring.req_prod_pvt++;
-
 	/* Keep a private copy so we can reissue requests when recovering. */
-	info->shadow[id].req = *ring_req;
+	rinfo->shadow[id].req = *ring_req;
+	if (unlikely(require_extra_req))
+		rinfo->shadow[extra_id].req = *extra_ring_req;
 
-	if (new_persistent_gnts)
+	if (max_grefs > 0)
 		gnttab_free_grant_references(setup.gref_head);
 
 	return 0;
@@ -711,27 +837,25 @@ static int blkif_queue_rw_req(struct request *req)
  *
  * @req: a request struct
  */
-static int blkif_queue_request(struct request *req)
+static int blkif_queue_request(struct request *req, struct blkfront_ring_info *rinfo)
 {
-	struct blkfront_info *info = req->rq_disk->private_data;
-
-	if (unlikely(info->connected != BLKIF_STATE_CONNECTED))
+	if (unlikely(rinfo->dev_info->connected != BLKIF_STATE_CONNECTED))
 		return 1;
 
 	if (unlikely(req->cmd_flags & (REQ_DISCARD | REQ_SECURE)))
-		return blkif_queue_discard_req(req);
+		return blkif_queue_discard_req(req, rinfo);
 	else
-		return blkif_queue_rw_req(req);
+		return blkif_queue_rw_req(req, rinfo);
 }
 
-static inline void flush_requests(struct blkfront_info *info)
+static inline void flush_requests(struct blkfront_ring_info *rinfo)
 {
 	int notify;
 
-	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&info->ring, notify);
+	RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&rinfo->ring, notify);
 
 	if (notify)
-		notify_remote_via_irq(info->irq);
+		notify_remote_via_irq(rinfo->irq);
 }
 
 static inline bool blkif_request_flush_invalid(struct request *req,
@@ -745,38 +869,50 @@ static inline bool blkif_request_flush_invalid(struct request *req,
 }
 
 static int blkif_queue_rq(struct blk_mq_hw_ctx *hctx,
-			   const struct blk_mq_queue_data *qd)
+			  const struct blk_mq_queue_data *qd)
 {
-	struct blkfront_info *info = qd->rq->rq_disk->private_data;
+	unsigned long flags;
+	struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)hctx->driver_data;
 
 	blk_mq_start_request(qd->rq);
-	spin_lock_irq(&info->io_lock);
-	if (RING_FULL(&info->ring))
+	spin_lock_irqsave(&rinfo->ring_lock, flags);
+	if (RING_FULL(&rinfo->ring))
 		goto out_busy;
 
-	if (blkif_request_flush_invalid(qd->rq, info))
+	if (blkif_request_flush_invalid(qd->rq, rinfo->dev_info))
 		goto out_err;
 
-	if (blkif_queue_request(qd->rq))
+	if (blkif_queue_request(qd->rq, rinfo))
 		goto out_busy;
 
-	flush_requests(info);
-	spin_unlock_irq(&info->io_lock);
+	flush_requests(rinfo);
+	spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 	return BLK_MQ_RQ_QUEUE_OK;
 
 out_err:
-	spin_unlock_irq(&info->io_lock);
+	spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 	return BLK_MQ_RQ_QUEUE_ERROR;
 
 out_busy:
-	spin_unlock_irq(&info->io_lock);
+	spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 	blk_mq_stop_hw_queue(hctx);
 	return BLK_MQ_RQ_QUEUE_BUSY;
 }
 
+static int blk_mq_init_hctx(struct blk_mq_hw_ctx *hctx, void *data,
+			    unsigned int index)
+{
+	struct blkfront_info *info = (struct blkfront_info *)data;
+
+	BUG_ON(info->nr_rings <= index);
+	hctx->driver_data = &info->rinfo[index];
+	return 0;
+}
+
 static struct blk_mq_ops blkfront_mq_ops = {
 	.queue_rq = blkif_queue_rq,
 	.map_queue = blk_mq_map_queue,
+	.init_hctx = blk_mq_init_hctx,
 };
 
 static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size,
@@ -788,19 +924,28 @@ static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size,
 
 	memset(&info->tag_set, 0, sizeof(info->tag_set));
 	info->tag_set.ops = &blkfront_mq_ops;
-	info->tag_set.nr_hw_queues = 1;
-	info->tag_set.queue_depth =  BLK_RING_SIZE(info);
+	info->tag_set.nr_hw_queues = info->nr_rings;
+	if (HAS_EXTRA_REQ && info->max_indirect_segments == 0) {
+		/*
+		 * When indirect descriptior is not supported, the I/O request
+		 * will be split between multiple request in the ring.
+		 * To avoid problems when sending the request, divide by
+		 * 2 the depth of the queue.
+		 */
+		info->tag_set.queue_depth =  BLK_RING_SIZE(info) / 2;
+	} else
+		info->tag_set.queue_depth = BLK_RING_SIZE(info);
 	info->tag_set.numa_node = NUMA_NO_NODE;
 	info->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_SG_MERGE;
 	info->tag_set.cmd_size = 0;
 	info->tag_set.driver_data = info;
 
 	if (blk_mq_alloc_tag_set(&info->tag_set))
-		return -1;
+		return -EINVAL;
 	rq = blk_mq_init_queue(&info->tag_set);
 	if (IS_ERR(rq)) {
 		blk_mq_free_tag_set(&info->tag_set);
-		return -1;
+		return PTR_ERR(rq);
 	}
 
 	queue_flag_set_unlocked(QUEUE_FLAG_VIRT, rq);
@@ -1028,7 +1173,7 @@ static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
 
 static void xlvbd_release_gendisk(struct blkfront_info *info)
 {
-	unsigned int minor, nr_minors;
+	unsigned int minor, nr_minors, i;
 
 	if (info->rq == NULL)
 		return;
@@ -1036,11 +1181,15 @@ static void xlvbd_release_gendisk(struct blkfront_info *info)
 	/* No more blkif_request(). */
 	blk_mq_stop_hw_queues(info->rq);
 
-	/* No more gnttab callback work. */
-	gnttab_cancel_free_callback(&info->callback);
+	for (i = 0; i < info->nr_rings; i++) {
+		struct blkfront_ring_info *rinfo = &info->rinfo[i];
 
-	/* Flush gnttab callback work. Must be done with no locks held. */
-	flush_work(&info->work);
+		/* No more gnttab callback work. */
+		gnttab_cancel_free_callback(&rinfo->callback);
+
+		/* Flush gnttab callback work. Must be done with no locks held. */
+		flush_work(&rinfo->work);
+	}
 
 	del_gendisk(info->gd);
 
@@ -1056,88 +1205,87 @@ static void xlvbd_release_gendisk(struct blkfront_info *info)
 	info->gd = NULL;
 }
 
-/* Must be called with io_lock holded */
-static void kick_pending_request_queues(struct blkfront_info *info)
+/* Already hold rinfo->ring_lock. */
+static inline void kick_pending_request_queues_locked(struct blkfront_ring_info *rinfo)
 {
-	if (!RING_FULL(&info->ring))
-		blk_mq_start_stopped_hw_queues(info->rq, true);
+	if (!RING_FULL(&rinfo->ring))
+		blk_mq_start_stopped_hw_queues(rinfo->dev_info->rq, true);
 }
 
-static void blkif_restart_queue(struct work_struct *work)
+static void kick_pending_request_queues(struct blkfront_ring_info *rinfo)
 {
-	struct blkfront_info *info = container_of(work, struct blkfront_info, work);
+	unsigned long flags;
 
-	spin_lock_irq(&info->io_lock);
-	if (info->connected == BLKIF_STATE_CONNECTED)
-		kick_pending_request_queues(info);
-	spin_unlock_irq(&info->io_lock);
+	spin_lock_irqsave(&rinfo->ring_lock, flags);
+	kick_pending_request_queues_locked(rinfo);
+	spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 }
 
-static void blkif_free(struct blkfront_info *info, int suspend)
+static void blkif_restart_queue(struct work_struct *work)
 {
-	struct grant *persistent_gnt;
-	struct grant *n;
-	int i, j, segs;
+	struct blkfront_ring_info *rinfo = container_of(work, struct blkfront_ring_info, work);
 
-	/* Prevent new requests being issued until we fix things up. */
-	spin_lock_irq(&info->io_lock);
-	info->connected = suspend ?
-		BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
-	/* No more blkif_request(). */
-	if (info->rq)
-		blk_mq_stop_hw_queues(info->rq);
+	if (rinfo->dev_info->connected == BLKIF_STATE_CONNECTED)
+		kick_pending_request_queues(rinfo);
+}
 
-	/* Remove all persistent grants */
-	if (!list_empty(&info->grants)) {
-		list_for_each_entry_safe(persistent_gnt, n,
-		                         &info->grants, node) {
-			list_del(&persistent_gnt->node);
-			if (persistent_gnt->gref != GRANT_INVALID_REF) {
-				gnttab_end_foreign_access(persistent_gnt->gref,
-				                          0, 0UL);
-				info->persistent_gnts_c--;
-			}
-			if (info->feature_persistent)
-				__free_page(persistent_gnt->page);
-			kfree(persistent_gnt);
-		}
-	}
-	BUG_ON(info->persistent_gnts_c != 0);
+static void blkif_free_ring(struct blkfront_ring_info *rinfo)
+{
+	struct grant *persistent_gnt, *n;
+	struct blkfront_info *info = rinfo->dev_info;
+	int i, j, segs;
 
 	/*
 	 * Remove indirect pages, this only happens when using indirect
 	 * descriptors but not persistent grants
 	 */
-	if (!list_empty(&info->indirect_pages)) {
+	if (!list_empty(&rinfo->indirect_pages)) {
 		struct page *indirect_page, *n;
 
 		BUG_ON(info->feature_persistent);
-		list_for_each_entry_safe(indirect_page, n, &info->indirect_pages, lru) {
+		list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
 			list_del(&indirect_page->lru);
 			__free_page(indirect_page);
 		}
 	}
 
+	/* Remove all persistent grants. */
+	if (!list_empty(&rinfo->grants)) {
+		list_for_each_entry_safe(persistent_gnt, n,
+					 &rinfo->grants, node) {
+			list_del(&persistent_gnt->node);
+			if (persistent_gnt->gref != GRANT_INVALID_REF) {
+				gnttab_end_foreign_access(persistent_gnt->gref,
+							  0, 0UL);
+				rinfo->persistent_gnts_c--;
+			}
+			if (info->feature_persistent)
+				__free_page(persistent_gnt->page);
+			kfree(persistent_gnt);
+		}
+	}
+	BUG_ON(rinfo->persistent_gnts_c != 0);
+
 	for (i = 0; i < BLK_RING_SIZE(info); i++) {
 		/*
 		 * Clear persistent grants present in requests already
 		 * on the shared ring
 		 */
-		if (!info->shadow[i].request)
+		if (!rinfo->shadow[i].request)
 			goto free_shadow;
 
-		segs = info->shadow[i].req.operation == BLKIF_OP_INDIRECT ?
-		       info->shadow[i].req.u.indirect.nr_segments :
-		       info->shadow[i].req.u.rw.nr_segments;
+		segs = rinfo->shadow[i].req.operation == BLKIF_OP_INDIRECT ?
+		       rinfo->shadow[i].req.u.indirect.nr_segments :
+		       rinfo->shadow[i].req.u.rw.nr_segments;
 		for (j = 0; j < segs; j++) {
-			persistent_gnt = info->shadow[i].grants_used[j];
+			persistent_gnt = rinfo->shadow[i].grants_used[j];
 			gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
 			if (info->feature_persistent)
 				__free_page(persistent_gnt->page);
 			kfree(persistent_gnt);
 		}
 
-		if (info->shadow[i].req.operation != BLKIF_OP_INDIRECT)
+		if (rinfo->shadow[i].req.operation != BLKIF_OP_INDIRECT)
 			/*
 			 * If this is not an indirect operation don't try to
 			 * free indirect segments
@@ -1145,42 +1293,59 @@ static void blkif_free(struct blkfront_info *info, int suspend)
 			goto free_shadow;
 
 		for (j = 0; j < INDIRECT_GREFS(segs); j++) {
-			persistent_gnt = info->shadow[i].indirect_grants[j];
+			persistent_gnt = rinfo->shadow[i].indirect_grants[j];
 			gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
 			__free_page(persistent_gnt->page);
 			kfree(persistent_gnt);
 		}
 
 free_shadow:
-		kfree(info->shadow[i].grants_used);
-		info->shadow[i].grants_used = NULL;
-		kfree(info->shadow[i].indirect_grants);
-		info->shadow[i].indirect_grants = NULL;
-		kfree(info->shadow[i].sg);
-		info->shadow[i].sg = NULL;
+		kfree(rinfo->shadow[i].grants_used);
+		rinfo->shadow[i].grants_used = NULL;
+		kfree(rinfo->shadow[i].indirect_grants);
+		rinfo->shadow[i].indirect_grants = NULL;
+		kfree(rinfo->shadow[i].sg);
+		rinfo->shadow[i].sg = NULL;
 	}
 
 	/* No more gnttab callback work. */
-	gnttab_cancel_free_callback(&info->callback);
-	spin_unlock_irq(&info->io_lock);
+	gnttab_cancel_free_callback(&rinfo->callback);
 
 	/* Flush gnttab callback work. Must be done with no locks held. */
-	flush_work(&info->work);
+	flush_work(&rinfo->work);
 
 	/* Free resources associated with old device channel. */
 	for (i = 0; i < info->nr_ring_pages; i++) {
-		if (info->ring_ref[i] != GRANT_INVALID_REF) {
-			gnttab_end_foreign_access(info->ring_ref[i], 0, 0);
-			info->ring_ref[i] = GRANT_INVALID_REF;
+		if (rinfo->ring_ref[i] != GRANT_INVALID_REF) {
+			gnttab_end_foreign_access(rinfo->ring_ref[i], 0, 0);
+			rinfo->ring_ref[i] = GRANT_INVALID_REF;
 		}
 	}
-	free_pages((unsigned long)info->ring.sring, get_order(info->nr_ring_pages * PAGE_SIZE));
-	info->ring.sring = NULL;
+	free_pages((unsigned long)rinfo->ring.sring, get_order(info->nr_ring_pages * PAGE_SIZE));
+	rinfo->ring.sring = NULL;
 
-	if (info->irq)
-		unbind_from_irqhandler(info->irq, info);
-	info->evtchn = info->irq = 0;
+	if (rinfo->irq)
+		unbind_from_irqhandler(rinfo->irq, rinfo);
+	rinfo->evtchn = rinfo->irq = 0;
+}
 
+static void blkif_free(struct blkfront_info *info, int suspend)
+{
+	unsigned int i;
+
+	/* Prevent new requests being issued until we fix things up. */
+	info->connected = suspend ?
+		BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
+	/* No more blkif_request(). */
+	if (info->rq)
+		blk_mq_stop_hw_queues(info->rq);
+
+	for (i = 0; i < info->nr_rings; i++)
+		blkif_free_ring(&info->rinfo[i]);
+
+	kfree(info->rinfo);
+	info->rinfo = NULL;
+	info->nr_rings = 0;
 }
 
 struct copy_from_grant {
@@ -1209,19 +1374,93 @@ static void blkif_copy_from_grant(unsigned long gfn, unsigned int offset,
 	kunmap_atomic(shared_data);
 }
 
-static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
+static enum blk_req_status blkif_rsp_to_req_status(int rsp)
+{
+	switch (rsp)
+	{
+	case BLKIF_RSP_OKAY:
+		return REQ_DONE;
+	case BLKIF_RSP_EOPNOTSUPP:
+		return REQ_EOPNOTSUPP;
+	case BLKIF_RSP_ERROR:
+		/* Fallthrough. */
+	default:
+		return REQ_ERROR;
+	}
+}
+
+/*
+ * Get the final status of the block request based on two ring response
+ */
+static int blkif_get_final_status(enum blk_req_status s1,
+				  enum blk_req_status s2)
+{
+	BUG_ON(s1 == REQ_WAITING);
+	BUG_ON(s2 == REQ_WAITING);
+
+	if (s1 == REQ_ERROR || s2 == REQ_ERROR)
+		return BLKIF_RSP_ERROR;
+	else if (s1 == REQ_EOPNOTSUPP || s2 == REQ_EOPNOTSUPP)
+		return BLKIF_RSP_EOPNOTSUPP;
+	return BLKIF_RSP_OKAY;
+}
+
+static bool blkif_completion(unsigned long *id,
+			     struct blkfront_ring_info *rinfo,
 			     struct blkif_response *bret)
 {
 	int i = 0;
 	struct scatterlist *sg;
 	int num_sg, num_grant;
+	struct blkfront_info *info = rinfo->dev_info;
+	struct blk_shadow *s = &rinfo->shadow[*id];
 	struct copy_from_grant data = {
-		.s = s,
 		.grant_idx = 0,
 	};
 
 	num_grant = s->req.operation == BLKIF_OP_INDIRECT ?
 		s->req.u.indirect.nr_segments : s->req.u.rw.nr_segments;
+
+	/* The I/O request may be split in two. */
+	if (unlikely(s->associated_id != NO_ASSOCIATED_ID)) {
+		struct blk_shadow *s2 = &rinfo->shadow[s->associated_id];
+
+		/* Keep the status of the current response in shadow. */
+		s->status = blkif_rsp_to_req_status(bret->status);
+
+		/* Wait the second response if not yet here. */
+		if (s2->status == REQ_WAITING)
+			return 0;
+
+		bret->status = blkif_get_final_status(s->status,
+						      s2->status);
+
+		/*
+		 * All the grants is stored in the first shadow in order
+		 * to make the completion code simpler.
+		 */
+		num_grant += s2->req.u.rw.nr_segments;
+
+		/*
+		 * The two responses may not come in order. Only the
+		 * first request will store the scatter-gather list.
+		 */
+		if (s2->num_sg != 0) {
+			/* Update "id" with the ID of the first response. */
+			*id = s->associated_id;
+			s = s2;
+		}
+
+		/*
+		 * We don't need anymore the second request, so recycling
+		 * it now.
+		 */
+		if (add_id_to_freelist(rinfo, s->associated_id))
+			WARN(1, "%s: can't recycle the second part (id = %ld) of the request\n",
+			     info->gd->disk_name, s->associated_id);
+	}
+
+	data.s = s;
 	num_sg = s->num_sg;
 
 	if (bret->operation == BLKIF_OP_READ && info->feature_persistent) {
@@ -1252,8 +1491,8 @@ static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
 			if (!info->feature_persistent)
 				pr_alert_ratelimited("backed has not unmapped grant: %u\n",
 						     s->grants_used[i]->gref);
-			list_add(&s->grants_used[i]->node, &info->grants);
-			info->persistent_gnts_c++;
+			list_add(&s->grants_used[i]->node, &rinfo->grants);
+			rinfo->persistent_gnts_c++;
 		} else {
 			/*
 			 * If the grant is not mapped by the backend we end the
@@ -1263,7 +1502,7 @@ static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
 			 */
 			gnttab_end_foreign_access(s->grants_used[i]->gref, 0, 0UL);
 			s->grants_used[i]->gref = GRANT_INVALID_REF;
-			list_add_tail(&s->grants_used[i]->node, &info->grants);
+			list_add_tail(&s->grants_used[i]->node, &rinfo->grants);
 		}
 	}
 	if (s->req.operation == BLKIF_OP_INDIRECT) {
@@ -1272,8 +1511,8 @@ static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
 				if (!info->feature_persistent)
 					pr_alert_ratelimited("backed has not unmapped grant: %u\n",
 							     s->indirect_grants[i]->gref);
-				list_add(&s->indirect_grants[i]->node, &info->grants);
-				info->persistent_gnts_c++;
+				list_add(&s->indirect_grants[i]->node, &rinfo->grants);
+				rinfo->persistent_gnts_c++;
 			} else {
 				struct page *indirect_page;
 
@@ -1284,13 +1523,15 @@ static void blkif_completion(struct blk_shadow *s, struct blkfront_info *info,
 				 */
 				if (!info->feature_persistent) {
 					indirect_page = s->indirect_grants[i]->page;
-					list_add(&indirect_page->lru, &info->indirect_pages);
+					list_add(&indirect_page->lru, &rinfo->indirect_pages);
 				}
 				s->indirect_grants[i]->gref = GRANT_INVALID_REF;
-				list_add_tail(&s->indirect_grants[i]->node, &info->grants);
+				list_add_tail(&s->indirect_grants[i]->node, &rinfo->grants);
 			}
 		}
 	}
+
+	return 1;
 }
 
 static irqreturn_t blkif_interrupt(int irq, void *dev_id)
@@ -1299,24 +1540,22 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 	struct blkif_response *bret;
 	RING_IDX i, rp;
 	unsigned long flags;
-	struct blkfront_info *info = (struct blkfront_info *)dev_id;
+	struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)dev_id;
+	struct blkfront_info *info = rinfo->dev_info;
 	int error;
 
-	spin_lock_irqsave(&info->io_lock, flags);
-
-	if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) {
-		spin_unlock_irqrestore(&info->io_lock, flags);
+	if (unlikely(info->connected != BLKIF_STATE_CONNECTED))
 		return IRQ_HANDLED;
-	}
 
+	spin_lock_irqsave(&rinfo->ring_lock, flags);
  again:
-	rp = info->ring.sring->rsp_prod;
+	rp = rinfo->ring.sring->rsp_prod;
 	rmb(); /* Ensure we see queued responses up to 'rp'. */
 
-	for (i = info->ring.rsp_cons; i != rp; i++) {
+	for (i = rinfo->ring.rsp_cons; i != rp; i++) {
 		unsigned long id;
 
-		bret = RING_GET_RESPONSE(&info->ring, i);
+		bret = RING_GET_RESPONSE(&rinfo->ring, i);
 		id   = bret->id;
 		/*
 		 * The backend has messed up and given us an id that we would
@@ -1330,12 +1569,18 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 			 * the id is busted. */
 			continue;
 		}
-		req  = info->shadow[id].request;
+		req  = rinfo->shadow[id].request;
 
-		if (bret->operation != BLKIF_OP_DISCARD)
-			blkif_completion(&info->shadow[id], info, bret);
+		if (bret->operation != BLKIF_OP_DISCARD) {
+			/*
+			 * We may need to wait for an extra response if the
+			 * I/O request is split in 2
+			 */
+			if (!blkif_completion(&id, rinfo, bret))
+				continue;
+		}
 
-		if (add_id_to_freelist(info, id)) {
+		if (add_id_to_freelist(rinfo, id)) {
 			WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n",
 			     info->gd->disk_name, op_name(bret->operation), id);
 			continue;
@@ -1364,7 +1609,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 				error = -EOPNOTSUPP;
 			}
 			if (unlikely(bret->status == BLKIF_RSP_ERROR &&
-				     info->shadow[id].req.u.rw.nr_segments == 0)) {
+				     rinfo->shadow[id].req.u.rw.nr_segments == 0)) {
 				printk(KERN_WARNING "blkfront: %s: empty %s op failed\n",
 				       info->gd->disk_name, op_name(bret->operation));
 				error = -EOPNOTSUPP;
@@ -1389,34 +1634,35 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 		}
 	}
 
-	info->ring.rsp_cons = i;
+	rinfo->ring.rsp_cons = i;
 
-	if (i != info->ring.req_prod_pvt) {
+	if (i != rinfo->ring.req_prod_pvt) {
 		int more_to_do;
-		RING_FINAL_CHECK_FOR_RESPONSES(&info->ring, more_to_do);
+		RING_FINAL_CHECK_FOR_RESPONSES(&rinfo->ring, more_to_do);
 		if (more_to_do)
 			goto again;
 	} else
-		info->ring.sring->rsp_event = i + 1;
+		rinfo->ring.sring->rsp_event = i + 1;
 
-	kick_pending_request_queues(info);
+	kick_pending_request_queues_locked(rinfo);
 
-	spin_unlock_irqrestore(&info->io_lock, flags);
+	spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 
 	return IRQ_HANDLED;
 }
 
 
 static int setup_blkring(struct xenbus_device *dev,
-			 struct blkfront_info *info)
+			 struct blkfront_ring_info *rinfo)
 {
 	struct blkif_sring *sring;
 	int err, i;
+	struct blkfront_info *info = rinfo->dev_info;
 	unsigned long ring_size = info->nr_ring_pages * XEN_PAGE_SIZE;
 	grant_ref_t gref[XENBUS_MAX_RING_GRANTS];
 
 	for (i = 0; i < info->nr_ring_pages; i++)
-		info->ring_ref[i] = GRANT_INVALID_REF;
+		rinfo->ring_ref[i] = GRANT_INVALID_REF;
 
 	sring = (struct blkif_sring *)__get_free_pages(GFP_NOIO | __GFP_HIGH,
 						       get_order(ring_size));
@@ -1425,29 +1671,29 @@ static int setup_blkring(struct xenbus_device *dev,
 		return -ENOMEM;
 	}
 	SHARED_RING_INIT(sring);
-	FRONT_RING_INIT(&info->ring, sring, ring_size);
+	FRONT_RING_INIT(&rinfo->ring, sring, ring_size);
 
-	err = xenbus_grant_ring(dev, info->ring.sring, info->nr_ring_pages, gref);
+	err = xenbus_grant_ring(dev, rinfo->ring.sring, info->nr_ring_pages, gref);
 	if (err < 0) {
 		free_pages((unsigned long)sring, get_order(ring_size));
-		info->ring.sring = NULL;
+		rinfo->ring.sring = NULL;
 		goto fail;
 	}
 	for (i = 0; i < info->nr_ring_pages; i++)
-		info->ring_ref[i] = gref[i];
+		rinfo->ring_ref[i] = gref[i];
 
-	err = xenbus_alloc_evtchn(dev, &info->evtchn);
+	err = xenbus_alloc_evtchn(dev, &rinfo->evtchn);
 	if (err)
 		goto fail;
 
-	err = bind_evtchn_to_irqhandler(info->evtchn, blkif_interrupt, 0,
-					"blkif", info);
+	err = bind_evtchn_to_irqhandler(rinfo->evtchn, blkif_interrupt, 0,
+					"blkif", rinfo);
 	if (err <= 0) {
 		xenbus_dev_fatal(dev, err,
 				 "bind_evtchn_to_irqhandler failed");
 		goto fail;
 	}
-	info->irq = err;
+	rinfo->irq = err;
 
 	return 0;
 fail:
@@ -1455,6 +1701,53 @@ fail:
 	return err;
 }
 
+/*
+ * Write out per-ring/queue nodes including ring-ref and event-channel, and each
+ * ring buffer may have multi pages depending on ->nr_ring_pages.
+ */
+static int write_per_ring_nodes(struct xenbus_transaction xbt,
+				struct blkfront_ring_info *rinfo, const char *dir)
+{
+	int err;
+	unsigned int i;
+	const char *message = NULL;
+	struct blkfront_info *info = rinfo->dev_info;
+
+	if (info->nr_ring_pages == 1) {
+		err = xenbus_printf(xbt, dir, "ring-ref", "%u", rinfo->ring_ref[0]);
+		if (err) {
+			message = "writing ring-ref";
+			goto abort_transaction;
+		}
+	} else {
+		for (i = 0; i < info->nr_ring_pages; i++) {
+			char ring_ref_name[RINGREF_NAME_LEN];
+
+			snprintf(ring_ref_name, RINGREF_NAME_LEN, "ring-ref%u", i);
+			err = xenbus_printf(xbt, dir, ring_ref_name,
+					    "%u", rinfo->ring_ref[i]);
+			if (err) {
+				message = "writing ring-ref";
+				goto abort_transaction;
+			}
+		}
+	}
+
+	err = xenbus_printf(xbt, dir, "event-channel", "%u", rinfo->evtchn);
+	if (err) {
+		message = "writing event-channel";
+		goto abort_transaction;
+	}
+
+	return 0;
+
+abort_transaction:
+	xenbus_transaction_end(xbt, 1);
+	if (message)
+		xenbus_dev_fatal(info->xbdev, err, "%s", message);
+
+	return err;
+}
 
 /* Common code used when first setting up, and when resuming. */
 static int talk_to_blkback(struct xenbus_device *dev,
@@ -1462,8 +1755,8 @@ static int talk_to_blkback(struct xenbus_device *dev,
 {
 	const char *message = NULL;
 	struct xenbus_transaction xbt;
-	int err, i;
-	unsigned int max_page_order = 0;
+	int err;
+	unsigned int i, max_page_order = 0;
 	unsigned int ring_page_order = 0;
 
 	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
@@ -1475,10 +1768,14 @@ static int talk_to_blkback(struct xenbus_device *dev,
 		info->nr_ring_pages = 1 << ring_page_order;
 	}
 
-	/* Create shared ring, alloc event channel. */
-	err = setup_blkring(dev, info);
-	if (err)
-		goto out;
+	for (i = 0; i < info->nr_rings; i++) {
+		struct blkfront_ring_info *rinfo = &info->rinfo[i];
+
+		/* Create shared ring, alloc event channel. */
+		err = setup_blkring(dev, rinfo);
+		if (err)
+			goto destroy_blkring;
+	}
 
 again:
 	err = xenbus_transaction_start(&xbt);
@@ -1487,38 +1784,49 @@ again:
 		goto destroy_blkring;
 	}
 
-	if (info->nr_ring_pages == 1) {
-		err = xenbus_printf(xbt, dev->nodename,
-				    "ring-ref", "%u", info->ring_ref[0]);
+	if (info->nr_ring_pages > 1) {
+		err = xenbus_printf(xbt, dev->nodename, "ring-page-order", "%u",
+				    ring_page_order);
 		if (err) {
-			message = "writing ring-ref";
+			message = "writing ring-page-order";
 			goto abort_transaction;
 		}
+	}
+
+	/* We already got the number of queues/rings in _probe */
+	if (info->nr_rings == 1) {
+		err = write_per_ring_nodes(xbt, &info->rinfo[0], dev->nodename);
+		if (err)
+			goto destroy_blkring;
 	} else {
-		err = xenbus_printf(xbt, dev->nodename,
-				    "ring-page-order", "%u", ring_page_order);
+		char *path;
+		size_t pathsize;
+
+		err = xenbus_printf(xbt, dev->nodename, "multi-queue-num-queues", "%u",
+				    info->nr_rings);
 		if (err) {
-			message = "writing ring-page-order";
+			message = "writing multi-queue-num-queues";
 			goto abort_transaction;
 		}
 
-		for (i = 0; i < info->nr_ring_pages; i++) {
-			char ring_ref_name[RINGREF_NAME_LEN];
+		pathsize = strlen(dev->nodename) + QUEUE_NAME_LEN;
+		path = kmalloc(pathsize, GFP_KERNEL);
+		if (!path) {
+			err = -ENOMEM;
+			message = "ENOMEM while writing ring references";
+			goto abort_transaction;
+		}
 
-			snprintf(ring_ref_name, RINGREF_NAME_LEN, "ring-ref%u", i);
-			err = xenbus_printf(xbt, dev->nodename, ring_ref_name,
-					    "%u", info->ring_ref[i]);
+		for (i = 0; i < info->nr_rings; i++) {
+			memset(path, 0, pathsize);
+			snprintf(path, pathsize, "%s/queue-%u", dev->nodename, i);
+			err = write_per_ring_nodes(xbt, &info->rinfo[i], path);
 			if (err) {
-				message = "writing ring-ref";
-				goto abort_transaction;
+				kfree(path);
+				goto destroy_blkring;
 			}
 		}
-	}
-	err = xenbus_printf(xbt, dev->nodename,
-			    "event-channel", "%u", info->evtchn);
-	if (err) {
-		message = "writing event-channel";
-		goto abort_transaction;
+		kfree(path);
 	}
 	err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
 			    XEN_IO_PROTO_ABI_NATIVE);
@@ -1540,9 +1848,14 @@ again:
 		goto destroy_blkring;
 	}
 
-	for (i = 0; i < BLK_RING_SIZE(info); i++)
-		info->shadow[i].req.u.rw.id = i+1;
-	info->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
+	for (i = 0; i < info->nr_rings; i++) {
+		unsigned int j;
+		struct blkfront_ring_info *rinfo = &info->rinfo[i];
+
+		for (j = 0; j < BLK_RING_SIZE(info); j++)
+			rinfo->shadow[j].req.u.rw.id = j + 1;
+		rinfo->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
+	}
 	xenbus_switch_state(dev, XenbusStateInitialised);
 
 	return 0;
@@ -1553,7 +1866,10 @@ again:
 		xenbus_dev_fatal(dev, err, "%s", message);
  destroy_blkring:
 	blkif_free(info, 0);
- out:
+
+	kfree(info);
+	dev_set_drvdata(&dev->dev, NULL);
+
 	return err;
 }
 
@@ -1567,7 +1883,9 @@ static int blkfront_probe(struct xenbus_device *dev,
 			  const struct xenbus_device_id *id)
 {
 	int err, vdevice;
+	unsigned int r_index;
 	struct blkfront_info *info;
+	unsigned int backend_max_queues = 0;
 
 	/* FIXME: Use dynamic device id if this is not set. */
 	err = xenbus_scanf(XBT_NIL, dev->nodename,
@@ -1617,15 +1935,39 @@ static int blkfront_probe(struct xenbus_device *dev,
 		return -ENOMEM;
 	}
 
-	mutex_init(&info->mutex);
-	spin_lock_init(&info->io_lock);
 	info->xbdev = dev;
+	/* Check if backend supports multiple queues. */
+	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+			   "multi-queue-max-queues", "%u", &backend_max_queues);
+	if (err < 0)
+		backend_max_queues = 1;
+
+	info->nr_rings = min(backend_max_queues, xen_blkif_max_queues);
+	/* We need at least one ring. */
+	if (!info->nr_rings)
+		info->nr_rings = 1;
+
+	info->rinfo = kzalloc(sizeof(struct blkfront_ring_info) * info->nr_rings, GFP_KERNEL);
+	if (!info->rinfo) {
+		xenbus_dev_fatal(dev, -ENOMEM, "allocating ring_info structure");
+		kfree(info);
+		return -ENOMEM;
+	}
+
+	for (r_index = 0; r_index < info->nr_rings; r_index++) {
+		struct blkfront_ring_info *rinfo;
+
+		rinfo = &info->rinfo[r_index];
+		INIT_LIST_HEAD(&rinfo->indirect_pages);
+		INIT_LIST_HEAD(&rinfo->grants);
+		rinfo->dev_info = info;
+		INIT_WORK(&rinfo->work, blkif_restart_queue);
+		spin_lock_init(&rinfo->ring_lock);
+	}
+
+	mutex_init(&info->mutex);
 	info->vdevice = vdevice;
-	INIT_LIST_HEAD(&info->grants);
-	INIT_LIST_HEAD(&info->indirect_pages);
-	info->persistent_gnts_c = 0;
 	info->connected = BLKIF_STATE_DISCONNECTED;
-	INIT_WORK(&info->work, blkif_restart_queue);
 
 	/* Front end dir is a number, which is used as the id. */
 	info->handle = simple_strtoul(strrchr(dev->nodename, '/')+1, NULL, 0);
@@ -1649,7 +1991,7 @@ static void split_bio_end(struct bio *bio)
 
 static int blkif_recover(struct blkfront_info *info)
 {
-	int i;
+	unsigned int i, r_index;
 	struct request *req, *n;
 	struct blk_shadow *copy;
 	int rc;
@@ -1660,64 +2002,73 @@ static int blkif_recover(struct blkfront_info *info)
 	struct split_bio *split_bio;
 	struct list_head requests;
 
-	/* Stage 1: Make a safe copy of the shadow state. */
-	copy = kmemdup(info->shadow, sizeof(info->shadow),
-		       GFP_NOIO | __GFP_REPEAT | __GFP_HIGH);
-	if (!copy)
-		return -ENOMEM;
-
-	/* Stage 2: Set up free list. */
-	memset(&info->shadow, 0, sizeof(info->shadow));
-	for (i = 0; i < BLK_RING_SIZE(info); i++)
-		info->shadow[i].req.u.rw.id = i+1;
-	info->shadow_free = info->ring.req_prod_pvt;
-	info->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
-
-	rc = blkfront_gather_backend_features(info);
-	if (rc) {
-		kfree(copy);
-		return rc;
-	}
-
+	blkfront_gather_backend_features(info);
 	segs = info->max_indirect_segments ? : BLKIF_MAX_SEGMENTS_PER_REQUEST;
 	blk_queue_max_segments(info->rq, segs);
 	bio_list_init(&bio_list);
 	INIT_LIST_HEAD(&requests);
-	for (i = 0; i < BLK_RING_SIZE(info); i++) {
-		/* Not in use? */
-		if (!copy[i].request)
-			continue;
 
-		/*
-		 * Get the bios in the request so we can re-queue them.
-		 */
-		if (copy[i].request->cmd_flags &
-		    (REQ_FLUSH | REQ_FUA | REQ_DISCARD | REQ_SECURE)) {
+	for (r_index = 0; r_index < info->nr_rings; r_index++) {
+		struct blkfront_ring_info *rinfo;
+
+		rinfo = &info->rinfo[r_index];
+		/* Stage 1: Make a safe copy of the shadow state. */
+		copy = kmemdup(rinfo->shadow, sizeof(rinfo->shadow),
+			       GFP_NOIO | __GFP_REPEAT | __GFP_HIGH);
+		if (!copy)
+			return -ENOMEM;
+
+		/* Stage 2: Set up free list. */
+		memset(&rinfo->shadow, 0, sizeof(rinfo->shadow));
+		for (i = 0; i < BLK_RING_SIZE(info); i++)
+			rinfo->shadow[i].req.u.rw.id = i+1;
+		rinfo->shadow_free = rinfo->ring.req_prod_pvt;
+		rinfo->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
+
+		rc = blkfront_setup_indirect(rinfo);
+		if (rc) {
+			kfree(copy);
+			return rc;
+		}
+
+		for (i = 0; i < BLK_RING_SIZE(info); i++) {
+			/* Not in use? */
+			if (!copy[i].request)
+				continue;
+
 			/*
-			 * Flush operations don't contain bios, so
-			 * we need to requeue the whole request
+			 * Get the bios in the request so we can re-queue them.
 			 */
-			list_add(&copy[i].request->queuelist, &requests);
-			continue;
+			if (copy[i].request->cmd_flags &
+			    (REQ_FLUSH | REQ_FUA | REQ_DISCARD | REQ_SECURE)) {
+				/*
+				 * Flush operations don't contain bios, so
+				 * we need to requeue the whole request
+				 */
+				list_add(&copy[i].request->queuelist, &requests);
+				continue;
+			}
+			merge_bio.head = copy[i].request->bio;
+			merge_bio.tail = copy[i].request->biotail;
+			bio_list_merge(&bio_list, &merge_bio);
+			copy[i].request->bio = NULL;
+			blk_end_request_all(copy[i].request, 0);
 		}
-		merge_bio.head = copy[i].request->bio;
-		merge_bio.tail = copy[i].request->biotail;
-		bio_list_merge(&bio_list, &merge_bio);
-		copy[i].request->bio = NULL;
-		blk_end_request_all(copy[i].request, 0);
-	}
-
-	kfree(copy);
 
+		kfree(copy);
+	}
 	xenbus_switch_state(info->xbdev, XenbusStateConnected);
 
-	spin_lock_irq(&info->io_lock);
-
 	/* Now safe for us to use the shared ring */
 	info->connected = BLKIF_STATE_CONNECTED;
 
-	/* Kick any other new requests queued since we resumed */
-	kick_pending_request_queues(info);
+	for (r_index = 0; r_index < info->nr_rings; r_index++) {
+		struct blkfront_ring_info *rinfo;
+
+		rinfo = &info->rinfo[r_index];
+		/* Kick any other new requests queued since we resumed */
+		kick_pending_request_queues(rinfo);
+	}
 
 	list_for_each_entry_safe(req, n, &requests, queuelist) {
 		/* Requeue pending requests (flush or discard) */
@@ -1725,7 +2076,6 @@ static int blkif_recover(struct blkfront_info *info)
 		BUG_ON(req->nr_phys_segments > segs);
 		blk_mq_requeue_request(req);
 	}
-	spin_unlock_irq(&info->io_lock);
 	blk_mq_kick_requeue_list(info->rq);
 
 	while ((bio = bio_list_pop(&bio_list)) != NULL) {
@@ -1790,8 +2140,7 @@ static int blkfront_resume(struct xenbus_device *dev)
 	return err;
 }
 
-static void
-blkfront_closing(struct blkfront_info *info)
+static void blkfront_closing(struct blkfront_info *info)
 {
 	struct xenbus_device *xbdev = info->xbdev;
 	struct block_device *bdev = NULL;
@@ -1851,18 +2200,29 @@ static void blkfront_setup_discard(struct blkfront_info *info)
 		info->feature_secdiscard = !!discard_secure;
 }
 
-static int blkfront_setup_indirect(struct blkfront_info *info)
+static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo)
 {
 	unsigned int psegs, grants;
 	int err, i;
+	struct blkfront_info *info = rinfo->dev_info;
 
-	if (info->max_indirect_segments == 0)
-		grants = BLKIF_MAX_SEGMENTS_PER_REQUEST;
+	if (info->max_indirect_segments == 0) {
+		if (!HAS_EXTRA_REQ)
+			grants = BLKIF_MAX_SEGMENTS_PER_REQUEST;
+		else {
+			/*
+			 * When an extra req is required, the maximum
+			 * grants supported is related to the size of the
+			 * Linux block segment.
+			 */
+			grants = GRANTS_PER_PSEG;
+		}
+	}
 	else
 		grants = info->max_indirect_segments;
 	psegs = grants / GRANTS_PER_PSEG;
 
-	err = fill_grant_buffer(info,
+	err = fill_grant_buffer(rinfo,
 				(grants + INDIRECT_GREFS(grants)) * BLK_RING_SIZE(info));
 	if (err)
 		goto out_of_memory;
@@ -1875,31 +2235,31 @@ static int blkfront_setup_indirect(struct blkfront_info *info)
 		 */
 		int num = INDIRECT_GREFS(grants) * BLK_RING_SIZE(info);
 
-		BUG_ON(!list_empty(&info->indirect_pages));
+		BUG_ON(!list_empty(&rinfo->indirect_pages));
 		for (i = 0; i < num; i++) {
 			struct page *indirect_page = alloc_page(GFP_NOIO);
 			if (!indirect_page)
 				goto out_of_memory;
-			list_add(&indirect_page->lru, &info->indirect_pages);
+			list_add(&indirect_page->lru, &rinfo->indirect_pages);
 		}
 	}
 
 	for (i = 0; i < BLK_RING_SIZE(info); i++) {
-		info->shadow[i].grants_used = kzalloc(
-			sizeof(info->shadow[i].grants_used[0]) * grants,
+		rinfo->shadow[i].grants_used = kzalloc(
+			sizeof(rinfo->shadow[i].grants_used[0]) * grants,
 			GFP_NOIO);
-		info->shadow[i].sg = kzalloc(sizeof(info->shadow[i].sg[0]) * psegs, GFP_NOIO);
+		rinfo->shadow[i].sg = kzalloc(sizeof(rinfo->shadow[i].sg[0]) * psegs, GFP_NOIO);
 		if (info->max_indirect_segments)
-			info->shadow[i].indirect_grants = kzalloc(
-				sizeof(info->shadow[i].indirect_grants[0]) *
+			rinfo->shadow[i].indirect_grants = kzalloc(
+				sizeof(rinfo->shadow[i].indirect_grants[0]) *
 				INDIRECT_GREFS(grants),
 				GFP_NOIO);
-		if ((info->shadow[i].grants_used == NULL) ||
-			(info->shadow[i].sg == NULL) ||
+		if ((rinfo->shadow[i].grants_used == NULL) ||
+			(rinfo->shadow[i].sg == NULL) ||
 		     (info->max_indirect_segments &&
-		     (info->shadow[i].indirect_grants == NULL)))
+		     (rinfo->shadow[i].indirect_grants == NULL)))
 			goto out_of_memory;
-		sg_init_table(info->shadow[i].sg, psegs);
+		sg_init_table(rinfo->shadow[i].sg, psegs);
 	}
 
 
@@ -1907,16 +2267,16 @@ static int blkfront_setup_indirect(struct blkfront_info *info)
 
 out_of_memory:
 	for (i = 0; i < BLK_RING_SIZE(info); i++) {
-		kfree(info->shadow[i].grants_used);
-		info->shadow[i].grants_used = NULL;
-		kfree(info->shadow[i].sg);
-		info->shadow[i].sg = NULL;
-		kfree(info->shadow[i].indirect_grants);
-		info->shadow[i].indirect_grants = NULL;
-	}
-	if (!list_empty(&info->indirect_pages)) {
+		kfree(rinfo->shadow[i].grants_used);
+		rinfo->shadow[i].grants_used = NULL;
+		kfree(rinfo->shadow[i].sg);
+		rinfo->shadow[i].sg = NULL;
+		kfree(rinfo->shadow[i].indirect_grants);
+		rinfo->shadow[i].indirect_grants = NULL;
+	}
+	if (!list_empty(&rinfo->indirect_pages)) {
 		struct page *indirect_page, *n;
-		list_for_each_entry_safe(indirect_page, n, &info->indirect_pages, lru) {
+		list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
 			list_del(&indirect_page->lru);
 			__free_page(indirect_page);
 		}
@@ -1927,7 +2287,7 @@ out_of_memory:
 /*
  * Gather all backend feature-*
  */
-static int blkfront_gather_backend_features(struct blkfront_info *info)
+static void blkfront_gather_backend_features(struct blkfront_info *info)
 {
 	int err;
 	int barrier, flush, discard, persistent;
@@ -1982,8 +2342,6 @@ static int blkfront_gather_backend_features(struct blkfront_info *info)
 	else
 		info->max_indirect_segments = min(indirect_segments,
 						  xen_blkif_max_segments);
-
-	return blkfront_setup_indirect(info);
 }
 
 /*
@@ -1996,7 +2354,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	unsigned long sector_size;
 	unsigned int physical_sector_size;
 	unsigned int binfo;
-	int err;
+	int err, i;
 
 	switch (info->connected) {
 	case BLKIF_STATE_CONNECTED:
@@ -2053,11 +2411,15 @@ static void blkfront_connect(struct blkfront_info *info)
 	if (err != 1)
 		physical_sector_size = sector_size;
 
-	err = blkfront_gather_backend_features(info);
-	if (err) {
-		xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
-				 info->xbdev->otherend);
-		return;
+	blkfront_gather_backend_features(info);
+	for (i = 0; i < info->nr_rings; i++) {
+		err = blkfront_setup_indirect(&info->rinfo[i]);
+		if (err) {
+			xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
+					 info->xbdev->otherend);
+			blkif_free(info, 0);
+			break;
+		}
 	}
 
 	err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size,
@@ -2071,10 +2433,9 @@ static void blkfront_connect(struct blkfront_info *info)
 	xenbus_switch_state(info->xbdev, XenbusStateConnected);
 
 	/* Kick pending requests. */
-	spin_lock_irq(&info->io_lock);
 	info->connected = BLKIF_STATE_CONNECTED;
-	kick_pending_request_queues(info);
-	spin_unlock_irq(&info->io_lock);
+	for (i = 0; i < info->nr_rings; i++)
+		kick_pending_request_queues(&info->rinfo[i]);
 
 	add_disk(info->gd);
 
@@ -2095,11 +2456,8 @@ static void blkback_changed(struct xenbus_device *dev,
 	case XenbusStateInitWait:
 		if (dev->state != XenbusStateInitialising)
 			break;
-		if (talk_to_blkback(dev, info)) {
-			kfree(info);
-			dev_set_drvdata(&dev->dev, NULL);
+		if (talk_to_blkback(dev, info))
 			break;
-		}
 	case XenbusStateInitialising:
 	case XenbusStateInitialised:
 	case XenbusStateReconfiguring:
@@ -2108,6 +2466,10 @@ static void blkback_changed(struct xenbus_device *dev,
 		break;
 
 	case XenbusStateConnected:
+		if (dev->state != XenbusStateInitialised) {
+			if (talk_to_blkback(dev, info))
+				break;
+		}
 		blkfront_connect(info);
 		break;
 
@@ -2281,6 +2643,7 @@ static struct xenbus_driver blkfront_driver = {
 static int __init xlblk_init(void)
 {
 	int ret;
+	int nr_cpus = num_online_cpus();
 
 	if (!xen_domain())
 		return -ENODEV;
@@ -2288,7 +2651,13 @@ static int __init xlblk_init(void)
 	if (xen_blkif_max_ring_order > XENBUS_MAX_RING_GRANT_ORDER) {
 		pr_info("Invalid max_ring_order (%d), will use default max: %d.\n",
 			xen_blkif_max_ring_order, XENBUS_MAX_RING_GRANT_ORDER);
-		xen_blkif_max_ring_order = 0;
+		xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
+	}
+
+	if (xen_blkif_max_queues > nr_cpus) {
+		pr_info("Invalid max_queues (%d), will use default max: %d.\n",
+			xen_blkif_max_queues, nr_cpus);
+		xen_blkif_max_queues = nr_cpus;
 	}
 
 	if (!xen_has_pv_disk_devices())

include/xen/interface/io/blkif.h

@@ -27,6 +27,54 @@
 typedef uint16_t blkif_vdev_t;
 typedef uint64_t blkif_sector_t;
 
+/*
+ * Multiple hardware queues/rings:
+ * If supported, the backend will write the key "multi-queue-max-queues" to
+ * the directory for that vbd, and set its value to the maximum supported
+ * number of queues.
+ * Frontends that are aware of this feature and wish to use it can write the
+ * key "multi-queue-num-queues" with the number they wish to use, which must be
+ * greater than zero, and no more than the value reported by the backend in
+ * "multi-queue-max-queues".
+ *
+ * For frontends requesting just one queue, the usual event-channel and
+ * ring-ref keys are written as before, simplifying the backend processing
+ * to avoid distinguishing between a frontend that doesn't understand the
+ * multi-queue feature, and one that does, but requested only one queue.
+ *
+ * Frontends requesting two or more queues must not write the toplevel
+ * event-channel and ring-ref keys, instead writing those keys under sub-keys
+ * having the name "queue-N" where N is the integer ID of the queue/ring for
+ * which those keys belong. Queues are indexed from zero.
+ * For example, a frontend with two queues must write the following set of
+ * queue-related keys:
+ *
+ * /local/domain/1/device/vbd/0/multi-queue-num-queues = "2"
+ * /local/domain/1/device/vbd/0/queue-0 = ""
+ * /local/domain/1/device/vbd/0/queue-0/ring-ref = "<ring-ref#0>"
+ * /local/domain/1/device/vbd/0/queue-0/event-channel = "<evtchn#0>"
+ * /local/domain/1/device/vbd/0/queue-1 = ""
+ * /local/domain/1/device/vbd/0/queue-1/ring-ref = "<ring-ref#1>"
+ * /local/domain/1/device/vbd/0/queue-1/event-channel = "<evtchn#1>"
+ *
+ * It is also possible to use multiple queues/rings together with
+ * feature multi-page ring buffer.
+ * For example, a frontend requests two queues/rings and the size of each ring
+ * buffer is two pages must write the following set of related keys:
+ *
+ * /local/domain/1/device/vbd/0/multi-queue-num-queues = "2"
+ * /local/domain/1/device/vbd/0/ring-page-order = "1"
+ * /local/domain/1/device/vbd/0/queue-0 = ""
+ * /local/domain/1/device/vbd/0/queue-0/ring-ref0 = "<ring-ref#0>"
+ * /local/domain/1/device/vbd/0/queue-0/ring-ref1 = "<ring-ref#1>"
+ * /local/domain/1/device/vbd/0/queue-0/event-channel = "<evtchn#0>"
+ * /local/domain/1/device/vbd/0/queue-1 = ""
+ * /local/domain/1/device/vbd/0/queue-1/ring-ref0 = "<ring-ref#2>"
+ * /local/domain/1/device/vbd/0/queue-1/ring-ref1 = "<ring-ref#3>"
+ * /local/domain/1/device/vbd/0/queue-1/event-channel = "<evtchn#1>"
+ *
+ */
+
 /*
  * REQUEST CODES.
  */