1 files changed, 256 insertions, 188 deletions

diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index fe7cd58c43d0..a7c2b04ab943 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -62,8 +62,8 @@
  * IO workloads.
  */
 
-static int xen_blkif_max_buffer_pages = 1024;
-module_param_named(max_buffer_pages, xen_blkif_max_buffer_pages, int, 0644);
+static int max_buffer_pages = 1024;
+module_param_named(max_buffer_pages, max_buffer_pages, int, 0644);
 MODULE_PARM_DESC(max_buffer_pages,
 "Maximum number of free pages to keep in each block backend buffer");
 
@@ -78,12 +78,24 @@ MODULE_PARM_DESC(max_buffer_pages,
  * algorithm.
  */
 
-static int xen_blkif_max_pgrants = 1056;
-module_param_named(max_persistent_grants, xen_blkif_max_pgrants, int, 0644);
+static int max_pgrants = 1056;
+module_param_named(max_persistent_grants, max_pgrants, int, 0644);
 MODULE_PARM_DESC(max_persistent_grants,
                  "Maximum number of grants to map persistently");
 
 /*
+ * How long a persistent grant is allowed to remain allocated without being in
+ * use. The time is in seconds, 0 means indefinitely long.
+ */
+
+static unsigned int pgrant_timeout = 60;
+module_param_named(persistent_grant_unused_seconds, pgrant_timeout,
+		   uint, 0644);
+MODULE_PARM_DESC(persistent_grant_unused_seconds,
+		 "Time in seconds an unused persistent grant is allowed to "
+		 "remain allocated. Default is 60, 0 means unlimited.");
+
+/*
  * Maximum number of rings/queues blkback supports, allow as many queues as there
  * are CPUs if user has not specified a value.
  */
@@ -98,7 +110,7 @@ MODULE_PARM_DESC(max_queues,
  * backend, 4KB page granularity is used.
  */
 unsigned int xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
-module_param_named(max_ring_page_order, xen_blkif_max_ring_order, int, S_IRUGO);
+module_param_named(max_ring_page_order, xen_blkif_max_ring_order, int, 0444);
 MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the shared ring");
 /*
  * The LRU mechanism to clean the lists of persistent grants needs to
@@ -120,70 +132,15 @@ module_param(log_stats, int, 0644);
 
 #define BLKBACK_INVALID_HANDLE (~0)
 
-/* 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_ring *ring, struct page **page)
-{
-	unsigned long 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(ring->free_pages_num == 0);
-	page[0] = list_first_entry(&ring->free_pages, struct page, lru);
-	list_del(&page[0]->lru);
-	ring->free_pages_num--;
-	spin_unlock_irqrestore(&ring->free_pages_lock, flags);
-
-	return 0;
-}
-
-static inline void put_free_pages(struct xen_blkif_ring *ring, struct page **page,
-                                  int num)
+static inline bool persistent_gnt_timeout(struct persistent_gnt *persistent_gnt)
 {
-	unsigned long flags;
-	int i;
-
-	spin_lock_irqsave(&ring->free_pages_lock, flags);
-	for (i = 0; i < num; i++)
-		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_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(&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);
-		ring->free_pages_num--;
-		if (++num_pages == NUM_BATCH_FREE_PAGES) {
-			spin_unlock_irqrestore(&ring->free_pages_lock, flags);
-			gnttab_free_pages(num_pages, page);
-			spin_lock_irqsave(&ring->free_pages_lock, flags);
-			num_pages = 0;
-		}
-	}
-	spin_unlock_irqrestore(&ring->free_pages_lock, flags);
-	if (num_pages != 0)
-		gnttab_free_pages(num_pages, page);
+	return pgrant_timeout && (jiffies - persistent_gnt->last_used >=
+			HZ * pgrant_timeout);
 }
 
 #define vaddr(page) ((unsigned long)pfn_to_kaddr(page_to_pfn(page)))
 
-static int do_block_io_op(struct xen_blkif_ring *ring);
+static int do_block_io_op(struct xen_blkif_ring *ring, unsigned int *eoi_flags);
 static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 				struct blkif_request *req,
 				struct pending_req *pending_req);
@@ -215,7 +172,7 @@ static int add_persistent_gnt(struct xen_blkif_ring *ring,
 	struct persistent_gnt *this;
 	struct xen_blkif *blkif = ring->blkif;
 
-	if (ring->persistent_gnt_c >= xen_blkif_max_pgrants) {
+	if (ring->persistent_gnt_c >= max_pgrants) {
 		if (!blkif->vbd.overflow_max_grants)
 			blkif->vbd.overflow_max_grants = 1;
 		return -EBUSY;
@@ -236,8 +193,7 @@ static int add_persistent_gnt(struct xen_blkif_ring *ring,
 		}
 	}
 
-	bitmap_zero(persistent_gnt->flags, PERSISTENT_GNT_FLAGS_SIZE);
-	set_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags);
+	persistent_gnt->active = true;
 	/* Add new node and rebalance tree. */
 	rb_link_node(&(persistent_gnt->node), parent, new);
 	rb_insert_color(&(persistent_gnt->node), &ring->persistent_gnts);
@@ -261,11 +217,11 @@ static struct persistent_gnt *get_persistent_gnt(struct xen_blkif_ring *ring,
 		else if (gref > data->gnt)
 			node = node->rb_right;
 		else {
-			if(test_bit(PERSISTENT_GNT_ACTIVE, data->flags)) {
+			if (data->active) {
 				pr_alert_ratelimited("requesting a grant already in use\n");
 				return NULL;
 			}
-			set_bit(PERSISTENT_GNT_ACTIVE, data->flags);
+			data->active = true;
 			atomic_inc(&ring->persistent_gnt_in_use);
 			return data;
 		}
@@ -276,16 +232,16 @@ static struct persistent_gnt *get_persistent_gnt(struct xen_blkif_ring *ring,
 static void put_persistent_gnt(struct xen_blkif_ring *ring,
                                struct persistent_gnt *persistent_gnt)
 {
-	if(!test_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags))
+	if (!persistent_gnt->active)
 		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);
+	persistent_gnt->last_used = jiffies;
+	persistent_gnt->active = false;
 	atomic_dec(&ring->persistent_gnt_in_use);
 }
 
-static void free_persistent_gnts(struct xen_blkif_ring *ring, struct rb_root *root,
-                                 unsigned int num)
+static void free_persistent_gnts(struct xen_blkif_ring *ring)
 {
+	struct rb_root *root = &ring->persistent_gnts;
 	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	struct persistent_gnt *persistent_gnt;
@@ -293,6 +249,9 @@ static void free_persistent_gnts(struct xen_blkif_ring *ring, struct rb_root *ro
 	int segs_to_unmap = 0;
 	struct gntab_unmap_queue_data unmap_data;
 
+	if (RB_EMPTY_ROOT(root))
+		return;
+
 	unmap_data.pages = pages;
 	unmap_data.unmap_ops = unmap;
 	unmap_data.kunmap_ops = NULL;
@@ -314,15 +273,18 @@ static void free_persistent_gnts(struct xen_blkif_ring *ring, struct rb_root *ro
 			unmap_data.count = segs_to_unmap;
 			BUG_ON(gnttab_unmap_refs_sync(&unmap_data));
 
-			put_free_pages(ring, pages, segs_to_unmap);
+			gnttab_page_cache_put(&ring->free_pages, pages,
+					      segs_to_unmap);
 			segs_to_unmap = 0;
 		}
 
 		rb_erase(&persistent_gnt->node, root);
 		kfree(persistent_gnt);
-		num--;
+		ring->persistent_gnt_c--;
 	}
-	BUG_ON(num != 0);
+
+	BUG_ON(!RB_EMPTY_ROOT(&ring->persistent_gnts));
+	BUG_ON(ring->persistent_gnt_c != 0);
 }
 
 void xen_blkbk_unmap_purged_grants(struct work_struct *work)
@@ -354,7 +316,8 @@ 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(ring, pages, segs_to_unmap);
+			gnttab_page_cache_put(&ring->free_pages, pages,
+					      segs_to_unmap);
 			segs_to_unmap = 0;
 		}
 		kfree(persistent_gnt);
@@ -362,7 +325,7 @@ 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(ring, pages, segs_to_unmap);
+		gnttab_page_cache_put(&ring->free_pages, pages, segs_to_unmap);
 	}
 }
 
@@ -371,26 +334,25 @@ static void purge_persistent_gnt(struct xen_blkif_ring *ring)
 	struct persistent_gnt *persistent_gnt;
 	struct rb_node *n;
 	unsigned int num_clean, total;
-	bool scan_used = false, clean_used = false;
+	bool scan_used = false;
 	struct rb_root *root;
 
-	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(&ring->persistent_purge_work)) {
 		pr_alert_ratelimited("Scheduled work from previous purge is still busy, cannot purge list\n");
 		goto out;
 	}
 
-	num_clean = (xen_blkif_max_pgrants / 100) * LRU_PERCENT_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 > (ring->persistent_gnt_c - atomic_read(&ring->persistent_gnt_in_use))))
-		goto out;
+	if (ring->persistent_gnt_c < max_pgrants ||
+	    (ring->persistent_gnt_c == max_pgrants &&
+	    !ring->blkif->vbd.overflow_max_grants)) {
+		num_clean = 0;
+	} else {
+		num_clean = (max_pgrants / 100) * LRU_PERCENT_CLEAN;
+		num_clean = ring->persistent_gnt_c - max_pgrants + num_clean;
+		num_clean = min(ring->persistent_gnt_c, num_clean);
+		pr_debug("Going to purge at least %u persistent grants\n",
+			 num_clean);
+	}
 
 	/*
 	 * At this point, we can assure that there will be no calls
@@ -401,9 +363,7 @@ static void purge_persistent_gnt(struct xen_blkif_ring *ring)
          * number of grants.
 	 */
 
-	total = num_clean;
-
-	pr_debug("Going to purge %u persistent grants\n", num_clean);
+	total = 0;
 
 	BUG_ON(!list_empty(&ring->persistent_purge_list));
 	root = &ring->persistent_gnts;
@@ -412,46 +372,37 @@ purge_list:
 		BUG_ON(persistent_gnt->handle ==
 			BLKBACK_INVALID_HANDLE);
 
-		if (clean_used) {
-			clear_bit(PERSISTENT_GNT_WAS_ACTIVE, persistent_gnt->flags);
+		if (persistent_gnt->active)
 			continue;
-		}
-
-		if (test_bit(PERSISTENT_GNT_ACTIVE, persistent_gnt->flags))
+		if (!scan_used && !persistent_gnt_timeout(persistent_gnt))
 			continue;
-		if (!scan_used &&
-		    (test_bit(PERSISTENT_GNT_WAS_ACTIVE, persistent_gnt->flags)))
+		if (scan_used && total >= num_clean)
 			continue;
 
 		rb_erase(&persistent_gnt->node, root);
 		list_add(&persistent_gnt->remove_node,
 			 &ring->persistent_purge_list);
-		if (--num_clean == 0)
-			goto finished;
+		total++;
 	}
 	/*
-	 * If we get here it means we also need to start cleaning
+	 * Check whether we also need to start cleaning
 	 * grants that were used since last purge in order to cope
 	 * with the requested num
 	 */
-	if (!scan_used && !clean_used) {
-		pr_debug("Still missing %u purged frames\n", num_clean);
+	if (!scan_used && total < num_clean) {
+		pr_debug("Still missing %u purged frames\n", num_clean - total);
 		scan_used = true;
 		goto purge_list;
 	}
-finished:
-	if (!clean_used) {
-		pr_debug("Finished scanning for grants to clean, removing used flag\n");
-		clean_used = true;
-		goto purge_list;
-	}
 
-	ring->persistent_gnt_c -= (total - num_clean);
-	ring->blkif->vbd.overflow_max_grants = 0;
+	if (total) {
+		ring->persistent_gnt_c -= total;
+		ring->blkif->vbd.overflow_max_grants = 0;
 
-	/* We can defer this work */
-	schedule_work(&ring->persistent_purge_work);
-	pr_debug("Purged %u/%u\n", (total - num_clean), total);
+		/* We can defer this work */
+		schedule_work(&ring->persistent_purge_work);
+		pr_debug("Purged %u/%u\n", num_clean, total);
+	}
 
 out:
 	return;
@@ -496,7 +447,7 @@ static void free_req(struct xen_blkif_ring *ring, struct pending_req *req)
  * Routines for managing virtual block devices (vbds).
  */
 static int xen_vbd_translate(struct phys_req *req, struct xen_blkif *blkif,
-			     int operation)
+			     enum req_op operation)
 {
 	struct xen_vbd *vbd = &blkif->vbd;
 	int rc = -EACCES;
@@ -514,7 +465,7 @@ static int xen_vbd_translate(struct phys_req *req, struct xen_blkif *blkif,
 	}
 
 	req->dev  = vbd->pdevice;
-	req->bdev = vbd->bdev;
+	req->bdev = file_bdev(vbd->bdev_file);
 	rc = 0;
 
  out:
@@ -592,9 +543,8 @@ static void print_stats(struct xen_blkif_ring *ring)
 		 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);
-	ring->st_print = jiffies + msecs_to_jiffies(10 * 1000);
+		 ring->persistent_gnt_c, max_pgrants);
+	ring->st_print = jiffies + secs_to_jiffies(10);
 	ring->st_rd_req = 0;
 	ring->st_wr_req = 0;
 	ring->st_oo_req = 0;
@@ -608,6 +558,8 @@ int xen_blkif_schedule(void *arg)
 	struct xen_vbd *vbd = &blkif->vbd;
 	unsigned long timeout;
 	int ret;
+	bool do_eoi;
+	unsigned int eoi_flags = XEN_EOI_FLAG_SPURIOUS;
 
 	set_freezable();
 	while (!kthread_should_stop()) {
@@ -632,16 +584,23 @@ int xen_blkif_schedule(void *arg)
 		if (timeout == 0)
 			goto purge_gnt_list;
 
+		do_eoi = ring->waiting_reqs;
+
 		ring->waiting_reqs = 0;
 		smp_mb(); /* clear flag *before* checking for work */
 
-		ret = do_block_io_op(ring);
+		ret = do_block_io_op(ring, &eoi_flags);
 		if (ret > 0)
 			ring->waiting_reqs = 1;
 		if (ret == -EACCES)
 			wait_event_interruptible(ring->shutdown_wq,
 						 kthread_should_stop());
 
+		if (do_eoi && !ring->waiting_reqs) {
+			xen_irq_lateeoi(ring->irq, eoi_flags);
+			eoi_flags |= XEN_EOI_FLAG_SPURIOUS;
+		}
+
 purge_gnt_list:
 		if (blkif->vbd.feature_gnt_persistent &&
 		    time_after(jiffies, ring->next_lru)) {
@@ -649,8 +608,12 @@ purge_gnt_list:
 			ring->next_lru = jiffies + msecs_to_jiffies(LRU_INTERVAL);
 		}
 
-		/* Shrink if we have more than xen_blkif_max_buffer_pages */
-		shrink_free_pagepool(ring, xen_blkif_max_buffer_pages);
+		/* Shrink the free pages pool if it is too large. */
+		if (time_before(jiffies, blkif->buffer_squeeze_end))
+			gnttab_page_cache_shrink(&ring->free_pages, 0);
+		else
+			gnttab_page_cache_shrink(&ring->free_pages,
+						 max_buffer_pages);
 
 		if (log_stats && time_after(jiffies, ring->st_print))
 			print_stats(ring);
@@ -673,15 +636,10 @@ purge_gnt_list:
 void xen_blkbk_free_caches(struct xen_blkif_ring *ring)
 {
 	/* Free all persistent grant pages */
-	if (!RB_EMPTY_ROOT(&ring->persistent_gnts))
-		free_persistent_gnts(ring, &ring->persistent_gnts,
-			ring->persistent_gnt_c);
-
-	BUG_ON(!RB_EMPTY_ROOT(&ring->persistent_gnts));
-	ring->persistent_gnt_c = 0;
+	free_persistent_gnts(ring);
 
 	/* Since we are shutting down remove all pages from the buffer */
-	shrink_free_pagepool(ring, 0 /* All */);
+	gnttab_page_cache_shrink(&ring->free_pages, 0 /* All */);
 }
 
 static unsigned int xen_blkbk_unmap_prepare(
@@ -705,9 +663,9 @@ static unsigned int xen_blkbk_unmap_prepare(
 				    GNTMAP_host_map, pages[i]->handle);
 		pages[i]->handle = BLKBACK_INVALID_HANDLE;
 		invcount++;
-       }
+	}
 
-       return invcount;
+	return invcount;
 }
 
 static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_queue_data *data)
@@ -720,7 +678,7 @@ static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_
 	   but is this the best way to deal with this? */
 	BUG_ON(result);
 
-	put_free_pages(ring, data->pages, data->count);
+	gnttab_page_cache_put(&ring->free_pages, data->pages, data->count);
 	make_response(ring, pending_req->id,
 		      pending_req->operation, pending_req->status);
 	free_req(ring, pending_req);
@@ -787,7 +745,8 @@ static void xen_blkbk_unmap(struct xen_blkif_ring *ring,
 		if (invcount) {
 			ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
 			BUG_ON(ret);
-			put_free_pages(ring, unmap_pages, invcount);
+			gnttab_page_cache_put(&ring->free_pages, unmap_pages,
+					      invcount);
 		}
 		pages += batch;
 		num -= batch;
@@ -834,8 +793,14 @@ again:
 			pages[i]->page = persistent_gnt->page;
 			pages[i]->persistent_gnt = persistent_gnt;
 		} else {
-			if (get_free_page(ring, &pages[i]->page))
-				goto out_of_memory;
+			if (gnttab_page_cache_get(&ring->free_pages,
+						  &pages[i]->page)) {
+				gnttab_page_cache_put(&ring->free_pages,
+						      pages_to_gnt,
+						      segs_to_map);
+				ret = -ENOMEM;
+				goto out;
+			}
 			addr = vaddr(pages[i]->page);
 			pages_to_gnt[segs_to_map] = pages[i]->page;
 			pages[i]->persistent_gnt = NULL;
@@ -851,10 +816,8 @@ again:
 			break;
 	}
 
-	if (segs_to_map) {
+	if (segs_to_map)
 		ret = gnttab_map_refs(map, NULL, pages_to_gnt, segs_to_map);
-		BUG_ON(ret);
-	}
 
 	/*
 	 * Now swizzle the MFN in our domain with the MFN from the other domain
@@ -867,9 +830,10 @@ 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(ring, &pages[seg_idx]->page, 1);
+				gnttab_page_cache_put(&ring->free_pages,
+						      &pages[seg_idx]->page, 1);
 				pages[seg_idx]->handle = BLKBACK_INVALID_HANDLE;
-				ret |= 1;
+				ret |= !ret;
 				goto next;
 			}
 			pages[seg_idx]->handle = map[new_map_idx].handle;
@@ -877,7 +841,7 @@ again:
 			continue;
 		}
 		if (use_persistent_gnts &&
-		    ring->persistent_gnt_c < xen_blkif_max_pgrants) {
+		    ring->persistent_gnt_c < max_pgrants) {
 			/*
 			 * We are using persistent grants, the grant is
 			 * not mapped but we might have room for it.
@@ -904,7 +868,7 @@ 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, ring->persistent_gnt_c,
-				 xen_blkif_max_pgrants);
+				 max_pgrants);
 			goto next;
 		}
 		if (use_persistent_gnts && !blkif->vbd.overflow_max_grants) {
@@ -921,15 +885,18 @@ next:
 	}
 	segs_to_map = 0;
 	last_map = map_until;
-	if (map_until != num)
+	if (!ret && map_until != num)
 		goto again;
 
-	return ret;
+out:
+	for (i = last_map; i < num; i++) {
+		/* Don't zap current batch's valid persistent grants. */
+		if (i >= map_until)
+			pages[i]->persistent_gnt = NULL;
+		pages[i]->handle = BLKBACK_INVALID_HANDLE;
+	}
 
-out_of_memory:
-	pr_alert("%s: out of memory\n", __func__);
-	put_free_pages(ring, pages_to_gnt, segs_to_map);
-	return -ENOMEM;
+	return ret;
 }
 
 static int xen_blkbk_map_seg(struct pending_req *pending_req)
@@ -964,7 +931,7 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req,
 	if (rc)
 		goto unmap;
 
-	for (n = 0, i = 0; n < nseg; n++) {
+	for (n = 0; n < nseg; n++) {
 		uint8_t first_sect, last_sect;
 
 		if ((n % SEGS_PER_INDIRECT_FRAME) == 0) {
@@ -1002,8 +969,7 @@ static int dispatch_discard_io(struct xen_blkif_ring *ring,
 	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 block_device *bdev = file_bdev(blkif->vbd.bdev_file);
 	struct phys_req preq;
 
 	xen_blkif_get(blkif);
@@ -1020,13 +986,15 @@ static int dispatch_discard_io(struct xen_blkif_ring *ring,
 	}
 	ring->st_ds_req++;
 
-	secure = (blkif->vbd.discard_secure &&
-		 (req->u.discard.flag & BLKIF_DISCARD_SECURE)) ?
-		 BLKDEV_DISCARD_SECURE : 0;
+	if (blkif->vbd.discard_secure &&
+	    (req->u.discard.flag & BLKIF_DISCARD_SECURE))
+		err = blkdev_issue_secure_erase(bdev,
+				req->u.discard.sector_number,
+				req->u.discard.nr_sectors, GFP_KERNEL);
+	else
+		err = blkdev_issue_discard(bdev, req->u.discard.sector_number,
+				req->u.discard.nr_sectors, GFP_KERNEL);
 
-	err = blkdev_issue_discard(bdev, req->u.discard.sector_number,
-				   req->u.discard.nr_sectors,
-				   GFP_KERNEL, secure);
 fail_response:
 	if (err == -EOPNOTSUPP) {
 		pr_debug("discard op failed, not supported\n");
@@ -1104,7 +1072,111 @@ static void end_block_io_op(struct bio *bio)
 	bio_put(bio);
 }
 
+static void blkif_get_x86_32_req(struct blkif_request *dst,
+				 const struct blkif_x86_32_request *src)
+{
+	unsigned int i, n;
+
+	dst->operation = READ_ONCE(src->operation);
+
+	switch (dst->operation) {
+	case BLKIF_OP_READ:
+	case BLKIF_OP_WRITE:
+	case BLKIF_OP_WRITE_BARRIER:
+	case BLKIF_OP_FLUSH_DISKCACHE:
+		dst->u.rw.nr_segments = READ_ONCE(src->u.rw.nr_segments);
+		dst->u.rw.handle = src->u.rw.handle;
+		dst->u.rw.id = src->u.rw.id;
+		dst->u.rw.sector_number = src->u.rw.sector_number;
+		n = min_t(unsigned int, BLKIF_MAX_SEGMENTS_PER_REQUEST,
+			  dst->u.rw.nr_segments);
+		for (i = 0; i < n; i++)
+			dst->u.rw.seg[i] = src->u.rw.seg[i];
+		break;
+
+	case BLKIF_OP_DISCARD:
+		dst->u.discard.flag = src->u.discard.flag;
+		dst->u.discard.id = src->u.discard.id;
+		dst->u.discard.sector_number = src->u.discard.sector_number;
+		dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
+		break;
+
+	case BLKIF_OP_INDIRECT:
+		dst->u.indirect.indirect_op = src->u.indirect.indirect_op;
+		dst->u.indirect.nr_segments =
+			READ_ONCE(src->u.indirect.nr_segments);
+		dst->u.indirect.handle = src->u.indirect.handle;
+		dst->u.indirect.id = src->u.indirect.id;
+		dst->u.indirect.sector_number = src->u.indirect.sector_number;
+		n = min(MAX_INDIRECT_PAGES,
+			INDIRECT_PAGES(dst->u.indirect.nr_segments));
+		for (i = 0; i < n; i++)
+			dst->u.indirect.indirect_grefs[i] =
+				src->u.indirect.indirect_grefs[i];
+		break;
+
+	default:
+		/*
+		 * Don't know how to translate this op. Only get the
+		 * ID so failure can be reported to the frontend.
+		 */
+		dst->u.other.id = src->u.other.id;
+		break;
+	}
+}
+
+static void blkif_get_x86_64_req(struct blkif_request *dst,
+				 const struct blkif_x86_64_request *src)
+{
+	unsigned int i, n;
+
+	dst->operation = READ_ONCE(src->operation);
+
+	switch (dst->operation) {
+	case BLKIF_OP_READ:
+	case BLKIF_OP_WRITE:
+	case BLKIF_OP_WRITE_BARRIER:
+	case BLKIF_OP_FLUSH_DISKCACHE:
+		dst->u.rw.nr_segments = READ_ONCE(src->u.rw.nr_segments);
+		dst->u.rw.handle = src->u.rw.handle;
+		dst->u.rw.id = src->u.rw.id;
+		dst->u.rw.sector_number = src->u.rw.sector_number;
+		n = min_t(unsigned int, BLKIF_MAX_SEGMENTS_PER_REQUEST,
+			  dst->u.rw.nr_segments);
+		for (i = 0; i < n; i++)
+			dst->u.rw.seg[i] = src->u.rw.seg[i];
+		break;
+
+	case BLKIF_OP_DISCARD:
+		dst->u.discard.flag = src->u.discard.flag;
+		dst->u.discard.id = src->u.discard.id;
+		dst->u.discard.sector_number = src->u.discard.sector_number;
+		dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
+		break;
 
+	case BLKIF_OP_INDIRECT:
+		dst->u.indirect.indirect_op = src->u.indirect.indirect_op;
+		dst->u.indirect.nr_segments =
+			READ_ONCE(src->u.indirect.nr_segments);
+		dst->u.indirect.handle = src->u.indirect.handle;
+		dst->u.indirect.id = src->u.indirect.id;
+		dst->u.indirect.sector_number = src->u.indirect.sector_number;
+		n = min(MAX_INDIRECT_PAGES,
+			INDIRECT_PAGES(dst->u.indirect.nr_segments));
+		for (i = 0; i < n; i++)
+			dst->u.indirect.indirect_grefs[i] =
+				src->u.indirect.indirect_grefs[i];
+		break;
+
+	default:
+		/*
+		 * Don't know how to translate this op. Only get the
+		 * ID so failure can be reported to the frontend.
+		 */
+		dst->u.other.id = src->u.other.id;
+		break;
+	}
+}
 
 /*
  * Function to copy the from the ring buffer the 'struct blkif_request'
@@ -1112,7 +1184,7 @@ 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_ring *ring)
+__do_block_io_op(struct xen_blkif_ring *ring, unsigned int *eoi_flags)
 {
 	union blkif_back_rings *blk_rings = &ring->blk_rings;
 	struct blkif_request req;
@@ -1135,6 +1207,9 @@ __do_block_io_op(struct xen_blkif_ring *ring)
 		if (RING_REQUEST_CONS_OVERFLOW(&blk_rings->common, rc))
 			break;
 
+		/* We've seen a request, so clear spurious eoi flag. */
+		*eoi_flags &= ~XEN_EOI_FLAG_SPURIOUS;
+
 		if (kthread_should_stop()) {
 			more_to_do = 1;
 			break;
@@ -1193,13 +1268,13 @@ done:
 }
 
 static int
-do_block_io_op(struct xen_blkif_ring *ring)
+do_block_io_op(struct xen_blkif_ring *ring, unsigned int *eoi_flags)
 {
 	union blkif_back_rings *blk_rings = &ring->blk_rings;
 	int more_to_do;
 
 	do {
-		more_to_do = __do_block_io_op(ring);
+		more_to_do = __do_block_io_op(ring, eoi_flags);
 		if (more_to_do)
 			break;
 
@@ -1222,8 +1297,8 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 	struct bio *bio = NULL;
 	struct bio **biolist = pending_req->biolist;
 	int i, nbio = 0;
-	int operation;
-	int operation_flags = 0;
+	enum req_op operation;
+	blk_opf_t operation_flags = 0;
 	struct blk_plug plug;
 	bool drain = false;
 	struct grant_page **pages = pending_req->segments;
@@ -1251,6 +1326,7 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 		break;
 	case BLKIF_OP_WRITE_BARRIER:
 		drain = true;
+		fallthrough;
 	case BLKIF_OP_FLUSH_DISKCACHE:
 		ring->st_f_req++;
 		operation = REQ_OP_WRITE;
@@ -1355,18 +1431,13 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 				     pages[i]->page,
 				     seg[i].nsec << 9,
 				     seg[i].offset) == 0)) {
-
-			int nr_iovecs = min_t(int, (nseg-i), BIO_MAX_PAGES);
-			bio = bio_alloc(GFP_KERNEL, nr_iovecs);
-			if (unlikely(bio == NULL))
-				goto fail_put_bio;
-
+			bio = bio_alloc(preq.bdev, bio_max_segs(nseg - i),
+					operation | operation_flags,
+					GFP_KERNEL);
 			biolist[nbio++] = bio;
-			bio->bi_bdev    = preq.bdev;
 			bio->bi_private = pending_req;
 			bio->bi_end_io  = end_block_io_op;
 			bio->bi_iter.bi_sector  = preq.sector_number;
-			bio_set_op_attrs(bio, operation, operation_flags);
 		}
 
 		preq.sector_number += seg[i].nsec;
@@ -1376,15 +1447,11 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 	if (!bio) {
 		BUG_ON(operation_flags != REQ_PREFLUSH);
 
-		bio = bio_alloc(GFP_KERNEL, 0);
-		if (unlikely(bio == NULL))
-			goto fail_put_bio;
-
+		bio = bio_alloc(preq.bdev, 0, operation | operation_flags,
+				GFP_KERNEL);
 		biolist[nbio++] = bio;
-		bio->bi_bdev    = preq.bdev;
 		bio->bi_private = pending_req;
 		bio->bi_end_io  = end_block_io_op;
-		bio_set_op_attrs(bio, operation, operation_flags);
 	}
 
 	atomic_set(&pending_req->pendcnt, nbio);
@@ -1412,14 +1479,6 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 	free_req(ring, pending_req);
 	msleep(1); /* back off a bit */
 	return -EIO;
-
- fail_put_bio:
-	for (i = 0; i < nbio; i++)
-		bio_put(biolist[i]);
-	atomic_set(&pending_req->pendcnt, 1);
-	__end_block_io_op(pending_req, BLK_STS_RESOURCE);
-	msleep(1); /* back off a bit */
-	return -EIO;
 }
 
 
@@ -1496,5 +1555,14 @@ static int __init xen_blkif_init(void)
 
 module_init(xen_blkif_init);
 
+static void __exit xen_blkif_fini(void)
+{
+	xen_blkif_xenbus_fini();
+	xen_blkif_interface_fini();
+}
+
+module_exit(xen_blkif_fini);
+
+MODULE_DESCRIPTION("Virtual block device back-end driver");
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_ALIAS("xen-backend:vbd");