diff --git a/Documentation/ABI/testing/sysfs-driver-xen-blkback b/Documentation/ABI/testing/sysfs-driver-xen-blkback
index 8bb43b66eb55a1f8c0cbac1917301b2e96a55bbc..4e7babb3ba1fecc673018253d4adbb860e0cb8a4 100644
--- a/Documentation/ABI/testing/sysfs-driver-xen-blkback
+++ b/Documentation/ABI/testing/sysfs-driver-xen-blkback
@@ -15,3 +15,13 @@ Description:
blkback. If the frontend tries to use more than
max_persistent_grants, the LRU kicks in and starts
removing 5% of max_persistent_grants every 100ms.
+
+What: /sys/module/xen_blkback/parameters/persistent_grant_unused_seconds
+Date: August 2018
+KernelVersion: 4.19
+Contact: Roger Pau Monné <roger.pau@citrix.com>
+Description:
+ How long a persistent grant is allowed to remain
+ allocated without being in use. The time is in
+ seconds, 0 means indefinitely long.
+ The default is 60 seconds.
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index b55b245e805205a07266c9d745b96e4716ed3009..fd1e19f1a49f5803ca6dd5d939ec41ec592af077 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -83,6 +83,18 @@ 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");
+/*
+ * 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 xen_blkif_pgrant_timeout = 60;
+module_param_named(persistent_grant_unused_seconds, xen_blkif_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.
@@ -123,6 +135,13 @@ 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 bool persistent_gnt_timeout(struct persistent_gnt *persistent_gnt)
+{
+ return xen_blkif_pgrant_timeout &&
+ (jiffies - persistent_gnt->last_used >=
+ HZ * xen_blkif_pgrant_timeout);
+}
+
static inline int get_free_page(struct xen_blkif_ring *ring, struct page **page)
{
unsigned long flags;
@@ -236,8 +255,7 @@ 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 +279,11 @@ 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;
}
@@ -276,10 +294,10 @@ 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);
}
@@ -371,26 +389,26 @@ 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 < xen_blkif_max_pgrants ||
+ (ring->persistent_gnt_c == xen_blkif_max_pgrants &&
+ !ring->blkif->vbd.overflow_max_grants)) {
+ num_clean = 0;
+ } else {
+ 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);
+ 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 +419,7 @@ 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 +428,37 @@ static void purge_persistent_gnt(struct xen_blkif_ring *ring)
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;
diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h
index ecb35fe8ca8dbb54f36a85513a09064819acd67a..1d3002d773f7adb151dcf79adf18f919e461290a 100644
--- a/drivers/block/xen-blkback/common.h
+++ b/drivers/block/xen-blkback/common.h
@@ -233,16 +233,6 @@ struct xen_vbd {
struct backend_info;
-/* Number of available flags */
-#define PERSISTENT_GNT_FLAGS_SIZE 2
-/* This persistent grant is currently in use */
-#define PERSISTENT_GNT_ACTIVE 0
-/*
- * This persistent grant has been used, this flag is set when we remove the
- * PERSISTENT_GNT_ACTIVE, to know that this grant has been used recently.
- */
-#define PERSISTENT_GNT_WAS_ACTIVE 1
-
/* Number of requests that we can fit in a ring */
#define XEN_BLKIF_REQS_PER_PAGE 32
@@ -250,7 +240,8 @@ struct persistent_gnt {
struct page *page;
grant_ref_t gnt;
grant_handle_t handle;
- DECLARE_BITMAP(flags, PERSISTENT_GNT_FLAGS_SIZE);
+ unsigned long last_used;
+ bool active;
struct rb_node node;
struct list_head remove_node;
};
@@ -278,7 +269,6 @@ struct xen_blkif_ring {
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;
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 8986adab9bf58540bd8d2ab0744d317fba6c23a0..a71d817e900ddc07ff45d240f0ae290ff408b6cd 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -46,6 +46,7 @@
#include <linux/scatterlist.h>
#include <linux/bitmap.h>
#include <linux/list.h>
+#include <linux/workqueue.h>
#include <xen/xen.h>
#include <xen/xenbus.h>
@@ -121,6 +122,8 @@ static inline struct blkif_req *blkif_req(struct request *rq)
static DEFINE_MUTEX(blkfront_mutex);
static const struct block_device_operations xlvbd_block_fops;
+static struct delayed_work blkfront_work;
+static LIST_HEAD(info_list);
/*
* Maximum number of segments in indirect requests, the actual value used by
@@ -216,6 +219,7 @@ struct blkfront_info
/* Save uncomplete reqs and bios for migration. */
struct list_head requests;
struct bio_list bio_list;
+ struct list_head info_list;
};
static unsigned int nr_minors;
@@ -1759,6 +1763,12 @@ static int write_per_ring_nodes(struct xenbus_transaction xbt,
return err;
}
+static void free_info(struct blkfront_info *info)
+{
+ list_del(&info->info_list);
+ kfree(info);
+}
+
/* Common code used when first setting up, and when resuming. */
static int talk_to_blkback(struct xenbus_device *dev,
struct blkfront_info *info)
@@ -1880,7 +1890,10 @@ static int talk_to_blkback(struct xenbus_device *dev,
destroy_blkring:
blkif_free(info, 0);
- kfree(info);
+ mutex_lock(&blkfront_mutex);
+ free_info(info);
+ mutex_unlock(&blkfront_mutex);
+
dev_set_drvdata(&dev->dev, NULL);
return err;
@@ -1991,6 +2004,10 @@ static int blkfront_probe(struct xenbus_device *dev,
info->handle = simple_strtoul(strrchr(dev->nodename, '/')+1, NULL, 0);
dev_set_drvdata(&dev->dev, info);
+ mutex_lock(&blkfront_mutex);
+ list_add(&info->info_list, &info_list);
+ mutex_unlock(&blkfront_mutex);
+
return 0;
}
@@ -2301,6 +2318,12 @@ static void blkfront_gather_backend_features(struct blkfront_info *info)
if (indirect_segments <= BLKIF_MAX_SEGMENTS_PER_REQUEST)
indirect_segments = 0;
info->max_indirect_segments = indirect_segments;
+
+ if (info->feature_persistent) {
+ mutex_lock(&blkfront_mutex);
+ schedule_delayed_work(&blkfront_work, HZ * 10);
+ mutex_unlock(&blkfront_mutex);
+ }
}
/*
@@ -2482,7 +2505,9 @@ static int blkfront_remove(struct xenbus_device *xbdev)
mutex_unlock(&info->mutex);
if (!bdev) {
- kfree(info);
+ mutex_lock(&blkfront_mutex);
+ free_info(info);
+ mutex_unlock(&blkfront_mutex);
return 0;
}
@@ -2502,7 +2527,9 @@ static int blkfront_remove(struct xenbus_device *xbdev)
if (info && !bdev->bd_openers) {
xlvbd_release_gendisk(info);
disk->private_data = NULL;
- kfree(info);
+ mutex_lock(&blkfront_mutex);
+ free_info(info);
+ mutex_unlock(&blkfront_mutex);
}
mutex_unlock(&bdev->bd_mutex);
@@ -2585,7 +2612,7 @@ static void blkif_release(struct gendisk *disk, fmode_t mode)
dev_info(disk_to_dev(bdev->bd_disk), "releasing disk\n");
xlvbd_release_gendisk(info);
disk->private_data = NULL;
- kfree(info);
+ free_info(info);
}
out:
@@ -2618,6 +2645,61 @@ static struct xenbus_driver blkfront_driver = {
.is_ready = blkfront_is_ready,
};
+static void purge_persistent_grants(struct blkfront_info *info)
+{
+ unsigned int i;
+ unsigned long flags;
+
+ for (i = 0; i < info->nr_rings; i++) {
+ struct blkfront_ring_info *rinfo = &info->rinfo[i];
+ struct grant *gnt_list_entry, *tmp;
+
+ spin_lock_irqsave(&rinfo->ring_lock, flags);
+
+ if (rinfo->persistent_gnts_c == 0) {
+ spin_unlock_irqrestore(&rinfo->ring_lock, flags);
+ continue;
+ }
+
+ list_for_each_entry_safe(gnt_list_entry, tmp, &rinfo->grants,
+ node) {
+ if (gnt_list_entry->gref == GRANT_INVALID_REF ||
+ gnttab_query_foreign_access(gnt_list_entry->gref))
+ continue;
+
+ list_del(&gnt_list_entry->node);
+ gnttab_end_foreign_access(gnt_list_entry->gref, 0, 0UL);
+ rinfo->persistent_gnts_c--;
+ __free_page(gnt_list_entry->page);
+ kfree(gnt_list_entry);
+ }
+
+ spin_unlock_irqrestore(&rinfo->ring_lock, flags);
+ }
+}
+
+static void blkfront_delay_work(struct work_struct *work)
+{
+ struct blkfront_info *info;
+ bool need_schedule_work = false;
+
+ mutex_lock(&blkfront_mutex);
+
+ list_for_each_entry(info, &info_list, info_list) {
+ if (info->feature_persistent) {
+ need_schedule_work = true;
+ mutex_lock(&info->mutex);
+ purge_persistent_grants(info);
+ mutex_unlock(&info->mutex);
+ }
+ }
+
+ if (need_schedule_work)
+ schedule_delayed_work(&blkfront_work, HZ * 10);
+
+ mutex_unlock(&blkfront_mutex);
+}
+
static int __init xlblk_init(void)
{
int ret;
@@ -2626,6 +2708,15 @@ static int __init xlblk_init(void)
if (!xen_domain())
return -ENODEV;
+ if (!xen_has_pv_disk_devices())
+ return -ENODEV;
+
+ if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
+ pr_warn("xen_blk: can't get major %d with name %s\n",
+ XENVBD_MAJOR, DEV_NAME);
+ return -ENODEV;
+ }
+
if (xen_blkif_max_segments < BLKIF_MAX_SEGMENTS_PER_REQUEST)
xen_blkif_max_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
@@ -2641,14 +2732,7 @@ static int __init xlblk_init(void)
xen_blkif_max_queues = nr_cpus;
}
- if (!xen_has_pv_disk_devices())
- return -ENODEV;
-
- if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
- printk(KERN_WARNING "xen_blk: can't get major %d with name %s\n",
- XENVBD_MAJOR, DEV_NAME);
- return -ENODEV;
- }
+ INIT_DELAYED_WORK(&blkfront_work, blkfront_delay_work);
ret = xenbus_register_frontend(&blkfront_driver);
if (ret) {
@@ -2663,6 +2747,8 @@ module_init(xlblk_init);
static void __exit xlblk_exit(void)
{
+ cancel_delayed_work_sync(&blkfront_work);
+
xenbus_unregister_driver(&blkfront_driver);
unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
kfree(minors);