xen/blkback: make pool of persistent grants and free pages per-queue
Make pool of persistent grants and free pages per-queue/ring instead of
per-device to get better scalability.
Test was done based on null_blk driver:
dom0: v4.2-rc8 16vcpus 10GB "modprobe null_blk"
domu: v4.2-rc8 16vcpus 10GB
[test]
rw=read
direct=1
ioengine=libaio
bs=4k
time_based
runtime=30
filename=/dev/xvdb
numjobs=16
iodepth=64
iodepth_batch=64
iodepth_batch_complete=64
group_reporting
Results:
iops1: After patch "xen/blkfront: make persistent grants per-queue".
iops2: After this patch.
Queues: 1 4 8 16
Iops orig(k): 810 1064 780 700
Iops1(k): 810 1230(~20%) 1024(~20%) 850(~20%)
Iops2(k): 810 1410(~35%) 1354(~75%) 1440(~100%)
With 4 queues after this commit we can get ~75% increase in IOPS, and
performance won't drop if increasing queue numbers.
Please find the respective chart in this link:
https://www.dropbox.com/s/agrcy2pbzbsvmwv/iops.png?dl=0
Signed-off-by: Bob Liu <bob.liu@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index 18b2777..a00d6c6 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -123,60 +123,60 @@
/* 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);
}
@@ -199,23 +199,29 @@
/*
- * pers_gnts_lock must be used around all the persistent grant helpers
- * because blkback may use multi-thread/queue for each backend.
+ * We don't need locking around the persistent grant helpers
+ * 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
+ * from interrupt context (by xen_blkbk_unmap), so we have to use atomic
+ * 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;
- BUG_ON(!spin_is_locked(&blkif->pers_gnts_lock));
- 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);
@@ -234,20 +240,19 @@
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;
- BUG_ON(!spin_is_locked(&blkif->pers_gnts_lock));
- node = blkif->persistent_gnts.rb_node;
+ node = ring->persistent_gnts.rb_node;
while (node) {
data = container_of(node, struct persistent_gnt, node);
@@ -261,25 +266,24 @@
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)
{
- BUG_ON(!spin_is_locked(&blkif->pers_gnts_lock));
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];
@@ -293,7 +297,6 @@
unmap_data.unmap_ops = unmap;
unmap_data.kunmap_ops = NULL;
- BUG_ON(!spin_is_locked(&blkif->pers_gnts_lock));
foreach_grant_safe(persistent_gnt, n, root, node) {
BUG_ON(persistent_gnt->handle ==
BLKBACK_INVALID_HANDLE);
@@ -311,7 +314,7 @@
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;
}
@@ -328,17 +331,15 @@
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;
- unsigned long flags;
unmap_data.pages = pages;
unmap_data.unmap_ops = unmap;
unmap_data.kunmap_ops = NULL;
- spin_lock_irqsave(&blkif->pers_gnts_lock, flags);
- 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);
@@ -353,45 +354,42 @@
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);
}
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, flags);
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;
unsigned int num_clean, total;
bool scan_used = false, clean_used = false;
struct rb_root *root;
- unsigned long flags;
- spin_lock_irqsave(&blkif->pers_gnts_lock, flags);
- if (blkif->persistent_gnt_c < xen_blkif_max_pgrants ||
- (blkif->persistent_gnt_c == xen_blkif_max_pgrants &&
- !blkif->vbd.overflow_max_grants)) {
+ 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");
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))))
+ (num_clean > (ring->persistent_gnt_c - atomic_read(&ring->persistent_gnt_in_use))))
goto out;
/*
@@ -407,8 +405,8 @@
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 ==
@@ -427,7 +425,7 @@
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;
}
@@ -448,18 +446,14 @@
goto purge_list;
}
- blkif->persistent_gnt_c -= (total - num_clean);
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, flags);
- 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);
- return;
out:
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, flags);
-
return;
}
@@ -591,14 +585,16 @@
* SCHEDULER FUNCTIONS
*/
-static void print_stats(struct xen_blkif *blkif)
+static void print_stats(struct xen_blkif_ring *ring)
{
+ struct xen_blkif *blkif = ring->blkif;
+
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,
+ ring->persistent_gnt_c,
xen_blkif_max_pgrants);
blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000);
blkif->st_rd_req = 0;
@@ -651,23 +647,23 @@
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->blkif->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);
ring->xenblkd = NULL;
xen_blkif_put(blkif);
@@ -680,21 +676,16 @@
*/
void xen_blkbk_free_caches(struct xen_blkif_ring *ring)
{
- struct xen_blkif *blkif = ring->blkif;
- unsigned long flags;
-
/* Free all persistent grant pages */
- spin_lock_irqsave(&blkif->pers_gnts_lock, flags);
- 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;
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, flags);
+ 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(
@@ -705,13 +696,10 @@
struct page **unmap_pages)
{
unsigned int i, invcount = 0;
- unsigned long flags;
for (i = 0; i < num; i++) {
if (pages[i]->persistent_gnt != NULL) {
- spin_lock_irqsave(&ring->blkif->pers_gnts_lock, flags);
- put_persistent_gnt(ring->blkif, pages[i]->persistent_gnt);
- spin_unlock_irqrestore(&ring->blkif->pers_gnts_lock, flags);
+ put_persistent_gnt(ring, pages[i]->persistent_gnt);
continue;
}
if (pages[i]->handle == BLKBACK_INVALID_HANDLE)
@@ -736,7 +724,7 @@
but is this the best way to deal with this? */
BUG_ON(result);
- put_free_pages(blkif, data->pages, data->count);
+ put_free_pages(ring, data->pages, data->count);
make_response(ring, pending_req->id,
pending_req->operation, pending_req->status);
free_req(ring, pending_req);
@@ -803,7 +791,7 @@
if (invcount) {
ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
BUG_ON(ret);
- put_free_pages(ring->blkif, unmap_pages, invcount);
+ put_free_pages(ring, unmap_pages, invcount);
}
pages += batch;
num -= batch;
@@ -824,7 +812,6 @@
int last_map = 0, map_until = 0;
int use_persistent_gnts;
struct xen_blkif *blkif = ring->blkif;
- unsigned long irq_flags;
use_persistent_gnts = (blkif->vbd.feature_gnt_persistent);
@@ -838,11 +825,9 @@
uint32_t flags;
if (use_persistent_gnts) {
- spin_lock_irqsave(&blkif->pers_gnts_lock, irq_flags);
persistent_gnt = get_persistent_gnt(
- blkif,
+ ring,
pages[i]->gref);
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, irq_flags);
}
if (persistent_gnt) {
@@ -853,7 +838,7 @@
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;
@@ -886,7 +871,7 @@
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;
@@ -896,7 +881,7 @@
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.
@@ -914,19 +899,16 @@
persistent_gnt->gnt = map[new_map_idx].ref;
persistent_gnt->handle = map[new_map_idx].handle;
persistent_gnt->page = pages[seg_idx]->page;
- spin_lock_irqsave(&blkif->pers_gnts_lock, irq_flags);
- if (add_persistent_gnt(blkif,
+ if (add_persistent_gnt(ring,
persistent_gnt)) {
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, irq_flags);
kfree(persistent_gnt);
persistent_gnt = NULL;
goto next;
}
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);
- spin_unlock_irqrestore(&blkif->pers_gnts_lock, irq_flags);
goto next;
}
if (use_persistent_gnts && !blkif->vbd.overflow_max_grants) {
@@ -950,7 +932,7 @@
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;
}
