mm: use pagevec to rotate reclaimable page
While running some memory intensive load, system response deteriorated just
after swap-out started.
The cause of this problem is that when a PG_reclaim page is moved to the tail
of the inactive LRU list in rotate_reclaimable_page(), lru_lock spin lock is
acquired every page writeback . This deteriorates system performance and
makes interrupt hold off time longer when swap-out started.
Following patch solves this problem. I use pagevec in rotating reclaimable
pages to mitigate LRU spin lock contention and reduce interrupt hold off time.
I did a test that allocating and touching pages in multiple processes, and
pinging to the test machine in flooding mode to measure response under memory
intensive load.
The test result is:
-2.6.23-rc5
--- testmachine ping statistics ---
3000 packets transmitted, 3000 received, 0% packet loss, time 53222ms
rtt min/avg/max/mdev = 0.074/0.652/172.228/7.176 ms, pipe 11, ipg/ewma
17.746/0.092 ms
-2.6.23-rc5-patched
--- testmachine ping statistics ---
3000 packets transmitted, 3000 received, 0% packet loss, time 51924ms
rtt min/avg/max/mdev = 0.072/0.108/3.884/0.114 ms, pipe 2, ipg/ewma
17.314/0.091 ms
Max round-trip-time was improved.
The test machine spec is that 4CPU(3.16GHz, Hyper-threading enabled)
8GB memory , 8GB swap.
I did ping test again to observe performance deterioration caused by taking
a ref.
-2.6.23-rc6-with-modifiedpatch
--- testmachine ping statistics ---
3000 packets transmitted, 3000 received, 0% packet loss, time 53386ms
rtt min/avg/max/mdev = 0.074/0.110/4.716/0.147 ms, pipe 2, ipg/ewma 17.801/0.129 ms
The result for my original patch is as follows.
-2.6.23-rc5-with-originalpatch
--- testmachine ping statistics ---
3000 packets transmitted, 3000 received, 0% packet loss, time 51924ms
rtt min/avg/max/mdev = 0.072/0.108/3.884/0.114 ms, pipe 2, ipg/ewma 17.314/0.091 ms
The influence to response was small.
[akpm@linux-foundation.org: fix uninitalised var warning]
[hugh@veritas.com: fix locking]
[randy.dunlap@oracle.com: fix function declaration]
[hugh@veritas.com: fix BUG at include/linux/mm.h:220!]
[hugh@veritas.com: kill redundancy in rotate_reclaimable_page]
[hugh@veritas.com: move_tail_pages into lru_add_drain]
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/mm/swap.c b/mm/swap.c
index d93fa1e..d034b21 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -32,6 +32,10 @@
/* How many pages do we try to swap or page in/out together? */
int page_cluster;
+static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs) = { 0, };
+static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs) = { 0, };
+static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs) = { 0, };
+
/*
* This path almost never happens for VM activity - pages are normally
* freed via pagevecs. But it gets used by networking.
@@ -92,23 +96,47 @@
EXPORT_SYMBOL(put_pages_list);
/*
+ * pagevec_move_tail() must be called with IRQ disabled.
+ * Otherwise this may cause nasty races.
+ */
+static void pagevec_move_tail(struct pagevec *pvec)
+{
+ int i;
+ int pgmoved = 0;
+ struct zone *zone = NULL;
+
+ for (i = 0; i < pagevec_count(pvec); i++) {
+ struct page *page = pvec->pages[i];
+ struct zone *pagezone = page_zone(page);
+
+ if (pagezone != zone) {
+ if (zone)
+ spin_unlock(&zone->lru_lock);
+ zone = pagezone;
+ spin_lock(&zone->lru_lock);
+ }
+ if (PageLRU(page) && !PageActive(page)) {
+ list_move_tail(&page->lru, &zone->inactive_list);
+ pgmoved++;
+ }
+ }
+ if (zone)
+ spin_unlock(&zone->lru_lock);
+ __count_vm_events(PGROTATED, pgmoved);
+ release_pages(pvec->pages, pvec->nr, pvec->cold);
+ pagevec_reinit(pvec);
+}
+
+/*
* Writeback is about to end against a page which has been marked for immediate
* reclaim. If it still appears to be reclaimable, move it to the tail of the
- * inactive list. The page still has PageWriteback set, which will pin it.
- *
- * We don't expect many pages to come through here, so don't bother batching
- * things up.
- *
- * To avoid placing the page at the tail of the LRU while PG_writeback is still
- * set, this function will clear PG_writeback before performing the page
- * motion. Do that inside the lru lock because once PG_writeback is cleared
- * we may not touch the page.
+ * inactive list.
*
* Returns zero if it cleared PG_writeback.
*/
int rotate_reclaimable_page(struct page *page)
{
- struct zone *zone;
+ struct pagevec *pvec;
unsigned long flags;
if (PageLocked(page))
@@ -120,15 +148,16 @@
if (!PageLRU(page))
return 1;
- zone = page_zone(page);
- spin_lock_irqsave(&zone->lru_lock, flags);
- if (PageLRU(page) && !PageActive(page)) {
- list_move_tail(&page->lru, &zone->inactive_list);
- __count_vm_event(PGROTATED);
- }
+ page_cache_get(page);
+ local_irq_save(flags);
+ pvec = &__get_cpu_var(lru_rotate_pvecs);
+ if (!pagevec_add(pvec, page))
+ pagevec_move_tail(pvec);
+ local_irq_restore(flags);
+
if (!test_clear_page_writeback(page))
BUG();
- spin_unlock_irqrestore(&zone->lru_lock, flags);
+
return 0;
}
@@ -172,9 +201,6 @@
* lru_cache_add: add a page to the page lists
* @page: the page to add
*/
-static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs) = { 0, };
-static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs) = { 0, };
-
void fastcall lru_cache_add(struct page *page)
{
struct pagevec *pvec = &get_cpu_var(lru_add_pvecs);
@@ -195,21 +221,37 @@
put_cpu_var(lru_add_active_pvecs);
}
-static void __lru_add_drain(int cpu)
+/*
+ * Drain pages out of the cpu's pagevecs.
+ * Either "cpu" is the current CPU, and preemption has already been
+ * disabled; or "cpu" is being hot-unplugged, and is already dead.
+ */
+static void drain_cpu_pagevecs(int cpu)
{
- struct pagevec *pvec = &per_cpu(lru_add_pvecs, cpu);
+ struct pagevec *pvec;
- /* CPU is dead, so no locking needed. */
+ pvec = &per_cpu(lru_add_pvecs, cpu);
if (pagevec_count(pvec))
__pagevec_lru_add(pvec);
+
pvec = &per_cpu(lru_add_active_pvecs, cpu);
if (pagevec_count(pvec))
__pagevec_lru_add_active(pvec);
+
+ pvec = &per_cpu(lru_rotate_pvecs, cpu);
+ if (pagevec_count(pvec)) {
+ unsigned long flags;
+
+ /* No harm done if a racing interrupt already did this */
+ local_irq_save(flags);
+ pagevec_move_tail(pvec);
+ local_irq_restore(flags);
+ }
}
void lru_add_drain(void)
{
- __lru_add_drain(get_cpu());
+ drain_cpu_pagevecs(get_cpu());
put_cpu();
}
@@ -256,6 +298,7 @@
int i;
struct pagevec pages_to_free;
struct zone *zone = NULL;
+ unsigned long uninitialized_var(flags);
pagevec_init(&pages_to_free, cold);
for (i = 0; i < nr; i++) {
@@ -263,7 +306,7 @@
if (unlikely(PageCompound(page))) {
if (zone) {
- spin_unlock_irq(&zone->lru_lock);
+ spin_unlock_irqrestore(&zone->lru_lock, flags);
zone = NULL;
}
put_compound_page(page);
@@ -277,9 +320,10 @@
struct zone *pagezone = page_zone(page);
if (pagezone != zone) {
if (zone)
- spin_unlock_irq(&zone->lru_lock);
+ spin_unlock_irqrestore(&zone->lru_lock,
+ flags);
zone = pagezone;
- spin_lock_irq(&zone->lru_lock);
+ spin_lock_irqsave(&zone->lru_lock, flags);
}
VM_BUG_ON(!PageLRU(page));
__ClearPageLRU(page);
@@ -288,7 +332,7 @@
if (!pagevec_add(&pages_to_free, page)) {
if (zone) {
- spin_unlock_irq(&zone->lru_lock);
+ spin_unlock_irqrestore(&zone->lru_lock, flags);
zone = NULL;
}
__pagevec_free(&pages_to_free);
@@ -296,7 +340,7 @@
}
}
if (zone)
- spin_unlock_irq(&zone->lru_lock);
+ spin_unlock_irqrestore(&zone->lru_lock, flags);
pagevec_free(&pages_to_free);
}
@@ -489,7 +533,7 @@
if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
atomic_add(*committed, &vm_committed_space);
*committed = 0;
- __lru_add_drain((long)hcpu);
+ drain_cpu_pagevecs((long)hcpu);
}
return NOTIFY_OK;
}