Skip to contentSkip to conversation

mm/swap.c: reduce lock contention in lru_cache_add

PATCH In review Alex Shi 6 participants 23 messages Nov 20, 2020 – Dec 25, 2020 Memory management

lore
Share & tools
lore

Alex Shi

author
#0 lore raw

PATCH 1/1 mm/swap.c: reduce lock contention in lru_cache_add

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

---
 mm/swap.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 49 insertions(+), 8 deletions(-)
mm/swap.c +49 -8
index 490553f3f9ef..c787b38bf9c0 100644
@@ -1009,24 +1009,65 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
 	trace_mm_lru_insertion(page, lru);
 }
 
+struct lruvecs {
+	struct list_head lists[PAGEVEC_SIZE];
+	struct lruvec *vecs[PAGEVEC_SIZE];
+};
+
+/* Sort pvec pages on their lruvec */
+int sort_page_lruvec(struct lruvecs *lruvecs, struct pagevec *pvec)
+{
+	int i, j, nr_lruvec;
+	struct page *page;
+	struct lruvec *lruvec = NULL;
+
+	lruvecs->vecs[0] = NULL;
+	for (i = nr_lruvec = 0; i < pagevec_count(pvec); i++) {
+		page = pvec->pages[i];
+		lruvec = mem_cgroup_page_lruvec(page, page_pgdat(page));
+
+		/* Try to find a same lruvec */
+		for (j = 0; j <= nr_lruvec; j++)
+			if (lruvec == lruvecs->vecs[j])
+				break;
+
+		/* A new lruvec */
+		if (j > nr_lruvec) {
+			INIT_LIST_HEAD(&lruvecs->lists[nr_lruvec]);
+			lruvecs->vecs[nr_lruvec] = lruvec;
+			j = nr_lruvec++;
+			lruvecs->vecs[nr_lruvec] = 0;
+		}
+
+		list_add_tail(&page->lru, &lruvecs->lists[j]);
+	}
+
+	return nr_lruvec;
+}
+
 /*
  * Add the passed pages to the LRU, then drop the caller's refcount
  * on them.  Reinitialises the caller's pagevec.
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);
 
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
+	for (i = 0; i < nr_lruvec; i++) {
+		spin_lock_irqsave(&lruvecs.vecs[i]->lru_lock, flags);
+		while (!list_empty(&lruvecs.lists[i])) {
+			page = lru_to_page(&lruvecs.lists[i]);
+			list_del(&page->lru);
+			__pagevec_lru_add_fn(page, lruvecs.vecs[i]);
+		}
+		spin_unlock_irqrestore(&lruvecs.vecs[i]->lru_lock, flags);
 	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
+
 	release_pages(pvec->pages, pvec->nr);
 	pagevec_reinit(pvec);
 }
Signature · 1 line
2.29.GIT
maintainer
#1 lore raw
Quoted 9 lines · On Fri, 20 Nov 2020 16:27:27 +0800 Alex Shi <[email protected]> wrote:

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

But what happens when all or most of the pages belong to the same lruvec? This sounds like the common case - won't it suffer?

author
#2 lore raw
Quoted 7 lines · Andrew Morton
Quoted 5 lines · On Fri, 20 Nov 2020 16:27:27 +0800 Alex Shi <[email protected]> wrote:

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

Quoted 8 lines
Quoted 4 lines

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

But what happens when all or most of the pages belong to the same lruvec? This sounds like the common case - won't it suffer?

Hi Andrew,

My testing show no regression on this situation, like original centos7, The most spending time is on lru_lock for lru sensitive case.

Thanks
Alex

#3 lore raw
In reply to Alex Shi
Quoted 9 lines · Alex Shi

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

Hm, once you sort the pages like this, it's a shame not to splice them instead of more list_del() + list_add() iterations. update_lru_size() could be also called once?

Quoted 92 lines
---
  mm/swap.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++--------
  1 file changed, 49 insertions(+), 8 deletions(-)
mm/swap.c +49 -8
index 490553f3f9ef..c787b38bf9c0 100644
@@ -1009,24 +1009,65 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
  	trace_mm_lru_insertion(page, lru);
  }
 
+struct lruvecs {
+	struct list_head lists[PAGEVEC_SIZE];
+	struct lruvec *vecs[PAGEVEC_SIZE];
+};
+
+/* Sort pvec pages on their lruvec */
+int sort_page_lruvec(struct lruvecs *lruvecs, struct pagevec *pvec)
+{
+	int i, j, nr_lruvec;
+	struct page *page;
+	struct lruvec *lruvec = NULL;
+
+	lruvecs->vecs[0] = NULL;
+	for (i = nr_lruvec = 0; i < pagevec_count(pvec); i++) {
+		page = pvec->pages[i];
+		lruvec = mem_cgroup_page_lruvec(page, page_pgdat(page));
+
+		/* Try to find a same lruvec */
+		for (j = 0; j <= nr_lruvec; j++)
+			if (lruvec == lruvecs->vecs[j])
+				break;
+
+		/* A new lruvec */
+		if (j > nr_lruvec) {
+			INIT_LIST_HEAD(&lruvecs->lists[nr_lruvec]);
+			lruvecs->vecs[nr_lruvec] = lruvec;
+			j = nr_lruvec++;
+			lruvecs->vecs[nr_lruvec] = 0;
+		}
+
+		list_add_tail(&page->lru, &lruvecs->lists[j]);
+	}
+
+	return nr_lruvec;
+}
+
  /*
   * Add the passed pages to the LRU, then drop the caller's refcount
   * on them.  Reinitialises the caller's pagevec.
   */
  void __pagevec_lru_add(struct pagevec *pvec)
  {
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
  	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);
 
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
+	for (i = 0; i < nr_lruvec; i++) {
+		spin_lock_irqsave(&lruvecs.vecs[i]->lru_lock, flags);
+		while (!list_empty(&lruvecs.lists[i])) {
+			page = lru_to_page(&lruvecs.lists[i]);
+			list_del(&page->lru);
+			__pagevec_lru_add_fn(page, lruvecs.vecs[i]);
+		}
+		spin_unlock_irqrestore(&lruvecs.vecs[i]->lru_lock, flags);
  	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
+
  	release_pages(pvec->pages, pvec->nr);
  	pagevec_reinit(pvec);
  }
author
#4 lore raw
Quoted 6 lines · Vlastimil Babka
Quoted 5 lines · Alex Shi

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

Quoted 6 lines
Quoted 4 lines

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

Hm, once you sort the pages like this, it's a shame not to splice them instead of more list_del() + list_add() iterations. update_lru_size() could be also called once?

Yes, looks it's a good idea to use splice instead of list_del/add, but pages may on different lru list in a same lruvec, and also may come from different zones. That could involve 5 cycles for different lists, and more for zones...

I give up the try.

#5 lore raw
Quoted 14 lines · Alex Shi
Quoted 11 lines · Vlastimil Babka
Quoted 9 lines · Alex Shi

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

Quoted 5 lines
Quoted 1 line

Hm, once you sort the pages like this, it's a shame not to splice them instead of more list_del() + list_add() iterations. update_lru_size() could be also called once?

Yes, looks it's a good idea to use splice instead of list_del/add, but pages may on different lru list in a same lruvec, and also may come from different zones. That could involve 5 cycles for different lists, and more for zones...

Hmm, zones wouldn't affect splicing (there's a per-node lru these days), but would affect accounting. And yeah, there are 5 lru lists, and we probably need to be under lru lock to stabilize where a page belongs, so pre-sorting without lock wouldn't be safe? Bummer.

Quoted 2 lines

I give up the try.

#6 lore raw
In reply to Alex Shi
Quoted 73 lines · Alex Shi

The current relock logical will change lru_lock when found a new lruvec, so if 2 memcgs are reading file or alloc page at same time, they could hold the lru_lock alternately, and wait for each other for fairness attribute of ticket spin lock.

This patch will sort that all lru_locks and only hold them once in above scenario. That could reduce fairness waiting for lock reget. Than, vm-scalability/case-lru-file-readtwice could get ~5% performance gain on my 2P*20core*HT machine.

---
 mm/swap.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 49 insertions(+), 8 deletions(-)
mm/swap.c +36 -0
index 490553f3f9ef..c787b38bf9c0 100644
@@ -1009,24 +1009,65 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
 	trace_mm_lru_insertion(page, lru);
 }
 
+struct lruvecs {
+	struct list_head lists[PAGEVEC_SIZE];
+	struct lruvec *vecs[PAGEVEC_SIZE];
+};
+
+/* Sort pvec pages on their lruvec */
+int sort_page_lruvec(struct lruvecs *lruvecs, struct pagevec *pvec)
+{
+	int i, j, nr_lruvec;
+	struct page *page;
+	struct lruvec *lruvec = NULL;
+
+	lruvecs->vecs[0] = NULL;
+	for (i = nr_lruvec = 0; i < pagevec_count(pvec); i++) {
+		page = pvec->pages[i];
+		lruvec = mem_cgroup_page_lruvec(page, page_pgdat(page));
+
+		/* Try to find a same lruvec */
+		for (j = 0; j <= nr_lruvec; j++)
+			if (lruvec == lruvecs->vecs[j])
+				break;
+
+		/* A new lruvec */
+		if (j > nr_lruvec) {
+			INIT_LIST_HEAD(&lruvecs->lists[nr_lruvec]);
+			lruvecs->vecs[nr_lruvec] = lruvec;
+			j = nr_lruvec++;
+			lruvecs->vecs[nr_lruvec] = 0;
+		}
+
+		list_add_tail(&page->lru, &lruvecs->lists[j]);
+	}
+
+	return nr_lruvec;
+}
+
 /*
  * Add the passed pages to the LRU, then drop the caller's refcount
  * on them.  Reinitialises the caller's pagevec.
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1059

Quoted 10 lines
mm/swap.c +4 -4
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);

Simply looping pvec multiple times (15 at most) for different lruvecs would be better because 1) it requires no extra data structures and therefore has better cache locality (theoretically faster) 2) it only loops once when !CONFIG_MEMCG and !CONFIG_NUMA and therefore has no impact on Android and Chrome OS.

Quoted 20 lines
mm/swap.c +9 -4
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
+	for (i = 0; i < nr_lruvec; i++) {
+		spin_lock_irqsave(&lruvecs.vecs[i]->lru_lock, flags);
+		while (!list_empty(&lruvecs.lists[i])) {
+			page = lru_to_page(&lruvecs.lists[i]);
+			list_del(&page->lru);
+			__pagevec_lru_add_fn(page, lruvecs.vecs[i]);
+		}
+		spin_unlock_irqrestore(&lruvecs.vecs[i]->lru_lock, flags);
 	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
+
 	release_pages(pvec->pages, pvec->nr);
 	pagevec_reinit(pvec);
 }
Signature · 1 line
2.29.GIT
author
#7 lore raw
Quoted 10 lines · Yu Zhao
Quoted 10 lines

*/
void __pagevec_lru_add(struct pagevec *pvec)
{

mm/swap.c +3 -2
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
Quoted 9 lines

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1059

Quoted 3 lines
mm/swap.c +1 -2
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);

Simply looping pvec multiple times (15 at most) for different lruvecs would be better because 1) it requires no extra data structures and therefore has better cache locality (theoretically faster) 2) it only loops once when !CONFIG_MEMCG and !CONFIG_NUMA and therefore has no impact on Android and Chrome OS.

With multiple memcgs, it do help a lot, I had gotten 30% grain on readtwice case. but yes, w/o MEMCG and NUMA, it's good to keep old behavior. So would you like has a proposal for this?

Thanks
Alex

#8 lore raw
Quoted 23 lines · Alex Shi
Quoted 10 lines · Yu Zhao
Quoted 10 lines

*/
void __pagevec_lru_add(struct pagevec *pvec)
{

(quoted diff) +3 -2
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
Quoted 9 lines

Review

Quoted 3 lines
(quoted diff) +1 -2
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);

Simply looping pvec multiple times (15 at most) for different lruvecs would be better because 1) it requires no extra data structures and therefore has better cache locality (theoretically faster) 2) it only loops once when !CONFIG_MEMCG and !CONFIG_NUMA and therefore has no impact on Android and Chrome OS.

Quoted 3 lines

With multiple memcgs, it do help a lot, I had gotten 30% grain on readtwice case. but yes, w/o MEMCG and NUMA, it's good to keep old behavior. So would you like has a proposal for this?

Oh, no, I'm not against your idea. I was saying it doesn't seem necessary to sort -- a nested loop would just do the job given pagevec is small.

mm/swap.c +13 -2
index cb3794e13b48..1d238edc2907 100644
@@ -996,15 +996,26 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {
-	int i;
+	int i, j;
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
 
 	for (i = 0; i < pagevec_count(pvec); i++) {
 		struct page *page = pvec->pages[i];
 
+		if (!page)
+			continue;
+
 		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
+
+		for (j = i; j < pagevec_count(pvec); j++) {
+			if (page_to_nid(pvec->pages[j]) != page_to_nid(page) ||
+			    page_memcg(pvec->pages[j]) != page_memcg(page))
+				continue;
+
+			__pagevec_lru_add_fn(pvec->pages[j], lruvec);
+			pvec->pages[j] = NULL;
+		}
 	}
 	if (lruvec)
 		unlock_page_lruvec_irqrestore(lruvec, flags);
author
#9 lore raw
Quoted 24 lines · Yu Zhao

Oh, no, I'm not against your idea. I was saying it doesn't seem necessary to sort -- a nested loop would just do the job given pagevec is small.

mm/swap.c +4 -1
index cb3794e13b48..1d238edc2907 100644
@@ -996,15 +996,26 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {
-	int i;
+	int i, j;
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
 
 	for (i = 0; i < pagevec_count(pvec); i++) {
 		struct page *page = pvec->pages[i];
 
+		if (!page)
+			continue;
+
 		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1018

Quoted 10 lines
mm/swap.c +9 -1
-		__pagevec_lru_add_fn(page, lruvec);
+
+		for (j = i; j < pagevec_count(pvec); j++) {
+			if (page_to_nid(pvec->pages[j]) != page_to_nid(page) ||
+			    page_memcg(pvec->pages[j]) != page_memcg(page))
+				continue;
+
+			__pagevec_lru_add_fn(pvec->pages[j], lruvec);
+			pvec->pages[j] = NULL;
+		}

Uh, I have to say your method is more better than mine.
And this could be reused for all relock_page_lruvec. I expect this could speed up lru performance a lot!

Quoted 3 lines
mm/swap.c +0 -0
 	}
 	if (lruvec)
 		unlock_page_lruvec_irqrestore(lruvec, flags);
#10 lore raw
In reply to Yu Zhao
Quoted 24 lines · Yu Zhao
Quoted 23 lines · Alex Shi
Quoted 19 lines · Yu Zhao

Review

Quoted 13 lines

*/
void __pagevec_lru_add(struct pagevec *pvec)
{

(quoted diff) +4 -4
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);

Simply looping pvec multiple times (15 at most) for different lruvecs would be better because 1) it requires no extra data structures and therefore has better cache locality (theoretically faster) 2) it only loops once when !CONFIG_MEMCG and !CONFIG_NUMA and therefore has no impact on Android and Chrome OS.

Quoted 6 lines
Quoted 2 lines

With multiple memcgs, it do help a lot, I had gotten 30% grain on readtwice
case. but yes, w/o MEMCG and NUMA, it's good to keep old behavior. So would you like has a proposal for this?

Oh, no, I'm not against your idea. I was saying it doesn't seem necessary to sort -- a nested loop would just do the job given pagevec is small.

Yeah that could work. The worst case doesn't look nice (O(n^2)) but it should be rather rare to have pages from really multiple memcgs/nodes?

Maybe with the following change? Avoids going through the nulls if we got lucky (or have !MEMCG !NUMA).

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 999

Quoted 10 lines
mm/swap.c +1 -1
index cb3794e13b48..1d238edc2907 100644
@@ -996,15 +996,26 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
   */
  void __pagevec_lru_add(struct pagevec *pvec)
  {
-	int i;
+	int i, j;

int i, j, n;

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1002

Quoted 3 lines
mm/swap.c +0 -0
  	struct lruvec *lruvec = NULL;
  	unsigned long flags = 0;
 

n = pagevec_count(pvec);

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1003

Quoted 1 line
mm/swap.c +0 -0
  	for (i = 0; i < pagevec_count(pvec); i++) {
   	for (i = 0; n; i++) {

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1007

Quoted 7 lines
mm/swap.c +3 -1
  		struct page *page = pvec->pages[i];
 
+		if (!page)
+			continue;
+
  		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
		n--;

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1017

Quoted 8 lines
mm/swap.c +8 -0
+
+		for (j = i; j < pagevec_count(pvec); j++) {
+			if (page_to_nid(pvec->pages[j]) != page_to_nid(page) ||
+			    page_memcg(pvec->pages[j]) != page_memcg(page))
+				continue;
+
+			__pagevec_lru_add_fn(pvec->pages[j], lruvec);
+			pvec->pages[j] = NULL;
			n--;
Quoted 5 lines
mm/swap.c +1 -0
+		}
  	}
  	if (lruvec)
  		unlock_page_lruvec_irqrestore(lruvec, flags);
#11 lore raw
Quoted 28 lines · Vlastimil Babka
Quoted 27 lines · Yu Zhao
Quoted 25 lines · Alex Shi
Quoted 10 lines · Yu Zhao
Quoted 10 lines

*/
void __pagevec_lru_add(struct pagevec *pvec)
{

(quoted diff) +3 -2
-	int i;
-	struct lruvec *lruvec = NULL;
+	int i, nr_lruvec;
 	unsigned long flags = 0;
+	struct page *page;
+	struct lruvecs lruvecs;
Quoted 9 lines

Review

Quoted 3 lines
(quoted diff) +1 -2
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	nr_lruvec = sort_page_lruvec(&lruvecs, pvec);

Simply looping pvec multiple times (15 at most) for different lruvecs would be better because 1) it requires no extra data structures and therefore has better cache locality (theoretically faster) 2) it only loops once when !CONFIG_MEMCG and !CONFIG_NUMA and therefore has no impact on Android and Chrome OS.

With multiple memcgs, it do help a lot, I had gotten 30% grain on readtwice
case. but yes, w/o MEMCG and NUMA, it's good to keep old behavior. So would you like has a proposal for this?

Quoted 6 lines
Quoted 3 lines

Oh, no, I'm not against your idea. I was saying it doesn't seem necessary to sort -- a nested loop would just do the job given pagevec is small.

Yeah that could work. The worst case doesn't look nice (O(n^2)) but it should be rather rare to have pages from really multiple memcgs/nodes?

However, Matthew wanted to increase pagevec size [1] and once 15^2 becomes 63^2, it starts to be somewhat more worrying.

[1] https://lore.kernel.org/linux-mm/[email protected]/

Quoted 51 lines

Maybe with the following change? Avoids going through the nulls if we got lucky (or have !MEMCG !NUMA).

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 999

Quoted 10 lines
mm/swap.c +1 -1
index cb3794e13b48..1d238edc2907 100644
@@ -996,15 +996,26 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
   */
  void __pagevec_lru_add(struct pagevec *pvec)
  {
-	int i;
+	int i, j;

int i, j, n;

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1002

Quoted 3 lines
mm/swap.c +0 -0
  	struct lruvec *lruvec = NULL;
  	unsigned long flags = 0;
 

n = pagevec_count(pvec);

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1003

Quoted 1 line
mm/swap.c +0 -0
  	for (i = 0; i < pagevec_count(pvec); i++) {
     	for (i = 0; n; i++) {

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1007

Quoted 7 lines
mm/swap.c +3 -1
  		struct page *page = pvec->pages[i];
 
+		if (!page)
+			continue;
+
  		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
		n--;

mm/swap.c static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec) · line 1017

Quoted 8 lines
mm/swap.c +8 -0
+
+		for (j = i; j < pagevec_count(pvec); j++) {
+			if (page_to_nid(pvec->pages[j]) != page_to_nid(page) ||
+			    page_memcg(pvec->pages[j]) != page_memcg(page))
+				continue;
+
+			__pagevec_lru_add_fn(pvec->pages[j], lruvec);
+			pvec->pages[j] = NULL;
			n--;
Quoted 5 lines
mm/swap.c +1 -0
+		}
  	}
  	if (lruvec)
  		unlock_page_lruvec_irqrestore(lruvec, flags);
author
#13 lore raw
Quoted 6 lines · Matthew Wilcox
Quoted 4 lines · Vlastimil Babka

However, Matthew wanted to increase pagevec size [1] and once 15^2 becomes 63^2, it starts to be somewhat more worrying.

[1] https://lore.kernel.org/linux-mm/[email protected]/

Quoted 8 lines

Well, Tim wanted it ;-)

I would suggest that rather than an insertion sort (or was it a bubble sort?), we should be using a Shell sort. It's ideal for these kinds of smallish arrays.

https://en.wikipedia.org/wiki/Shellsort

Uh, looks perfect good!. I gonna look into it. :)

Thanks!

author
#14 lore raw

PATCH 1/3 mm/swap.c: pre-sort pages in pagevec for pagevec_lru_move_fn

In reply to Matthew Wilcox

Pages in pagevec may have different lruvec, so we have to do relock in function pagevec_lru_move_fn(), but a relock may cause current cpu wait for long time on the same lock for spinlock fairness reason.

Before per memcg lru_lock, we have to bear the relock since the spinlock is the only way to serialize page's memcg/lruvec. Now TestClearPageLRU could be used to isolate pages exculsively, and stable the page's lruvec/memcg. So it gives us a chance to sort the page's lruvec before moving action in pagevec_lru_move_fn. Then we don't suffer from the spinlock's fairness wait.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 92 +++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 79 insertions(+), 13 deletions(-)
mm/swap.c +79 -13
index 490553f3f9ef..17d8990e5ca7 100644
@@ -201,29 +201,95 @@ int get_kernel_page(unsigned long start, int write, struct page **pages)
 }
 EXPORT_SYMBOL_GPL(get_kernel_page);
 
+/* Pratt's gaps for shell sort, https://en.wikipedia.org/wiki/Shellsort */
+static int gaps[] = { 6, 4, 3, 2, 1, 0};
+
+/* Shell sort pagevec[] on page's lruvec.*/
+static void shell_sort(struct pagevec *pvec, unsigned long *lvaddr)
+{
+	int g, i, j, n = pagevec_count(pvec);
+
+	for (g=0; gaps[g] > 0 && gaps[g] <= n/2; g++) {
+		int gap = gaps[g];
+
+		for (i = gap; i < n; i++) {
+			unsigned long tmp = lvaddr[i];
+			struct page *page = pvec->pages[i];
+
+			for (j = i - gap; j >= 0 && lvaddr[j] > tmp; j -= gap) {
+				lvaddr[j + gap] = lvaddr[j];
+				pvec->pages[j + gap] = pvec->pages[j];
+			}
+			lvaddr[j + gap] = tmp;
+			pvec->pages[j + gap] = page;
+		}
+	}
+}
+
+/* Get lru bit cleared page and their lruvec address, release the others */
+void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
+		unsigned long *lvaddr)
+{
+	int i, j;
+	struct pagevec busypv;
+
+	pagevec_init(&busypv);
+
+	for (i = 0, j = 0; i < pagevec_count(pvec); i++) {
+		struct page *page = pvec->pages[i];
+
+		pvec->pages[i] = NULL;
+
+		/* block memcg migration during page moving between lru */
+		if (!TestClearPageLRU(page)) {
+			pagevec_add(&busypv, page);
+			continue;
+		}
+		lvaddr[j++] = (unsigned long)
+				mem_cgroup_page_lruvec(page, page_pgdat(page));
+		pagevec_add(isopv, page);
+	}
+	pagevec_reinit(pvec);
+	if (pagevec_count(&busypv))
+		release_pages(busypv.pages, busypv.nr);
+
+	shell_sort(isopv, lvaddr);
+}
+
 static void pagevec_lru_move_fn(struct pagevec *pvec,
 	void (*move_fn)(struct page *page, struct lruvec *lruvec))
 {
-	int i;
+	int i, n;
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
+	unsigned long lvaddr[PAGEVEC_SIZE];
+	struct pagevec isopv;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	pagevec_init(&isopv);
 
-		/* block memcg migration during page moving between lru */
-		if (!TestClearPageLRU(page))
-			continue;
+	sort_isopv(pvec, &isopv, lvaddr);
 
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		(*move_fn)(page, lruvec);
+	n = pagevec_count(&isopv);
+	if (!n)
+		return;
 
-		SetPageLRU(page);
+	lruvec = (struct lruvec *)lvaddr[0];
+	spin_lock_irqsave(&lruvec->lru_lock, flags);
+
+	for (i = 0; i < n; i++) {
+		/* lock new lruvec if lruvec changes, we have sorted them */
+		if (lruvec != (struct lruvec *)lvaddr[i]) {
+			spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+			lruvec = (struct lruvec *)lvaddr[i];
+			spin_lock_irqsave(&lruvec->lru_lock, flags);
+		}
+
+		(*move_fn)(isopv.pages[i], lruvec);
+
+		SetPageLRU(isopv.pages[i]);
 	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
-	release_pages(pvec->pages, pvec->nr);
-	pagevec_reinit(pvec);
+	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+	release_pages(isopv.pages, isopv.nr);
 }
 
 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec)
Signature · 1 line
2.29.GIT
author
#15 lore raw

PATCH 2/3 mm/swap.c: bail out early for no memcg and no numa

If a system has memcg disabled and no numa node, like a embedded system, there is no needs to do the pagevec sort, since only just one lruvec in system. In this situation, we could skip the pagevec sorting.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
mm/swap.c +12 -7
index 17d8990e5ca7..814809845700 100644
@@ -264,12 +264,17 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 	unsigned long flags = 0;
 	unsigned long lvaddr[PAGEVEC_SIZE];
 	struct pagevec isopv;
+	struct pagevec *pv;
 
-	pagevec_init(&isopv);
-
-	sort_isopv(pvec, &isopv, lvaddr);
+	if (!mem_cgroup_disabled() || num_online_nodes() > 1) {
+		pagevec_init(&isopv);
+		sort_isopv(pvec, &isopv, lvaddr);
+		pv = &isopv;
+	} else {
+		pv = pvec;
+	}
 
-	n = pagevec_count(&isopv);
+	n = pagevec_count(pv);
 	if (!n)
 		return;
 
@@ -284,12 +289,12 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 			spin_lock_irqsave(&lruvec->lru_lock, flags);
 		}
 
-		(*move_fn)(isopv.pages[i], lruvec);
+		(*move_fn)(pv->pages[i], lruvec);
 
-		SetPageLRU(isopv.pages[i]);
+		SetPageLRU(pv->pages[i]);
 	}
 	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
-	release_pages(isopv.pages, isopv.nr);
+	release_pages(pv->pages, pv->nr);
 }
 
 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec)
Signature · 1 line
2.29.GIT
author
#16 lore raw

PATCH 3/3 mm/swap.c: extend the usage to pagevec_lru_add

In reply to Alex Shi

The only different for __pagevec_lru_add and other page moving between lru lists is page to add lru list has no need to do TestClearPageLRU and set the lru bit back. So we could combound them with a clear lru bit switch in sort function parameter.

Than all lru list operation functions could be united.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 30 +++++++++++-------------------
 1 file changed, 11 insertions(+), 19 deletions(-)
mm/swap.c +11 -19
index 814809845700..6a7920b2937f 100644
@@ -12,6 +12,7 @@
  * Started 18.12.91
  * Swap aging added 23.2.95, Stephen Tweedie.
  * Buffermem limits added 12.3.98, Rik van Riel.
+ * Pre-sort pagevec added 1.12.20, Alex Shi.
  */
 
 #include <linux/mm.h>
@@ -227,8 +228,8 @@ static void shell_sort(struct pagevec *pvec, unsigned long *lvaddr)
 }
 
 /* Get lru bit cleared page and their lruvec address, release the others */
-void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
-		unsigned long *lvaddr)
+static void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
+		unsigned long *lvaddr, bool clearlru)
 {
 	int i, j;
 	struct pagevec busypv;
@@ -241,7 +242,7 @@ void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
 		pvec->pages[i] = NULL;
 
 		/* block memcg migration during page moving between lru */
-		if (!TestClearPageLRU(page)) {
+		if (clearlru && !TestClearPageLRU(page)) {
 			pagevec_add(&busypv, page);
 			continue;
 		}
@@ -265,10 +266,13 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 	unsigned long lvaddr[PAGEVEC_SIZE];
 	struct pagevec isopv;
 	struct pagevec *pv;
+	bool clearlru;
+
+	clearlru = pvec != this_cpu_ptr(&lru_pvecs.lru_add);
 
 	if (!mem_cgroup_disabled() || num_online_nodes() > 1) {
 		pagevec_init(&isopv);
-		sort_isopv(pvec, &isopv, lvaddr);
+		sort_isopv(pvec, &isopv, lvaddr, clearlru);
 		pv = &isopv;
 	} else {
 		pv = pvec;
@@ -291,7 +295,8 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 
 		(*move_fn)(pv->pages[i], lruvec);
 
-		SetPageLRU(pv->pages[i]);
+		if (clearlru)
+			SetPageLRU(pv->pages[i]);
 	}
 	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
 	release_pages(pv->pages, pv->nr);
@@ -1086,20 +1091,7 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {
-	int i;
-	struct lruvec *lruvec = NULL;
-	unsigned long flags = 0;
-
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
-
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
-	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
-	release_pages(pvec->pages, pvec->nr);
-	pagevec_reinit(pvec);
+	pagevec_lru_move_fn(pvec, __pagevec_lru_add_fn);
 }
 
 /**
Signature · 1 line
2.29.GIT
#17 lore raw

Re: [PATCH 1/3] mm/swap.c: pre-sort pages in pagevec for pagevec_lru_move_fn

In reply to Alex Shi
Quoted 10 lines · Alex Shi

Pages in pagevec may have different lruvec, so we have to do relock in function pagevec_lru_move_fn(), but a relock may cause current cpu wait for long time on the same lock for spinlock fairness reason.

Before per memcg lru_lock, we have to bear the relock since the spinlock is the only way to serialize page's memcg/lruvec. Now TestClearPageLRU could be used to isolate pages exculsively, and stable the page's lruvec/memcg. So it gives us a chance to sort the page's lruvec before moving action in pagevec_lru_move_fn. Then we don't suffer from the spinlock's fairness wait.

Do you have any data to show any improvements from this?

Quoted 129 lines
Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 92 +++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 79 insertions(+), 13 deletions(-)
mm/swap.c +79 -13
index 490553f3f9ef..17d8990e5ca7 100644
@@ -201,29 +201,95 @@ int get_kernel_page(unsigned long start, int write, struct page **pages)
 }
 EXPORT_SYMBOL_GPL(get_kernel_page);
 
+/* Pratt's gaps for shell sort, https://en.wikipedia.org/wiki/Shellsort */
+static int gaps[] = { 6, 4, 3, 2, 1, 0};
+
+/* Shell sort pagevec[] on page's lruvec.*/
+static void shell_sort(struct pagevec *pvec, unsigned long *lvaddr)
+{
+	int g, i, j, n = pagevec_count(pvec);
+
+	for (g=0; gaps[g] > 0 && gaps[g] <= n/2; g++) {
+		int gap = gaps[g];
+
+		for (i = gap; i < n; i++) {
+			unsigned long tmp = lvaddr[i];
+			struct page *page = pvec->pages[i];
+
+			for (j = i - gap; j >= 0 && lvaddr[j] > tmp; j -= gap) {
+				lvaddr[j + gap] = lvaddr[j];
+				pvec->pages[j + gap] = pvec->pages[j];
+			}
+			lvaddr[j + gap] = tmp;
+			pvec->pages[j + gap] = page;
+		}
+	}
+}
+
+/* Get lru bit cleared page and their lruvec address, release the others */
+void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
+		unsigned long *lvaddr)
+{
+	int i, j;
+	struct pagevec busypv;
+
+	pagevec_init(&busypv);
+
+	for (i = 0, j = 0; i < pagevec_count(pvec); i++) {
+		struct page *page = pvec->pages[i];
+
+		pvec->pages[i] = NULL;
+
+		/* block memcg migration during page moving between lru */
+		if (!TestClearPageLRU(page)) {
+			pagevec_add(&busypv, page);
+			continue;
+		}
+		lvaddr[j++] = (unsigned long)
+				mem_cgroup_page_lruvec(page, page_pgdat(page));
+		pagevec_add(isopv, page);
+	}
+	pagevec_reinit(pvec);
+	if (pagevec_count(&busypv))
+		release_pages(busypv.pages, busypv.nr);
+
+	shell_sort(isopv, lvaddr);
+}
+
 static void pagevec_lru_move_fn(struct pagevec *pvec,
 	void (*move_fn)(struct page *page, struct lruvec *lruvec))
 {
-	int i;
+	int i, n;
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
+	unsigned long lvaddr[PAGEVEC_SIZE];
+	struct pagevec isopv;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	pagevec_init(&isopv);
 
-		/* block memcg migration during page moving between lru */
-		if (!TestClearPageLRU(page))
-			continue;
+	sort_isopv(pvec, &isopv, lvaddr);
 
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		(*move_fn)(page, lruvec);
+	n = pagevec_count(&isopv);
+	if (!n)
+		return;
 
-		SetPageLRU(page);
+	lruvec = (struct lruvec *)lvaddr[0];
+	spin_lock_irqsave(&lruvec->lru_lock, flags);
+
+	for (i = 0; i < n; i++) {
+		/* lock new lruvec if lruvec changes, we have sorted them */
+		if (lruvec != (struct lruvec *)lvaddr[i]) {
+			spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+			lruvec = (struct lruvec *)lvaddr[i];
+			spin_lock_irqsave(&lruvec->lru_lock, flags);
+		}
+
+		(*move_fn)(isopv.pages[i], lruvec);
+
+		SetPageLRU(isopv.pages[i]);
 	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
-	release_pages(pvec->pages, pvec->nr);
-	pagevec_reinit(pvec);
+	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+	release_pages(isopv.pages, isopv.nr);
 }
 
 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec)
Signature · 1 line
2.29.GIT
Signature · 2 lines
Michal Hocko
SUSE Labs
author
#18 lore raw

Re: [PATCH 1/3] mm/swap.c: pre-sort pages in pagevec for pagevec_lru_move_fn

Quoted 5 lines · Michal Hocko
Quoted 4 lines · Alex Shi

Pages in pagevec may have different lruvec, so we have to do relock in function pagevec_lru_move_fn(), but a relock may cause current cpu wait for long time on the same lock for spinlock fairness reason.

Quoted 8 lines
Quoted 6 lines

Before per memcg lru_lock, we have to bear the relock since the spinlock is the only way to serialize page's memcg/lruvec. Now TestClearPageLRU could be used to isolate pages exculsively, and stable the page's lruvec/memcg. So it gives us a chance to sort the page's lruvec before moving action in pagevec_lru_move_fn. Then we don't suffer from the spinlock's fairness wait.

Do you have any data to show any improvements from this?

Hi Michal,

Thanks for quick response.

Not yet. I am running for data. but according to the lru_add result, there should be a big gain for multiple memcgs scenario.

Also I don't except a quick accept, just send out the idea for comments when the thread is still warm. :)

Thanks
Alex

author
#19 lore raw

cover pre sort pages on lruvec in pagevec

In reply to Matthew Wilcox

This idea was tried on per memcg lru lock patchset v18, and had a good result, about 5%~20+% performance gain on lru lock busy benchmarks, like case-lru-file-readtwice.

But on the latest kernel, I can not reproduce the result on my box. Also I can not reproduce Tim's performance gain too on my box.

So I don't know if it's workable in some scenario, just sent out if someone has interesting...

Alex Shi (4):
mm/swap.c: pre-sort pages in pagevec for pagevec_lru_move_fn mm/swap.c: bail out early for no memcg and no numa mm/swap.c: extend the usage to pagevec_lru_add mm/swap.c: no sort if all page's lruvec are same

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
 mm/swap.c | 118 +++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 91 insertions(+), 27 deletions(-)
Signature · 1 line
2.29.GIT
author
#20 lore raw

RFC PATCH 1/4 mm/swap.c: pre-sort pages in pagevec for pagevec_lru_move_fn

Pages in pagevec may have different lruvec, so we have to do relock in function pagevec_lru_move_fn(), but a relock may cause current cpu wait for long time on the same lock for spinlock fairness reason.

Before per memcg lru_lock, we have to bear the relock since the spinlock is the only way to serialize page's memcg/lruvec. Now TestClearPageLRU could be used to isolate pages exculsively, and stable the page's lruvec/memcg. So it gives us a chance to sort the page's lruvec before moving action in pagevec_lru_move_fn. Then we don't suffer from the spinlock's fairness wait.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 92 +++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 79 insertions(+), 13 deletions(-)
mm/swap.c +79 -13
index c5363bdebe67..994641331bf7 100644
@@ -201,29 +201,95 @@ int get_kernel_page(unsigned long start, int write, struct page **pages)
 }
 EXPORT_SYMBOL_GPL(get_kernel_page);
 
+/* Pratt's gaps for shell sort, https://en.wikipedia.org/wiki/Shellsort */
+static int gaps[] = { 6, 4, 3, 2, 1, 0};
+
+/* Shell sort pagevec[] on page's lruvec.*/
+static void shell_sort(struct pagevec *pvec, unsigned long *lvaddr)
+{
+	int g, i, j, n = pagevec_count(pvec);
+
+	for (g=0; gaps[g] > 0 && gaps[g] <= n/2; g++) {
+		int gap = gaps[g];
+
+		for (i = gap; i < n; i++) {
+			unsigned long tmp = lvaddr[i];
+			struct page *page = pvec->pages[i];
+
+			for (j = i - gap; j >= 0 && lvaddr[j] > tmp; j -= gap) {
+				lvaddr[j + gap] = lvaddr[j];
+				pvec->pages[j + gap] = pvec->pages[j];
+			}
+			lvaddr[j + gap] = tmp;
+			pvec->pages[j + gap] = page;
+		}
+	}
+}
+
+/* Get lru bit cleared page and their lruvec address, release the others */
+void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
+		unsigned long *lvaddr)
+{
+	int i, j;
+	struct pagevec busypv;
+
+	pagevec_init(&busypv);
+
+	for (i = 0, j = 0; i < pagevec_count(pvec); i++) {
+		struct page *page = pvec->pages[i];
+
+		pvec->pages[i] = NULL;
+
+		/* block memcg migration during page moving between lru */
+		if (!TestClearPageLRU(page)) {
+			pagevec_add(&busypv, page);
+			continue;
+		}
+		lvaddr[j++] = (unsigned long)
+				mem_cgroup_page_lruvec(page, page_pgdat(page));
+		pagevec_add(isopv, page);
+	}
+	pagevec_reinit(pvec);
+	if (pagevec_count(&busypv))
+		release_pages(busypv.pages, busypv.nr);
+
+	shell_sort(isopv, lvaddr);
+}
+
 static void pagevec_lru_move_fn(struct pagevec *pvec,
 	void (*move_fn)(struct page *page, struct lruvec *lruvec))
 {
-	int i;
+	int i, n;
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
+	unsigned long lvaddr[PAGEVEC_SIZE];
+	struct pagevec isopv;
 
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
+	pagevec_init(&isopv);
 
-		/* block memcg migration during page moving between lru */
-		if (!TestClearPageLRU(page))
-			continue;
+	sort_isopv(pvec, &isopv, lvaddr);
 
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		(*move_fn)(page, lruvec);
+	n = pagevec_count(&isopv);
+	if (!n)
+		return;
 
-		SetPageLRU(page);
+	lruvec = (struct lruvec *)lvaddr[0];
+	spin_lock_irqsave(&lruvec->lru_lock, flags);
+
+	for (i = 0; i < n; i++) {
+		/* lock new lruvec if lruvec changes, we have sorted them */
+		if (lruvec != (struct lruvec *)lvaddr[i]) {
+			spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+			lruvec = (struct lruvec *)lvaddr[i];
+			spin_lock_irqsave(&lruvec->lru_lock, flags);
+		}
+
+		(*move_fn)(isopv.pages[i], lruvec);
+
+		SetPageLRU(isopv.pages[i]);
 	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
-	release_pages(pvec->pages, pvec->nr);
-	pagevec_reinit(pvec);
+	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
+	release_pages(isopv.pages, isopv.nr);
 }
 
 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec)
Signature · 1 line
2.29.GIT
author
#21 lore raw

RFC PATCH 2/4 mm/swap.c: bail out early for no memcg and no numa

In reply to Alex Shi

If a system has memcg disabled and no numa node, like a embedded system, there is no needs to do the pagevec sort, since only just one lruvec in system. In this situation, we could skip the pagevec sorting.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)
mm/swap.c +10 -9
index 994641331bf7..bb5300b7e321 100644
@@ -235,6 +235,7 @@ void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
 
 	pagevec_init(&busypv);
 
+
 	for (i = 0, j = 0; i < pagevec_count(pvec); i++) {
 		struct page *page = pvec->pages[i];
 
@@ -253,7 +254,8 @@ void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
 	if (pagevec_count(&busypv))
 		release_pages(busypv.pages, busypv.nr);
 
-	shell_sort(isopv, lvaddr);
+	if (!mem_cgroup_disabled() || num_online_nodes() > 1)
+		shell_sort(isopv, lvaddr);
 }
 
 static void pagevec_lru_move_fn(struct pagevec *pvec,
@@ -263,13 +265,12 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 	struct lruvec *lruvec = NULL;
 	unsigned long flags = 0;
 	unsigned long lvaddr[PAGEVEC_SIZE];
-	struct pagevec isopv;
-
-	pagevec_init(&isopv);
+	struct pagevec sortedpv;
 
-	sort_isopv(pvec, &isopv, lvaddr);
+	pagevec_init(&sortedpv);
+	sort_isopv(pvec, &sortedpv, lvaddr);
 
-	n = pagevec_count(&isopv);
+	n = pagevec_count(&sortedpv);
 	if (!n)
 		return;
 
@@ -284,12 +285,12 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 			spin_lock_irqsave(&lruvec->lru_lock, flags);
 		}
 
-		(*move_fn)(isopv.pages[i], lruvec);
+		(*move_fn)(sortedpv.pages[i], lruvec);
 
-		SetPageLRU(isopv.pages[i]);
+		SetPageLRU(sortedpv.pages[i]);
 	}
 	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
-	release_pages(isopv.pages, isopv.nr);
+	release_pages(sortedpv.pages, sortedpv.nr);
 }
 
 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec)
Signature · 1 line
2.29.GIT
author
#22 lore raw

RFC PATCH 3/4 mm/swap.c: extend the usage to pagevec_lru_add

In reply to Alex Shi

The only different for __pagevec_lru_add and other page moving between lru lists is page to add lru list has no need to do TestClearPageLRU and set the lru bit back. So we could combound them with a clear lru bit switch in sort function parameter.

Than all lru list operation functions could be united.

Cc Konstantin Khlebnikov <[email protected]>
Cc Hugh Dickins <[email protected]>
Cc Yu Zhao <[email protected]>
Cc Michal Hocko <[email protected]>
Cc Matthew Wilcox (Oracle) <[email protected]>
Cc Andrew Morton <[email protected]>
---
 mm/swap.c | 31 ++++++++++++-------------------
 1 file changed, 12 insertions(+), 19 deletions(-)
mm/swap.c +12 -19
index bb5300b7e321..9a2269e5099b 100644
@@ -12,6 +12,7 @@
  * Started 18.12.91
  * Swap aging added 23.2.95, Stephen Tweedie.
  * Buffermem limits added 12.3.98, Rik van Riel.
+ * Pre-sort pagevec added 12.1.20, Alex Shi.
  */
 
 #include <linux/mm.h>
@@ -227,8 +228,8 @@ static void shell_sort(struct pagevec *pvec, unsigned long *lvaddr)
 }
 
 /* Get lru bit cleared page and their lruvec address, release the others */
-void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
-		unsigned long *lvaddr)
+static void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
+		unsigned long *lvaddr, bool clearlru)
 {
 	int i, j;
 	struct pagevec busypv;
@@ -242,7 +243,7 @@ void sort_isopv(struct pagevec *pvec, struct pagevec *isopv,
 		pvec->pages[i] = NULL;
 
 		/* block memcg migration during page moving between lru */
-		if (!TestClearPageLRU(page)) {
+		if (clearlru && !TestClearPageLRU(page)) {
 			pagevec_add(&busypv, page);
 			continue;
 		}
@@ -266,9 +267,13 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 	unsigned long flags = 0;
 	unsigned long lvaddr[PAGEVEC_SIZE];
 	struct pagevec sortedpv;
+	bool clearlru;
+
+	/* don't clear lru bit for new page adding to lru */
+	clearlru = pvec != this_cpu_ptr(&lru_pvecs.lru_add);
 
 	pagevec_init(&sortedpv);
-	sort_isopv(pvec, &sortedpv, lvaddr);
+	sort_isopv(pvec, &sortedpv, lvaddr, clearlru);
 
 	n = pagevec_count(&sortedpv);
 	if (!n)
@@ -287,7 +292,8 @@ static void pagevec_lru_move_fn(struct pagevec *pvec,
 
 		(*move_fn)(sortedpv.pages[i], lruvec);
 
-		SetPageLRU(sortedpv.pages[i]);
+		if (clearlru)
+			SetPageLRU(sortedpv.pages[i]);
 	}
 	spin_unlock_irqrestore(&lruvec->lru_lock, flags);
 	release_pages(sortedpv.pages, sortedpv.nr);
@@ -1111,20 +1117,7 @@ static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec)
  */
 void __pagevec_lru_add(struct pagevec *pvec)
 {
-	int i;
-	struct lruvec *lruvec = NULL;
-	unsigned long flags = 0;
-
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
-
-		lruvec = relock_page_lruvec_irqsave(page, lruvec, &flags);
-		__pagevec_lru_add_fn(page, lruvec);
-	}
-	if (lruvec)
-		unlock_page_lruvec_irqrestore(lruvec, flags);
-	release_pages(pvec->pages, pvec->nr);
-	pagevec_reinit(pvec);
+	pagevec_lru_move_fn(pvec, __pagevec_lru_add_fn);
 }
 
 /**
Signature · 1 line
2.29.GIT