linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2] mm/page_alloc: speeding up the iteration of max_order
@ 2020-12-04 12:56 Muchun Song
  2020-12-04 15:28 ` Vlastimil Babka
  0 siblings, 1 reply; 3+ messages in thread
From: Muchun Song @ 2020-12-04 12:56 UTC (permalink / raw)
  To: akpm, vbabka; +Cc: linux-mm, linux-kernel, Muchun Song

When we free a page whose order is very close to MAX_ORDER and greater
than pageblock_order, it wastes some CPU cycles to increase max_order
to MAX_ORDER one by one and check the pageblock migratetype of that page
repeatedly especially when MAX_ORDER is much larger than pageblock_order.

Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
Changes in v2:
 - Rework the code suggested by Vlastimil. Thanks.

 mm/page_alloc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f91df593bf71..56e603eea1dd 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1002,7 +1002,7 @@ static inline void __free_one_page(struct page *page,
 	struct page *buddy;
 	bool to_tail;
 
-	max_order = min_t(unsigned int, MAX_ORDER, pageblock_order + 1);
+	max_order = min_t(unsigned int, MAX_ORDER - 1, pageblock_order);
 
 	VM_BUG_ON(!zone_is_initialized(zone));
 	VM_BUG_ON_PAGE(page->flags & PAGE_FLAGS_CHECK_AT_PREP, page);
@@ -1015,7 +1015,7 @@ static inline void __free_one_page(struct page *page,
 	VM_BUG_ON_PAGE(bad_range(zone, page), page);
 
 continue_merging:
-	while (order < max_order - 1) {
+	while (order < max_order) {
 		if (compaction_capture(capc, page, order, migratetype)) {
 			__mod_zone_freepage_state(zone, -(1 << order),
 								migratetype);
@@ -1041,7 +1041,7 @@ static inline void __free_one_page(struct page *page,
 		pfn = combined_pfn;
 		order++;
 	}
-	if (max_order < MAX_ORDER) {
+	if (order < MAX_ORDER - 1) {
 		/* If we are here, it means order is >= pageblock_order.
 		 * We want to prevent merge between freepages on isolate
 		 * pageblock and normal pageblock. Without this, pageblock
@@ -1062,7 +1062,7 @@ static inline void __free_one_page(struct page *page,
 						is_migrate_isolate(buddy_mt)))
 				goto done_merging;
 		}
-		max_order++;
+		max_order = order + 1;
 		goto continue_merging;
 	}
 
-- 
2.11.0



^ permalink raw reply related	[flat|nested] 3+ messages in thread
* Re: [PATCH v2] mm/page_alloc: speeding up the iteration of max_order
  2020-12-04 12:56 [PATCH v2] mm/page_alloc: speeding up the iteration of max_order Muchun Song
@ 2020-12-04 15:28 ` Vlastimil Babka
  2020-12-04 15:45   ` [External] " Muchun Song
  0 siblings, 1 reply; 3+ messages in thread
From: Vlastimil Babka @ 2020-12-04 15:28 UTC (permalink / raw)
  To: Muchun Song, akpm; +Cc: linux-mm, linux-kernel

On 12/4/20 1:56 PM, Muchun Song wrote:
> When we free a page whose order is very close to MAX_ORDER and greater
> than pageblock_order, it wastes some CPU cycles to increase max_order
> to MAX_ORDER one by one and check the pageblock migratetype of that page
> repeatedly especially when MAX_ORDER is much larger than pageblock_order.

I would add:

We also should not be checking migratetype of buddy when "order == MAX_ORDER -
1" as the buddy pfn may be invalid, so adjust the condition. With the new check,
we don't need the max_order check anymore, so we replace it.

Also adjust max_order initialization so that it's lower by one than previously,
which makes the code hopefully more clear.

> Signed-off-by: Muchun Song <songmuchun@bytedance.com>

Fixes: d9dddbf55667 ("mm/page_alloc: prevent merging between isolated and other
pageblocks")
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Thanks!

> ---
> Changes in v2:
>  - Rework the code suggested by Vlastimil. Thanks.
> 
>  mm/page_alloc.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index f91df593bf71..56e603eea1dd 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1002,7 +1002,7 @@ static inline void __free_one_page(struct page *page,
>  	struct page *buddy;
>  	bool to_tail;
>  
> -	max_order = min_t(unsigned int, MAX_ORDER, pageblock_order + 1);
> +	max_order = min_t(unsigned int, MAX_ORDER - 1, pageblock_order);
>  
>  	VM_BUG_ON(!zone_is_initialized(zone));
>  	VM_BUG_ON_PAGE(page->flags & PAGE_FLAGS_CHECK_AT_PREP, page);
> @@ -1015,7 +1015,7 @@ static inline void __free_one_page(struct page *page,
>  	VM_BUG_ON_PAGE(bad_range(zone, page), page);
>  
>  continue_merging:
> -	while (order < max_order - 1) {
> +	while (order < max_order) {
>  		if (compaction_capture(capc, page, order, migratetype)) {
>  			__mod_zone_freepage_state(zone, -(1 << order),
>  								migratetype);
> @@ -1041,7 +1041,7 @@ static inline void __free_one_page(struct page *page,
>  		pfn = combined_pfn;
>  		order++;
>  	}
> -	if (max_order < MAX_ORDER) {
> +	if (order < MAX_ORDER - 1) {
>  		/* If we are here, it means order is >= pageblock_order.
>  		 * We want to prevent merge between freepages on isolate
>  		 * pageblock and normal pageblock. Without this, pageblock
> @@ -1062,7 +1062,7 @@ static inline void __free_one_page(struct page *page,
>  						is_migrate_isolate(buddy_mt)))
>  				goto done_merging;
>  		}
> -		max_order++;
> +		max_order = order + 1;
>  		goto continue_merging;
>  	}
>  
> 



^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: [External] Re: [PATCH v2] mm/page_alloc: speeding up the iteration of max_order
  2020-12-04 15:28 ` Vlastimil Babka
@ 2020-12-04 15:45   ` Muchun Song
  0 siblings, 0 replies; 3+ messages in thread
From: Muchun Song @ 2020-12-04 15:45 UTC (permalink / raw)
  To: Vlastimil Babka; +Cc: Andrew Morton, Linux Memory Management List, LKML

On Fri, Dec 4, 2020 at 11:28 PM Vlastimil Babka <vbabka@suse.cz> wrote:
>
> On 12/4/20 1:56 PM, Muchun Song wrote:
> > When we free a page whose order is very close to MAX_ORDER and greater
> > than pageblock_order, it wastes some CPU cycles to increase max_order
> > to MAX_ORDER one by one and check the pageblock migratetype of that page
> > repeatedly especially when MAX_ORDER is much larger than pageblock_order.
>
> I would add:
>
> We also should not be checking migratetype of buddy when "order == MAX_ORDER -
> 1" as the buddy pfn may be invalid, so adjust the condition. With the new check,
> we don't need the max_order check anymore, so we replace it.
>
> Also adjust max_order initialization so that it's lower by one than previously,
> which makes the code hopefully more clear.

Got it. Thanks.

>
> > Signed-off-by: Muchun Song <songmuchun@bytedance.com>
>
> Fixes: d9dddbf55667 ("mm/page_alloc: prevent merging between isolated and other
> pageblocks")
> Acked-by: Vlastimil Babka <vbabka@suse.cz>
> Thanks!
>
> > ---
> > Changes in v2:
> >  - Rework the code suggested by Vlastimil. Thanks.
> >
> >  mm/page_alloc.c | 8 ++++----
> >  1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index f91df593bf71..56e603eea1dd 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -1002,7 +1002,7 @@ static inline void __free_one_page(struct page *page,
> >       struct page *buddy;
> >       bool to_tail;
> >
> > -     max_order = min_t(unsigned int, MAX_ORDER, pageblock_order + 1);
> > +     max_order = min_t(unsigned int, MAX_ORDER - 1, pageblock_order);
> >
> >       VM_BUG_ON(!zone_is_initialized(zone));
> >       VM_BUG_ON_PAGE(page->flags & PAGE_FLAGS_CHECK_AT_PREP, page);
> > @@ -1015,7 +1015,7 @@ static inline void __free_one_page(struct page *page,
> >       VM_BUG_ON_PAGE(bad_range(zone, page), page);
> >
> >  continue_merging:
> > -     while (order < max_order - 1) {
> > +     while (order < max_order) {
> >               if (compaction_capture(capc, page, order, migratetype)) {
> >                       __mod_zone_freepage_state(zone, -(1 << order),
> >                                                               migratetype);
> > @@ -1041,7 +1041,7 @@ static inline void __free_one_page(struct page *page,
> >               pfn = combined_pfn;
> >               order++;
> >       }
> > -     if (max_order < MAX_ORDER) {
> > +     if (order < MAX_ORDER - 1) {
> >               /* If we are here, it means order is >= pageblock_order.
> >                * We want to prevent merge between freepages on isolate
> >                * pageblock and normal pageblock. Without this, pageblock
> > @@ -1062,7 +1062,7 @@ static inline void __free_one_page(struct page *page,
> >                                               is_migrate_isolate(buddy_mt)))
> >                               goto done_merging;
> >               }
> > -             max_order++;
> > +             max_order = order + 1;
> >               goto continue_merging;
> >       }
> >
> >
>


-- 
Yours,
Muchun


^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-12-04 15:46 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-12-04 12:56 [PATCH v2] mm/page_alloc: speeding up the iteration of max_order Muchun Song
2020-12-04 15:28 ` Vlastimil Babka
2020-12-04 15:45   ` [External] " Muchun Song

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).