Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-04 Thread Mel Gorman
On Tue, Dec 04, 2012 at 06:37:41AM -0800, Michel Lespinasse wrote: > On Mon, Dec 3, 2012 at 6:17 AM, Mel Gorman wrote: > > On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: > >> @@ -732,7 +732,7 @@ static int page_referenced_anon(struct p > >> struct anon_vma_chain *avc; > >>

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-04 Thread Michel Lespinasse
On Mon, Dec 3, 2012 at 6:17 AM, Mel Gorman wrote: > On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: >> @@ -732,7 +732,7 @@ static int page_referenced_anon(struct p >> struct anon_vma_chain *avc; >> int referenced = 0; >> >> - anon_vma = page_lock_anon_vma(page); >> +

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-04 Thread Michel Lespinasse
On Mon, Dec 3, 2012 at 6:17 AM, Mel Gorman mgor...@suse.de wrote: On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: @@ -732,7 +732,7 @@ static int page_referenced_anon(struct p struct anon_vma_chain *avc; int referenced = 0; - anon_vma = page_lock_anon_vma(page);

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-04 Thread Mel Gorman
On Tue, Dec 04, 2012 at 06:37:41AM -0800, Michel Lespinasse wrote: On Mon, Dec 3, 2012 at 6:17 AM, Mel Gorman mgor...@suse.de wrote: On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: @@ -732,7 +732,7 @@ static int page_referenced_anon(struct p struct anon_vma_chain *avc;

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-03 Thread Mel Gorman
On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: > > Note, with this optimization I went a farther than the > boundaries of the migration code - it seemed worthwile to do and > I've reviewed all the other users of page_lock_anon_vma() as > well and none seemed to be modifying the

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-03 Thread Mel Gorman
On Sat, Dec 01, 2012 at 09:15:38PM +0100, Ingo Molnar wrote: Note, with this optimization I went a farther than the boundaries of the migration code - it seemed worthwile to do and I've reviewed all the other users of page_lock_anon_vma() as well and none seemed to be modifying the list

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-01 Thread Rik van Riel
On 12/01/2012 03:15 PM, Ingo Molnar wrote: Index: linux/include/linux/rmap.h === --- linux.orig/include/linux/rmap.h +++ linux/include/linux/rmap.h @@ -128,6 +128,17 @@ static inline void anon_vma_unlock(struc

[PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-01 Thread Ingo Molnar
Note, with this optimization I went a farther than the boundaries of the migration code - it seemed worthwile to do and I've reviewed all the other users of page_lock_anon_vma() as well and none seemed to be modifying the list inside that lock. Please review this patch carefully - in

[PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-01 Thread Ingo Molnar
Note, with this optimization I went a farther than the boundaries of the migration code - it seemed worthwile to do and I've reviewed all the other users of page_lock_anon_vma() as well and none seemed to be modifying the list inside that lock. Please review this patch carefully - in

Re: [PATCH 2/2] mm/migration: Make rmap_walk_anon() and try_to_unmap_anon() more scalable

2012-12-01 Thread Rik van Riel
On 12/01/2012 03:15 PM, Ingo Molnar wrote: Index: linux/include/linux/rmap.h === --- linux.orig/include/linux/rmap.h +++ linux/include/linux/rmap.h @@ -128,6 +128,17 @@ static inline void anon_vma_unlock(struc