* [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-04-15 7:32 ` Brice Goglin
0 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-04-15 7:32 UTC (permalink / raw)
To: Christoph Lameter; +Cc: linux-mm, LKML
migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
throughput by breaking it into chunks, but it also made migrate_prep()
be called once per chunk (every 128pages or so) instead of once per
move_pages().
This patch reverts to calling migrate_prep() only once per chunk
as we did before 2.6.29.
It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
mm: move migrate_prep out from under mmap_sem
This improves migration throughput on the above machine from 600MB/s
to 750MB/s.
Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
diff --git a/mm/migrate.c b/mm/migrate.c
index 068655d..a2d3e83 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -820,7 +820,6 @@ static int do_move_page_to_node_array(struct mm_struct *mm,
struct page_to_node *pp;
LIST_HEAD(pagelist);
- migrate_prep();
down_read(&mm->mmap_sem);
/*
@@ -907,6 +906,9 @@ static int do_pages_move(struct mm_struct *mm, struct task_struct *task,
pm = (struct page_to_node *)__get_free_page(GFP_KERNEL);
if (!pm)
goto out;
+
+ migrate_prep();
+
/*
* Store a chunk of page_to_node array in a page,
* but keep the last one as a marker
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-04-15 7:32 ` Brice Goglin
0 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-04-15 7:32 UTC (permalink / raw)
To: Christoph Lameter; +Cc: linux-mm, LKML
migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
throughput by breaking it into chunks, but it also made migrate_prep()
be called once per chunk (every 128pages or so) instead of once per
move_pages().
This patch reverts to calling migrate_prep() only once per chunk
as we did before 2.6.29.
It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
mm: move migrate_prep out from under mmap_sem
This improves migration throughput on the above machine from 600MB/s
to 750MB/s.
Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
diff --git a/mm/migrate.c b/mm/migrate.c
index 068655d..a2d3e83 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -820,7 +820,6 @@ static int do_move_page_to_node_array(struct mm_struct *mm,
struct page_to_node *pp;
LIST_HEAD(pagelist);
- migrate_prep();
down_read(&mm->mmap_sem);
/*
@@ -907,6 +906,9 @@ static int do_pages_move(struct mm_struct *mm, struct task_struct *task,
pm = (struct page_to_node *)__get_free_page(GFP_KERNEL);
if (!pm)
goto out;
+
+ migrate_prep();
+
/*
* Store a chunk of page_to_node array in a page,
* but keep the last one as a marker
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
2009-04-15 7:32 ` Brice Goglin
@ 2009-04-15 7:49 ` KAMEZAWA Hiroyuki
-1 siblings, 0 replies; 12+ messages in thread
From: KAMEZAWA Hiroyuki @ 2009-04-15 7:49 UTC (permalink / raw)
To: Brice Goglin; +Cc: Christoph Lameter, linux-mm, LKML
On Wed, 15 Apr 2009 09:32:10 +0200
Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
> throughput by breaking it into chunks, but it also made migrate_prep()
> be called once per chunk (every 128pages or so) instead of once per
> move_pages().
>
> This patch reverts to calling migrate_prep() only once per chunk
> as we did before 2.6.29.
> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
> mm: move migrate_prep out from under mmap_sem
>
> This improves migration throughput on the above machine from 600MB/s
> to 750MB/s.
>
> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
I think this patch is good. page migration is best-effort syscall ;)
BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
Thanks,
-Kame
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 068655d..a2d3e83 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -820,7 +820,6 @@ static int do_move_page_to_node_array(struct mm_struct *mm,
> struct page_to_node *pp;
> LIST_HEAD(pagelist);
>
> - migrate_prep();
> down_read(&mm->mmap_sem);
>
> /*
> @@ -907,6 +906,9 @@ static int do_pages_move(struct mm_struct *mm, struct task_struct *task,
> pm = (struct page_to_node *)__get_free_page(GFP_KERNEL);
> if (!pm)
> goto out;
> +
> + migrate_prep();
> +
> /*
> * Store a chunk of page_to_node array in a page,
> * but keep the last one as a marker
>
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-04-15 7:49 ` KAMEZAWA Hiroyuki
0 siblings, 0 replies; 12+ messages in thread
From: KAMEZAWA Hiroyuki @ 2009-04-15 7:49 UTC (permalink / raw)
To: Brice Goglin; +Cc: Christoph Lameter, linux-mm, LKML
On Wed, 15 Apr 2009 09:32:10 +0200
Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
> throughput by breaking it into chunks, but it also made migrate_prep()
> be called once per chunk (every 128pages or so) instead of once per
> move_pages().
>
> This patch reverts to calling migrate_prep() only once per chunk
> as we did before 2.6.29.
> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
> mm: move migrate_prep out from under mmap_sem
>
> This improves migration throughput on the above machine from 600MB/s
> to 750MB/s.
>
> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
I think this patch is good. page migration is best-effort syscall ;)
BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
Thanks,
-Kame
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 068655d..a2d3e83 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -820,7 +820,6 @@ static int do_move_page_to_node_array(struct mm_struct *mm,
> struct page_to_node *pp;
> LIST_HEAD(pagelist);
>
> - migrate_prep();
> down_read(&mm->mmap_sem);
>
> /*
> @@ -907,6 +906,9 @@ static int do_pages_move(struct mm_struct *mm, struct task_struct *task,
> pm = (struct page_to_node *)__get_free_page(GFP_KERNEL);
> if (!pm)
> goto out;
> +
> + migrate_prep();
> +
> /*
> * Store a chunk of page_to_node array in a page,
> * but keep the last one as a marker
>
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
2009-04-15 7:49 ` KAMEZAWA Hiroyuki
@ 2009-04-15 9:33 ` Brice Goglin
-1 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-04-15 9:33 UTC (permalink / raw)
To: KAMEZAWA Hiroyuki; +Cc: Christoph Lameter, linux-mm, LKML
KAMEZAWA Hiroyuki wrote:
> On Wed, 15 Apr 2009 09:32:10 +0200
> Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
>
>
>> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
>> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
>> throughput by breaking it into chunks, but it also made migrate_prep()
>> be called once per chunk (every 128pages or so) instead of once per
>> move_pages().
>>
>> This patch reverts to calling migrate_prep() only once per chunk
>> as we did before 2.6.29.
>> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
>> mm: move migrate_prep out from under mmap_sem
>>
>> This improves migration throughput on the above machine from 600MB/s
>> to 750MB/s.
>>
>> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>>
>>
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>
> I think this patch is good. page migration is best-effort syscall ;)
>
My next feeling now is about improving migrate_prep() itself. It makes
the move_pages() startup overhead very high.
But lru_add_drain_all() touches some code that I am far from
understanding :/ Can we imagine using IPI instead of a deferred
work_struct for this kind of things? Or maybe, for each processor, check
whether drain_cpu_pagevecs() would have something to do before actually
scheduling the local work_struct? It's racy, but migrate_prep() doesn't
guarantee anyway that pages won't be moved out of the LRU before the
actual migration, so...
Also I don't see why the cost of lru_add_drain_all() seems to increase
linearly with the number of cores in the machine. There may be some lock
contention, but it should scale better when there's pretty-much nothing
in the CPU lists...
> BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
>
I'd say they ignore it since it doesn't happen often :)
Brice
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-04-15 9:33 ` Brice Goglin
0 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-04-15 9:33 UTC (permalink / raw)
To: KAMEZAWA Hiroyuki; +Cc: Christoph Lameter, linux-mm, LKML
KAMEZAWA Hiroyuki wrote:
> On Wed, 15 Apr 2009 09:32:10 +0200
> Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
>
>
>> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
>> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
>> throughput by breaking it into chunks, but it also made migrate_prep()
>> be called once per chunk (every 128pages or so) instead of once per
>> move_pages().
>>
>> This patch reverts to calling migrate_prep() only once per chunk
>> as we did before 2.6.29.
>> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
>> mm: move migrate_prep out from under mmap_sem
>>
>> This improves migration throughput on the above machine from 600MB/s
>> to 750MB/s.
>>
>> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>>
>>
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>
> I think this patch is good. page migration is best-effort syscall ;)
>
My next feeling now is about improving migrate_prep() itself. It makes
the move_pages() startup overhead very high.
But lru_add_drain_all() touches some code that I am far from
understanding :/ Can we imagine using IPI instead of a deferred
work_struct for this kind of things? Or maybe, for each processor, check
whether drain_cpu_pagevecs() would have something to do before actually
scheduling the local work_struct? It's racy, but migrate_prep() doesn't
guarantee anyway that pages won't be moved out of the LRU before the
actual migration, so...
Also I don't see why the cost of lru_add_drain_all() seems to increase
linearly with the number of cores in the machine. There may be some lock
contention, but it should scale better when there's pretty-much nothing
in the CPU lists...
> BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
>
I'd say they ignore it since it doesn't happen often :)
Brice
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
2009-04-15 9:33 ` Brice Goglin
@ 2009-04-15 12:41 ` Christoph Lameter
-1 siblings, 0 replies; 12+ messages in thread
From: Christoph Lameter @ 2009-04-15 12:41 UTC (permalink / raw)
To: Brice Goglin; +Cc: KAMEZAWA Hiroyuki, linux-mm, LKML
On Wed, 15 Apr 2009, Brice Goglin wrote:
> But lru_add_drain_all() touches some code that I am far from
> understanding :/ Can we imagine using IPI instead of a deferred
> work_struct for this kind of things? Or maybe, for each processor, check
> whether drain_cpu_pagevecs() would have something to do before actually
> scheduling the local work_struct? It's racy, but migrate_prep() doesn't
> guarantee anyway that pages won't be moved out of the LRU before the
> actual migration, so...
IPI means that code must run with interrupts disabled.
> > BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
> >
>
> I'd say they ignore it since it doesn't happen often :)
Right.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-04-15 12:41 ` Christoph Lameter
0 siblings, 0 replies; 12+ messages in thread
From: Christoph Lameter @ 2009-04-15 12:41 UTC (permalink / raw)
To: Brice Goglin; +Cc: KAMEZAWA Hiroyuki, linux-mm, LKML
On Wed, 15 Apr 2009, Brice Goglin wrote:
> But lru_add_drain_all() touches some code that I am far from
> understanding :/ Can we imagine using IPI instead of a deferred
> work_struct for this kind of things? Or maybe, for each processor, check
> whether drain_cpu_pagevecs() would have something to do before actually
> scheduling the local work_struct? It's racy, but migrate_prep() doesn't
> guarantee anyway that pages won't be moved out of the LRU before the
> actual migration, so...
IPI means that code must run with interrupts disabled.
> > BTW, current users of sys_move_pages() does retry when it gets -EBUSY ?
> >
>
> I'd say they ignore it since it doesn't happen often :)
Right.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
2009-04-15 7:49 ` KAMEZAWA Hiroyuki
@ 2009-05-13 5:52 ` Brice Goglin
-1 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-05-13 5:52 UTC (permalink / raw)
To: KAMEZAWA Hiroyuki; +Cc: Christoph Lameter, linux-mm, LKML, Andrew Morton
KAMEZAWA Hiroyuki wrote:
> On Wed, 15 Apr 2009 09:32:10 +0200
> Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
>
>
>> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
>> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
>> throughput by breaking it into chunks, but it also made migrate_prep()
>> be called once per chunk (every 128pages or so) instead of once per
>> move_pages().
>>
>> This patch reverts to calling migrate_prep() only once per chunk
>> as we did before 2.6.29.
>> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
>> mm: move migrate_prep out from under mmap_sem
>>
>> This improves migration throughput on the above machine from 600MB/s
>> to 750MB/s.
>>
>> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>>
>>
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>
> I think this patch is good. page migration is best-effort syscall ;)
>
Since nobody complained about this patch, may I get a Ack and get the
patch merged for 2.6.31?
thanks,
Brice
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-05-13 5:52 ` Brice Goglin
0 siblings, 0 replies; 12+ messages in thread
From: Brice Goglin @ 2009-05-13 5:52 UTC (permalink / raw)
To: KAMEZAWA Hiroyuki; +Cc: Christoph Lameter, linux-mm, LKML, Andrew Morton
KAMEZAWA Hiroyuki wrote:
> On Wed, 15 Apr 2009 09:32:10 +0200
> Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
>
>
>> migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz).
>> Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages()
>> throughput by breaking it into chunks, but it also made migrate_prep()
>> be called once per chunk (every 128pages or so) instead of once per
>> move_pages().
>>
>> This patch reverts to calling migrate_prep() only once per chunk
>> as we did before 2.6.29.
>> It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d
>> mm: move migrate_prep out from under mmap_sem
>>
>> This improves migration throughput on the above machine from 600MB/s
>> to 750MB/s.
>>
>> Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
>>
>>
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>
> I think this patch is good. page migration is best-effort syscall ;)
>
Since nobody complained about this patch, may I get a Ack and get the
patch merged for 2.6.31?
thanks,
Brice
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
2009-05-13 5:52 ` Brice Goglin
@ 2009-05-14 20:11 ` Christoph Lameter
-1 siblings, 0 replies; 12+ messages in thread
From: Christoph Lameter @ 2009-05-14 20:11 UTC (permalink / raw)
To: Brice Goglin; +Cc: KAMEZAWA Hiroyuki, linux-mm, LKML, Andrew Morton
I acked it.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] migration: only migrate_prep() once per move_pages()
@ 2009-05-14 20:11 ` Christoph Lameter
0 siblings, 0 replies; 12+ messages in thread
From: Christoph Lameter @ 2009-05-14 20:11 UTC (permalink / raw)
To: Brice Goglin; +Cc: KAMEZAWA Hiroyuki, linux-mm, LKML, Andrew Morton
I acked it.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2009-05-14 20:29 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-04-15 7:32 [PATCH] migration: only migrate_prep() once per move_pages() Brice Goglin
2009-04-15 7:32 ` Brice Goglin
2009-04-15 7:49 ` KAMEZAWA Hiroyuki
2009-04-15 7:49 ` KAMEZAWA Hiroyuki
2009-04-15 9:33 ` Brice Goglin
2009-04-15 9:33 ` Brice Goglin
2009-04-15 12:41 ` Christoph Lameter
2009-04-15 12:41 ` Christoph Lameter
2009-05-13 5:52 ` Brice Goglin
2009-05-13 5:52 ` Brice Goglin
2009-05-14 20:11 ` Christoph Lameter
2009-05-14 20:11 ` Christoph Lameter
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.