All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@suse.de>
To: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	James Bottomley <James.Bottomley@hansenpartnership.com>,
	Colin King <colin.king@canonical.com>,
	Raghavendra D Prabhu <raghu.prabhu13@gmail.com>,
	Jan Kara <jack@suse.cz>, Chris Mason <chris.mason@oracle.com>,
	Christoph Lameter <cl@linux.com>,
	Pekka Enberg <penberg@kernel.org>, Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	linux-mm <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	linux-ext4 <linux-ext4@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm: slub: Do not take expensive steps for SLUBs speculative high-order allocations
Date: Wed, 11 May 2011 22:10:43 +0100	[thread overview]
Message-ID: <20110511211043.GB17898@suse.de> (raw)
In-Reply-To: <alpine.DEB.2.00.1105111312020.9346@chino.kir.corp.google.com>

On Wed, May 11, 2011 at 01:38:44PM -0700, David Rientjes wrote:
> On Wed, 11 May 2011, Mel Gorman wrote:
> 
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 9f8a97b..057f1e2 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -1972,6 +1972,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
> >  {
> >  	int alloc_flags = ALLOC_WMARK_MIN | ALLOC_CPUSET;
> >  	const gfp_t wait = gfp_mask & __GFP_WAIT;
> > +	const gfp_t can_wake_kswapd = !(gfp_mask & __GFP_NO_KSWAPD);
> >  
> >  	/* __GFP_HIGH is assumed to be the same as ALLOC_HIGH to save a branch. */
> >  	BUILD_BUG_ON(__GFP_HIGH != (__force gfp_t) ALLOC_HIGH);
> > @@ -1984,7 +1985,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
> >  	 */
> >  	alloc_flags |= (__force int) (gfp_mask & __GFP_HIGH);
> >  
> > -	if (!wait) {
> > +	if (!wait && can_wake_kswapd) {
> >  		/*
> >  		 * Not worth trying to allocate harder for
> >  		 * __GFP_NOMEMALLOC even if it can't schedule.
> > diff --git a/mm/slub.c b/mm/slub.c
> > index 98c358d..1071723 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -1170,7 +1170,8 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
> >  	 * Let the initial higher-order allocation fail under memory pressure
> >  	 * so we fall-back to the minimum order allocation.
> >  	 */
> > -	alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY | __GFP_NO_KSWAPD) & ~__GFP_NOFAIL;
> > +	alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY | __GFP_NO_KSWAPD) &
> > +			~(__GFP_NOFAIL | __GFP_WAIT);
> 
> __GFP_NORETRY is a no-op without __GFP_WAIT.
> 

True. I'll remove it in a V2 but I won't respin just yet.

> >  
> >  	page = alloc_slab_page(alloc_gfp, node, oo);
> >  	if (unlikely(!page)) {

-- 
Mel Gorman
SUSE Labs

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Mel Gorman <mgorman@suse.de>
To: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	James Bottomley <James.Bottomley@hansenpartnership.com>,
	Colin King <colin.king@canonical.com>,
	Raghavendra D Prabhu <raghu.prabhu13@gmail.com>,
	Jan Kara <jack@suse.cz>, Chris Mason <chris.mason@oracle.com>,
	Christoph Lameter <cl@linux.com>,
	Pekka Enberg <penberg@kernel.org>, Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	linux-mm <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	linux-ext4 <linux-ext4@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm: slub: Do not take expensive steps for SLUBs speculative high-order allocations
Date: Wed, 11 May 2011 22:10:43 +0100	[thread overview]
Message-ID: <20110511211043.GB17898@suse.de> (raw)
In-Reply-To: <alpine.DEB.2.00.1105111312020.9346@chino.kir.corp.google.com>

On Wed, May 11, 2011 at 01:38:44PM -0700, David Rientjes wrote:
> On Wed, 11 May 2011, Mel Gorman wrote:
> 
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 9f8a97b..057f1e2 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -1972,6 +1972,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
> >  {
> >  	int alloc_flags = ALLOC_WMARK_MIN | ALLOC_CPUSET;
> >  	const gfp_t wait = gfp_mask & __GFP_WAIT;
> > +	const gfp_t can_wake_kswapd = !(gfp_mask & __GFP_NO_KSWAPD);
> >  
> >  	/* __GFP_HIGH is assumed to be the same as ALLOC_HIGH to save a branch. */
> >  	BUILD_BUG_ON(__GFP_HIGH != (__force gfp_t) ALLOC_HIGH);
> > @@ -1984,7 +1985,7 @@ gfp_to_alloc_flags(gfp_t gfp_mask)
> >  	 */
> >  	alloc_flags |= (__force int) (gfp_mask & __GFP_HIGH);
> >  
> > -	if (!wait) {
> > +	if (!wait && can_wake_kswapd) {
> >  		/*
> >  		 * Not worth trying to allocate harder for
> >  		 * __GFP_NOMEMALLOC even if it can't schedule.
> > diff --git a/mm/slub.c b/mm/slub.c
> > index 98c358d..1071723 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -1170,7 +1170,8 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
> >  	 * Let the initial higher-order allocation fail under memory pressure
> >  	 * so we fall-back to the minimum order allocation.
> >  	 */
> > -	alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY | __GFP_NO_KSWAPD) & ~__GFP_NOFAIL;
> > +	alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY | __GFP_NO_KSWAPD) &
> > +			~(__GFP_NOFAIL | __GFP_WAIT);
> 
> __GFP_NORETRY is a no-op without __GFP_WAIT.
> 

True. I'll remove it in a V2 but I won't respin just yet.

> >  
> >  	page = alloc_slab_page(alloc_gfp, node, oo);
> >  	if (unlikely(!page)) {

-- 
Mel Gorman
SUSE Labs

  reply	other threads:[~2011-05-11 21:10 UTC|newest]

Thread overview: 157+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-11 15:29 [PATCH 0/3] Reduce impact to overall system of SLUB using high-order allocations Mel Gorman
2011-05-11 15:29 ` Mel Gorman
2011-05-11 15:29 ` [PATCH 1/3] mm: slub: Do not wake kswapd for SLUBs speculative " Mel Gorman
2011-05-11 15:29   ` Mel Gorman
2011-05-11 20:38   ` David Rientjes
2011-05-11 20:38     ` David Rientjes
2011-05-11 15:29 ` [PATCH 2/3] mm: slub: Do not take expensive steps " Mel Gorman
2011-05-11 15:29   ` Mel Gorman
2011-05-11 20:38   ` David Rientjes
2011-05-11 20:38     ` David Rientjes
2011-05-11 21:10     ` Mel Gorman [this message]
2011-05-11 21:10       ` Mel Gorman
2011-05-12 17:25       ` Andrea Arcangeli
2011-05-12 17:25         ` Andrea Arcangeli
2011-05-11 15:29 ` [PATCH 3/3] mm: slub: Default slub_max_order to 0 Mel Gorman
2011-05-11 15:29   ` Mel Gorman
2011-05-11 20:38   ` David Rientjes
2011-05-11 20:38     ` David Rientjes
2011-05-11 20:53     ` James Bottomley
2011-05-11 20:53       ` James Bottomley
2011-05-11 21:09     ` Mel Gorman
2011-05-11 21:09       ` Mel Gorman
2011-05-11 22:27       ` David Rientjes
2011-05-11 22:27         ` David Rientjes
2011-05-13 10:14         ` Mel Gorman
2011-05-13 10:14           ` Mel Gorman
2011-05-12 17:36     ` Andrea Arcangeli
2011-05-12 17:36       ` Andrea Arcangeli
2011-05-16 21:03       ` David Rientjes
2011-05-16 21:03         ` David Rientjes
2011-05-17  9:48         ` Mel Gorman
2011-05-17  9:48           ` Mel Gorman
2011-05-17 19:25           ` David Rientjes
2011-05-17 19:25             ` David Rientjes
2011-05-12 14:43   ` Christoph Lameter
2011-05-12 14:43     ` Christoph Lameter
2011-05-12 15:15     ` James Bottomley
2011-05-12 15:15       ` James Bottomley
2011-05-12 15:27       ` Christoph Lameter
2011-05-12 15:27         ` Christoph Lameter
2011-05-12 15:43         ` James Bottomley
2011-05-12 15:43           ` James Bottomley
2011-05-12 15:46           ` Dave Jones
2011-05-12 15:46             ` Dave Jones
2011-05-12 16:00             ` James Bottomley
2011-05-12 16:00               ` James Bottomley
2011-05-12 16:08               ` Dave Jones
2011-05-12 16:08                 ` Dave Jones
2011-05-12 16:27               ` Christoph Lameter
2011-05-12 16:27                 ` Christoph Lameter
2011-05-12 16:30                 ` James Bottomley
2011-05-12 16:30                   ` James Bottomley
2011-05-12 16:48                   ` Christoph Lameter
2011-05-12 16:48                     ` Christoph Lameter
2011-05-12 17:46                     ` Andrea Arcangeli
2011-05-12 17:46                       ` Andrea Arcangeli
2011-05-12 18:00                       ` Christoph Lameter
2011-05-12 18:00                         ` Christoph Lameter
2011-05-12 18:18                         ` Andrea Arcangeli
2011-05-12 18:18                           ` Andrea Arcangeli
2011-05-12 17:06                   ` Pekka Enberg
2011-05-12 17:06                     ` Pekka Enberg
2011-05-12 17:11                     ` Pekka Enberg
2011-05-12 17:11                       ` Pekka Enberg
2011-05-12 17:38                       ` Christoph Lameter
2011-05-12 17:38                         ` Christoph Lameter
2011-05-12 18:00                         ` Andrea Arcangeli
2011-05-12 18:00                           ` Andrea Arcangeli
2011-05-13  9:49                           ` Mel Gorman
2011-05-13  9:49                             ` Mel Gorman
2011-05-15 16:39                             ` Andrea Arcangeli
2011-05-15 16:39                               ` Andrea Arcangeli
2011-05-16  8:42                               ` Mel Gorman
2011-05-16  8:42                                 ` Mel Gorman
2011-05-12 17:51                       ` Andrea Arcangeli
2011-05-12 17:51                         ` Andrea Arcangeli
2011-05-12 18:03                         ` Christoph Lameter
2011-05-12 18:03                           ` Christoph Lameter
2011-05-12 18:09                           ` Andrea Arcangeli
2011-05-12 18:09                             ` Andrea Arcangeli
2011-05-12 18:16                             ` Christoph Lameter
2011-05-12 18:16                               ` Christoph Lameter
2011-05-12 18:36                       ` James Bottomley
2011-05-12 18:36                         ` James Bottomley
2011-05-12 17:40                 ` Andrea Arcangeli
2011-05-12 17:40                   ` Andrea Arcangeli
2011-05-12 15:55           ` Pekka Enberg
2011-05-12 15:55             ` Pekka Enberg
2011-05-12 18:37             ` James Bottomley
2011-05-12 18:37               ` James Bottomley
2011-05-12 18:46               ` Christoph Lameter
2011-05-12 18:46                 ` Christoph Lameter
2011-05-12 19:21                 ` James Bottomley
2011-05-12 19:21                   ` James Bottomley
2011-05-12 19:44               ` James Bottomley
2011-05-12 19:44                 ` James Bottomley
2011-05-12 20:04                 ` James Bottomley
2011-05-12 20:04                   ` James Bottomley
2011-05-12 20:29                   ` Johannes Weiner
2011-05-12 20:29                     ` Johannes Weiner
2011-05-12 20:31                     ` Johannes Weiner
2011-05-12 20:31                       ` Johannes Weiner
2011-05-12 20:31                     ` James Bottomley
2011-05-12 20:31                       ` James Bottomley
2011-05-12 22:04                   ` James Bottomley
2011-05-12 22:04                     ` James Bottomley
2011-05-12 22:15                     ` Johannes Weiner
2011-05-12 22:15                       ` Johannes Weiner
2011-05-12 22:58                       ` Minchan Kim
2011-05-12 22:58                         ` Minchan Kim
2011-05-12 22:58                         ` Minchan Kim
2011-05-13  5:39                         ` Minchan Kim
2011-05-13  5:39                           ` Minchan Kim
2011-05-13  0:47                       ` James Bottomley
2011-05-13  0:47                         ` James Bottomley
2011-05-13  4:12                         ` James Bottomley
2011-05-13  4:12                           ` James Bottomley
2011-05-13 10:55                         ` Mel Gorman
2011-05-13 10:55                           ` Mel Gorman
2011-05-13 14:16                           ` James Bottomley
2011-05-13 14:16                             ` James Bottomley
2011-05-13 10:30                       ` Mel Gorman
2011-05-13 10:30                         ` Mel Gorman
2011-05-13  6:16                   ` Pekka Enberg
2011-05-13  6:16                     ` Pekka Enberg
2011-05-13  6:16                     ` Pekka Enberg
2011-05-13 10:05                     ` Mel Gorman
2011-05-13 10:05                       ` Mel Gorman
2011-05-13 10:05                       ` Mel Gorman
2011-05-12 16:01           ` Christoph Lameter
2011-05-12 16:01             ` Christoph Lameter
2011-05-12 16:10             ` Eric Dumazet
2011-05-12 16:10               ` Eric Dumazet
2011-05-12 16:10               ` Eric Dumazet
2011-05-12 17:37               ` Andrew Morton
2011-05-12 17:37                 ` Andrew Morton
2011-05-12 15:45         ` Dave Jones
2011-05-12 15:45           ` Dave Jones
2011-05-11 21:39 ` [PATCH 0/3] Reduce impact to overall system of SLUB using high-order allocations James Bottomley
2011-05-11 21:39   ` James Bottomley
2011-05-11 22:28   ` David Rientjes
2011-05-11 22:28     ` David Rientjes
2011-05-11 22:34     ` James Bottomley
2011-05-11 22:34       ` James Bottomley
2011-05-12 11:13       ` Pekka Enberg
2011-05-12 11:13         ` Pekka Enberg
2011-05-12 13:19         ` Mel Gorman
2011-05-12 13:19           ` Mel Gorman
2011-05-12 14:04         ` James Bottomley
2011-05-12 14:04           ` James Bottomley
2011-05-12 15:53           ` James Bottomley
2011-05-13 11:25             ` Mel Gorman
2011-05-13 11:25               ` Mel Gorman
2011-05-12 18:04       ` Andrea Arcangeli
2011-05-12 18:04         ` Andrea Arcangeli
2011-05-13 11:24         ` Mel Gorman
2011-05-13 11:24           ` Mel Gorman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110511211043.GB17898@suse.de \
    --to=mgorman@suse.de \
    --cc=James.Bottomley@hansenpartnership.com \
    --cc=akpm@linux-foundation.org \
    --cc=chris.mason@oracle.com \
    --cc=cl@linux.com \
    --cc=colin.king@canonical.com \
    --cc=hannes@cmpxchg.org \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=penberg@kernel.org \
    --cc=raghu.prabhu13@gmail.com \
    --cc=riel@redhat.com \
    --cc=rientjes@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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.