linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: "Alex,Shi" <alex.shi@intel.com>
To: Christoph Lameter <cl@linux.com>
Cc: "penberg@kernel.org" <penberg@kernel.org>,
	"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
	linux-mm@kvack.org, Andi Kleen <andi@firstfloor.org>
Subject: [rfc ] slub: unfreeze full page if it's in node partial
Date: Wed, 07 Sep 2011 10:56:12 +0800	[thread overview]
Message-ID: <1315364172.31737.174.camel@debian> (raw)
In-Reply-To: <1315362396.31737.151.camel@debian>

In the per cpu partial slub, we may add a full page into node partial
list. like the following scenario:

        cpu1                                    cpu2 
    in unfreeze_partials                   in __slab_alloc
        ...
   add_partial(n, page, 1);
                                        alloced from cpu partial, and 
                                        set frozen = 1.
   second cmpxchg_double_slab()
   set frozen = 0


At that time, maybe we'd better to unfreeze it in acquire_slab(). That
let it in 'full list' mode, frozen=0 and freelist = NULL, same as we did
in __slab_alloc()


Signed-off-by: Alex Shi <alex.shi@intel.com>
---
 mm/slub.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/slub.c b/mm/slub.c
index 6fca71c..7846951 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1579,7 +1579,7 @@ static inline void *acquire_slab(struct kmem_cache *s,
 			new.inuse = page->objects;
 
 		VM_BUG_ON(new.frozen);
-		new.frozen = 1;
+		new.frozen = freelist != NULL;
 
 	} while (!__cmpxchg_double_slab(s, page,
 			freelist, counters,
-- 
1.7.0



--
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>

  parent reply	other threads:[~2011-09-07  2:50 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1315188460.31737.5.camel@debian>
     [not found] ` <alpine.DEB.2.00.1109061914440.18646@router.home>
2011-09-07  1:03   ` [PATCH] slub Discard slab page only when node partials > minimum setting Alex,Shi
2011-09-07  2:26     ` [PATCH] slub: code optimze in get_partial_node() Alex,Shi
2011-09-07  2:45       ` [PATCH 2/2] slub: continue to seek slab in node partial if met a null page Alex,Shi
2011-09-07 15:01         ` Christoph Lameter
2011-09-08  8:38           ` Alex,Shi
2011-09-08 18:41             ` Christoph Lameter
2011-09-07  2:56       ` Alex,Shi [this message]
2011-09-07  3:06         ` [rfc ] slub: unfreeze full page if it's in node partial Alex,Shi
2011-09-07 14:56       ` [PATCH] slub: code optimze in get_partial_node() Christoph Lameter
     [not found]     ` <alpine.DEB.2.00.1109062022100.20474@router.home>
     [not found]       ` <4E671E5C.7010405@cs.helsinki.fi>
     [not found]         ` <6E3BC7F7C9A4BF4286DD4C043110F30B5D00DA333C@shsmsx502.ccr.corp.intel.com>
     [not found]           ` <alpine.DEB.2.00.1109071003240.9406@router.home>
2011-09-08  0:43             ` [PATCH] slub Discard slab page only when node partials > minimum setting Alex,Shi
2011-09-08  1:34               ` Shaohua Li
2011-09-08  2:24                 ` Alex,Shi
2011-09-15  5:40                   ` Pekka Enberg
2011-09-15  6:03                     ` Alex,Shi
2011-09-08 18:37               ` Christoph Lameter
2011-09-09  8:45                 ` Alex,Shi
2011-09-11 11:41                   ` Christoph Lameter
2011-09-13  8:29                   ` Alex,Shi
2011-09-13 15:04                     ` Christoph Lameter
2011-09-15  1:32                       ` Alex,Shi
2011-09-15  1:51                         ` Christoph Lameter
2011-09-15  2:00                           ` Alex,Shi
     [not found]                             ` <1316765880.4188.34.camel@debian>
     [not found]                               ` <alpine.DEB.2.00.1109231500580.15559@router.home>
2011-09-29  9:53                                 ` Alex,Shi
2011-09-29 14:32                                   ` Christoph Lameter
2011-10-02 12:47                                     ` Shi, Alex
2011-10-03 15:21                                       ` Christoph Lameter
2011-10-09  6:28                                         ` Alex,Shi
2011-10-10 17:12                                           ` Christoph Lameter
2011-09-14 15:38     ` Christoph Lameter
2011-09-15  5:48     ` Pekka Enberg
2011-09-15  6:16       ` Alex,Shi
2011-09-07  3:14   ` [PATCH] slub: correct comments error for per cpu partial Alex,Shi

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=1315364172.31737.174.camel@debian \
    --to=alex.shi@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=cl@linux.com \
    --cc=linux-mm@kvack.org \
    --cc=penberg@kernel.org \
    /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 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).