From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S935123AbYEIGIp (ORCPT ); Fri, 9 May 2008 02:08:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1764925AbYEIGIi (ORCPT ); Fri, 9 May 2008 02:08:38 -0400 Received: from cn.fujitsu.com ([222.73.24.84]:51078 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1759582AbYEIGIh (ORCPT ); Fri, 9 May 2008 02:08:37 -0400 Message-ID: <4823EA11.3060907@cn.fujitsu.com> Date: Fri, 09 May 2008 14:07:13 +0800 From: Li Zefan User-Agent: Thunderbird 2.0.0.9 (X11/20071115) MIME-Version: 1.0 To: KAMEZAWA Hiroyuki CC: LKML , "linux-mm@kvack.org" , Andrew Morton , "balbir@linux.vnet.ibm.com" , "xemul@openvz.org" , "yamamoto@valinux.co.jp" Subject: Re: [PATCH] memcg: avoid unnecessary initialization References: <20080509145941.f68e8f66.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: <20080509145941.f68e8f66.kamezawa.hiroyu@jp.fujitsu.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org KAMEZAWA Hiroyuki wrote: > An easy cut out from memcg: performance improvement patch set. > Tested on: x86-64/linux-2.6.26-rc1-git6 > > Thanks, > -Kame > == > * remove over-killing initialization (in fast path) > * makeing the condition for PAGE_CGROUP_FLAG_ACTIVE be more obvious. > > Signed-off-by: KAMEAZAWA Hiroyuki > Yes we don't need to init or del_init pc->lru. Reviewed-by: Li Zefan > --- > mm/memcontrol.c | 11 ++++++++--- > 1 file changed, 8 insertions(+), 3 deletions(-) > > Index: linux-2.6.26-rc1/mm/memcontrol.c > =================================================================== > --- linux-2.6.26-rc1.orig/mm/memcontrol.c > +++ linux-2.6.26-rc1/mm/memcontrol.c > @@ -296,7 +296,7 @@ static void __mem_cgroup_remove_list(str > MEM_CGROUP_ZSTAT(mz, MEM_CGROUP_ZSTAT_INACTIVE) -= 1; > > mem_cgroup_charge_statistics(pc->mem_cgroup, pc->flags, false); > - list_del_init(&pc->lru); > + list_del(&pc->lru); > } > > static void __mem_cgroup_add_list(struct mem_cgroup_per_zone *mz, > @@ -559,7 +559,7 @@ retry: > } > unlock_page_cgroup(page); > > - pc = kmem_cache_zalloc(page_cgroup_cache, gfp_mask); > + pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask); > if (pc == NULL) > goto err; > > @@ -606,9 +606,14 @@ retry: > pc->ref_cnt = 1; > pc->mem_cgroup = mem; > pc->page = page; > - pc->flags = PAGE_CGROUP_FLAG_ACTIVE; > + /* > + * If a page is accounted as a page cache, insert to inactive list. > + * If anon, insert to active list. > + */ > if (ctype == MEM_CGROUP_CHARGE_TYPE_CACHE) > pc->flags = PAGE_CGROUP_FLAG_CACHE; > + else > + pc->flags = PAGE_CGROUP_FLAG_ACTIVE; > > lock_page_cgroup(page); > if (page_get_page_cgroup(page)) { > > >