From mboxrd@z Thu Jan 1 00:00:00 1970 From: Glauber Costa Subject: Re: [PATCH v3 15/28] slub: always get the cache from its page in kfree Date: Tue, 29 May 2012 19:59:47 +0400 Message-ID: <4FC4F273.6060807@parallels.com> References: <1337951028-3427-1-git-send-email-glommer@parallels.com> <1337951028-3427-16-git-send-email-glommer@parallels.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: Christoph Lameter Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org, Tejun Heo , Li Zefan , Greg Thelen , Suleiman Souhlal , Michal Hocko , Johannes Weiner , devel-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org, David Rientjes , Pekka Enberg On 05/29/2012 06:42 PM, Christoph Lameter wrote: > On Fri, 25 May 2012, Glauber Costa wrote: > >> struct page already have this information. If we start chaining >> caches, this information will always be more trustworthy than >> whatever is passed into the function > > Yes but the lookup of the page struct also costs some cycles. SLAB in > !NUMA mode and SLOB avoid these lookups and can improve their freeing > speed because of that. But for our case, I don't really see a way around. What I can do, is wrap it further, so when we're not using it, code goes exactly the same way as before, instead of always calculating the page. Would it be better? >> diff --git a/mm/slub.c b/mm/slub.c >> index 0eb9e72..640872f 100644 >> --- a/mm/slub.c >> +++ b/mm/slub.c >> @@ -2598,10 +2598,14 @@ redo: >> void kmem_cache_free(struct kmem_cache *s, void *x) >> { >> struct page *page; >> + bool slab_match; >> >> page = virt_to_head_page(x); >> >> - slab_free(s, page, x, _RET_IP_); >> + slab_match = (page->slab == s) | slab_is_parent(page->slab, s); >> + VM_BUG_ON(!slab_match); > > Why add a slab_match bool if you do not really need it? style. I find aux variables a very human readable way to deal with the 80-col limitation. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx205.postini.com [74.125.245.205]) by kanga.kvack.org (Postfix) with SMTP id 29EDA6B0075 for ; Tue, 29 May 2012 12:02:05 -0400 (EDT) Message-ID: <4FC4F273.6060807@parallels.com> Date: Tue, 29 May 2012 19:59:47 +0400 From: Glauber Costa MIME-Version: 1.0 Subject: Re: [PATCH v3 15/28] slub: always get the cache from its page in kfree References: <1337951028-3427-1-git-send-email-glommer@parallels.com> <1337951028-3427-16-git-send-email-glommer@parallels.com> In-Reply-To: Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Christoph Lameter Cc: linux-kernel@vger.kernel.org, cgroups@vger.kernel.org, linux-mm@kvack.org, kamezawa.hiroyu@jp.fujitsu.com, Tejun Heo , Li Zefan , Greg Thelen , Suleiman Souhlal , Michal Hocko , Johannes Weiner , devel@openvz.org, David Rientjes , Pekka Enberg On 05/29/2012 06:42 PM, Christoph Lameter wrote: > On Fri, 25 May 2012, Glauber Costa wrote: > >> struct page already have this information. If we start chaining >> caches, this information will always be more trustworthy than >> whatever is passed into the function > > Yes but the lookup of the page struct also costs some cycles. SLAB in > !NUMA mode and SLOB avoid these lookups and can improve their freeing > speed because of that. But for our case, I don't really see a way around. What I can do, is wrap it further, so when we're not using it, code goes exactly the same way as before, instead of always calculating the page. Would it be better? >> diff --git a/mm/slub.c b/mm/slub.c >> index 0eb9e72..640872f 100644 >> --- a/mm/slub.c >> +++ b/mm/slub.c >> @@ -2598,10 +2598,14 @@ redo: >> void kmem_cache_free(struct kmem_cache *s, void *x) >> { >> struct page *page; >> + bool slab_match; >> >> page = virt_to_head_page(x); >> >> - slab_free(s, page, x, _RET_IP_); >> + slab_match = (page->slab == s) | slab_is_parent(page->slab, s); >> + VM_BUG_ON(!slab_match); > > Why add a slab_match bool if you do not really need it? style. I find aux variables a very human readable way to deal with the 80-col limitation. -- 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: email@kvack.org From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754143Ab2E2QCF (ORCPT ); Tue, 29 May 2012 12:02:05 -0400 Received: from mx2.parallels.com ([64.131.90.16]:44005 "EHLO mx2.parallels.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752617Ab2E2QCD (ORCPT ); Tue, 29 May 2012 12:02:03 -0400 Message-ID: <4FC4F273.6060807@parallels.com> Date: Tue, 29 May 2012 19:59:47 +0400 From: Glauber Costa User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.1) Gecko/20120216 Thunderbird/10.0.1 MIME-Version: 1.0 To: Christoph Lameter CC: , , , , Tejun Heo , Li Zefan , Greg Thelen , Suleiman Souhlal , Michal Hocko , Johannes Weiner , , David Rientjes , Pekka Enberg Subject: Re: [PATCH v3 15/28] slub: always get the cache from its page in kfree References: <1337951028-3427-1-git-send-email-glommer@parallels.com> <1337951028-3427-16-git-send-email-glommer@parallels.com> In-Reply-To: Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [188.255.67.70] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 05/29/2012 06:42 PM, Christoph Lameter wrote: > On Fri, 25 May 2012, Glauber Costa wrote: > >> struct page already have this information. If we start chaining >> caches, this information will always be more trustworthy than >> whatever is passed into the function > > Yes but the lookup of the page struct also costs some cycles. SLAB in > !NUMA mode and SLOB avoid these lookups and can improve their freeing > speed because of that. But for our case, I don't really see a way around. What I can do, is wrap it further, so when we're not using it, code goes exactly the same way as before, instead of always calculating the page. Would it be better? >> diff --git a/mm/slub.c b/mm/slub.c >> index 0eb9e72..640872f 100644 >> --- a/mm/slub.c >> +++ b/mm/slub.c >> @@ -2598,10 +2598,14 @@ redo: >> void kmem_cache_free(struct kmem_cache *s, void *x) >> { >> struct page *page; >> + bool slab_match; >> >> page = virt_to_head_page(x); >> >> - slab_free(s, page, x, _RET_IP_); >> + slab_match = (page->slab == s) | slab_is_parent(page->slab, s); >> + VM_BUG_ON(!slab_match); > > Why add a slab_match bool if you do not really need it? style. I find aux variables a very human readable way to deal with the 80-col limitation.