From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1946036Ab2EKRvY (ORCPT ); Fri, 11 May 2012 13:51:24 -0400 Received: from mailhub.sw.ru ([195.214.232.25]:14777 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1945930Ab2EKRvO (ORCPT ); Fri, 11 May 2012 13:51:14 -0400 From: Glauber Costa To: Cc: , , , Tejun Heo , Li Zefan , Greg Thelen , Suleiman Souhlal , Michal Hocko , Johannes Weiner , , Glauber Costa , Christoph Lameter , Pekka Enberg Subject: [PATCH v2 17/29] slab: create duplicate cache Date: Fri, 11 May 2012 14:44:19 -0300 Message-Id: <1336758272-24284-18-git-send-email-glommer@parallels.com> X-Mailer: git-send-email 1.7.7.6 In-Reply-To: <1336758272-24284-1-git-send-email-glommer@parallels.com> References: <1336758272-24284-1-git-send-email-glommer@parallels.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This patch provides kmem_cache_dup(), that duplicates a cache for a memcg, preserving its creation properties. Object size, alignment and flags are all respected. An exception is the SLAB_PANIC flag, since cache creation inside a memcg should not be fatal. This code is mostly written by Suleiman Souhlal, with some adaptations and simplifications by me. Signed-off-by: Glauber Costa CC: Christoph Lameter CC: Pekka Enberg CC: Michal Hocko CC: Kamezawa Hiroyuki CC: Johannes Weiner CC: Suleiman Souhlal --- mm/slab.c | 32 ++++++++++++++++++++++++++++++++ 1 files changed, 32 insertions(+), 0 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index d05a326..985714a 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -301,6 +301,8 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int len, int node); static int enable_cpucache(struct kmem_cache *cachep, gfp_t gfp); static void cache_reap(struct work_struct *unused); +static int do_tune_cpucache(struct kmem_cache *cachep, int limit, + int batchcount, int shared, gfp_t gfp); /* * This function must be completely optimized away if a constant is passed to @@ -2598,6 +2600,36 @@ kmem_cache_create(const char *name, size_t size, size_t align, EXPORT_SYMBOL(kmem_cache_create); #ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +struct kmem_cache *kmem_cache_dup(struct mem_cgroup *memcg, + struct kmem_cache *cachep) +{ + struct kmem_cache *new; + unsigned long flags; + char *name; + + name = mem_cgroup_cache_name(memcg, cachep); + if (!name) + return NULL; + + flags = cachep->flags & ~(SLAB_PANIC|CFLGS_OFF_SLAB); + mutex_lock(&cache_chain_mutex); + new = __kmem_cache_create(memcg, name, obj_size(cachep), + cachep->memcg_params.orig_align, flags, cachep->ctor); + + if (new == NULL) + goto out; + + if ((cachep->limit != new->limit) || + (cachep->batchcount != new->batchcount) || + (cachep->shared != new->shared)) + do_tune_cpucache(new, cachep->limit, cachep->batchcount, + cachep->shared, GFP_KERNEL); +out: + mutex_unlock(&cache_chain_mutex); + kfree(name); + return new; +} + static int __init memcg_slab_register_all(void) { struct kmem_cache *cachep; -- 1.7.7.6 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx160.postini.com [74.125.245.160]) by kanga.kvack.org (Postfix) with SMTP id 809298D0017 for ; Fri, 11 May 2012 13:50:05 -0400 (EDT) From: Glauber Costa Subject: [PATCH v2 17/29] slab: create duplicate cache Date: Fri, 11 May 2012 14:44:19 -0300 Message-Id: <1336758272-24284-18-git-send-email-glommer@parallels.com> In-Reply-To: <1336758272-24284-1-git-send-email-glommer@parallels.com> References: <1336758272-24284-1-git-send-email-glommer@parallels.com> Sender: owner-linux-mm@kvack.org List-ID: To: linux-kernel@vger.kernel.org Cc: 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, Glauber Costa , Christoph Lameter , Pekka Enberg This patch provides kmem_cache_dup(), that duplicates a cache for a memcg, preserving its creation properties. Object size, alignment and flags are all respected. An exception is the SLAB_PANIC flag, since cache creation inside a memcg should not be fatal. This code is mostly written by Suleiman Souhlal, with some adaptations and simplifications by me. Signed-off-by: Glauber Costa CC: Christoph Lameter CC: Pekka Enberg CC: Michal Hocko CC: Kamezawa Hiroyuki CC: Johannes Weiner CC: Suleiman Souhlal --- mm/slab.c | 32 ++++++++++++++++++++++++++++++++ 1 files changed, 32 insertions(+), 0 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index d05a326..985714a 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -301,6 +301,8 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int len, int node); static int enable_cpucache(struct kmem_cache *cachep, gfp_t gfp); static void cache_reap(struct work_struct *unused); +static int do_tune_cpucache(struct kmem_cache *cachep, int limit, + int batchcount, int shared, gfp_t gfp); /* * This function must be completely optimized away if a constant is passed to @@ -2598,6 +2600,36 @@ kmem_cache_create(const char *name, size_t size, size_t align, EXPORT_SYMBOL(kmem_cache_create); #ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +struct kmem_cache *kmem_cache_dup(struct mem_cgroup *memcg, + struct kmem_cache *cachep) +{ + struct kmem_cache *new; + unsigned long flags; + char *name; + + name = mem_cgroup_cache_name(memcg, cachep); + if (!name) + return NULL; + + flags = cachep->flags & ~(SLAB_PANIC|CFLGS_OFF_SLAB); + mutex_lock(&cache_chain_mutex); + new = __kmem_cache_create(memcg, name, obj_size(cachep), + cachep->memcg_params.orig_align, flags, cachep->ctor); + + if (new == NULL) + goto out; + + if ((cachep->limit != new->limit) || + (cachep->batchcount != new->batchcount) || + (cachep->shared != new->shared)) + do_tune_cpucache(new, cachep->limit, cachep->batchcount, + cachep->shared, GFP_KERNEL); +out: + mutex_unlock(&cache_chain_mutex); + kfree(name); + return new; +} + static int __init memcg_slab_register_all(void) { struct kmem_cache *cachep; -- 1.7.7.6 -- 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 From: Glauber Costa Subject: [PATCH v2 17/29] slab: create duplicate cache Date: Fri, 11 May 2012 14:44:19 -0300 Message-ID: <1336758272-24284-18-git-send-email-glommer@parallels.com> References: <1336758272-24284-1-git-send-email-glommer@parallels.com> Return-path: In-Reply-To: <1336758272-24284-1-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org> Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org Cc: 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, Glauber Costa , Christoph Lameter , Pekka Enberg This patch provides kmem_cache_dup(), that duplicates a cache for a memcg, preserving its creation properties. Object size, alignment and flags are all respected. An exception is the SLAB_PANIC flag, since cache creation inside a memcg should not be fatal. This code is mostly written by Suleiman Souhlal, with some adaptations and simplifications by me. Signed-off-by: Glauber Costa CC: Christoph Lameter CC: Pekka Enberg CC: Michal Hocko CC: Kamezawa Hiroyuki CC: Johannes Weiner CC: Suleiman Souhlal --- mm/slab.c | 32 ++++++++++++++++++++++++++++++++ 1 files changed, 32 insertions(+), 0 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index d05a326..985714a 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -301,6 +301,8 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int len, int node); static int enable_cpucache(struct kmem_cache *cachep, gfp_t gfp); static void cache_reap(struct work_struct *unused); +static int do_tune_cpucache(struct kmem_cache *cachep, int limit, + int batchcount, int shared, gfp_t gfp); /* * This function must be completely optimized away if a constant is passed to @@ -2598,6 +2600,36 @@ kmem_cache_create(const char *name, size_t size, size_t align, EXPORT_SYMBOL(kmem_cache_create); #ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +struct kmem_cache *kmem_cache_dup(struct mem_cgroup *memcg, + struct kmem_cache *cachep) +{ + struct kmem_cache *new; + unsigned long flags; + char *name; + + name = mem_cgroup_cache_name(memcg, cachep); + if (!name) + return NULL; + + flags = cachep->flags & ~(SLAB_PANIC|CFLGS_OFF_SLAB); + mutex_lock(&cache_chain_mutex); + new = __kmem_cache_create(memcg, name, obj_size(cachep), + cachep->memcg_params.orig_align, flags, cachep->ctor); + + if (new == NULL) + goto out; + + if ((cachep->limit != new->limit) || + (cachep->batchcount != new->batchcount) || + (cachep->shared != new->shared)) + do_tune_cpucache(new, cachep->limit, cachep->batchcount, + cachep->shared, GFP_KERNEL); +out: + mutex_unlock(&cache_chain_mutex); + kfree(name); + return new; +} + static int __init memcg_slab_register_all(void) { struct kmem_cache *cachep; -- 1.7.7.6