From: Kamezawa Hiroyuki <kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org>
To: Glauber Costa <glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
Cc: Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
Dave Chinner <dchinner-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Dave Shrinnker <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org>,
Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
Mel Gorman <mgorman-l3A5Bk7waGM@public.gmane.org>,
Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>,
linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Subject: Re: [PATCH v2 22/28] memcg, list_lru: duplicate LRUs upon kmemcg creation
Date: Mon, 01 Apr 2013 17:05:36 +0900 [thread overview]
Message-ID: <51593FD0.9080502@jp.fujitsu.com> (raw)
In-Reply-To: <1364548450-28254-23-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
(2013/03/29 18:14), Glauber Costa wrote:
> When a new memcg is created, we need to open up room for its descriptors
> in all of the list_lrus that are marked per-memcg. The process is quite
> similar to the one we are using for the kmem caches: we initialize the
> new structures in an array indexed by kmemcg_id, and grow the array if
> needed. Key data like the size of the array will be shared between the
> kmem cache code and the list_lru code (they basically describe the same
> thing)
>
> Signed-off-by: Glauber Costa <glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
> Cc: Dave Chinner <dchinner-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> Cc: Mel Gorman <mgorman-l3A5Bk7waGM@public.gmane.org>
> Cc: Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> Cc: Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
> Cc: Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>
> Cc: Hugh Dickins <hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
> Cc: Kamezawa Hiroyuki <kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org>
> Cc: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
> ---
> include/linux/list_lru.h | 37 ++++++++++-
> include/linux/memcontrol.h | 12 ++++
> lib/list_lru.c | 101 +++++++++++++++++++++++++++---
> mm/memcontrol.c | 151 +++++++++++++++++++++++++++++++++++++++++++--
> mm/slab_common.c | 1 -
> 5 files changed, 285 insertions(+), 17 deletions(-)
>
> diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
> index 02796da..d6cf126 100644
> --- a/include/linux/list_lru.h
> +++ b/include/linux/list_lru.h
> @@ -16,12 +16,47 @@ struct list_lru_node {
> long nr_items;
> } ____cacheline_aligned_in_smp;
>
> +/*
> + * This is supposed to be M x N matrix, where M is kmem-limited memcg,
> + * and N is the number of nodes.
> + */
Could you add a comment that M can be changed and the array can be resized.
> +struct list_lru_array {
> + struct list_lru_node node[1];
> +};
> +
> struct list_lru {
> struct list_lru_node node[MAX_NUMNODES];
> nodemask_t active_nodes;
> +#ifdef CONFIG_MEMCG_KMEM
> + struct list_head lrus;
> + struct list_lru_array **memcg_lrus;
> +#endif
please add comments, for what ....
> };
>
> -int list_lru_init(struct list_lru *lru);
> +struct mem_cgroup;
> +#ifdef CONFIG_MEMCG_KMEM
> +struct list_lru_array *lru_alloc_array(void);
> +int memcg_update_all_lrus(unsigned long num);
> +void list_lru_destroy(struct list_lru *lru);
> +void list_lru_destroy_memcg(struct mem_cgroup *memcg);
> +int __memcg_init_lru(struct list_lru *lru);
> +#else
> +static inline void list_lru_destroy(struct list_lru *lru)
> +{
> +}
> +#endif
> +
> +int __list_lru_init(struct list_lru *lru, bool memcg_enabled);
> +static inline int list_lru_init(struct list_lru *lru)
> +{
> + return __list_lru_init(lru, false);
> +}
> +
> +static inline int list_lru_init_memcg(struct list_lru *lru)
> +{
> + return __list_lru_init(lru, true);
> +}
> +
> int list_lru_add(struct list_lru *lru, struct list_head *item);
> int list_lru_del(struct list_lru *lru, struct list_head *item);
> long list_lru_count_nodemask(struct list_lru *lru, nodemask_t *nodes_to_count);
> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
> index 4c24249..ee3199d 100644
> --- a/include/linux/memcontrol.h
> +++ b/include/linux/memcontrol.h
> @@ -23,6 +23,7 @@
> #include <linux/vm_event_item.h>
> #include <linux/hardirq.h>
> #include <linux/jump_label.h>
> +#include <linux/list_lru.h>
>
> struct mem_cgroup;
> struct page_cgroup;
> @@ -469,6 +470,12 @@ void memcg_update_array_size(int num_groups);
> struct kmem_cache *
> __memcg_kmem_get_cache(struct kmem_cache *cachep, gfp_t gfp);
>
> +int memcg_new_lru(struct list_lru *lru);
> +int memcg_init_lru(struct list_lru *lru);
> +
> +int memcg_kmem_update_lru_size(struct list_lru *lru, int num_groups,
> + bool new_lru);
> +
> void mem_cgroup_destroy_cache(struct kmem_cache *cachep);
> void kmem_cache_destroy_memcg_children(struct kmem_cache *s);
>
> @@ -632,6 +639,11 @@ memcg_kmem_get_cache(struct kmem_cache *cachep, gfp_t gfp)
> static inline void kmem_cache_destroy_memcg_children(struct kmem_cache *s)
> {
> }
> +
> +static inline int memcg_init_lru(struct list_lru *lru)
> +{
> + return 0;
> +}
> #endif /* CONFIG_MEMCG_KMEM */
> #endif /* _LINUX_MEMCONTROL_H */
>
> diff --git a/lib/list_lru.c b/lib/list_lru.c
> index 0f08ed6..a9616a0 100644
> --- a/lib/list_lru.c
> +++ b/lib/list_lru.c
> @@ -8,6 +8,7 @@
> #include <linux/module.h>
> #include <linux/mm.h>
> #include <linux/list_lru.h>
> +#include <linux/memcontrol.h>
>
> int
> list_lru_add(
> @@ -184,18 +185,100 @@ list_lru_dispose_all(
> return total;
> }
>
> -int
> -list_lru_init(
> - struct list_lru *lru)
> +/*
> + * This protects the list of all LRU in the system. One only needs
> + * to take when registering an LRU, or when duplicating the list of lrus.
> + * Transversing an LRU can and should be done outside the lock
> + */
> +static DEFINE_MUTEX(all_memcg_lrus_mutex);
> +static LIST_HEAD(all_memcg_lrus);
> +
> +static void list_lru_init_one(struct list_lru_node *lru)
> {
> + spin_lock_init(&lru->lock);
> + INIT_LIST_HEAD(&lru->list);
> + lru->nr_items = 0;
> +}
> +
> +struct list_lru_array *lru_alloc_array(void)
> +{
> + struct list_lru_array *lru_array;
> int i;
>
> - nodes_clear(lru->active_nodes);
> - for (i = 0; i < MAX_NUMNODES; i++) {
> - spin_lock_init(&lru->node[i].lock);
> - INIT_LIST_HEAD(&lru->node[i].list);
> - lru->node[i].nr_items = 0;
> + lru_array = kzalloc(nr_node_ids * sizeof(struct list_lru_node),
> + GFP_KERNEL);
A nitpick...you can use kmalloc() here. All field will be overwritten.
> + if (!lru_array)
> + return NULL;
> +
> + for (i = 0; i < nr_node_ids ; i++)
> + list_lru_init_one(&lru_array->node[i]);
> +
> + return lru_array;
> +}
> +
> +#ifdef CONFIG_MEMCG_KMEM
> +int __memcg_init_lru(struct list_lru *lru)
> +{
> + int ret;
> +
> + INIT_LIST_HEAD(&lru->lrus);
> + mutex_lock(&all_memcg_lrus_mutex);
> + list_add(&lru->lrus, &all_memcg_lrus);
> + ret = memcg_new_lru(lru);
> + mutex_unlock(&all_memcg_lrus_mutex);
> + return ret;
> +}
returns 0 at success ? what kind of error can be shown here ?
> +
> +int memcg_update_all_lrus(unsigned long num)
> +{
> + int ret = 0;
> + struct list_lru *lru;
> +
> + mutex_lock(&all_memcg_lrus_mutex);
> + list_for_each_entry(lru, &all_memcg_lrus, lrus) {
> + ret = memcg_kmem_update_lru_size(lru, num, false);
> + if (ret)
> + goto out;
> + }
> +out:
> + mutex_unlock(&all_memcg_lrus_mutex);
> + return ret;
> +}
> +
> +void list_lru_destroy(struct list_lru *lru)
> +{
> + if (!lru->memcg_lrus)
> + return;
> +
> + mutex_lock(&all_memcg_lrus_mutex);
> + list_del(&lru->lrus);
> + mutex_unlock(&all_memcg_lrus_mutex);
> +}
> +
> +void list_lru_destroy_memcg(struct mem_cgroup *memcg)
> +{
> + struct list_lru *lru;
> + mutex_lock(&all_memcg_lrus_mutex);
> + list_for_each_entry(lru, &all_memcg_lrus, lrus) {
> + kfree(lru->memcg_lrus[memcg_cache_id(memcg)]);
> + lru->memcg_lrus[memcg_cache_id(memcg)] = NULL;
> + /* everybody must beaware that this memcg is no longer valid */
> + wmb();
> }
> + mutex_unlock(&all_memcg_lrus_mutex);
> +}
> +#endif
> +
> +int __list_lru_init(struct list_lru *lru, bool memcg_enabled)
> +{
> + int i;
> +
> + nodes_clear(lru->active_nodes);
> + for (i = 0; i < MAX_NUMNODES; i++)
> + list_lru_init_one(&lru->node[i]);
> +
> + if (memcg_enabled)
> + return memcg_init_lru(lru);
> return 0;
> }
> -EXPORT_SYMBOL_GPL(list_lru_init);
> +EXPORT_SYMBOL_GPL(__list_lru_init);
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index ecdae39..c6c90d8 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -2988,16 +2988,30 @@ int memcg_update_cache_sizes(struct mem_cgroup *memcg)
> memcg_kmem_set_activated(memcg);
>
> ret = memcg_update_all_caches(num+1);
> - if (ret) {
> - ida_simple_remove(&kmem_limited_groups, num);
> - memcg_kmem_clear_activated(memcg);
> - return ret;
> - }
> + if (ret)
> + goto out;
> +
> + /*
> + * We should make sure that the array size is not updated until we are
> + * done; otherwise we have no easy way to know whether or not we should
> + * grow the array.
> + */
> + ret = memcg_update_all_lrus(num + 1);
> + if (ret)
> + goto out;
>
> memcg->kmemcg_id = num;
> +
> + memcg_update_array_size(num + 1);
> +
> INIT_LIST_HEAD(&memcg->memcg_slab_caches);
> mutex_init(&memcg->slab_caches_mutex);
> +
> return 0;
> +out:
> + ida_simple_remove(&kmem_limited_groups, num);
> + memcg_kmem_clear_activated(memcg);
> + return ret;
When this failure can happens ? This happens only when the user
tries to set kmem_limit and doesn't affect kernel internal logic ?
Thanks,
-Kame
next prev parent reply other threads:[~2013-04-01 8:05 UTC|newest]
Thread overview: 95+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-29 9:13 [PATCH v2 00/28] memcg-aware slab shrinking Glauber Costa
[not found] ` <1364548450-28254-1-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-03-29 9:13 ` [PATCH v2 01/28] super: fix calculation of shrinkable objects for small numbers Glauber Costa
[not found] ` <1364548450-28254-2-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-01 7:16 ` Kamezawa Hiroyuki
2013-03-29 9:13 ` [PATCH v2 02/28] vmscan: take at least one pass with shrinkers Glauber Costa
2013-04-01 7:26 ` Kamezawa Hiroyuki
2013-04-01 8:10 ` Glauber Costa
[not found] ` <515940E4.8050704-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-10 5:09 ` Ric Mason
2013-04-10 7:32 ` Glauber Costa
2013-04-10 9:19 ` Dave Chinner
2013-04-08 8:42 ` Joonsoo Kim
2013-04-08 8:47 ` Glauber Costa
2013-04-08 9:01 ` Joonsoo Kim
2013-04-08 9:05 ` Glauber Costa
2013-04-09 0:55 ` Joonsoo Kim
[not found] ` <20130409005547.GC21654-Hm3cg6mZ9cc@public.gmane.org>
2013-04-09 1:29 ` Dave Chinner
2013-04-09 2:05 ` Joonsoo Kim
2013-04-09 7:43 ` Glauber Costa
[not found] ` <5163C6A5.5050307-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-09 9:08 ` Joonsoo Kim
[not found] ` <20130409020505.GA4218-Hm3cg6mZ9cc@public.gmane.org>
2013-04-09 12:30 ` Dave Chinner
2013-04-10 2:51 ` Joonsoo Kim
2013-04-10 7:30 ` Glauber Costa
[not found] ` <51651518.4010007-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-10 8:19 ` Joonsoo Kim
2013-04-10 8:46 ` Wanpeng Li
[not found] ` <20130410025115.GA5872-Hm3cg6mZ9cc@public.gmane.org>
2013-04-10 8:46 ` Wanpeng Li
[not found] ` <20130410084606.GA10235@hacker.(null)>
2013-04-10 10:07 ` Dave Chinner
2013-04-10 14:03 ` JoonSoo Kim
[not found] ` <CAAmzW4OMyZ=nVbHK_AiifPK5LVxvhOQUXmsD5NGfo33CBjf=eA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-04-11 0:41 ` Dave Chinner
2013-04-11 7:27 ` Wanpeng Li
2013-04-11 7:27 ` Wanpeng Li
[not found] ` <20130411072729.GA3605@hacker.(null)>
2013-04-11 9:25 ` Dave Chinner
2013-03-29 9:13 ` [PATCH v2 03/28] dcache: convert dentry_stat.nr_unused to per-cpu counters Glauber Costa
2013-04-05 1:09 ` Greg Thelen
[not found] ` <xr93r4ipkcl0.fsf-aSPv4SP+Du0KgorLzL7FmE7CuiCeIGUxQQ4Iyu8u01E@public.gmane.org>
2013-04-05 1:15 ` Dave Chinner
2013-04-08 9:14 ` Glauber Costa
2013-04-08 13:18 ` Glauber Costa
[not found] ` <51628A88.2090002-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-08 23:26 ` Dave Chinner
2013-04-09 8:02 ` Glauber Costa
[not found] ` <5163CB0D.1040000-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-09 12:47 ` Dave Chinner
2013-03-29 9:13 ` [PATCH v2 04/28] dentry: move to per-sb LRU locks Glauber Costa
2013-03-29 9:13 ` [PATCH v2 05/28] dcache: remove dentries from LRU before putting on dispose list Glauber Costa
[not found] ` <1364548450-28254-6-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-03 6:51 ` Sha Zhengju
2013-04-03 8:55 ` Glauber Costa
2013-04-04 6:19 ` Dave Chinner
2013-04-04 6:56 ` Glauber Costa
2013-03-29 9:13 ` [PATCH v2 06/28] mm: new shrinker API Glauber Costa
2013-04-05 1:09 ` Greg Thelen
2013-03-29 9:13 ` [PATCH v2 07/28] shrinker: convert superblock shrinkers to new API Glauber Costa
2013-03-29 9:13 ` [PATCH v2 08/28] list: add a new LRU list type Glauber Costa
2013-04-04 21:53 ` Greg Thelen
2013-04-05 1:20 ` Dave Chinner
2013-04-05 8:01 ` Glauber Costa
2013-04-06 0:04 ` Dave Chinner
2013-03-29 9:13 ` [PATCH v2 09/28] inode: convert inode lru list to generic lru list code Glauber Costa
2013-03-29 9:13 ` [PATCH v2 10/28] dcache: convert to use new lru list infrastructure Glauber Costa
2013-04-08 13:14 ` Glauber Costa
[not found] ` <5162C2C4.7010807-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-08 23:28 ` Dave Chinner
2013-03-29 9:13 ` [PATCH v2 11/28] list_lru: per-node " Glauber Costa
2013-03-29 9:13 ` [PATCH v2 12/28] shrinker: add node awareness Glauber Costa
2013-03-29 9:13 ` [PATCH v2 13/28] fs: convert inode and dentry shrinking to be node aware Glauber Costa
2013-03-29 9:13 ` [PATCH v2 14/28] xfs: convert buftarg LRU to generic code Glauber Costa
2013-03-29 9:13 ` [PATCH v2 15/28] xfs: convert dquot cache lru to list_lru Glauber Costa
2013-03-29 9:13 ` [PATCH v2 16/28] fs: convert fs shrinkers to new scan/count API Glauber Costa
2013-03-29 9:13 ` [PATCH v2 17/28] drivers: convert shrinkers to new count/scan API Glauber Costa
2013-03-29 9:14 ` [PATCH v2 18/28] shrinker: convert remaining shrinkers to " Glauber Costa
2013-03-29 9:14 ` [PATCH v2 19/28] hugepage: convert huge zero page shrinker to new shrinker API Glauber Costa
2013-03-29 9:14 ` [PATCH v2 20/28] shrinker: Kill old ->shrink API Glauber Costa
2013-03-29 9:14 ` [PATCH v2 21/28] vmscan: also shrink slab in memcg pressure Glauber Costa
2013-04-01 7:46 ` Kamezawa Hiroyuki
2013-04-01 8:51 ` Glauber Costa
[not found] ` <1364548450-28254-22-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-03 10:11 ` Sha Zhengju
2013-04-03 10:43 ` Glauber Costa
2013-04-04 9:35 ` Sha Zhengju
2013-04-05 8:25 ` Glauber Costa
2013-03-29 9:14 ` [PATCH v2 22/28] memcg,list_lru: duplicate LRUs upon kmemcg creation Glauber Costa
[not found] ` <1364548450-28254-23-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-01 8:05 ` Kamezawa Hiroyuki [this message]
2013-04-01 8:22 ` Glauber Costa
2013-03-29 9:14 ` [PATCH v2 23/28] lru: add an element to a memcg list Glauber Costa
2013-04-01 8:18 ` Kamezawa Hiroyuki
2013-04-01 8:29 ` Glauber Costa
2013-03-29 9:14 ` [PATCH v2 24/28] list_lru: also include memcg lists in counts and scans Glauber Costa
2013-03-29 9:14 ` [PATCH v2 25/28] list_lru: per-memcg walks Glauber Costa
2013-03-29 9:14 ` [PATCH v2 26/28] memcg: per-memcg kmem shrinking Glauber Costa
2013-04-01 8:31 ` Kamezawa Hiroyuki
2013-04-01 8:48 ` Glauber Costa
[not found] ` <515949EB.7020400-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-01 9:01 ` Kamezawa Hiroyuki
2013-04-01 9:14 ` Glauber Costa
2013-04-01 9:35 ` Kamezawa Hiroyuki
2013-03-29 9:14 ` [PATCH v2 27/28] list_lru: reclaim proportionaly between memcgs and nodes Glauber Costa
2013-03-29 9:14 ` [PATCH v2 28/28] super: targeted memcg reclaim Glauber Costa
2013-04-01 12:38 ` [PATCH v2 00/28] memcg-aware slab shrinking Serge Hallyn
2013-04-01 12:45 ` Glauber Costa
[not found] ` <51598168.4050404-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2013-04-01 14:12 ` Serge Hallyn
2013-04-08 8:11 ` Glauber Costa
2013-04-02 4:58 ` Dave Chinner
2013-04-02 7:55 ` Glauber Costa
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=51593FD0.9080502@jp.fujitsu.com \
--to=kamezawa.hiroyu-+cum20s59erqfuhtdcdx3a@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org \
--cc=dchinner-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org \
--cc=hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org \
--cc=hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=mgorman-l3A5Bk7waGM@public.gmane.org \
--cc=mhocko-AlSwsSmVLrQ@public.gmane.org \
--cc=riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.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).