From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1031338Ab2CFVKJ (ORCPT ); Tue, 6 Mar 2012 16:10:09 -0500 Received: from mx1.redhat.com ([209.132.183.28]:1089 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1031121Ab2CFVKF (ORCPT ); Tue, 6 Mar 2012 16:10:05 -0500 Date: Tue, 6 Mar 2012 16:09:55 -0500 From: Vivek Goyal To: Tejun Heo Cc: axboe@kernel.dk, hughd@google.com, avi@redhat.com, nate@cpanel.net, cl@linux-foundation.org, linux-kernel@vger.kernel.org, dpshah@google.com, ctalbott@google.com, rni@google.com, Andrew Morton Subject: Re: [PATCHSET] mempool, percpu, blkcg: fix percpu stat allocation and remove stats_lock Message-ID: <20120306210954.GF32148@redhat.com> References: <20120223144336.58742e1b.akpm@linux-foundation.org> <20120223230123.GL22536@google.com> <20120223231204.GM22536@google.com> <20120224142033.GA5095@redhat.com> <20120225214421.GA3401@dhcp-172-17-108-109.mtv.corp.google.com> <20120227031146.GA25187@redhat.com> <20120227091141.GG3401@dhcp-172-17-108-109.mtv.corp.google.com> <20120227194321.GF27677@redhat.com> <20120229173639.GB5930@redhat.com> <20120305221321.GF1263@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120305221321.GF1263@google.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Mar 05, 2012 at 02:13:21PM -0800, Tejun Heo wrote: > Hello, Vivek. > > On Wed, Feb 29, 2012 at 12:36:39PM -0500, Vivek Goyal wrote: > > Index: tejun-misc/block/blk-cgroup.h > > =================================================================== > > --- tejun-misc.orig/block/blk-cgroup.h 2012-02-28 01:29:09.238256494 -0500 > > +++ tejun-misc/block/blk-cgroup.h 2012-02-28 01:29:12.000000000 -0500 > > @@ -180,6 +180,8 @@ struct blkio_group { > > struct request_queue *q; > > struct list_head q_node; > > struct hlist_node blkcg_node; > > + /* List of blkg waiting for per cpu stats memory to be allocated */ > > + struct list_head pending_alloc_node; > > Can we move this right on top of rcu_head? It's one of the coldest > entries. Also, long field names tend to be a bit painful. How about > just alloc_node? Hi Tejun, Here is the new version (lets call it v2) of the patch. Please let me know if some more changes are required. Thanks Vivek blk-cgroup: Alloc per cpu stats from worker thread in a delayed manner Current per cpu stat allocation assumes GFP_KERNEL allocation flag. But in IO path there are times when we want GFP_NOIO semantics. As there is no way to pass the allocation flags to alloc_percpu(), this patch delays the allocation of stats using a worker thread. v2-> tejun suggested following changes. Changed the patch accordingly. - move alloc_node location in structure - reduce the size of names of some of the fields - Reduce the scope of locking of alloc_list_lock - Simplified stat_alloc_fn() by allocating stats for all policies in one go and then assigning these to a group. Signed-off-by: Vivek Goyal --- block/blk-cgroup.c | 128 ++++++++++++++++++++++++++++++++++++----------------- block/blk-cgroup.h | 2 2 files changed, 90 insertions(+), 40 deletions(-) Index: tejun-misc/block/blk-cgroup.c =================================================================== --- tejun-misc.orig/block/blk-cgroup.c 2012-03-07 02:33:14.761787970 -0500 +++ tejun-misc/block/blk-cgroup.c 2012-03-07 02:33:51.042788674 -0500 @@ -30,6 +30,15 @@ static LIST_HEAD(blkio_list); static DEFINE_MUTEX(all_q_mutex); static LIST_HEAD(all_q_list); +/* List of groups pending per cpu stats allocation */ +static DEFINE_SPINLOCK(alloc_list_lock); +static LIST_HEAD(alloc_list); + +/* Array of per cpu stat pointers allocated for blk groups */ +static void *pcpu_stats[BLKIO_NR_POLICIES]; +static void blkio_stat_alloc_fn(struct work_struct *); +static DECLARE_WORK(blkio_stat_alloc_work, blkio_stat_alloc_fn); + struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT }; EXPORT_SYMBOL_GPL(blkio_root_cgroup); @@ -391,6 +400,9 @@ void blkiocg_update_dispatch_stats(struc struct blkio_group_stats_cpu *stats_cpu; unsigned long flags; + if (pd->stats_cpu == NULL) + return; + /* * Disabling interrupts to provide mutual exclusion between two * writes on same cpu. It probably is not needed for 64bit. Not @@ -443,6 +455,9 @@ void blkiocg_update_io_merged_stats(stru struct blkio_group_stats_cpu *stats_cpu; unsigned long flags; + if (pd->stats_cpu == NULL) + return; + /* * Disabling interrupts to provide mutual exclusion between two * writes on same cpu. It probably is not needed for 64bit. Not @@ -460,6 +475,59 @@ void blkiocg_update_io_merged_stats(stru } EXPORT_SYMBOL_GPL(blkiocg_update_io_merged_stats); +static void blkio_stat_alloc_fn(struct work_struct *work) +{ + + struct blkio_group *blkg, *n; + int i; + +alloc_stats: + spin_lock_irq(&alloc_list_lock); + if (list_empty(&alloc_list)) { + /* Nothing to do */ + spin_unlock_irq(&alloc_list_lock); + return; + } + spin_unlock_irq(&alloc_list_lock); + + for (i = 0; i < BLKIO_NR_POLICIES; i++) { + if (pcpu_stats[i] != NULL) + continue; + + pcpu_stats[i] = alloc_percpu(struct blkio_group_stats_cpu); + if (pcpu_stats[i] == NULL) + goto alloc_stats; + } + + spin_lock_irq(&blkio_list_lock); + spin_lock(&alloc_list_lock); + + list_for_each_entry_safe(blkg, n, &alloc_list, alloc_node) { + for (i = 0; i < BLKIO_NR_POLICIES; i++) { + struct blkio_policy_type *pol = blkio_policy[i]; + struct blkg_policy_data *pd; + + if (!pol) + continue; + + if (!blkg->pd[i]) + continue; + + pd = blkg->pd[i]; + if (pd->stats_cpu) + continue; + + pd->stats_cpu = pcpu_stats[i]; + pcpu_stats[i] = NULL; + } + list_del_init(&blkg->alloc_node); + break; + } + spin_unlock(&alloc_list_lock); + spin_unlock_irq(&blkio_list_lock); + goto alloc_stats; +} + /** * blkg_free - free a blkg * @blkg: blkg to free @@ -491,9 +559,6 @@ static void blkg_free(struct blkio_group * @q: request_queue the new blkg is associated with * * Allocate a new blkg assocating @blkcg and @q. - * - * FIXME: Should be called with queue locked but currently isn't due to - * percpu stat breakage. */ static struct blkio_group *blkg_alloc(struct blkio_cgroup *blkcg, struct request_queue *q) @@ -509,6 +574,7 @@ static struct blkio_group *blkg_alloc(st spin_lock_init(&blkg->stats_lock); blkg->q = q; INIT_LIST_HEAD(&blkg->q_node); + INIT_LIST_HEAD(&blkg->alloc_node); blkg->blkcg = blkcg; blkg->refcnt = 1; cgroup_path(blkcg->css.cgroup, blkg->path, sizeof(blkg->path)); @@ -530,13 +596,6 @@ static struct blkio_group *blkg_alloc(st blkg->pd[i] = pd; pd->blkg = blkg; - - /* broken, read comment in the callsite */ - pd->stats_cpu = alloc_percpu(struct blkio_group_stats_cpu); - if (!pd->stats_cpu) { - blkg_free(blkg); - return NULL; - } } /* invoke per-policy init */ @@ -556,7 +615,7 @@ struct blkio_group *blkg_lookup_create(s bool for_root) __releases(q->queue_lock) __acquires(q->queue_lock) { - struct blkio_group *blkg, *new_blkg; + struct blkio_group *blkg; WARN_ON_ONCE(!rcu_read_lock_held()); lockdep_assert_held(q->queue_lock); @@ -580,48 +639,27 @@ struct blkio_group *blkg_lookup_create(s /* * Allocate and initialize. - * - * FIXME: The following is broken. Percpu memory allocation - * requires %GFP_KERNEL context and can't be performed from IO - * path. Allocation here should inherently be atomic and the - * following lock dancing can be removed once the broken percpu - * allocation is fixed. */ - spin_unlock_irq(q->queue_lock); - rcu_read_unlock(); - - new_blkg = blkg_alloc(blkcg, q); - - rcu_read_lock(); - spin_lock_irq(q->queue_lock); - - /* did bypass get turned on inbetween? */ - if (unlikely(blk_queue_bypass(q)) && !for_root) { - blkg = ERR_PTR(blk_queue_dead(q) ? -EINVAL : -EBUSY); - goto out; - } - - /* did someone beat us to it? */ - blkg = blkg_lookup(blkcg, q); - if (unlikely(blkg)) - goto out; + blkg = blkg_alloc(blkcg, q); /* did alloc fail? */ - if (unlikely(!new_blkg)) { + if (unlikely(!blkg)) { blkg = ERR_PTR(-ENOMEM); goto out; } /* insert */ spin_lock(&blkcg->lock); - swap(blkg, new_blkg); - hlist_add_head_rcu(&blkg->blkcg_node, &blkcg->blkg_list); list_add(&blkg->q_node, &q->blkg_list); - spin_unlock(&blkcg->lock); + + spin_lock(&alloc_list_lock); + list_add(&blkg->alloc_node, &alloc_list); + /* Queue per cpu stat allocation from worker thread. */ + queue_work(system_nrt_wq, &blkio_stat_alloc_work); + spin_unlock(&alloc_list_lock); out: - blkg_free(new_blkg); return blkg; } EXPORT_SYMBOL_GPL(blkg_lookup_create); @@ -654,6 +692,10 @@ static void blkg_destroy(struct blkio_gr list_del_init(&blkg->q_node); hlist_del_init_rcu(&blkg->blkcg_node); + spin_lock(&alloc_list_lock); + list_del_init(&blkg->alloc_node); + spin_unlock(&alloc_list_lock); + /* * Put the reference taken at the time of creation so that when all * queues are gone, group can be destroyed. @@ -752,6 +794,9 @@ static void blkio_reset_stats_cpu(struct struct blkg_policy_data *pd = blkg->pd[plid]; struct blkio_group_stats_cpu *stats_cpu; int i, j, k; + + if (pd->stats_cpu == NULL) + return; /* * Note: On 64 bit arch this should not be an issue. This has the * possibility of returning some inconsistent value on 32bit arch @@ -883,6 +928,9 @@ static uint64_t blkio_read_stat_cpu(stru struct blkio_group_stats_cpu *stats_cpu; u64 val = 0, tval; + if (pd->stats_cpu == NULL) + return val; + for_each_possible_cpu(cpu) { unsigned int start; stats_cpu = per_cpu_ptr(pd->stats_cpu, cpu); Index: tejun-misc/block/blk-cgroup.h =================================================================== --- tejun-misc.orig/block/blk-cgroup.h 2012-03-07 02:33:14.761787970 -0500 +++ tejun-misc/block/blk-cgroup.h 2012-03-07 02:33:51.042788674 -0500 @@ -190,6 +190,8 @@ struct blkio_group { spinlock_t stats_lock; struct blkg_policy_data *pd[BLKIO_NR_POLICIES]; + /* List of blkg waiting for per cpu stats memory to be allocated */ + struct list_head alloc_node; struct rcu_head rcu_head; };