From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail172.messagelabs.com (mail172.messagelabs.com [216.82.254.3]) by kanga.kvack.org (Postfix) with ESMTP id 98ED4900086 for ; Thu, 14 Apr 2011 02:32:57 -0400 (EDT) Received: from wpaz13.hot.corp.google.com (wpaz13.hot.corp.google.com [172.24.198.77]) by smtp-out.google.com with ESMTP id p3E6Wrea016334 for ; Wed, 13 Apr 2011 23:32:53 -0700 Received: from gwj15 (gwj15.prod.google.com [10.200.10.15]) by wpaz13.hot.corp.google.com with ESMTP id p3E6Wp4W023820 (version=TLSv1/SSLv3 cipher=RC4-SHA bits=128 verify=NOT) for ; Wed, 13 Apr 2011 23:32:52 -0700 Received: by gwj15 with SMTP id 15so720397gwj.39 for ; Wed, 13 Apr 2011 23:32:51 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: <1302678187-24154-1-git-send-email-yinghan@google.com> <1302678187-24154-5-git-send-email-yinghan@google.com> Date: Wed, 13 Apr 2011 23:32:51 -0700 Message-ID: Subject: Re: [PATCH V3 4/7] Infrastructure to support per-memcg reclaim. From: Ying Han Content-Type: multipart/alternative; boundary=000e0cd37a7e37a35004a0db184f Sender: owner-linux-mm@kvack.org List-ID: To: Zhu Yanhai Cc: KAMEZAWA Hiroyuki , Pavel Emelyanov , Balbir Singh , Daisuke Nishimura , Li Zefan , Mel Gorman , Christoph Lameter , Johannes Weiner , Rik van Riel , Hugh Dickins , KOSAKI Motohiro , Tejun Heo , Michal Hocko , Andrew Morton , Dave Hansen , linux-mm@kvack.org --000e0cd37a7e37a35004a0db184f Content-Type: text/plain; charset=ISO-8859-1 On Wed, Apr 13, 2011 at 8:57 PM, Zhu Yanhai wrote: > Hi Ying, > > 2011/4/13 Ying Han : > > -extern int kswapd_run(int nid); > > -extern void kswapd_stop(int nid); > > +extern int kswapd_run(int nid, struct mem_cgroup *mem); > > +extern void kswapd_stop(int nid, struct mem_cgroup *mem); > > This breaks online_pages() and offline_pages(), which are also > the caller of kswaped_run() and kswaped_stop(). > Thanks, that will be fixed in the next post. --Ying > > Thanks, > Zhu Yanhai > > > > > #ifdef CONFIG_MMU > > /* linux/mm/shmem.c */ > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > index 36ae377..acd84a8 100644 > > --- a/mm/memcontrol.c > > +++ b/mm/memcontrol.c > > @@ -274,6 +274,8 @@ struct mem_cgroup { > > spinlock_t pcp_counter_lock; > > > > int wmark_ratio; > > + > > + wait_queue_head_t *kswapd_wait; > > }; > > > > /* Stuffs for move charges at task migration. */ > > @@ -4622,6 +4624,33 @@ int mem_cgroup_watermark_ok(struct mem_cgroup > *mem, > > return ret; > > } > > > > +int mem_cgroup_init_kswapd(struct mem_cgroup *mem, struct kswapd > *kswapd_p) > > +{ > > + if (!mem || !kswapd_p) > > + return 0; > > + > > + mem->kswapd_wait = &kswapd_p->kswapd_wait; > > + kswapd_p->kswapd_mem = mem; > > + > > + return css_id(&mem->css); > > +} > > + > > +void mem_cgroup_clear_kswapd(struct mem_cgroup *mem) > > +{ > > + if (mem) > > + mem->kswapd_wait = NULL; > > + > > + return; > > +} > > + > > +wait_queue_head_t *mem_cgroup_kswapd_wait(struct mem_cgroup *mem) > > +{ > > + if (!mem) > > + return NULL; > > + > > + return mem->kswapd_wait; > > +} > > + > > static int mem_cgroup_soft_limit_tree_init(void) > > { > > struct mem_cgroup_tree_per_node *rtpn; > > diff --git a/mm/vmscan.c b/mm/vmscan.c > > index 77ac74f..a1a1211 100644 > > --- a/mm/vmscan.c > > +++ b/mm/vmscan.c > > @@ -2242,6 +2242,7 @@ static bool pgdat_balanced(pg_data_t *pgdat, > unsigned long balanced_pages, > > } > > > > static DEFINE_SPINLOCK(kswapds_spinlock); > > +#define is_node_kswapd(kswapd_p) (!(kswapd_p)->kswapd_mem) > > > > /* is kswapd sleeping prematurely? */ > > static int sleeping_prematurely(struct kswapd *kswapd, int order, > > @@ -2251,11 +2252,16 @@ static int sleeping_prematurely(struct kswapd > *kswapd, int order, > > unsigned long balanced = 0; > > bool all_zones_ok = true; > > pg_data_t *pgdat = kswapd->kswapd_pgdat; > > + struct mem_cgroup *mem = kswapd->kswapd_mem; > > > > /* If a direct reclaimer woke kswapd within HZ/10, it's premature > */ > > if (remaining) > > return true; > > > > + /* Doesn't support for per-memcg reclaim */ > > + if (mem) > > + return false; > > + > > /* Check the watermark levels */ > > for (i = 0; i < pgdat->nr_zones; i++) { > > struct zone *zone = pgdat->node_zones + i; > > @@ -2598,19 +2604,25 @@ static void kswapd_try_to_sleep(struct kswapd > *kswapd_p, int order, > > * go fully to sleep until explicitly woken up. > > */ > > if (!sleeping_prematurely(kswapd_p, order, remaining, > classzone_idx)) { > > - trace_mm_vmscan_kswapd_sleep(pgdat->node_id); > > + if (is_node_kswapd(kswapd_p)) { > > + trace_mm_vmscan_kswapd_sleep(pgdat->node_id); > > > > - /* > > - * vmstat counters are not perfectly accurate and the > estimated > > - * value for counters such as NR_FREE_PAGES can deviate > from the > > - * true value by nr_online_cpus * threshold. To avoid the > zone > > - * watermarks being breached while under pressure, we > reduce the > > - * per-cpu vmstat threshold while kswapd is awake and > restore > > - * them before going back to sleep. > > - */ > > - set_pgdat_percpu_threshold(pgdat, > calculate_normal_threshold); > > - schedule(); > > - set_pgdat_percpu_threshold(pgdat, > calculate_pressure_threshold); > > + /* > > + * vmstat counters are not perfectly accurate and > the > > + * estimated value for counters such as > NR_FREE_PAGES > > + * can deviate from the true value by > nr_online_cpus * > > + * threshold. To avoid the zone watermarks being > > + * breached while under pressure, we reduce the > per-cpu > > + * vmstat threshold while kswapd is awake and > restore > > + * them before going back to sleep. > > + */ > > + set_pgdat_percpu_threshold(pgdat, > > + > calculate_normal_threshold); > > + schedule(); > > + set_pgdat_percpu_threshold(pgdat, > > + > calculate_pressure_threshold); > > + } else > > + schedule(); > > } else { > > if (remaining) > > count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY); > > @@ -2620,6 +2632,12 @@ static void kswapd_try_to_sleep(struct kswapd > *kswapd_p, int order, > > finish_wait(wait_h, &wait); > > } > > > > +static unsigned long balance_mem_cgroup_pgdat(struct mem_cgroup > *mem_cont, > > + int order) > > +{ > > + return 0; > > +} > > + > > /* > > * The background pageout daemon, started as a kernel thread > > * from the init process. > > @@ -2639,6 +2657,7 @@ int kswapd(void *p) > > int classzone_idx; > > struct kswapd *kswapd_p = (struct kswapd *)p; > > pg_data_t *pgdat = kswapd_p->kswapd_pgdat; > > + struct mem_cgroup *mem = kswapd_p->kswapd_mem; > > wait_queue_head_t *wait_h = &kswapd_p->kswapd_wait; > > struct task_struct *tsk = current; > > > > @@ -2649,10 +2668,12 @@ int kswapd(void *p) > > > > lockdep_set_current_reclaim_state(GFP_KERNEL); > > > > - BUG_ON(pgdat->kswapd_wait != wait_h); > > - cpumask = cpumask_of_node(pgdat->node_id); > > - if (!cpumask_empty(cpumask)) > > - set_cpus_allowed_ptr(tsk, cpumask); > > + if (is_node_kswapd(kswapd_p)) { > > + BUG_ON(pgdat->kswapd_wait != wait_h); > > + cpumask = cpumask_of_node(pgdat->node_id); > > + if (!cpumask_empty(cpumask)) > > + set_cpus_allowed_ptr(tsk, cpumask); > > + } > > current->reclaim_state = &reclaim_state; > > > > /* > > @@ -2677,24 +2698,29 @@ int kswapd(void *p) > > int new_classzone_idx; > > int ret; > > > > - new_order = pgdat->kswapd_max_order; > > - new_classzone_idx = pgdat->classzone_idx; > > - pgdat->kswapd_max_order = 0; > > - pgdat->classzone_idx = MAX_NR_ZONES - 1; > > - if (order < new_order || classzone_idx > > new_classzone_idx) { > > - /* > > - * Don't sleep if someone wants a larger 'order' > > - * allocation or has tigher zone constraints > > - */ > > - order = new_order; > > - classzone_idx = new_classzone_idx; > > - } else { > > - kswapd_try_to_sleep(kswapd_p, order, > classzone_idx); > > - order = pgdat->kswapd_max_order; > > - classzone_idx = pgdat->classzone_idx; > > + if (is_node_kswapd(kswapd_p)) { > > + new_order = pgdat->kswapd_max_order; > > + new_classzone_idx = pgdat->classzone_idx; > > pgdat->kswapd_max_order = 0; > > pgdat->classzone_idx = MAX_NR_ZONES - 1; > > - } > > + if (order < new_order || > > + classzone_idx > > new_classzone_idx) { > > + /* > > + * Don't sleep if someone wants a larger > 'order' > > + * allocation or has tigher zone > constraints > > + */ > > + order = new_order; > > + classzone_idx = new_classzone_idx; > > + } else { > > + kswapd_try_to_sleep(kswapd_p, order, > > + classzone_idx); > > + order = pgdat->kswapd_max_order; > > + classzone_idx = pgdat->classzone_idx; > > + pgdat->kswapd_max_order = 0; > > + pgdat->classzone_idx = MAX_NR_ZONES - 1; > > + } > > + } else > > + kswapd_try_to_sleep(kswapd_p, order, > classzone_idx); > > > > ret = try_to_freeze(); > > if (kthread_should_stop()) > > @@ -2705,8 +2731,13 @@ int kswapd(void *p) > > * after returning from the refrigerator > > */ > > if (!ret) { > > - trace_mm_vmscan_kswapd_wake(pgdat->node_id, > order); > > - order = balance_pgdat(pgdat, order, > &classzone_idx); > > + if (is_node_kswapd(kswapd_p)) { > > + > trace_mm_vmscan_kswapd_wake(pgdat->node_id, > > + order); > > + order = balance_pgdat(pgdat, order, > > + &classzone_idx); > > + } else > > + balance_mem_cgroup_pgdat(mem, order); > > } > > } > > return 0; > > @@ -2853,30 +2884,53 @@ static int __devinit cpu_callback(struct > notifier_block *nfb, > > * This kswapd start function will be called by init and node-hot-add. > > * On node-hot-add, kswapd will moved to proper cpus if cpus are > hot-added. > > */ > > -int kswapd_run(int nid) > > +int kswapd_run(int nid, struct mem_cgroup *mem) > > { > > - pg_data_t *pgdat = NODE_DATA(nid); > > struct task_struct *kswapd_thr; > > + pg_data_t *pgdat = NULL; > > struct kswapd *kswapd_p; > > + static char name[TASK_COMM_LEN]; > > + int memcg_id; > > int ret = 0; > > > > - if (pgdat->kswapd_wait) > > - return 0; > > + if (!mem) { > > + pgdat = NODE_DATA(nid); > > + if (pgdat->kswapd_wait) > > + return ret; > > + } > > > > kswapd_p = kzalloc(sizeof(struct kswapd), GFP_KERNEL); > > if (!kswapd_p) > > return -ENOMEM; > > > > init_waitqueue_head(&kswapd_p->kswapd_wait); > > - pgdat->kswapd_wait = &kswapd_p->kswapd_wait; > > - kswapd_p->kswapd_pgdat = pgdat; > > > > - kswapd_thr = kthread_run(kswapd, kswapd_p, "kswapd%d", nid); > > + if (!mem) { > > + pgdat->kswapd_wait = &kswapd_p->kswapd_wait; > > + kswapd_p->kswapd_pgdat = pgdat; > > + snprintf(name, TASK_COMM_LEN, "kswapd_%d", nid); > > + } else { > > + memcg_id = mem_cgroup_init_kswapd(mem, kswapd_p); > > + if (!memcg_id) { > > + kfree(kswapd_p); > > + return ret; > > + } > > + snprintf(name, TASK_COMM_LEN, "memcg_%d", memcg_id); > > + } > > + > > + kswapd_thr = kthread_run(kswapd, kswapd_p, name); > > if (IS_ERR(kswapd_thr)) { > > /* failure at boot is fatal */ > > BUG_ON(system_state == SYSTEM_BOOTING); > > - printk("Failed to start kswapd on node %d\n",nid); > > - pgdat->kswapd_wait = NULL; > > + if (!mem) { > > + printk(KERN_ERR "Failed to start kswapd on node > %d\n", > > + nid); > > + pgdat->kswapd_wait = NULL; > > + } else { > > + printk(KERN_ERR "Failed to start kswapd on memcg > %d\n", > > + > memcg_id); > > + mem_cgroup_clear_kswapd(mem); > > + } > > kfree(kswapd_p); > > ret = -1; > > } else > > @@ -2887,16 +2941,18 @@ int kswapd_run(int nid) > > /* > > * Called by memory hotplug when all memory in a node is offlined. > > */ > > -void kswapd_stop(int nid) > > +void kswapd_stop(int nid, struct mem_cgroup *mem) > > { > > struct task_struct *kswapd_thr = NULL; > > struct kswapd *kswapd_p = NULL; > > wait_queue_head_t *wait; > > > > - pg_data_t *pgdat = NODE_DATA(nid); > > - > > spin_lock(&kswapds_spinlock); > > - wait = pgdat->kswapd_wait; > > + if (!mem) > > + wait = NODE_DATA(nid)->kswapd_wait; > > + else > > + wait = mem_cgroup_kswapd_wait(mem); > > + > > if (wait) { > > kswapd_p = container_of(wait, struct kswapd, kswapd_wait); > > kswapd_thr = kswapd_p->kswapd_task; > > @@ -2916,7 +2972,7 @@ static int __init kswapd_init(void) > > > > swap_setup(); > > for_each_node_state(nid, N_HIGH_MEMORY) > > - kswapd_run(nid); > > + kswapd_run(nid, NULL); > > hotcpu_notifier(cpu_callback, 0); > > return 0; > > } > > -- > > 1.7.3.1 > > > > -- > > 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 > > > --000e0cd37a7e37a35004a0db184f Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

On Wed, Apr 13, 2011 at 8:57 PM, Zhu Yan= hai <zhu.yanha= i@gmail.com> wrote:
Hi Ying,

2011/4/13 Ying Han <yinghan@google= .com>:
> -extern int kswapd_run(int nid);
> -extern void kswapd_stop(int nid);
> +extern int kswapd_run(int nid, struct mem_cgroup *mem);
> +extern void kswapd_stop(int nid, struct mem_cgroup *mem);

This breaks online_pages() and offline_pages(), which are also
the caller of kswaped_run() and kswaped_stop().

Thanks, that will be fixed in the next post.

<= div>--Ying=A0

Thanks,
Zhu Yanhai

>
> =A0#ifdef CONFIG_MMU
> =A0/* linux/mm/shmem.c */
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 36ae377..acd84a8 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -274,6 +274,8 @@ struct mem_cgroup {
> =A0 =A0 =A0 =A0spinlock_t pcp_counter_lock;
>
> =A0 =A0 =A0 =A0int wmark_ratio;
> +
> + =A0 =A0 =A0 wait_queue_head_t *kswapd_wait;
> =A0};
>
> =A0/* Stuffs for move charges at task migration. */
> @@ -4622,6 +4624,33 @@ int mem_cgroup_watermark_ok(struct mem_cgroup *= mem,
> =A0 =A0 =A0 =A0return ret;
> =A0}
>
> +int mem_cgroup_init_kswapd(struct mem_cgroup *mem, struct kswapd *ksw= apd_p)
> +{
> + =A0 =A0 =A0 if (!mem || !kswapd_p)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 return 0;
> +
> + =A0 =A0 =A0 mem->kswapd_wait =3D &kswapd_p->kswapd_wait; > + =A0 =A0 =A0 kswapd_p->kswapd_mem =3D mem;
> +
> + =A0 =A0 =A0 return css_id(&mem->css);
> +}
> +
> +void mem_cgroup_clear_kswapd(struct mem_cgroup *mem)
> +{
> + =A0 =A0 =A0 if (mem)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 mem->kswapd_wait =3D NULL;
> +
> + =A0 =A0 =A0 return;
> +}
> +
> +wait_queue_head_t *mem_cgroup_kswapd_wait(struct mem_cgroup *mem)
> +{
> + =A0 =A0 =A0 if (!mem)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 return NULL;
> +
> + =A0 =A0 =A0 return mem->kswapd_wait;
> +}
> +
> =A0static int mem_cgroup_soft_limit_tree_init(void)
> =A0{
> =A0 =A0 =A0 =A0struct mem_cgroup_tree_per_node *rtpn;
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 77ac74f..a1a1211 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2242,6 +2242,7 @@ static bool pgdat_balanced(pg_data_t *pgdat, uns= igned long balanced_pages,
> =A0}
>
> =A0static DEFINE_SPINLOCK(kswapds_spinlock);
> +#define is_node_kswapd(kswapd_p) (!(kswapd_p)->kswapd_mem)
>
> =A0/* is kswapd sleeping prematurely? */
> =A0static int sleeping_prematurely(struct kswapd *kswapd, int order, > @@ -2251,11 +2252,16 @@ static int sleeping_prematurely(struct kswapd = *kswapd, int order,
> =A0 =A0 =A0 =A0unsigned long balanced =3D 0;
> =A0 =A0 =A0 =A0bool all_zones_ok =3D true;
> =A0 =A0 =A0 =A0pg_data_t *pgdat =3D kswapd->kswapd_pgdat;
> + =A0 =A0 =A0 struct mem_cgroup *mem =3D kswapd->kswapd_mem;
>
> =A0 =A0 =A0 =A0/* If a direct reclaimer woke kswapd within HZ/10, it&#= 39;s premature */
> =A0 =A0 =A0 =A0if (remaining)
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0return true;
>
> + =A0 =A0 =A0 /* Doesn't support for per-memcg reclaim */
> + =A0 =A0 =A0 if (mem)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 return false;
> +
> =A0 =A0 =A0 =A0/* Check the watermark levels */
> =A0 =A0 =A0 =A0for (i =3D 0; i < pgdat->nr_zones; i++) {
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0struct zone *zone =3D pgdat->node_zo= nes + i;
> @@ -2598,19 +2604,25 @@ static void kswapd_try_to_sleep(struct kswapd = *kswapd_p, int order,
> =A0 =A0 =A0 =A0 * go fully to sleep until explicitly woken up.
> =A0 =A0 =A0 =A0 */
> =A0 =A0 =A0 =A0if (!sleeping_prematurely(kswapd_p, order, remaining, c= lasszone_idx)) {
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 trace_mm_vmscan_kswapd_sleep(pgdat->n= ode_id);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (is_node_kswapd(kswapd_p)) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 trace_mm_vmscan_kswapd_s= leep(pgdat->node_id);
>
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* vmstat counters are not perfectly a= ccurate and the estimated
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* value for counters such as NR_FREE_= PAGES can deviate from the
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* true value by nr_online_cpus * thre= shold. To avoid the zone
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* watermarks being breached while und= er pressure, we reduce the
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* per-cpu vmstat threshold while kswa= pd is awake and restore
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* them before going back to sleep. > - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0*/
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_pgdat_percpu_threshold(pgdat, calcul= ate_normal_threshold);
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 schedule();
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_pgdat_percpu_threshold(pgdat, calcul= ate_pressure_threshold);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* vmstat counters are= not perfectly accurate and the
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* estimated value for= counters such as NR_FREE_PAGES
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* can deviate from th= e true value by nr_online_cpus *
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* threshold. To avoid= the zone watermarks being
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* breached while unde= r pressure, we reduce the per-cpu
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* vmstat threshold wh= ile kswapd is awake and restore
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* them before going b= ack to sleep.
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0*/
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_pgdat_percpu_thresho= ld(pgdat,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0calculate_normal_threshold);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 schedule();
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_pgdat_percpu_thresho= ld(pgdat,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 calculate_pressure_threshold);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 schedule();
> =A0 =A0 =A0 =A0} else {
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0if (remaining)
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0count_vm_event(KSWAPD_L= OW_WMARK_HIT_QUICKLY);
> @@ -2620,6 +2632,12 @@ static void kswapd_try_to_sleep(struct kswapd *= kswapd_p, int order,
> =A0 =A0 =A0 =A0finish_wait(wait_h, &wait);
> =A0}
>
> +static unsigned long balance_mem_cgroup_pgdat(struct mem_cgroup *mem_= cont,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 int order)
> +{
> + =A0 =A0 =A0 return 0;
> +}
> +
> =A0/*
> =A0* The background pageout daemon, started as a kernel thread
> =A0* from the init process.
> @@ -2639,6 +2657,7 @@ int kswapd(void *p)
> =A0 =A0 =A0 =A0int classzone_idx;
> =A0 =A0 =A0 =A0struct kswapd *kswapd_p =3D (struct kswapd *)p;
> =A0 =A0 =A0 =A0pg_data_t *pgdat =3D kswapd_p->kswapd_pgdat;
> + =A0 =A0 =A0 struct mem_cgroup *mem =3D kswapd_p->kswapd_mem;
> =A0 =A0 =A0 =A0wait_queue_head_t *wait_h =3D &kswapd_p->kswapd_= wait;
> =A0 =A0 =A0 =A0struct task_struct *tsk =3D current;
>
> @@ -2649,10 +2668,12 @@ int kswapd(void *p)
>
> =A0 =A0 =A0 =A0lockdep_set_current_reclaim_state(GFP_KERNEL);
>
> - =A0 =A0 =A0 BUG_ON(pgdat->kswapd_wait !=3D wait_h);
> - =A0 =A0 =A0 cpumask =3D cpumask_of_node(pgdat->node_id);
> - =A0 =A0 =A0 if (!cpumask_empty(cpumask))
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_cpus_allowed_ptr(tsk, cpumask);
> + =A0 =A0 =A0 if (is_node_kswapd(kswapd_p)) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 BUG_ON(pgdat->kswapd_wait !=3D wait_h= );
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 cpumask =3D cpumask_of_node(pgdat->no= de_id);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (!cpumask_empty(cpumask))
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_cpus_allowed_ptr(tsk= , cpumask);
> + =A0 =A0 =A0 }
> =A0 =A0 =A0 =A0current->reclaim_state =3D &reclaim_state;
>
> =A0 =A0 =A0 =A0/*
> @@ -2677,24 +2698,29 @@ int kswapd(void *p)
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0int new_classzone_idx;
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0int ret;
>
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 new_order =3D pgdat->kswapd_max_order= ;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 new_classzone_idx =3D pgdat->classzon= e_idx;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat->kswapd_max_order =3D 0;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat->classzone_idx =3D MAX_NR_ZONES= - 1;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (order < new_order || classzone_id= x > new_classzone_idx) {
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* Don't sleep if = someone wants a larger 'order'
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* allocation or has t= igher zone constraints
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0*/
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order =3D new_order;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 classzone_idx =3D new_cl= asszone_idx;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else {
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_try_to_sleep(kswa= pd_p, order, classzone_idx);
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order =3D pgdat->kswa= pd_max_order;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 classzone_idx =3D pgdat-= >classzone_idx;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (is_node_kswapd(kswapd_p)) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 new_order =3D pgdat->= kswapd_max_order;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 new_classzone_idx =3D pg= dat->classzone_idx;
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0pgdat->kswapd_max_or= der =3D 0;
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0pgdat->classzone_idx= =3D MAX_NR_ZONES - 1;
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 }
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (order < new_order= ||
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 classzone_idx > new_classzone_idx) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* Don= 't sleep if someone wants a larger 'order'
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* all= ocation or has tigher zone constraints
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0*/ > + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order = =3D new_order;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 classzon= e_idx =3D new_classzone_idx;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_t= ry_to_sleep(kswapd_p, order,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 classzone_idx);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order = =3D pgdat->kswapd_max_order;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 classzon= e_idx =3D pgdat->classzone_idx;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat-&g= t;kswapd_max_order =3D 0;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat-&g= t;classzone_idx =3D MAX_NR_ZONES - 1;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 }
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_try_to_sleep(kswa= pd_p, order, classzone_idx);
>
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0ret =3D try_to_freeze();
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0if (kthread_should_stop())
> @@ -2705,8 +2731,13 @@ int kswapd(void *p)
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 * after returning from the refrigerato= r
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 */
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0if (!ret) {
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 trace_mm_vmscan_kswapd_w= ake(pgdat->node_id, order);
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order =3D balance_pgdat(= pgdat, order, &classzone_idx);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (is_node_kswapd(kswap= d_p)) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 trace_mm= _vmscan_kswapd_wake(pgdat->node_id,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 order = =3D balance_pgdat(pgdat, order,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 &classzone_idx);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 balance_= mem_cgroup_pgdat(mem, order);
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0}
> =A0 =A0 =A0 =A0}
> =A0 =A0 =A0 =A0return 0;
> @@ -2853,30 +2884,53 @@ static int __devinit cpu_callback(struct notif= ier_block *nfb,
> =A0* This kswapd start function will be called by init and node-hot-ad= d.
> =A0* On node-hot-add, kswapd will moved to proper cpus if cpus are hot= -added.
> =A0*/
> -int kswapd_run(int nid)
> +int kswapd_run(int nid, struct mem_cgroup *mem)
> =A0{
> - =A0 =A0 =A0 pg_data_t *pgdat =3D NODE_DATA(nid);
> =A0 =A0 =A0 =A0struct task_struct *kswapd_thr;
> + =A0 =A0 =A0 pg_data_t *pgdat =3D NULL;
> =A0 =A0 =A0 =A0struct kswapd *kswapd_p;
> + =A0 =A0 =A0 static char name[TASK_COMM_LEN];
> + =A0 =A0 =A0 int memcg_id;
> =A0 =A0 =A0 =A0int ret =3D 0;
>
> - =A0 =A0 =A0 if (pgdat->kswapd_wait)
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 return 0;
> + =A0 =A0 =A0 if (!mem) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat =3D NODE_DATA(nid);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (pgdat->kswapd_wait)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 return ret;
> + =A0 =A0 =A0 }
>
> =A0 =A0 =A0 =A0kswapd_p =3D kzalloc(sizeof(struct kswapd), GFP_KERNEL)= ;
> =A0 =A0 =A0 =A0if (!kswapd_p)
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0return -ENOMEM;
>
> =A0 =A0 =A0 =A0init_waitqueue_head(&kswapd_p->kswapd_wait);
> - =A0 =A0 =A0 pgdat->kswapd_wait =3D &kswapd_p->kswapd_wait;=
> - =A0 =A0 =A0 kswapd_p->kswapd_pgdat =3D pgdat;
>
> - =A0 =A0 =A0 kswapd_thr =3D kthread_run(kswapd, kswapd_p, "kswap= d%d", nid);
> + =A0 =A0 =A0 if (!mem) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat->kswapd_wait =3D &kswapd_p-= >kswapd_wait;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_p->kswapd_pgdat =3D pgdat;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 snprintf(name, TASK_COMM_LEN, "kswa= pd_%d", nid);
> + =A0 =A0 =A0 } else {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 memcg_id =3D mem_cgroup_init_kswapd(mem,= kswapd_p);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (!memcg_id) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 kfree(kswapd_p);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 return ret;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 }
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 snprintf(name, TASK_COMM_LEN, "memc= g_%d", memcg_id);
> + =A0 =A0 =A0 }
> +
> + =A0 =A0 =A0 kswapd_thr =3D kthread_run(kswapd, kswapd_p, name);
> =A0 =A0 =A0 =A0if (IS_ERR(kswapd_thr)) {
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0/* failure at boot is fatal */
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0BUG_ON(system_state =3D=3D SYSTEM_BOOTI= NG);
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 printk("Failed to start kswapd on n= ode %d\n",nid);
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat->kswapd_wait =3D NULL;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (!mem) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 printk(KERN_ERR "Fa= iled to start kswapd on node %d\n",
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 nid);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 pgdat->kswapd_wait = =3D NULL;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 printk(KERN_ERR "Fa= iled to start kswapd on memcg %d\n",
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 memcg_id);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 mem_cgroup_clear_kswapd(= mem);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 }
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0kfree(kswapd_p);
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0ret =3D -1;
> =A0 =A0 =A0 =A0} else
> @@ -2887,16 +2941,18 @@ int kswapd_run(int nid)
> =A0/*
> =A0* Called by memory hotplug when all memory in a node is offlined. > =A0*/
> -void kswapd_stop(int nid)
> +void kswapd_stop(int nid, struct mem_cgroup *mem)
> =A0{
> =A0 =A0 =A0 =A0struct task_struct *kswapd_thr =3D NULL;
> =A0 =A0 =A0 =A0struct kswapd *kswapd_p =3D NULL;
> =A0 =A0 =A0 =A0wait_queue_head_t *wait;
>
> - =A0 =A0 =A0 pg_data_t *pgdat =3D NODE_DATA(nid);
> -
> =A0 =A0 =A0 =A0spin_lock(&kswapds_spinlock);
> - =A0 =A0 =A0 wait =3D pgdat->kswapd_wait;
> + =A0 =A0 =A0 if (!mem)
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 wait =3D NODE_DATA(nid)->kswapd_wait;=
> + =A0 =A0 =A0 else
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 wait =3D mem_cgroup_kswapd_wait(mem); > +
> =A0 =A0 =A0 =A0if (wait) {
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0kswapd_p =3D container_of(wait, struct = kswapd, kswapd_wait);
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0kswapd_thr =3D kswapd_p->kswapd_task= ;
> @@ -2916,7 +2972,7 @@ static int __init kswapd_init(void)
>
> =A0 =A0 =A0 =A0swap_setup();
> =A0 =A0 =A0 =A0for_each_node_state(nid, N_HIGH_MEMORY)
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_run(nid);
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 kswapd_run(nid, NULL);
> =A0 =A0 =A0 =A0hotcpu_notifier(cpu_callback, 0);
> =A0 =A0 =A0 =A0return 0;
> =A0}
> --
> 1.7.3.1
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in<= br> > the body to majordomo@kvack.org= . =A0For 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: <a href=3Dmailto:"dont@kvack.org"> emai= l@kvack.org </a>
>

--000e0cd37a7e37a35004a0db184f-- -- 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