From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id D2A0BC00140 for ; Mon, 15 Aug 2022 20:19:24 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 4B0858D0002; Mon, 15 Aug 2022 16:19:24 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 45F8F8D0001; Mon, 15 Aug 2022 16:19:24 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 326A08D0002; Mon, 15 Aug 2022 16:19:24 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0013.hostedemail.com [216.40.44.13]) by kanga.kvack.org (Postfix) with ESMTP id 225EE8D0001 for ; Mon, 15 Aug 2022 16:19:24 -0400 (EDT) Received: from smtpin26.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id E8A6F160619 for ; Mon, 15 Aug 2022 20:19:23 +0000 (UTC) X-FDA: 79802941806.26.A14F622 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by imf18.hostedemail.com (Postfix) with ESMTP id 147DA1C001E for ; Mon, 15 Aug 2022 20:19:22 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1660594762; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=ovMy9UBv9rqtIlUqp5wGD+98w6B8X4YaLrAOR51fWzw=; b=LqtcW2D4W8mljHRFSv938Hi2eLuSKADB3REhE2Vaw5eUicL7UxRTauEPnrb337hn89Pnwn dh94nCb2JXv6S1doPJlqybscc1xImxfsK8w6PtlsnvAQ4UFKfkRcQfYJoE7JgVuXKbaH0n fqT3E4cjNvdNoK3+SQ76kD5MkDn289Y= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-358-jBoGdbQzNaeIU7lXcta-5A-1; Mon, 15 Aug 2022 16:19:19 -0400 X-MC-Unique: jBoGdbQzNaeIU7lXcta-5A-1 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.rdu2.redhat.com [10.11.54.1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id BC95F1019C95; Mon, 15 Aug 2022 20:19:18 +0000 (UTC) Received: from fuller.cnet (ovpn-112-2.gru2.redhat.com [10.97.112.2]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 32AE440CFD0A; Mon, 15 Aug 2022 20:19:18 +0000 (UTC) Received: by fuller.cnet (Postfix, from userid 1000) id 89043416D5C1; Mon, 15 Aug 2022 17:18:59 -0300 (-03) Date: Mon, 15 Aug 2022 17:18:59 -0300 From: Marcelo Tosatti To: Aaron Tomlin Cc: frederic@kernel.org, cl@linux.com, tglx@linutronix.de, mingo@kernel.org, peterz@infradead.org, pauld@redhat.com, neelx@redhat.com, oleksandr@natalenko.name, atomlin@atomlin.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH v6 1/2] mm/vmstat: Use per cpu variable to track a vmstat discrepancy Message-ID: References: <20220808194820.676246-1-atomlin@redhat.com> <20220808194820.676246-2-atomlin@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20220808194820.676246-2-atomlin@redhat.com> X-Scanned-By: MIMEDefang 2.84 on 10.11.54.1 ARC-Authentication-Results: i=1; imf18.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=LqtcW2D4; spf=pass (imf18.hostedemail.com: domain of mtosatti@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=mtosatti@redhat.com; dmarc=pass (policy=none) header.from=redhat.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1660594763; a=rsa-sha256; cv=none; b=D+6k2pohvyt8PywdtQzJ8mfQ1Gek/seRMUDYcW9kH3vsuJorjR/tlJzIQX5FPPDkxA8Pbz vzFHzuSQM83+oTplRj+TarFoykenEZdLHHiWWMSdJIjMq9V+DWSeULh8ZWz1WLFK5uFcxq 3dN4QtJNtF7nsXj864EZJtsHqpGxzsE= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1660594763; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=ovMy9UBv9rqtIlUqp5wGD+98w6B8X4YaLrAOR51fWzw=; b=a3RdgFZEOXKzwgxwnZXzZGEyBIeA3McNTAaJMxnuWQoEXKWmN05vFG/7KdPeHkL9yYB9g8 mu6taLGcilt2PNfbbx5dR7w3qRtLROuq/wiOzV9kAqnG0s0RJKBkNd0ZUqvEiKjewTYp4r qZ+HzqVNPxjkIUdx0eiHtFH6f2Gp7mI= Authentication-Results: imf18.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=LqtcW2D4; spf=pass (imf18.hostedemail.com: domain of mtosatti@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=mtosatti@redhat.com; dmarc=pass (policy=none) header.from=redhat.com X-Stat-Signature: gqigfpxufpf9r54ro43b4kgti9j1rdup X-Rspamd-Server: rspam09 X-Rspam-User: X-Rspamd-Queue-Id: 147DA1C001E X-HE-Tag: 1660594762-260024 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: Hi Aaron, On Mon, Aug 08, 2022 at 08:48:19PM +0100, Aaron Tomlin wrote: > This patch incorporates an idea from Marcelo's patch [1] where a > CPU-specific variable namely vmstat_dirty is used to indicate if > a vmstat imbalance is present for a given CPU. Therefore, at the > appropriate time, we can fold all the remaining differentials. > > [1]: https://lore.kernel.org/lkml/20220204173554.763888172@fedora.localdomain/ > > Signed-off-by: Aaron Tomlin > --- > mm/vmstat.c | 46 +++++++++++++++------------------------------- > 1 file changed, 15 insertions(+), 31 deletions(-) > > diff --git a/mm/vmstat.c b/mm/vmstat.c > index 373d2730fcf2..51564b7c85fe 100644 > --- a/mm/vmstat.c > +++ b/mm/vmstat.c > @@ -195,6 +195,12 @@ void fold_vm_numa_events(void) > #endif > > #ifdef CONFIG_SMP > +static DEFINE_PER_CPU_ALIGNED(bool, vmstat_dirty); > + > +static inline void mark_vmstat_dirty(void) > +{ > + this_cpu_write(vmstat_dirty, true); > +} > > int calculate_pressure_threshold(struct zone *zone) > { > @@ -367,6 +373,7 @@ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item, > x = 0; > } > __this_cpu_write(*p, x); > + mark_vmstat_dirty(); > > if (IS_ENABLED(CONFIG_PREEMPT_RT)) > preempt_enable(); > @@ -405,6 +412,7 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, > x = 0; > } > __this_cpu_write(*p, x); > + mark_vmstat_dirty(); > > if (IS_ENABLED(CONFIG_PREEMPT_RT)) > preempt_enable(); > @@ -603,6 +611,7 @@ static inline void mod_zone_state(struct zone *zone, > > if (z) > zone_page_state_add(z, zone, item); > + mark_vmstat_dirty(); > } > > void mod_zone_page_state(struct zone *zone, enum zone_stat_item item, > @@ -671,6 +680,7 @@ static inline void mod_node_state(struct pglist_data *pgdat, > > if (z) > node_page_state_add(z, pgdat, item); > + mark_vmstat_dirty(); > } > > void mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, > @@ -1873,6 +1883,7 @@ int sysctl_stat_interval __read_mostly = HZ; > static void refresh_vm_stats(struct work_struct *work) > { > refresh_cpu_vm_stats(true); Another task can be scheduled here, set data in zone->per_cpu_zonestats->vm_stat_diff (after refresh_cpu_vm_stats has cleared it to zero), and you'll end up marking per-CPU vmstat_dirty bool as false incorrectly. Guess it is necessary to disable preemption here. > + this_cpu_write(vmstat_dirty, false); > } > > int vmstat_refresh(struct ctl_table *table, int write, > @@ -1937,6 +1948,7 @@ int vmstat_refresh(struct ctl_table *table, int write, > static void vmstat_update(struct work_struct *w) > { > if (refresh_cpu_vm_stats(true)) { > + this_cpu_write(vmstat_dirty, false); Same problem. > /* > * Counters were updated so we expect more updates > * to occur in the future. Keep on running the > @@ -1948,35 +1960,6 @@ static void vmstat_update(struct work_struct *w) > } > } > > -/* > - * Check if the diffs for a certain cpu indicate that > - * an update is needed. > - */ > -static bool need_update(int cpu) > -{ > - pg_data_t *last_pgdat = NULL; > - struct zone *zone; > - > - for_each_populated_zone(zone) { > - struct per_cpu_zonestat *pzstats = per_cpu_ptr(zone->per_cpu_zonestats, cpu); > - struct per_cpu_nodestat *n; > - > - /* > - * The fast way of checking if there are any vmstat diffs. > - */ > - if (memchr_inv(pzstats->vm_stat_diff, 0, sizeof(pzstats->vm_stat_diff))) > - return true; > - > - if (last_pgdat == zone->zone_pgdat) > - continue; > - last_pgdat = zone->zone_pgdat; > - n = per_cpu_ptr(zone->zone_pgdat->per_cpu_nodestats, cpu); > - if (memchr_inv(n->vm_node_stat_diff, 0, sizeof(n->vm_node_stat_diff))) > - return true; > - } > - return false; > -} > - > /* > * Switch off vmstat processing and then fold all the remaining differentials > * until the diffs stay at zero. The function is used by NOHZ and can only be > @@ -1990,7 +1973,7 @@ void quiet_vmstat(void) > if (!delayed_work_pending(this_cpu_ptr(&vmstat_work))) > return; > > - if (!need_update(smp_processor_id())) > + if (!__this_cpu_read(vmstat_dirty)) > return; > > /* > @@ -2000,6 +1983,7 @@ void quiet_vmstat(void) > * vmstat_shepherd will take care about that for us. > */ > refresh_cpu_vm_stats(false); > + __this_cpu_write(vmstat_dirty, false); > } >From this context, irqs should be disabled. > /* > @@ -2021,7 +2005,7 @@ static void vmstat_shepherd(struct work_struct *w) > for_each_online_cpu(cpu) { > struct delayed_work *dw = &per_cpu(vmstat_work, cpu); > > - if (!delayed_work_pending(dw) && need_update(cpu)) > + if (!delayed_work_pending(dw) && per_cpu(vmstat_dirty, cpu)) > queue_delayed_work_on(cpu, mm_percpu_wq, dw, 0); > > cond_resched(); > -- > 2.37.1 > >