From mboxrd@z Thu Jan 1 00:00:00 1970 From: Shakeel Butt Subject: Re: [PATCH v2 1/2] memcg: flush stats only if updated Date: Thu, 14 Oct 2021 09:31:46 -0700 Message-ID: <20211014163146.2177266-1-shakeelb@google.com> References: <20211013180130.GB22036@blackbody.suse.cz> Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=date:in-reply-to:message-id:mime-version:references:subject:from:to :cc:content-transfer-encoding; bh=5c2jin9rhCwDmbScWWGB5EKaPi0t1d9qwSSLJb9JZsg=; b=I3cPTF8SIEgtc1ITsYhyR4jjKtKaIy5AuU9x5dsANrtt1TFE8Pt9H2bU2qV8pZP0Y+ iW5cOuNg86tSRYu6pNhMdMY+Q8lxtPwRKVUGYUsPK1C2KrAuWc7J4aSPvVpZksupkqSC JBCn2hrqSeMWclLfLWz1CTneP6GO3TyKg0EvNIvg+lJENLbrmcbE5S/paSMIxNr9SR5C Fw+ueGSjfyi/rSu4HCn05sX4bLasZnsx0wPVYmQN/BwQWIJPb9I/UAbY1bsb+U5oxggi 2kWwq+I56+jD+Ri8j1dSPUhr8bLDHdElkJLfAuB01MPNYY/2wdSnC/XNJWPTtddJGzxP inqA== In-Reply-To: <20211013180130.GB22036-9OudH3eul5jcvrawFnH+a6VXKuFTiq87@public.gmane.org> List-ID: Content-Type: text/plain; charset="utf-8" To: mkoutny-IBi9RG/b67k@public.gmane.org Cc: akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org Hi Michal, On Wed, Oct 13, 2021 at 11:01 AM Michal Koutn=C3=BD wrot= e: > > On Fri, Oct 01, 2021 at 12:00:39PM -0700, Shakeel Butt wrote: > > In this patch we kept the stats update codepath very minimal and let th= e > > stats reader side to flush the stats only when the updates are over a > > specific threshold. =C2=A0For now the threshold is (nr_cpus * CHARGE_BA= TCH). > > BTW, a noob question -- are the updates always single page sized? > > This is motivated by apples vs oranges comparison since the > =C2=A0 =C2=A0 =C2=A0 =C2=A0 nr_cpus * MEMCG_CHARGE_BATCH > suggests what could the expected error be in pages (bytes). But it's most= ly > wrong since: a) uncertain single-page updates, b) various counter > updates summed together. I wonder whether the formula can serve to > provide at least some (upper) estimate. > Thanks for your review. This forces me to think more on this because each update does not necessarily be a single page sized update e.g. adding a hug= epage to an LRU. Though I think the error is time bounded by 2 seconds but in those 2 second= s mathematically the error can be large. What do you think of the following change? It will bound the error better within the 2 seconds window. >From e87a36eedd02b0d10d8f66f83833bd6e2bae17b8 Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Thu, 14 Oct 2021 08:49:06 -0700 Subject: [PATCH] Better bounds on the stats error --- mm/memcontrol.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 8f1d9c028897..e5d5c850a521 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -626,14 +626,20 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_= tree_per_node *mctz) static void flush_memcg_stats_dwork(struct work_struct *w); static DECLARE_DEFERRABLE_WORK(stats_flush_dwork, flush_memcg_stats_dwork)= ; static DEFINE_SPINLOCK(stats_flush_lock); -static DEFINE_PER_CPU(unsigned int, stats_updates); +static DEFINE_PER_CPU(int, stats_diff); static atomic_t stats_flush_threshold =3D ATOMIC_INIT(0); =20 -static inline void memcg_rstat_updated(struct mem_cgroup *memcg) +static inline void memcg_rstat_updated(struct mem_cgroup *memcg, int val) { + unsigned int x; + cgroup_rstat_updated(memcg->css.cgroup, smp_processor_id()); - if (!(__this_cpu_inc_return(stats_updates) % MEMCG_CHARGE_BATCH)) - atomic_inc(&stats_flush_threshold); + + x =3D abs(__this_cpu_add_return(stats_diff, val)); + if (x > MEMCG_CHARGE_BATCH) { + atomic_add(x / MEMCG_CHARGE_BATCH, &stats_flush_threshold); + __this_cpu_write(stats_diff, 0); + } } =20 static void __mem_cgroup_flush_stats(void) @@ -672,7 +678,7 @@ void __mod_memcg_state(struct mem_cgroup *memcg, int id= x, int val) return; =20 __this_cpu_add(memcg->vmstats_percpu->state[idx], val); - memcg_rstat_updated(memcg); + memcg_rstat_updated(memcg, val); } =20 /* idx can be of type enum memcg_stat_item or node_stat_item. */ @@ -705,7 +711,7 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, en= um node_stat_item idx, /* Update lruvec */ __this_cpu_add(pn->lruvec_stats_percpu->state[idx], val); =20 - memcg_rstat_updated(memcg); + memcg_rstat_updated(memcg, val); } =20 /** @@ -807,7 +813,7 @@ void __count_memcg_events(struct mem_cgroup *memcg, enu= m vm_event_item idx, return; =20 __this_cpu_add(memcg->vmstats_percpu->events[idx], count); - memcg_rstat_updated(memcg); + memcg_rstat_updated(memcg, val); } =20 static unsigned long memcg_events(struct mem_cgroup *memcg, int event) --=20 2.33.0.882.g93a45727a2-goog