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 X-Spam-Level: X-Spam-Status: No, score=-15.1 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C59BAC4708C for ; Fri, 28 May 2021 10:28:02 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 4E4D8613B5 for ; Fri, 28 May 2021 10:28:02 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 4E4D8613B5 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=suse.cz Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id D03B06B006C; Fri, 28 May 2021 06:28:01 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id CB4026B006E; Fri, 28 May 2021 06:28:01 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id B2D196B0070; Fri, 28 May 2021 06:28:01 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0247.hostedemail.com [216.40.44.247]) by kanga.kvack.org (Postfix) with ESMTP id 840B76B006C for ; Fri, 28 May 2021 06:28:01 -0400 (EDT) Received: from smtpin33.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id 1CE83A2B8 for ; Fri, 28 May 2021 10:28:01 +0000 (UTC) X-FDA: 78190264362.33.E569933 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by imf01.hostedemail.com (Postfix) with ESMTP id D6CD6503BD26 for ; Fri, 28 May 2021 10:27:51 +0000 (UTC) Received: from imap.suse.de (imap-alt.suse-dmz.suse.de [192.168.254.47]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id 45F891FD2F; Fri, 28 May 2021 10:27:59 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_rsa; t=1622197679; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=MeSbALZmIII2yBXVbZ7zlQ01so3hhDxLJ3HUHGlfodo=; b=Iv5gaDbQ+fBLXM2H7Tl9O2Q0jFLvN0DMmwocQlOEJUg+76X3o0LPbz2dQ6qpC0I+L3h0Fk R2uNaDYi/rsXcK2OjmtEQGLZSgTs4/DjdxEC964QMqGxWrLhzAtMFqcbc5L29I8GI50c/R DFDKHcoN/4HH5k0kCib7g/5gTemQ+j8= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_ed25519; t=1622197679; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=MeSbALZmIII2yBXVbZ7zlQ01so3hhDxLJ3HUHGlfodo=; b=ZapyTHsk/d3QTM7phAtdC7lp9ZjEBUDQ1UDlWpV/LRxkzoFzXtczfpTUgdAk7HWnQCtnWV zMYp0n4vBar8laBA== Received: from director2.suse.de (director2.suse-dmz.suse.de [192.168.254.72]) by imap.suse.de (Postfix) with ESMTPSA id 2E6C611A98; Fri, 28 May 2021 10:27:59 +0000 (UTC) To: Mel Gorman Cc: Andrew Morton , Hillf Danton , Dave Hansen , Michal Hocko , LKML , Linux-MM References: <20210525080119.5455-1-mgorman@techsingularity.net> <20210525080119.5455-3-mgorman@techsingularity.net> <10cb326c-b4ad-3a82-a38b-aba7d2192736@suse.cz> <20210527105241.GB30378@techsingularity.net> From: Vlastimil Babka Subject: Re: [PATCH 2/6] mm/page_alloc: Disassociate the pcp->high from pcp->batch Message-ID: <9ccf113a-d292-2b34-2470-5a4e2ed4276e@suse.cz> Date: Fri, 28 May 2021 12:27:58 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.10.2 MIME-Version: 1.0 In-Reply-To: <20210527105241.GB30378@techsingularity.net> Content-Type: text/plain; charset=iso-8859-15 Content-Language: en-US X-Rspamd-Queue-Id: D6CD6503BD26 Authentication-Results: imf01.hostedemail.com; dkim=pass header.d=suse.cz header.s=susede2_rsa header.b=Iv5gaDbQ; dkim=pass header.d=suse.cz header.s=susede2_ed25519 header.b=ZapyTHsk; dmarc=none; spf=pass (imf01.hostedemail.com: domain of vbabka@suse.cz designates 195.135.220.29 as permitted sender) smtp.mailfrom=vbabka@suse.cz X-Rspamd-Server: rspam03 X-Stat-Signature: 9qk663tc1kj7ezc6mhqaw57tjpua4woj X-HE-Tag: 1622197671-586967 Content-Transfer-Encoding: quoted-printable 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: On 5/27/21 12:52 PM, Mel Gorman wrote: > On Wed, May 26, 2021 at 08:14:13PM +0200, Vlastimil Babka wrote: >> > @@ -6698,11 +6717,10 @@ static void __zone_set_pageset_high_and_batc= h(struct zone *zone, unsigned long h >> > */ >> > static void zone_set_pageset_high_and_batch(struct zone *zone) >> > { >> > - unsigned long new_high, new_batch; >> > + int new_high, new_batch; >> > =20 >> > - new_batch =3D zone_batchsize(zone); >> > - new_high =3D 6 * new_batch; >> > - new_batch =3D max(1UL, 1 * new_batch); >> > + new_batch =3D max(1, zone_batchsize(zone)); >> > + new_high =3D zone_highsize(zone, new_batch); >> > =20 >> > if (zone->pageset_high =3D=3D new_high && >> > zone->pageset_batch =3D=3D new_batch) >> > @@ -8170,6 +8188,12 @@ static void __setup_per_zone_wmarks(void) >> > zone->_watermark[WMARK_LOW] =3D min_wmark_pages(zone) + tmp; >> > zone->_watermark[WMARK_HIGH] =3D min_wmark_pages(zone) + tmp * 2; >> > =20 >> > + /* >> > + * The watermark size have changed so update the pcpu batch >> > + * and high limits or the limits may be inappropriate. >> > + */ >> > + zone_set_pageset_high_and_batch(zone); >>=20 >> Hm so this puts the call in the path of various watermark related sysc= tl >> handlers, but it's not protected by pcp_batch_high_lock. The zone lock= won't >> help against zone_pcp_update() from a hotplug handler. On the other ha= nd, since >> hotplug handlers also call __setup_per_zone_wmarks(), the zone_pcp_upd= ate() >> calls there are now redundant and could be removed, no? >> But later there will be a new sysctl in patch 6/6 using pcp_batch_high= _lock, >> thus that one will not be protected against the watermark related sysc= tl >> handlers that reach here. >>=20 >> To solve all this, seems like the static lock in setup_per_zone_wmarks= () could >> become a top-level visible lock and pcp high/batch updates could switc= h to that >> one instead of own pcp_batch_high_lock. And zone_pcp_update() calls fr= om hotplug >> handlers could be removed. >>=20 >=20 > Hmm, the locking has very different hold times. The static lock in > setup_per_zone_wmarks is a spinlock that protects against parallel upda= tes > of watermarks and is held for a short duration. The pcp_batch_high_lock > is a mutex that is held for a relatively long time while memory is bein= g > offlined and can sleep. Memory hotplug updates the watermarks without > pcp_batch_high_lock held so overall, unifying the locking there should > be a separate series. >=20 > How about this as a fix for this patch? >=20 > ---8<--- > mm/page_alloc: Disassociate the pcp->high from pcp->batch -fix >=20 > Vlastimil Babka noted that __setup_per_zone_wmarks updating pcp->high > did not protect watermark-related sysctl handlers from a parallel > memory hotplug operations. This patch moves the PCP update to > setup_per_zone_wmarks and updates the PCP high value while protected > by the pcp_batch_high_lock mutex. >=20 > This is a fix to the mmotm patch mm-page_alloc-disassociate-the-pcp-hig= h-from-pcp-batch.patch. > It'll cause a conflict with mm-page_alloc-adjust-pcp-high-after-cpu-hot= plug-events.patch > but the resolution is simply to change the caller in setup_per_zone_wma= rks > to zone_pcp_update(zone, 0) >=20 > Signed-off-by: Mel Gorman Looks fine. But I would also remove the redudancy introduced by this patc= h+fix, as part of the patch: online_pages() zone_pcp_update(zone); <- this predates the patch init_per_zone_wmark_min() setup_per_zone_wmarks() for_each_zone(zone) zone_pcp_update(zone); <- new in this patch offline_pages() similarly In any case, for the fixed version: Acked-by: Vlastimil Babka > --- > mm/page_alloc.c | 14 ++++++++------ > 1 file changed, 8 insertions(+), 6 deletions(-) >=20 > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 329b71e41db4..b1b3c66e9d88 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -8199,12 +8199,6 @@ static void __setup_per_zone_wmarks(void) > zone->_watermark[WMARK_LOW] =3D min_wmark_pages(zone) + tmp; > zone->_watermark[WMARK_HIGH] =3D min_wmark_pages(zone) + tmp * 2; > =20 > - /* > - * The watermark size have changed so update the pcpu batch > - * and high limits or the limits may be inappropriate. > - */ > - zone_set_pageset_high_and_batch(zone); > - > spin_unlock_irqrestore(&zone->lock, flags); > } > =20 > @@ -8221,11 +8215,19 @@ static void __setup_per_zone_wmarks(void) > */ > void setup_per_zone_wmarks(void) > { > + struct zone *zone; > static DEFINE_SPINLOCK(lock); > =20 > spin_lock(&lock); > __setup_per_zone_wmarks(); > spin_unlock(&lock); > + > + /* > + * The watermark size have changed so update the pcpu batch > + * and high limits or the limits may be inappropriate. > + */ > + for_each_zone(zone) > + zone_pcp_update(zone); > } > =20 > /* >=20