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=-0.9 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, SPF_HELO_NONE,SPF_PASS autolearn=no 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 99605C35249 for ; Wed, 5 Feb 2020 06:57:56 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 559C72085B for ; Wed, 5 Feb 2020 06:57:56 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="XBjf+Igg" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 559C72085B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id CBE936B000A; Wed, 5 Feb 2020 01:57:55 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id C4A5D6B000C; Wed, 5 Feb 2020 01:57:55 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id B0FC66B000D; Wed, 5 Feb 2020 01:57:55 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0025.hostedemail.com [216.40.44.25]) by kanga.kvack.org (Postfix) with ESMTP id 927506B000A for ; Wed, 5 Feb 2020 01:57:55 -0500 (EST) Received: from smtpin12.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id 4879C249E for ; Wed, 5 Feb 2020 06:57:55 +0000 (UTC) X-FDA: 76455168510.12.bomb73_7f52114255142 X-HE-Tag: bomb73_7f52114255142 X-Filterd-Recvd-Size: 8455 Received: from us-smtp-1.mimecast.com (us-smtp-delivery-1.mimecast.com [205.139.110.120]) by imf36.hostedemail.com (Postfix) with ESMTP for ; Wed, 5 Feb 2020 06:57:54 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1580885874; 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: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Xni7UVgQ/ooHAbX5cF9eiaIafJT935OiivgoDQ343cM=; b=XBjf+Iggbb6s517K5h3BJl/eYb1lpjBn8SkeVUJ/Q/p5dPeQsmdBN4G5mLFksEKYrcFjCc hIfea/fXy4mMaDpKvPnNfYRsVjH3ZkYcfkVepaxWS9XUq/5PDVMKgttbfBWZygs/nLiHW8 8Kuqb3HvYFbYNUbPDrXEInUhAePpIF8= Received: from mail-qk1-f198.google.com (mail-qk1-f198.google.com [209.85.222.198]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-174--2n3EbXlPRCICIQa3RocLQ-1; Wed, 05 Feb 2020 01:57:52 -0500 Received: by mail-qk1-f198.google.com with SMTP id b23so646395qkg.17 for ; Tue, 04 Feb 2020 22:57:52 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:from:to:cc:subject:message-id:references :mime-version:content-disposition:content-transfer-encoding :in-reply-to; bh=7YGEo6r1q7a3oPDK4t6yazkdvlNaHSSuGWyn/XwzK5Y=; b=DI+CXEA2aOvwxc0VkQgiGIiwl9DAYrmHlVO9qVrM0wCL5GYM+YVSuayUuxlBJhhmaB ++nPh4WfKGzAU+3tjPrShjSwDowbhfsG+yFp1ZYNEVeKIjWCcXr1T7t7T55fF3LcF7Q7 GhNT1M82K3pnaCurHdedRt59QbBzHloOFfceU3O46gCzlyp7faPjB1iXeuJJZMWgWIW+ hADuALUt379WINMH24VUpGe98Q2QZxm4lBWuYLhZzxTaHJONlxa9G+v8ujihM3q3YOub tTPNZIl2Mp4YkH6rJtf1bRRJ8KeqsF5wnP+YhQHx/JugeNz3gYFGMM48TtEnlfBNFFKs /tvA== X-Gm-Message-State: APjAAAVsZsdXLcZDOruuIPv/1e//NZveiN/BmurLPSLW4t+Ckp0RaajV nGPwZ3IHz7+sUKFm9rbuuMPqet3vnDMLMx3crE+55DEqM355eJ0f+TY/Ym0+R5oOptnxK1JLbY9 MSj0c4sghakY= X-Received: by 2002:a05:620a:557:: with SMTP id o23mr32943454qko.246.1580885871724; Tue, 04 Feb 2020 22:57:51 -0800 (PST) X-Google-Smtp-Source: APXvYqyvwexytbZapMU9lmm33cMGazuF+4PG7l7eRe7Qx04MX+AWz5t2Gc8YiWF7CKjVQsiMg2YOgQ== X-Received: by 2002:a05:620a:557:: with SMTP id o23mr32943432qko.246.1580885871460; Tue, 04 Feb 2020 22:57:51 -0800 (PST) Received: from redhat.com (bzq-79-176-41-183.red.bezeqint.net. [79.176.41.183]) by smtp.gmail.com with ESMTPSA id x126sm12593612qkc.42.2020.02.04.22.57.48 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 04 Feb 2020 22:57:50 -0800 (PST) Date: Wed, 5 Feb 2020 01:57:45 -0500 From: "Michael S. Tsirkin" To: Tyler Sanderson Cc: David Hildenbrand , Alexander Duyck , "Wang, Wei W" , "virtualization@lists.linux-foundation.org" , David Rientjes , "linux-mm@kvack.org" , Michal Hocko , namit@vmware.com Subject: Re: Balloon pressuring page cache Message-ID: <20200205015057-mutt-send-email-mst@kernel.org> References: <20200203080520-mutt-send-email-mst@kernel.org> <5ac131de8e3b7fc1fafd05a61feb5f6889aeb917.camel@linux.intel.com> <20200203120225-mutt-send-email-mst@kernel.org> <74cc25a6-cefb-c580-8e59-5b76fb680bf4@redhat.com> MIME-Version: 1.0 In-Reply-To: X-MC-Unique: -2n3EbXlPRCICIQa3RocLQ-1 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline 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 Tue, Feb 04, 2020 at 03:58:51PM -0800, Tyler Sanderson wrote: > >=A0 =A0 =A0> > >=A0 =A0 =A0>=A0 1. It is last-resort, which means the system has alr= eady gone through > >=A0 =A0 =A0>=A0 =A0 =A0heroics to prevent OOM. Those heroic reclaim = efforts are expensive > >=A0 =A0 =A0>=A0 =A0 =A0and impact application performance. > > > >=A0 =A0 =A0That's *exactly* what "deflate on OOM" suggests. > > > > > > It seems there are some use cases where "deflate on OOM" is desired= and > > others where "deflate on pressure" is desired. > > This suggests adding a new feature bit "DEFLATE_ON_PRESSURE" that > > registers the shrinker, and reverting DEFLATE_ON_OOM to use the OOM > > notifier callback. > > > > This lets users configure the balloon for their use case. >=20 > You want the old behavior back, so why should we introduce a new one?= Or > am I missing something? (you did want us to revert to old handling, n= o?) >=20 > Reverting actually doesn't help me because this has been the behavior sin= ce > Linux 4.19 which is already widely in use. So my device implementation ne= eds to > handle the shrinker behavior anyways. I started this conversation to ask = what > the intended device implementation was. >=20 > I think there are reasonable device implementations that would prefer the > shrinker=A0behavior (it turns out that mine doesn't). > For example, an implementation that slowly inflates the balloon for the p= urpose > of memory overcommit. It might leave the balloon inflated and expect any = memory > pressure (including page cache usage) to deflate the balloon as a way to > dynamically right-size the balloon. So just to make sure we understand, what exactly does your implementation do? > Two reasons I didn't go with the above implementation: > 1. I need to support guests before Linux 4.19 which don't have the shrink= er > behavior. > 2. Memory in the balloon does not appear as "available" in /proc/meminfo = even > though it is freeable. This is confusing to users, but isn't a deal break= er. >=20 > If we added a DEFLATE_ON_PRESSURE feature bit that indicated shrinker API > support then that would resolve reason=A0#1 (ideally we would backport th= e bit to > 4.19). We could declare lack of pagecache pressure with DEFLATE_ON_OOM a regression and backport the revert but not I think the new DEFLATE_ON_PRESSURE. > In any case, the shrinker=A0behavior when pressuring page cache is more o= f an > inefficiency than a bug. It's not clear to me that it necessitates revert= ing. > If there were/are reasons to be on the shrinker=A0interface then I think = those > carry similar weight as the problem itself. > =A0 >=20 >=20 > I consider virtio-balloon to this very day a big hack. And I don't se= e > it getting better with new config knobs. Having that said, the > technologies that are candidates to replace it (free page reporting, > taming the guest page cache, etc.) are still not ready - so we'll hav= e > to stick with it for now :( . >=20 > > > > I'm actually not sure how you would safely do memory overcommit wit= hout > > DEFLATE_ON_OOM. So I think it unlocks a huge use case. >=20 > Using better suited technologies that are not ready yet (well, some f= orm > of free page reporting is available under IBM z already but in a > proprietary form) ;) Anyhow, I remember that DEFLATE_ON_OOM only make= s > it less likely to crash your guest, but not that you are safe to sque= eze > the last bit out of your guest VM. >=20 > Can you elaborate on the danger of DEFLATE_ON_OOM? I haven't seen any pro= blems > in testing but I'd really like to know about the dangers. > Is there a difference in safety between the OOM notifier callback and the > shrinker API? It's not about dangers as such. It's just that when linux hits OOM all kind of error paths are being hit, latent bugs start triggering, latency goes up drastically. >=20 >=20 > -- > Thanks, >=20 > David / dhildenb >=20 >=20 From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: Balloon pressuring page cache Date: Wed, 5 Feb 2020 01:57:45 -0500 Message-ID: <20200205015057-mutt-send-email-mst@kernel.org> References: <20200203080520-mutt-send-email-mst@kernel.org> <5ac131de8e3b7fc1fafd05a61feb5f6889aeb917.camel@linux.intel.com> <20200203120225-mutt-send-email-mst@kernel.org> <74cc25a6-cefb-c580-8e59-5b76fb680bf4@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: Content-Disposition: inline List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: virtualization-bounces@lists.linux-foundation.org Sender: "Virtualization" To: Tyler Sanderson Cc: "virtualization@lists.linux-foundation.org" , "linux-mm@kvack.org" , namit@vmware.com, David Rientjes , Alexander Duyck , Michal Hocko List-Id: virtualization@lists.linuxfoundation.org On Tue, Feb 04, 2020 at 03:58:51PM -0800, Tyler Sanderson wrote: > >=C2=A0 =C2=A0 =C2=A0> > >=C2=A0 =C2=A0 =C2=A0>=C2=A0 1. It is last-resort, which means the sy= stem has already gone through > >=C2=A0 =C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A0heroics to prevent OOM. Tho= se heroic reclaim efforts are expensive > >=C2=A0 =C2=A0 =C2=A0>=C2=A0 =C2=A0 =C2=A0and impact application perf= ormance. > > > >=C2=A0 =C2=A0 =C2=A0That's *exactly* what "deflate on OOM" suggests. > > > > > > It seems there are some use cases where "deflate on OOM" is desired= and > > others where "deflate on pressure" is desired. > > This suggests adding a new feature bit "DEFLATE_ON_PRESSURE" that > > registers the shrinker, and reverting DEFLATE_ON_OOM to use the OOM > > notifier callback. > > > > This lets users configure the balloon for their use case. > = > You want the old behavior back, so why should we introduce a new one?= Or > am I missing something? (you did want us to revert to old handling, n= o?) > = > Reverting actually doesn't help me because this has been the behavior sin= ce > Linux 4.19 which is already widely in use. So my device implementation ne= eds to > handle the shrinker behavior anyways. I started this conversation to ask = what > the intended device implementation was. > = > I think there are reasonable device implementations that would prefer the > shrinker=C2=A0behavior (it turns out that mine doesn't). > For example, an implementation that slowly inflates the balloon for the p= urpose > of memory overcommit. It might leave the balloon inflated and expect any = memory > pressure (including page cache usage) to deflate the balloon as a way to > dynamically right-size the balloon. So just to make sure we understand, what exactly does your implementation do? > Two reasons I didn't go with the above implementation: > 1. I need to support guests before Linux 4.19 which don't have the shrink= er > behavior. > 2. Memory in the balloon does not appear as "available" in /proc/meminfo = even > though it is freeable. This is confusing to users, but isn't a deal break= er. > = > If we added a DEFLATE_ON_PRESSURE feature bit that indicated shrinker API > support then that would resolve reason=C2=A0#1 (ideally we would backport= the bit to > 4.19). We could declare lack of pagecache pressure with DEFLATE_ON_OOM a regression and backport the revert but not I think the new DEFLATE_ON_PRESSURE. > In any case, the shrinker=C2=A0behavior when pressuring page cache is mor= e of an > inefficiency than a bug. It's not clear to me that it necessitates revert= ing. > If there were/are reasons to be on the shrinker=C2=A0interface then I thi= nk those > carry similar weight as the problem itself. > =C2=A0 > = > = > I consider virtio-balloon to this very day a big hack. And I don't see > it getting better with new config knobs. Having that said, the > technologies that are candidates to replace it (free page reporting, > taming the guest page cache, etc.) are still not ready - so we'll have > to stick with it for now :( . > = > > > > I'm actually not sure how you would safely do memory overcommit wit= hout > > DEFLATE_ON_OOM. So I think it unlocks a huge use case. > = > Using better suited technologies that are not ready yet (well, some f= orm > of free page reporting is available under IBM z already but in a > proprietary form) ;) Anyhow, I remember that DEFLATE_ON_OOM only makes > it less likely to crash your guest, but not that you are safe to sque= eze > the last bit out of your guest VM. > = > Can you elaborate on the danger of DEFLATE_ON_OOM? I haven't seen any pro= blems > in testing but I'd really like to know about the dangers. > Is there a difference in safety between the OOM notifier callback and the > shrinker API? It's not about dangers as such. It's just that when linux hits OOM all kind of error paths are being hit, latent bugs start triggering, latency goes up drastically. > = > = > -- > Thanks, > = > David / dhildenb > = > =