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=-16.5 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, MENTIONS_GIT_HOSTING,NICE_REPLY_A,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS, UNPARSEABLE_RELAY,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 330CCC433DF for ; Thu, 13 Aug 2020 01:47:39 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id D8E5A21775 for ; Thu, 13 Aug 2020 01:47:38 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D8E5A21775 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 5E7536B0031; Wed, 12 Aug 2020 21:47:38 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 570476B0032; Wed, 12 Aug 2020 21:47:38 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 437B98D0002; Wed, 12 Aug 2020 21:47:38 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0141.hostedemail.com [216.40.44.141]) by kanga.kvack.org (Postfix) with ESMTP id 23FD76B0031 for ; Wed, 12 Aug 2020 21:47:38 -0400 (EDT) Received: from smtpin05.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id D0A3C180AD804 for ; Thu, 13 Aug 2020 01:47:37 +0000 (UTC) X-FDA: 77143858554.05.sofa67_121064a26ff0 Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin05.hostedemail.com (Postfix) with ESMTP id 99E4E1802E6CF for ; Thu, 13 Aug 2020 01:47:37 +0000 (UTC) X-HE-Tag: sofa67_121064a26ff0 X-Filterd-Recvd-Size: 18074 Received: from out30-133.freemail.mail.aliyun.com (out30-133.freemail.mail.aliyun.com [115.124.30.133]) by imf28.hostedemail.com (Postfix) with ESMTP for ; Thu, 13 Aug 2020 01:47:35 +0000 (UTC) X-Alimail-AntiSpam:AC=PASS;BC=-1|-1;BR=01201311R211e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01f04397;MF=alex.shi@linux.alibaba.com;NM=1;PH=DS;RN=19;SR=0;TI=SMTPD_---0U5bewDD_1597283247; Received: from IT-FVFX43SYHV2H.local(mailfrom:alex.shi@linux.alibaba.com fp:SMTPD_---0U5bewDD_1597283247) by smtp.aliyun-inc.com(127.0.0.1); Thu, 13 Aug 2020 09:47:30 +0800 Subject: Re: [PATCH v17 14/21] mm/compaction: do page isolation first in compaction To: Alexander Duyck Cc: Andrew Morton , Mel Gorman , Tejun Heo , Hugh Dickins , Konstantin Khlebnikov , Daniel Jordan , Yang Shi , Matthew Wilcox , Johannes Weiner , kbuild test robot , linux-mm , LKML , cgroups@vger.kernel.org, Shakeel Butt , Joonsoo Kim , Wei Yang , "Kirill A. Shutemov" , Rong Chen References: <1595681998-19193-1-git-send-email-alex.shi@linux.alibaba.com> <1595681998-19193-15-git-send-email-alex.shi@linux.alibaba.com> <241ca157-104f-4f0d-7d5b-de394443788d@linux.alibaba.com> <8dbd004e-8eba-f1ec-a5eb-5dc551978936@linux.alibaba.com> <9581db48-cef3-788a-7f5a-8548fee56c13@linux.alibaba.com> From: Alex Shi Message-ID: <3828d045-17e4-16aa-f0e6-d5dda7ad6b1b@linux.alibaba.com> Date: Thu, 13 Aug 2020 09:46:51 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:68.0) Gecko/20100101 Thunderbird/68.7.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 X-Rspamd-Queue-Id: 99E4E1802E6CF X-Spamd-Result: default: False [0.00 / 100.00] X-Rspamd-Server: rspam04 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: =E5=9C=A8 2020/8/13 =E4=B8=8A=E5=8D=8812:51, Alexander Duyck =E5=86=99=E9= =81=93: > On Wed, Aug 12, 2020 at 4:44 AM Alex Shi w= rote: >> >> >> >> =E5=9C=A8 2020/8/11 =E4=B8=8B=E5=8D=8810:47, Alexander Duyck =E5=86=99= =E9=81=93: >>> On Tue, Aug 11, 2020 at 1:23 AM Alex Shi = wrote: >>>> >>>> >>>> >>>> =E5=9C=A8 2020/8/10 =E4=B8=8B=E5=8D=8810:41, Alexander Duyck =E5=86=99= =E9=81=93: >>>>> On Mon, Aug 10, 2020 at 6:10 AM Alex Shi wrote: >>>>>> >>>>>> >>>>>> >>>>>> =E5=9C=A8 2020/8/7 =E4=B8=8B=E5=8D=8810:51, Alexander Duyck =E5=86= =99=E9=81=93: >>>>>>> I wonder if this entire section shouldn't be restructured. This i= s the >>>>>>> only spot I can see where we are resetting the LRU flag instead o= f >>>>>>> pulling the page from the LRU list with the lock held. Looking ov= er >>>>>>> the code it seems like something like that should be possible. I = am >>>>>>> not sure the LRU lock is really protecting us in either the >>>>>>> PageCompound check nor the skip bits. It seems like holding a >>>>>>> reference on the page should prevent it from switching between >>>>>>> compound or not, and the skip bits are per pageblock with the LRU= bits >>>>>>> being per node/memcg which I would think implies that we could ha= ve >>>>>>> multiple LRU locks that could apply to a single skip bit. >>>>>> >>>>>> Hi Alexander, >>>>>> >>>>>> I don't find problem yet on compound or skip bit usage. Would you = clarify the >>>>>> issue do you concerned? >>>>>> >>>>>> Thanks! >>>>> >>>>> The point I was getting at is that the LRU lock is being used to >>>>> protect these and with your changes I don't think that makes sense >>>>> anymore. >>>>> >>>>> The skip bits are per-pageblock bits. With your change the LRU lock= is >>>>> now per memcg first and then per node. As such I do not believe it >>>>> really provides any sort of exclusive access to the skip bits. I st= ill >>>>> have to look into this more, but it seems like you need a lock per >>>>> either section or zone that can be used to protect those bits and d= eal >>>>> with this sooner rather than waiting until you have found an LRU pa= ge. >>>>> The one part that is confusing though is that the definition of the >>>>> skip bits seems to call out that they are a hint since they are not >>>>> protected by a lock, but that is exactly what has been happening he= re. >>>>> >>>> >>>> The skip bits are safe here, since even it race with other skip acti= on, >>>> It will still skip out. The skip action is try not to compaction too= much, >>>> not a exclusive action needs avoid race. >>> >>> That would be the case if it didn't have the impact that they >>> currently do on the compaction process. What I am getting at is that = a >>> race was introduced when you placed this test between the clearing of >>> the LRU flag and the actual pulling of the page from the LRU list. So >>> if you tested the skip bits before clearing the LRU flag then I would >>> be okay with the code, however because it is triggering an abort afte= r >> >> Hi Alexander, >> >> Thanks a lot for comments and suggestions! >> >> I have tried your suggestion: >> >> Signed-off-by: Alex Shi >> --- >> mm/compaction.c | 14 +++++++------- >> 1 file changed, 7 insertions(+), 7 deletions(-) >> >> diff --git a/mm/compaction.c b/mm/compaction.c >> index b99c96c4862d..6c881dee8c9a 100644 >> --- a/mm/compaction.c >> +++ b/mm/compaction.c >> @@ -988,6 +988,13 @@ static bool too_many_isolated(pg_data_t *pgdat) >> if (__isolate_lru_page_prepare(page, isolate_mode) !=3D= 0) >> goto isolate_fail_put; >> >> + /* Try get exclusive access under lock */ >> + if (!skip_updated) { >> + skip_updated =3D true; >> + if (test_and_set_skip(cc, page, low_pfn)) >> + goto isolate_fail_put; >> + } >> + >> /* Try isolate the page */ >> if (!TestClearPageLRU(page)) >> goto isolate_fail_put; >=20 > I would have made this much sooner. Probably before you call > get_page_unless_zero so as to avoid the unnecessary atomic operations. >=20 >> @@ -1006,13 +1013,6 @@ static bool too_many_isolated(pg_data_t *pgdat) >> >> lruvec_memcg_debug(lruvec, page); >> >> - /* Try get exclusive access under lock */ >> - if (!skip_updated) { >> - skip_updated =3D true; >> - if (test_and_set_skip(cc, page, low_pf= n)) >> - goto isolate_abort; >> - } >> - >> /* >> * Page become compound since the non-locked c= heck, >> * and it's on LRU. It can only be a THP so th= e order >> -- >> >> Performance of case-lru-file-mmap-read in vm-scalibity is dropped a bi= t. not >> helpful >=20 > So one issue with this change is that it is still too late to be of > much benefit. Really you should probably be doing this much sooner, > for example somewhere before the get_page_unless_zero(). Also the > thing that still has me scratching my head is the "Try get exclusive > access under lock" comment. The function declaration says this is > supposed to be a hint, but we were using the LRU lock to synchronize > it. I'm wondering if we should really be protecting this with the zone > lock since we are modifying the pageblock flags which also contain the > migration type value for the pageblock and are only modified while > holding the zone lock. zone lock is probability better. you can try and test. >=20 >>> the LRU flag is cleared then you are creating a situation where >>> multiple processes will be stomping all over each other as you can >>> have each thread essentially take a page via the LRU flag, but only >>> one thread will process a page and it could skip over all other pages >>> that preemptively had their LRU flag cleared. >> >> It increase a bit crowd here, but lru_lock do reduce some them, and sk= ip_bit >> could stop each other in a array check(bitmap). So compare to whole no= de >> lru_lock, the net profit is clear in patch 17. >=20 > My concern is that what you can end up with is multiple threads all > working over the same pageblock for isolation. With the old code the > LRU lock was used to make certain that test_and_set_skip was being > synchronized on the first page in the pageblock so you would only have > one thread going through and working a single pageblock. However after > your changes it doesn't seem like the test_and_set_skip has that > protection since only one thread will ever be able to successfully > call it for the first page in the pageblock assuming that the LRU flag > is set on the first page in the pageblock block. >=20 >>> >>> If you take a look at the test_and_set_skip the function only acts on >>> the pageblock aligned PFN for a given range. WIth the changes you hav= e >>> in place now that would mean that only one thread would ever actually >>> call this function anyway since the first PFN would take the LRU flag >>> so no other thread could follow through and test or set the bit as >> >> Is this good for only one process could do test_and_set_skip? is that >> the 'skip' meaning to be? >=20 > So only one thread really getting to fully use test_and_set_skip is > good, however the issue is that there is nothing to synchronize the > testing from the other threads. As a result the other threads could > have isolated other pages within the pageblock before the thread that > is calling test_and_set_skip will get to complete the setting of the > skip bit. This will result in isolation failures for the thread that > set the skip bit which may be undesirable behavior. >=20 > With the old code the threads were all synchronized on testing the > first PFN in the pageblock while holding the LRU lock and that is what > we lost. My concern is the cases where skip_on_failure =3D=3D true are > going to fail much more often now as the threads can easily interfere > with each other. I have a patch to fix this, which is on=20 https://github.com/alexshi/linux.git lrunext >=20 >>> well. The expectation before was that all threads would encounter thi= s >>> test and either proceed after setting the bit for the first PFN or >>> abort after testing the first PFN. With you changes only the first >>> thread actually runs this test and then it and the others will likely >>> encounter multiple failures as they are all clearing LRU bits >>> simultaneously and tripping each other up. That is why the skip bit >>> must have a test and set done before you even get to the point of >>> clearing the LRU flag. >> >> It make the things warse in my machine, would you like to have a try b= y yourself? >=20 > I plan to do that. I have already been working on a few things to > clean up and optimize your patch set further. I will try to submit an > RFC this evening so we can discuss. >=20 Glad to see your new code soon. Would you like do it base on=20 https://github.com/alexshi/linux.git lrunext >>> >>>>> The point I was getting at with the PageCompound check is that inst= ead >>>>> of needing the LRU lock you should be able to look at PageCompound = as >>>>> soon as you call get_page_unless_zero() and preempt the need to set >>>>> the LRU bit again. Instead of trying to rely on the LRU lock to >>>>> guarantee that the page hasn't been merged you could just rely on t= he >>>>> fact that you are holding a reference to it so it isn't going to >>>>> switch between being compound or order 0 since it cannot be freed. = It >>>>> spoils the idea I originally had of combining the logic for >>>>> get_page_unless_zero and TestClearPageLRU into a single function, b= ut >>>>> the advantage is you aren't clearing the LRU flag unless you are >>>>> actually going to pull the page from the LRU list. >>>> >>>> Sorry, I still can not follow you here. Compound code part is unchan= ged >>>> and follow the original logical. So would you like to pose a new cod= e to >>>> see if its works? >>> >>> No there are significant changes as you reordered all of the >>> operations. Prior to your change the LRU bit was checked, but not >>> cleared before testing for PageCompound. Now you are clearing it >>> before you are testing if it is a compound page. So if compaction is >>> running we will be seeing the pages in the LRU stay put, but the >>> compound bit flickering off and on if the compound page is encountere= d >>> with the wrong or NULL lruvec. What I was suggesting is that the >> >> The lruvec could be wrong or NULL here, that is the base stone of whol= e >> patchset. >=20 > Sorry I had a typo in my comment as well as it is the LRU bit that > will be flickering, not the compound. The goal here is to avoid > clearing the LRU bit unless we are sure we are going to take the > lruvec lock and pull the page from the list. >=20 >>> PageCompound test probably doesn't need to be concerned with the lock >>> after your changes. You could test it after you call >>> get_page_unless_zero() and before you call >>> __isolate_lru_page_prepare(). Instead of relying on the LRU lock to >>> protect us from the page switching between compound and not we would >>> be relying on the fact that we are holding a reference to the page so >>> it should not be freed and transition between compound or not. >>> >> >> I have tried the patch as your suggested, it has no clear help on perf= ormance >> on above vm-scaliblity case. Maybe it's due to we checked the same thi= ng >> before lock already. >> >> diff --git a/mm/compaction.c b/mm/compaction.c >> index b99c96c4862d..cf2ac5148001 100644 >> --- a/mm/compaction.c >> +++ b/mm/compaction.c >> @@ -985,6 +985,16 @@ static bool too_many_isolated(pg_data_t *pgdat) >> if (unlikely(!get_page_unless_zero(page))) >> goto isolate_fail; >> >> + /* >> + * Page become compound since the non-locked c= heck, >> + * and it's on LRU. It can only be a THP so th= e order >> + * is safe to read and it's 0 for tail pages. >> + */ >> + if (unlikely(PageCompound(page) && !cc->alloc_= contig)) { >> + low_pfn +=3D compound_nr(page) - 1; >> + goto isolate_fail_put; >> + } >> + >> if (__isolate_lru_page_prepare(page, isolate_mode) !=3D= 0) >> goto isolate_fail_put; >> >> @@ -1013,16 +1023,6 @@ static bool too_many_isolated(pg_data_t *pgdat) >> goto isolate_abort; >> } >> >> - /* >> - * Page become compound since the non-locked c= heck, >> - * and it's on LRU. It can only be a THP so th= e order >> - * is safe to read and it's 0 for tail pages. >> - */ >> - if (unlikely(PageCompound(page) && !cc->alloc_= contig)) { >> - low_pfn +=3D compound_nr(page) - 1; >> - SetPageLRU(page); >> - goto isolate_fail_put; >> - } >> } else >> rcu_read_unlock(); >> >=20 > So actually there is more we could do than just this. Specifically a > few lines below the rcu_read_lock there is yet another PageCompound > check that sets low_pfn yet again. So in theory we could combine both > of those and modify the code so you end up with something more like: > @@ -968,6 +974,16 @@ isolate_migratepages_block(struct compact_control > *cc, unsigned long low_pfn, > if (unlikely(!get_page_unless_zero(page))) > goto isolate_fail; >=20 > + if (PageCompound(page)) { > + const unsigned int order =3D compound_order(pag= e); > + > + if (likely(order < MAX_ORDER)) > + low_pfn +=3D (1UL << order) - 1; > + > + if (unlikely(!cc->alloc_contig)) > + goto isolate_fail_put; >=20 The current don't check this unless locked changed. But anyway check it every page may have no performance impact. + } > + > if (__isolate_lru_page_prepare(page, isolate_mode) !=3D= 0) > goto isolate_fail_put; >=20 > Doing this you would be more likely to skip over the entire compound > page in a single jump should you not be able to either take the LRU > bit or encounter a busy page in __isolate_Lru_page_prepare. I had > copied this bit from an earlier check and modified it as I was not > sure I can guarantee that this is a THP since we haven't taken the LRU > lock yet. However I believe the page cannot be split up while we are > holding the extra reference so the PageCompound flag and order should > not change until we call put_page. >=20 It looks like the lock_page protect this instead of get_page that just wo= rks after split func called. Thanks Alex