linux-kernel - Re: [PATCH v8 03/10] mm/lru: replace pgdat lru

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <54af0662-cbb4-88c7-7eae-f969684025dd@linux.alibaba.com>
Date:   Wed, 15 Apr 2020 21:42:26 +0800
From:   Alex Shi <alex.shi@...ux.alibaba.com>
To:     Johannes Weiner <hannes@...xchg.org>
Cc:     cgroups@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, akpm@...ux-foundation.org,
        mgorman@...hsingularity.net, tj@...nel.org, hughd@...gle.com,
        khlebnikov@...dex-team.ru, daniel.m.jordan@...cle.com,
        yang.shi@...ux.alibaba.com, willy@...radead.org,
        shakeelb@...gle.com, Michal Hocko <mhocko@...nel.org>,
        Vladimir Davydov <vdavydov.dev@...il.com>,
        Roman Gushchin <guro@...com>,
        Chris Down <chris@...isdown.name>,
        Thomas Gleixner <tglx@...utronix.de>,
        Vlastimil Babka <vbabka@...e.cz>, Qian Cai <cai@....pw>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
        Jérôme Glisse <jglisse@...hat.com>,
        Andrea Arcangeli <aarcange@...hat.com>,
        David Rientjes <rientjes@...gle.com>,
        "Aneesh Kumar K.V" <aneesh.kumar@...ux.ibm.com>,
        swkhack <swkhack@...il.com>,
        "Potyra, Stefan" <Stefan.Potyra@...ktrobit.com>,
        Mike Rapoport <rppt@...ux.vnet.ibm.com>,
        Stephen Rothwell <sfr@...b.auug.org.au>,
        Colin Ian King <colin.king@...onical.com>,
        Jason Gunthorpe <jgg@...pe.ca>,
        Mauro Carvalho Chehab <mchehab+samsung@...nel.org>,
        Peng Fan <peng.fan@....com>,
        Nikolay Borisov <nborisov@...e.com>,
        Ira Weiny <ira.weiny@...el.com>,
        Kirill Tkhai <ktkhai@...tuozzo.com>,
        Yafang Shao <laoar.shao@...il.com>
Subject: Re: [PATCH v8 03/10] mm/lru: replace pgdat lru_lock with lruvec lock



在 2020/4/15 上午12:31, Johannes Weiner 写道:
> On Tue, Apr 14, 2020 at 12:52:30PM +0800, Alex Shi wrote:
>> 在 2020/4/14 上午2:07, Johannes Weiner 写道:
>>> Plus, the overhead of tracking is tiny - 512k per G of swap (0.04%).
>>>
>>> Maybe we should just delete MEMCG_SWAP and unconditionally track swap
>>> entry ownership when the memory controller is enabled. I don't see a
>>> good reason not to, and it would simplify the entire swapin path, the
>>> LRU locking, and the page->mem_cgroup stabilization rules.
>>>
>>
>> Sorry for not follow you up, did you mean just remove the MEMCG_SWAP configuration
>> and keep the feature in default memcg? 
> 
> Yes.
> 
>> That does can remove lrucare, but PageLRU lock scheme still fails since
>> we can not isolate the page during commit_charge, is that right?
> 
> No, without lrucare the scheme works. Charges usually do:
> 
> page->mem_cgroup = new;
> SetPageLRU(page);
> 
> And so if you can TestClearPageLRU(), page->mem_cgroup is stable.
> 
> lrucare charging is the exception: it changes page->mem_cgroup AFTER
> PageLRU has already been set, and even when it CANNOT acquire the
> PageLRU lock itself. It violates the rules.
> 
> If we make MEMCG_SWAP mandatory, we always have cgroup records for
> swapped out pages. That means we can charge all swapin pages
> (incl. readahead pages) directly in __read_swap_cache_async(), before
> setting PageLRU on the new pages.
> 
> Then we can delete lrucare.
> 
> And then TestClearPageLRU() guarantees page->mem_cgroup is stable.
> 

Hi Johannes,

Thanks a lot for point out!

Charging in __read_swap_cache_async would ask for 3 layers function arguments
pass, that would be a bit ugly. Compare to this, could we move out the
lru_cache add after commit_charge, like ksm copied pages?

That give a bit extra non lru list time, but the page just only be used only
after add_anon_rmap setting. Could it cause troubles?

I tried to track down the reason of lru_cache_add here, but no explanation
till first git kernel commit.

Thanks
Alex