From: Johannes Weiner <hannes@cmpxchg.org>
To: Alex Shi <alex.shi@linux.alibaba.com>
Cc: cgroups@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, akpm@linux-foundation.org,
mgorman@techsingularity.net, tj@kernel.org, hughd@google.com,
khlebnikov@yandex-team.ru, daniel.m.jordan@oracle.com,
yang.shi@linux.alibaba.com, willy@infradead.org,
shakeelb@google.com, "Michal Hocko" <mhocko@kernel.org>,
"Vladimir Davydov" <vdavydov.dev@gmail.com>,
"Roman Gushchin" <guro@fb.com>,
"Chris Down" <chris@chrisdown.name>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Vlastimil Babka" <vbabka@suse.cz>, "Qian Cai" <cai@lca.pw>,
"Andrey Ryabinin" <aryabinin@virtuozzo.com>,
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
"Jérôme Glisse" <jglisse@redhat.com>,
"Andrea Arcangeli" <aarcange@redhat.com>,
"David Rientjes" <rientjes@google.com>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
swkhack <swkhack@gmail.com>,
"Potyra, Stefan" <Stefan.Potyra@elektrobit.com>,
"Mike Rapoport" <rppt@linux.vnet.ibm.com>,
"Stephen Rothwell" <sfr@canb.auug.org.au>,
"Colin Ian King" <colin.king@canonical.com>,
"Jason Gunthorpe" <jgg@ziepe.ca>,
"Mauro Carvalho Chehab" <mchehab+samsung@kernel.org>,
"Peng Fan" <peng.fan@nxp.com>,
"Nikolay Borisov" <nborisov@suse.com>,
"Ira Weiny" <ira.weiny@intel.com>,
"Kirill Tkhai" <ktkhai@virtuozzo.com>,
"Yafang Shao" <laoar.shao@gmail.com>
Subject: Re: [PATCH v8 03/10] mm/lru: replace pgdat lru_lock with lruvec lock
Date: Tue, 21 Jan 2020 11:00:05 -0500 [thread overview]
Message-ID: <20200121160005.GA69293@cmpxchg.org> (raw)
In-Reply-To: <9ee80b68-a78f-714a-c727-1f6d2b4f87ea@linux.alibaba.com>
On Mon, Jan 20, 2020 at 08:58:09PM +0800, Alex Shi wrote:
>
>
> 在 2020/1/17 上午5:52, Johannes Weiner 写道:
>
> > You simply cannot serialize on page->mem_cgroup->lruvec when
> > page->mem_cgroup isn't stable. You need to serialize on the page
> > itself, one way or another, to make this work.
> >
> >
> > So here is a crazy idea that may be worth exploring:
> >
> > Right now, pgdat->lru_lock protects both PageLRU *and* the lruvec's
> > linked list.
> >
> > Can we make PageLRU atomic and use it to stabilize the lru_lock
> > instead, and then use the lru_lock only serialize list operations?
> >
>
> Hi Johannes,
>
> I am trying to figure out the solution of atomic PageLRU, but is
> blocked by the following sitations, when PageLRU and lru list was protected
> together under lru_lock, the PageLRU could be a indicator if page on lru list
> But now seems it can't be the indicator anymore.
> Could you give more clues of stabilization usage of PageLRU?
There are two types of PageLRU checks: optimistic and deterministic.
The check in activate_page() for example is optimistic and the result
unstable, but that's okay, because if we miss a page here and there
it's not the end of the world.
But the check in __activate_page() is deterministic, because we need
to be sure before del_page_from_lru_list(). Currently it's made
deterministic by testing under the lock: whoever acquires the lock
first gets to touch the LRU state. The same can be done with an atomic
TestClearPagLRU: whoever clears the flag first gets to touch the LRU
state (the lock is then only acquired to not corrupt the linked list,
in case somebody adds or removes a different page at the same time).
I.e. in my proposal, if you want to get a stable read of PageLRU, you
have to clear it atomically. But AFAICS, everybody who currently does
need a stable read either already clears it or can easily be converted
to clear it and then set it again (like __activate_page and friends).
> __page_cache_release/release_pages/compaction __pagevec_lru_add
> if (TestClearPageLRU(page)) if (!PageLRU())
> lruvec_lock();
> list_add();
> lruvec_unlock();
> SetPageLRU() //position 1
> lock_page_lruvec_irqsave(page, &flags);
> del_page_from_lru_list(page, lruvec, ..);
> unlock_page_lruvec_irqrestore(lruvec, flags);
> SetPageLRU() //position 2
Hm, that's not how __pagevec_lru_add() looks. In fact,
__pagevec_lru_add_fn() has a BUG_ON(PageLRU).
That's because only one thread can own the isolation state at a time.
If PageLRU is set, only one thread can claim it. Right now, whoever
takes the lock first and clears it wins. When we replace it with
TestClearPageLRU, it's the same thing: only one thread can win.
And you cannot set PageLRU, unless you own it. Either you isolated the
page using TestClearPageLRU, or you allocated a new page.
So you can have multiple threads trying to isolate a page from the LRU
list, hence the atomic testclear. But no two threads should ever be
racing to add a page to the LRU list, because only one thread can own
the isolation state.
With the atomic PageLRU flag, the sequence would be this:
__pagevec_lru_add:
BUG_ON(PageLRU()) // Caller *must* own the isolation state
lruvec_lock() // The lruvec is stable, because changing
// page->mem_cgroup requires owning the
// isolation state (PageLRU) and we own it
list_add() // Linked list protected by lru_lock
lruvec_unlock()
SetPageLRU() // The page has been added to the linked
// list, give up our isolation state. Once
// this flag becomes visible, other threads
// can isolate the page from the LRU list
next prev parent reply other threads:[~2020-01-21 16:00 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-16 3:04 [PATCH v8 00/10] per lruvec lru_lock for memcg Alex Shi
2020-01-16 3:05 ` [PATCH v8 01/10] mm/vmscan: remove unnecessary lruvec adding Alex Shi
2020-01-16 3:05 ` [PATCH v8 02/10] mm/memcg: fold lock_page_lru into commit_charge Alex Shi
2020-01-16 3:05 ` [PATCH v8 03/10] mm/lru: replace pgdat lru_lock with lruvec lock Alex Shi
2020-01-16 21:52 ` Johannes Weiner
2020-01-19 11:32 ` Alex Shi
2020-01-20 12:58 ` Alex Shi
2020-01-21 16:00 ` Johannes Weiner [this message]
2020-01-22 12:01 ` Alex Shi
2020-01-22 18:31 ` Johannes Weiner
2020-04-13 10:48 ` Alex Shi
2020-04-13 18:07 ` Johannes Weiner
2020-04-14 4:52 ` Alex Shi
2020-04-14 16:31 ` Johannes Weiner
2020-04-15 13:42 ` Alex Shi
2020-04-16 8:01 ` Alex Shi
2020-04-16 15:28 ` Johannes Weiner
2020-04-16 17:47 ` Shakeel Butt
2020-04-17 13:18 ` Alex Shi
2020-04-17 14:39 ` Alex Shi
2020-04-14 8:19 ` Alex Shi
2020-04-14 16:36 ` Johannes Weiner
2020-01-16 3:05 ` [PATCH v8 04/10] mm/lru: introduce the relock_page_lruvec function Alex Shi
2020-01-16 3:05 ` [PATCH v8 05/10] mm/mlock: optimize munlock_pagevec by relocking Alex Shi
2020-01-16 3:05 ` [PATCH v8 06/10] mm/swap: only change the lru_lock iff page's lruvec is different Alex Shi
2020-01-16 3:05 ` [PATCH v8 07/10] mm/pgdat: remove pgdat lru_lock Alex Shi
2020-01-16 3:05 ` [PATCH v8 08/10] mm/lru: revise the comments of lru_lock Alex Shi
2020-01-16 3:05 ` [PATCH v8 09/10] mm/lru: add debug checking for page memcg moving Alex Shi
2020-01-16 3:05 ` [PATCH v8 10/10] mm/memcg: add debug checking in lock_page_memcg Alex Shi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200121160005.GA69293@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=Stefan.Potyra@elektrobit.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.shi@linux.alibaba.com \
--cc=aneesh.kumar@linux.ibm.com \
--cc=aryabinin@virtuozzo.com \
--cc=cai@lca.pw \
--cc=cgroups@vger.kernel.org \
--cc=chris@chrisdown.name \
--cc=colin.king@canonical.com \
--cc=daniel.m.jordan@oracle.com \
--cc=guro@fb.com \
--cc=hughd@google.com \
--cc=ira.weiny@intel.com \
--cc=jgg@ziepe.ca \
--cc=jglisse@redhat.com \
--cc=khlebnikov@yandex-team.ru \
--cc=kirill.shutemov@linux.intel.com \
--cc=ktkhai@virtuozzo.com \
--cc=laoar.shao@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mchehab+samsung@kernel.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@kernel.org \
--cc=nborisov@suse.com \
--cc=peng.fan@nxp.com \
--cc=rientjes@google.com \
--cc=rppt@linux.vnet.ibm.com \
--cc=sfr@canb.auug.org.au \
--cc=shakeelb@google.com \
--cc=swkhack@gmail.com \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=vbabka@suse.cz \
--cc=vdavydov.dev@gmail.com \
--cc=willy@infradead.org \
--cc=yang.shi@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).