From: Matthew Wilcox <willy@infradead.org>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: linux-mm@kvack.org, cgroups@vger.kernel.org,
Michal Hocko <mhocko@kernel.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [PATCH v3 13/18] mm/memcg: Add folio_memcg_lock() and folio_memcg_unlock()
Date: Fri, 9 Jul 2021 20:37:27 +0100 [thread overview]
Message-ID: <YOild62fNH5TwEpf@casper.infradead.org> (raw)
In-Reply-To: <YOYRYXATm2gHoGGq@cmpxchg.org>
On Wed, Jul 07, 2021 at 04:41:05PM -0400, Johannes Weiner wrote:
> On Wed, Jul 07, 2021 at 08:28:39PM +0100, Matthew Wilcox wrote:
> > On Wed, Jul 07, 2021 at 01:08:51PM -0400, Johannes Weiner wrote:
> > > On Wed, Jun 30, 2021 at 05:00:29AM +0100, Matthew Wilcox (Oracle) wrote:
> > > > -static void __unlock_page_memcg(struct mem_cgroup *memcg)
> > > > +static void __memcg_unlock(struct mem_cgroup *memcg)
> > >
> > > This is too generic a name. There are several locks in the memcg, and
> > > this one only locks the page->memcg bindings in the group.
> >
> > Fair. __memcg_move_unlock looks like the right name to me?
>
> Could you please elaborate what the problem with the current name is?
>
> mem_cgroup_move_account() does this:
>
> lock_page_memcg(page);
> page->memcg = to;
> __unlock_page_memcg(from);
>
> It locks and unlocks the page->memcg binding which can be done coming
> from the page or the memcg. The current names are symmetrical to
> reflect that it's the same lock.
OK, so in the prerequisite series to this patch, lock_page() becomes
folio_lock(). This series turns lock_page_memcg() into
folio_memcg_lock(). As a minimum, then, this needs to turn into
__folio_memcg_unlock().
> We could switch them both to move_lock, but as per the other email,
> lock_page_memcg() was chosen to resemble lock_page(). Because from a
> memcg POV they're interchangeable - the former is just a more narrowly
> scoped version for contexts that don't hold the page lock. It used to
> be called something else and we had several contexts taking redundant
> locks on accident because this hierarchy wasn't clear.
Unfortunately, it's still not clear. I've answered questions from
people who think that they have the page locked because they called
lock_page_memcg() ;-(
next prev parent reply other threads:[~2021-07-09 19:37 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-30 4:00 [PATCH v3 00/18] Folio conversion of memcg Matthew Wilcox (Oracle)
2021-06-30 4:00 ` [PATCH v3 01/18] mm: Add folio_nid() Matthew Wilcox (Oracle)
2021-07-01 6:56 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 02/18] mm/memcg: Remove 'page' parameter to mem_cgroup_charge_statistics() Matthew Wilcox (Oracle)
2021-06-30 14:17 ` Johannes Weiner
2021-06-30 4:00 ` [PATCH v3 03/18] mm/memcg: Use the node id in mem_cgroup_update_tree() Matthew Wilcox (Oracle)
2021-06-30 6:55 ` Michal Hocko
2021-06-30 14:18 ` Johannes Weiner
2021-07-01 6:57 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 04/18] mm/memcg: Remove soft_limit_tree_node() Matthew Wilcox (Oracle)
2021-06-30 6:56 ` Michal Hocko
2021-06-30 14:19 ` Johannes Weiner
2021-07-01 7:09 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 05/18] mm/memcg: Convert memcg_check_events to take a node ID Matthew Wilcox (Oracle)
2021-06-30 6:58 ` Michal Hocko
2021-06-30 6:59 ` Michal Hocko
2021-07-01 7:09 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 06/18] mm/memcg: Add folio_memcg() and related functions Matthew Wilcox (Oracle)
2021-06-30 6:53 ` kernel test robot
2021-07-01 7:12 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 07/18] mm/memcg: Convert commit_charge() to take a folio Matthew Wilcox (Oracle)
2021-06-30 4:00 ` [PATCH v3 08/18] mm/memcg: Convert mem_cgroup_charge() " Matthew Wilcox (Oracle)
2021-06-30 7:17 ` kernel test robot
2021-07-01 7:13 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 09/18] mm/memcg: Convert uncharge_page() to uncharge_folio() Matthew Wilcox (Oracle)
2021-07-01 7:15 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 10/18] mm/memcg: Convert mem_cgroup_uncharge() to take a folio Matthew Wilcox (Oracle)
2021-06-30 8:46 ` kernel test robot
2021-07-01 7:17 ` Christoph Hellwig
2021-07-07 12:09 ` Matthew Wilcox
2021-06-30 4:00 ` [PATCH v3 11/18] mm/memcg: Convert mem_cgroup_migrate() to take folios Matthew Wilcox (Oracle)
2021-07-01 7:20 ` Christoph Hellwig
2021-06-30 4:00 ` [PATCH v3 12/18] mm/memcg: Convert mem_cgroup_track_foreign_dirty_slowpath() to folio Matthew Wilcox (Oracle)
2021-06-30 4:00 ` [PATCH v3 13/18] mm/memcg: Add folio_memcg_lock() and folio_memcg_unlock() Matthew Wilcox (Oracle)
2021-06-30 8:32 ` Michal Hocko
2021-07-07 15:10 ` Matthew Wilcox
2021-07-08 7:28 ` Michal Hocko
2021-07-07 17:08 ` Johannes Weiner
2021-07-07 19:28 ` Matthew Wilcox
2021-07-07 20:41 ` Johannes Weiner
2021-07-09 19:37 ` Matthew Wilcox [this message]
2021-06-30 4:00 ` [PATCH v3 14/18] mm/memcg: Convert mem_cgroup_move_account() to use a folio Matthew Wilcox (Oracle)
2021-06-30 8:30 ` Michal Hocko
2021-06-30 11:22 ` Matthew Wilcox
2021-06-30 12:20 ` Michal Hocko
2021-06-30 12:31 ` Matthew Wilcox
2021-06-30 12:45 ` Michal Hocko
2021-07-07 15:25 ` Matthew Wilcox
2021-07-08 7:30 ` Michal Hocko
2021-06-30 4:00 ` [PATCH v3 15/18] mm/memcg: Add mem_cgroup_folio_lruvec() Matthew Wilcox (Oracle)
2021-06-30 8:12 ` kernel test robot
2021-06-30 19:18 ` Matthew Wilcox
2021-06-30 21:21 ` Johannes Weiner
2021-06-30 4:00 ` [PATCH v3 16/18] mm/memcg: Add folio_lruvec_lock() and similar functions Matthew Wilcox (Oracle)
2021-06-30 8:36 ` Michal Hocko
2021-06-30 4:00 ` [PATCH v3 17/18] mm/memcg: Add folio_lruvec_relock_irq() and folio_lruvec_relock_irqsave() Matthew Wilcox (Oracle)
2021-06-30 8:39 ` Michal Hocko
2021-06-30 4:00 ` [PATCH v3 18/18] mm/workingset: Convert workingset_activation to take a folio Matthew Wilcox (Oracle)
2021-06-30 8:44 ` [PATCH v3 00/18] Folio conversion of memcg Michal Hocko
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=YOild62fNH5TwEpf@casper.infradead.org \
--to=willy@infradead.org \
--cc=cgroups@vger.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=vdavydov.dev@gmail.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).