linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Johannes Weiner <hannes@cmpxchg.org>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.cz>,
	Vladimir Davydov <vdavydov@parallels.com>,
	cgroups@vger.kernel.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: [patch 0/3] mm: memcontrol: lockless page counters v3
Date: Mon, 13 Oct 2014 21:46:00 -0400	[thread overview]
Message-ID: <1413251163-8517-1-git-send-email-hannes@cmpxchg.org> (raw)

Hi,

this series replaces the spinlock_irq-protected 64-bit res_counters
with lockless word-sized page counters.  This improves memory cgroup
scalability on the higher end and gets rid of 64-bit math on 32-bit
machines in the core accounting, but also generally simplifies the
counter code and interface.

Version 3 has a few more touch-ups based on reviews from Michal and
Vladimir:

    - documented ordering between limit() and try_charge() [vladimir]
    - removed charge revert during uncharge underflow [michal]
    - reworked limit update code flow [michal]
    - removed rounding-up in limit setting [michal]
    - fixed mem_cgroup_margin() memsw calculation [michal]
    - updated page_counter raciness comments [vladimir]

Version 2:

    cleanups:
    - moved new page_counter API to its own file [vladimir, michal]
    - documented page counter API [vladimir]
    - documented acceptable race conditions [vladimir]
    - split out res_counter removal to reduce patch size [vladimir]
    - split out hugetlb controller conversion to reduce patch size
    - split page_counter_charge and page_counter_try_charge [vladimir]
    - wrapped signed-to-unsigned read in page_counter_read() [vladimir]
    - wrapped watermark reset in page_counter_reset_watermark() [vladimir]
    - reverted counter->limited back to counter->failcnt [vladimir]
    - changed underflow to WARN_ON_ONCE and counter revert [kame, vladimir]

    fixes:
    - fixed kmem's value for unlimited [vladimir]
    - fixed page_counter_cancel() return value [vladimir]
    - based page counter range on atomic_long_t's max [vladimir]
    - fixed tcp memcontrol's usage reporting [vladimir]
    - fixed hugepage limit page alignment [vladimir]
    - fixed page_counter_limit() serialization [vladimir]

    optimizations:
    - converted page_counter_try_charge() from CAS to FAA [vladimir]

 Documentation/cgroups/hugetlb.txt          |   2 +-
 Documentation/cgroups/memory.txt           |   4 +-
 Documentation/cgroups/resource_counter.txt | 197 ---------
 include/linux/hugetlb_cgroup.h             |   1 -
 include/linux/memcontrol.h                 |   5 +-
 include/linux/page_counter.h               |  49 +++
 include/linux/res_counter.h                | 223 ----------
 include/net/sock.h                         |  26 +-
 init/Kconfig                               |  12 +-
 kernel/Makefile                            |   1 -
 kernel/res_counter.c                       | 211 ---------
 mm/Makefile                                |   1 +
 mm/hugetlb_cgroup.c                        | 103 +++--
 mm/memcontrol.c                            | 633 +++++++++++++--------------
 mm/page_counter.c                          | 203 +++++++++
 net/ipv4/tcp_memcontrol.c                  |  87 ++--
 16 files changed, 669 insertions(+), 1089 deletions(-)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

             reply	other threads:[~2014-10-14  1:46 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-14  1:46 Johannes Weiner [this message]
2014-10-14  1:46 ` [patch 1/3] mm: memcontrol: lockless page counters Johannes Weiner
2014-10-14 15:56   ` Michal Hocko
2014-10-14 16:33     ` Johannes Weiner
2014-10-15  9:40       ` Michal Hocko
2014-10-17  7:47   ` Vladimir Davydov
2014-10-14  1:46 ` [patch 2/3] mm: hugetlb_cgroup: convert to " Johannes Weiner
2014-10-14  1:46 ` [patch 3/3] kernel: res_counter: remove the unused API Johannes Weiner
2014-10-16  7:20   ` Paul Bolle
2014-10-16 11:20     ` Michal Hocko
2014-10-16 14:46       ` Johannes Weiner
2014-10-16 15:05         ` 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=1413251163-8517-1-git-send-email-hannes@cmpxchg.org \
    --to=hannes@cmpxchg.org \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.cz \
    --cc=vdavydov@parallels.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).