Message-ID: <alpine.LSU.2.11.2006072100390.2001@eggly.anvils>
Date:   Sun, 7 Jun 2020 21:15:21 -0700 (PDT)
From:   Hugh Dickins <hughd@...gle.com>
To:     Alex Shi <alex.shi@...ux.alibaba.com>
cc:     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, hannes@...xchg.org, lkp@...el.com,
        linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        cgroups@...r.kernel.org, shakeelb@...gle.com,
        iamjoonsoo.kim@....com, richard.weiyang@...il.com
Subject: Re: [PATCH v11 00/16] per memcg lru lock

On Thu, 28 May 2020, Alex Shi wrote:

> This is a new version which bases on linux-next 
> 
> Johannes Weiner has suggested:
> "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?
> ..."
> 
> With new memcg charge path and this solution, we could isolate
> LRU pages to exclusive visit them in compaction, page migration, reclaim,
> memcg move_accunt, huge page split etc scenarios while keeping pages' 
> memcg stable. Then possible to change per node lru locking to per memcg
> lru locking. As to pagevec_lru_move_fn funcs, it would be safe to let
> pages remain on lru list, lru lock could guard them for list integrity.
> 
> The patchset includes 3 parts:
> 1, some code cleanup and minimum optimization as a preparation.
> 2, use TestCleanPageLRU as page isolation's precondition
> 3, replace per node lru_lock with per memcg per node lru_lock
> 
> The 3rd part moves per node lru_lock into lruvec, thus bring a lru_lock for
> each of memcg per node. So on a large machine, each of memcg don't
> have to suffer from per node pgdat->lru_lock competition. They could go
> fast with their self lru_lock
> 
> Following Daniel Jordan's suggestion, I have run 208 'dd' with on 104
> containers on a 2s * 26cores * HT box with a modefied case:
> https://git.kernel.org/pub/scm/linux/kernel/git/wfg/vm-scalability.git/tree/case-lru-file-readtwice
> 
> With this patchset, the readtwice performance increased about 80%
> in concurrent containers.
> 
> Thanks Hugh Dickins and Konstantin Khlebnikov, they both brought this
> idea 8 years ago, and others who give comments as well: Daniel Jordan, 
> Mel Gorman, Shakeel Butt, Matthew Wilcox etc.
> 
> Thanks for Testing support from Intel 0day and Rong Chen, Fengguang Wu,
> and Yun Wang. Hugh Dickins also shared his kbuild-swap case. Thanks!
> 
> 
> Alex Shi (14):
>   mm/vmscan: remove unnecessary lruvec adding
>   mm/page_idle: no unlikely double check for idle page counting
>   mm/compaction: correct the comments of compact_defer_shift
>   mm/compaction: rename compact_deferred as compact_should_defer
>   mm/thp: move lru_add_page_tail func to huge_memory.c
>   mm/thp: clean up lru_add_page_tail
>   mm/thp: narrow lru locking
>   mm/memcg: add debug checking in lock_page_memcg
>   mm/lru: introduce TestClearPageLRU
>   mm/compaction: do page isolation first in compaction
>   mm/mlock: reorder isolation sequence during munlock
>   mm/lru: replace pgdat lru_lock with lruvec lock
>   mm/lru: introduce the relock_page_lruvec function
>   mm/pgdat: remove pgdat lru_lock
> 
> Hugh Dickins (2):
>   mm/vmscan: use relock for move_pages_to_lru
>   mm/lru: revise the comments of lru_lock
> 
>  Documentation/admin-guide/cgroup-v1/memcg_test.rst |  15 +-
>  Documentation/admin-guide/cgroup-v1/memory.rst     |   8 +-
>  Documentation/trace/events-kmem.rst                |   2 +-
>  Documentation/vm/unevictable-lru.rst               |  22 +--
>  include/linux/compaction.h                         |   4 +-
>  include/linux/memcontrol.h                         |  92 +++++++++++
>  include/linux/mm_types.h                           |   2 +-
>  include/linux/mmzone.h                             |   6 +-
>  include/linux/page-flags.h                         |   1 +
>  include/linux/swap.h                               |   4 +-
>  include/trace/events/compaction.h                  |   2 +-
>  mm/compaction.c                                    | 104 ++++++++-----
>  mm/filemap.c                                       |   4 +-
>  mm/huge_memory.c                                   |  51 +++++--
>  mm/memcontrol.c                                    |  87 ++++++++++-
>  mm/mlock.c                                         |  93 ++++++------
>  mm/mmzone.c                                        |   1 +
>  mm/page_alloc.c                                    |   1 -
>  mm/page_idle.c                                     |   8 -
>  mm/rmap.c                                          |   2 +-
>  mm/swap.c                                          | 112 ++++----------
>  mm/swap_state.c                                    |   6 +-
>  mm/vmscan.c                                        | 168 +++++++++++----------
>  mm/workingset.c                                    |   4 +-
>  24 files changed, 487 insertions(+), 312 deletions(-)

Hi Alex,

I didn't get to try v10 at all, waited until Johannes's preparatory
memcg swap cleanup was in mmotm; but I have spent a while thrashing
this v11, and can happily report that it is much better than v9 etc:
I believe this memcg lru_lock work will soon be ready for v5.9.

I've not yet found any flaw at the swapping end, but fixes are needed
for isolate_migratepages_block() and mem_cgroup_move_account(): I've
got a series of 4 fix patches to send you (I guess two to fold into
existing patches of yours, and two to keep as separate from me).

I haven't yet written the patch descriptions, will return to that
tomorrow.  I expect you will be preparing a v12 rebased on v5.8-rc1
or v5.8-rc2, and will be able to include these fixes in that.

Tomorrow...
Hugh

Powered by blists - more mailing lists