linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
To: Arun KS <arunks@codeaurora.org>
Cc: keescook@chromium.org, minchan@kernel.org, getarunks@gmail.com,
	gregkh@linuxfoundation.org, akpm@linux-foundation.org,
	mhocko@kernel.org, vbabka@suse.cz, linux-kernel@vger.kernel.org,
	linux-mm@kvack.org, julia.lawall@lip6.fr
Subject: Re: [PATCH v1 0/4]mm: convert totalram_pages, totalhigh_pages and managed pages to atomic
Date: Wed, 7 Nov 2018 10:02:55 +0300	[thread overview]
Message-ID: <c3b0edf9-e6a2-c1ab-8490-d94b9830c8ae@yandex-team.ru> (raw)
In-Reply-To: <22fa2222012341a54f6b0b6aea341aa2@codeaurora.org>

On 06.11.2018 11:43, Arun KS wrote:
> On 2018-11-06 14:07, Konstantin Khlebnikov wrote:
>> On 06.11.2018 11:30, Arun KS wrote:
>>> On 2018-11-06 13:47, Konstantin Khlebnikov wrote:
>>>> On 06.11.2018 8:38, Arun KS wrote:
>>>>> Any comments?
>>>>
>>>> Looks good.
>>>> Except unclear motivation behind this change.
>>>> This should be in comment of one of patch.
>>>
>>> totalram_pages, zone->managed_pages and totalhigh_pages are sometimes modified outside managed_page_count_lock. Hence convert these 
>>> variable to atomic to avoid readers potentially seeing a store tear.
>>
>> So, this is just theoretical issue or splat from sanitizer.
>> After boot memory online\offline are strictly serialized by rw-semaphore.
> 
> Few instances which can race with hot add. Please see below,
> https://patchwork.kernel.org/patch/10627521/
Could you point what exactly are you fixing with this set?

from v2:

 > totalram_pages, zone->managed_pages and totalhigh_pages updates
 > are protected by managed_page_count_lock, but readers never care
 > about it. Convert these variables to atomic to avoid readers
 > potentially seeing a store tear.

This?


Aligned unsigned long almost always stored at once.

To make it completely correct you could replace

a += b;

with

WRITE_ONCE(a, a + b);

> 
> Regards,
> Arun
> 
>>
>>>
>>> Will update the comment.
>>>
>>> Regards,
>>> Arun
>>>
>>>>
>>>> Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
>>>>
>>>>>
>>>>> Regards,
>>>>> Arun
>>>>>
>>>>> On 2018-10-26 16:30, Arun KS wrote:
>>>>>> This series convert totalram_pages, totalhigh_pages and
>>>>>> zone->managed_pages to atomic variables.
>>>>>>
>>>>>> The patch was comiple tested on x86(x86_64_defconfig & i386_defconfig)
>>>>>> on tip of linux-mmotm. And memory hotplug tested on arm64, but on an
>>>>>> older version of kernel.
>>>>>>
>>>>>> Arun KS (4):
>>>>>>   mm: Fix multiple evaluvations of totalram_pages and managed_pages
>>>>>>   mm: Convert zone->managed_pages to atomic variable
>>>>>>   mm: convert totalram_pages and totalhigh_pages variables to atomic
>>>>>>   mm: Remove managed_page_count spinlock
>>>>>>
>>>>>>  arch/csky/mm/init.c                           |  4 +-
>>>>>>  arch/powerpc/platforms/pseries/cmm.c          | 10 ++--
>>>>>>  arch/s390/mm/init.c                           |  2 +-
>>>>>>  arch/um/kernel/mem.c                          |  3 +-
>>>>>>  arch/x86/kernel/cpu/microcode/core.c          |  5 +-
>>>>>>  drivers/char/agp/backend.c                    |  4 +-
>>>>>>  drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  2 +-
>>>>>>  drivers/gpu/drm/i915/i915_gem.c               |  2 +-
>>>>>>  drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  4 +-
>>>>>>  drivers/hv/hv_balloon.c                       | 19 +++----
>>>>>>  drivers/md/dm-bufio.c                         |  2 +-
>>>>>>  drivers/md/dm-crypt.c                         |  2 +-
>>>>>>  drivers/md/dm-integrity.c                     |  2 +-
>>>>>>  drivers/md/dm-stats.c                         |  2 +-
>>>>>>  drivers/media/platform/mtk-vpu/mtk_vpu.c      |  2 +-
>>>>>>  drivers/misc/vmw_balloon.c                    |  2 +-
>>>>>>  drivers/parisc/ccio-dma.c                     |  4 +-
>>>>>>  drivers/parisc/sba_iommu.c                    |  4 +-
>>>>>>  drivers/staging/android/ion/ion_system_heap.c |  2 +-
>>>>>>  drivers/xen/xen-selfballoon.c                 |  6 +--
>>>>>>  fs/ceph/super.h                               |  2 +-
>>>>>>  fs/file_table.c                               |  7 +--
>>>>>>  fs/fuse/inode.c                               |  2 +-
>>>>>>  fs/nfs/write.c                                |  2 +-
>>>>>>  fs/nfsd/nfscache.c                            |  2 +-
>>>>>>  fs/ntfs/malloc.h                              |  2 +-
>>>>>>  fs/proc/base.c                                |  2 +-
>>>>>>  include/linux/highmem.h                       | 28 ++++++++++-
>>>>>>  include/linux/mm.h                            | 27 +++++++++-
>>>>>>  include/linux/mmzone.h                        | 15 +++---
>>>>>>  include/linux/swap.h                          |  1 -
>>>>>>  kernel/fork.c                                 |  5 +-
>>>>>>  kernel/kexec_core.c                           |  5 +-
>>>>>>  kernel/power/snapshot.c                       |  2 +-
>>>>>>  lib/show_mem.c                                |  2 +-
>>>>>>  mm/highmem.c                                  |  4 +-
>>>>>>  mm/huge_memory.c                              |  2 +-
>>>>>>  mm/kasan/quarantine.c                         |  2 +-
>>>>>>  mm/memblock.c                                 |  6 +--
>>>>>>  mm/memory_hotplug.c                           |  4 +-
>>>>>>  mm/mm_init.c                                  |  2 +-
>>>>>>  mm/oom_kill.c                                 |  2 +-
>>>>>>  mm/page_alloc.c                               | 71 +++++++++++++--------------
>>>>>>  mm/shmem.c                                    |  7 +--
>>>>>>  mm/slab.c                                     |  2 +-
>>>>>>  mm/swap.c                                     |  2 +-
>>>>>>  mm/util.c                                     |  2 +-
>>>>>>  mm/vmalloc.c                                  |  4 +-
>>>>>>  mm/vmstat.c                                   |  4 +-
>>>>>>  mm/workingset.c                               |  2 +-
>>>>>>  mm/zswap.c                                    |  4 +-
>>>>>>  net/dccp/proto.c                              |  7 +--
>>>>>>  net/decnet/dn_route.c                         |  2 +-
>>>>>>  net/ipv4/tcp_metrics.c                        |  2 +-
>>>>>>  net/netfilter/nf_conntrack_core.c             |  7 +--
>>>>>>  net/netfilter/xt_hashlimit.c                  |  5 +-
>>>>>>  net/sctp/protocol.c                           |  7 +--
>>>>>>  security/integrity/ima/ima_kexec.c            |  2 +-
>>>>>>  58 files changed, 195 insertions(+), 144 deletions(-)

  reply	other threads:[~2018-11-07  7:03 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-26 11:00 [PATCH v1 0/4]mm: convert totalram_pages, totalhigh_pages and managed pages to atomic Arun KS
2018-10-26 11:00 ` [PATCH v1 1/4] mm: Fix multiple evaluvations of totalram_pages and managed_pages Arun KS
2018-10-26 11:01 ` [PATCH v1 2/4] mm: Convert zone->managed_pages to atomic variable Arun KS
2018-10-26 11:01 ` [PATCH v1 3/4] mm: convert totalram_pages and totalhigh_pages variables to atomic Arun KS
     [not found]   ` <20181106141438.GQ27423@dhcp22.suse.cz>
2018-11-07 10:38     ` Michal Hocko
2018-10-26 11:01 ` [PATCH v1 4/4] mm: Remove managed_page_count spinlock Arun KS
     [not found]   ` <20181106141732.GR27423@dhcp22.suse.cz>
2018-11-07 10:36     ` Michal Hocko
2018-11-07 11:50   ` Vlastimil Babka
2018-11-06  5:38 ` [PATCH v1 0/4]mm: convert totalram_pages, totalhigh_pages and managed pages to atomic Arun KS
2018-11-06  8:17   ` Konstantin Khlebnikov
2018-11-06  8:30     ` Arun KS
2018-11-06  8:37       ` Konstantin Khlebnikov
2018-11-06  8:43         ` Arun KS
2018-11-07  7:02           ` Konstantin Khlebnikov [this message]
2018-11-07  8:50             ` Vlastimil Babka
2018-11-07 10:28               ` Michal Hocko
2018-11-07 10:39                 ` Michal Hocko
2018-11-07  0:22 ` Andrew Morton
2018-11-07  6:19   ` Arun KS
  -- strict thread matches above, loose matches on Subject: below --
2018-10-26 11:00 Arun KS

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=c3b0edf9-e6a2-c1ab-8490-d94b9830c8ae@yandex-team.ru \
    --to=khlebnikov@yandex-team.ru \
    --cc=akpm@linux-foundation.org \
    --cc=arunks@codeaurora.org \
    --cc=getarunks@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=julia.lawall@lip6.fr \
    --cc=keescook@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=minchan@kernel.org \
    --cc=vbabka@suse.cz \
    /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).