linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Kairui Song <ryncsn@gmail.com>
To: Chris Li <chrisl@kernel.org>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
	 "Huang, Ying" <ying.huang@intel.com>,
	David Hildenbrand <david@redhat.com>,
	Hugh Dickins <hughd@google.com>,
	 Johannes Weiner <hannes@cmpxchg.org>,
	Matthew Wilcox <willy@infradead.org>,
	Michal Hocko <mhocko@suse.com>,
	 linux-kernel@vger.kernel.org
Subject: Re: [PATCH 04/24] mm/swap: avoid setting page lock bit and doing extra unlock check
Date: Fri, 24 Nov 2023 16:14:41 +0800	[thread overview]
Message-ID: <CAMgjq7Bax-EOiXXDRxpHFoyrQWRQdSWbZ0KN7-W8YsocV5C97w@mail.gmail.com> (raw)
In-Reply-To: <CAF8kJuM05apUTd9Mpv6s4sZ6SfSAWD+hM_KLiDfatoXu5QwFbg@mail.gmail.com>

Chris Li <chrisl@kernel.org> 于2023年11月23日周四 04:57写道:
>
> Hi Kairui,
>
> On Wed, Nov 22, 2023 at 9:33 AM Kairui Song <ryncsn@gmail.com> wrote:
>
> > > There are two different charges. Memcg charging and memcg swapin charging.
> > > The folio_put will do the memcg discharge, the corresponding memcg
> > > charge is in follio allocation.
> >
> > Hi Chris,
> >
> > I didn't get your idea here... By "memcg swapin charge", do you mean
> > "memory.swap.*"? And "memcg charging" means "memory.*"?. There is no
>
> Sorry I should have used the function name then there is no ambiguity.
> "memcg swapin charge" I mean function mem_cgroup_swapin_charge_folio().
> This function will look up the swap entry and find the memcg by swap entry then
> charge to that memcg.
>
> > memcg charge related code in folio allocation (alloc_pages_mpol),
> > actually the mem_cgroup_swapin_charge_folio here is doing memcg charge
> > not memcg swapin charge. Swapin path actually need to uncharge
> > "memory.swap" by mem_cgroup_swapin_uncharge_swap in later part of this
> > function.
>
> I still think you have a bug there.
>
> Take this make up example:
> Let say the for loop runs 3 times and the 3rd time breaks out the for loop.
> The original code will call:
> filemap_get_folio() 3 times
> folio_put() 2 times
> mem_cgroup_swapin_charge_folio() 1 time.
>
> With your patch, it will call:
> filemap_get_folio() 3 times
> folio_put() 2 times
> mem_cgroup_swapin_charge_folio() 3 times.
>
> Do you see the behavior difference there?

Hi Chris.

folio_put will discharge a page if it's charged, in original code the
2 folio_put call simply free the page since it's not charged. But in
this patch, folio_put will cancel previous
mem_cgroup_swapin_charge_folio call, so actually the 3
mem_cgroup_swapin_charge_folio calls will only charge once. (2 calls
was cancelled by folio_put).

I think this is making it confusing indeed and causing more trouble in
error path (the uncharge could be more expensive than unlock check),
will rework this part.


  reply	other threads:[~2023-11-24  8:15 UTC|newest]

Thread overview: 93+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-19 19:47 [PATCH 00/24] Swapin path refactor for optimization and bugfix Kairui Song
2023-11-19 19:47 ` [PATCH 01/24] mm/swap: fix a potential undefined behavior issue Kairui Song
2023-11-19 20:55   ` Matthew Wilcox
2023-11-20  3:35     ` Chris Li
2023-11-20 11:14       ` Kairui Song
2023-11-20 17:34         ` Chris Li
2023-11-19 19:47 ` [PATCH 02/24] mm/swapfile.c: add back some comment Kairui Song
2023-11-19 19:47 ` [PATCH 03/24] mm/swap: move no readahead swapin code to a stand alone helper Kairui Song
2023-11-19 21:00   ` Matthew Wilcox
2023-11-20 11:14     ` Kairui Song
2023-11-20 14:55   ` Dan Carpenter
2023-11-21  5:34   ` Chris Li
2023-11-22 17:33     ` Kairui Song
2023-11-19 19:47 ` [PATCH 04/24] mm/swap: avoid setting page lock bit and doing extra unlock check Kairui Song
2023-11-20  4:17   ` Chris Li
2023-11-20 11:15     ` Kairui Song
2023-11-20 17:44       ` Chris Li
2023-11-22 17:32         ` Kairui Song
2023-11-22 20:57           ` Chris Li
2023-11-24  8:14             ` Kairui Song [this message]
2023-11-24  8:37               ` Christopher Li
2023-11-19 19:47 ` [PATCH 05/24] mm/swap: move readahead policy checking into swapin_readahead Kairui Song
2023-11-21  6:15   ` Chris Li
2023-11-21  6:35     ` Kairui Song
2023-11-21  7:41       ` Chris Li
2023-11-21  8:32         ` Kairui Song
2023-11-21 15:24           ` Chris Li
2023-11-19 19:47 ` [PATCH 06/24] swap: rework swapin_no_readahead arguments Kairui Song
2023-11-20  0:20   ` kernel test robot
2023-11-21  6:44   ` Chris Li
2023-11-23 10:51     ` Kairui Song
2023-11-19 19:47 ` [PATCH 07/24] mm/swap: move swap_count to header to be shared Kairui Song
2023-11-21  6:51   ` Chris Li
2023-11-21  7:03     ` Kairui Song
2023-11-19 19:47 ` [PATCH 08/24] mm/swap: check readahead policy per entry Kairui Song
2023-11-20  6:04   ` Huang, Ying
2023-11-20 11:17     ` Kairui Song
2023-11-21  1:10       ` Huang, Ying
2023-11-21  5:20         ` Chris Li
2023-11-21  5:13       ` Chris Li
2023-11-21  7:54   ` Chris Li
2023-11-23 10:52     ` Kairui Song
2023-11-19 19:47 ` [PATCH 09/24] mm/swap: inline __swap_count Kairui Song
2023-11-20  7:41   ` Huang, Ying
2023-11-21  8:02     ` Chris Li
2023-11-19 19:47 ` [PATCH 10/24] mm/swap: remove nr_rotate_swap and related code Kairui Song
2023-11-21 15:45   ` Chris Li
2023-11-19 19:47 ` [PATCH 11/24] mm/swap: also handle swapcache lookup in swapin_readahead Kairui Song
2023-11-20  0:47   ` kernel test robot
2023-11-21 16:06   ` Chris Li
2023-11-24  8:42     ` Kairui Song
2023-11-24  9:10       ` Chris Li
2023-11-19 19:47 ` [PATCH 12/24] mm/swap: simplify arguments for swap_cache_get_folio Kairui Song
2023-11-21 16:36   ` Chris Li
2023-11-19 19:47 ` [PATCH 13/24] swap: simplify swap_cache_get_folio Kairui Song
2023-11-21 16:50   ` Chris Li
2023-11-19 19:47 ` [PATCH 14/24] mm/swap: do shadow lookup as well when doing swap cache lookup Kairui Song
2023-11-21 16:55   ` Chris Li
2023-11-19 19:47 ` [PATCH 15/24] mm/swap: avoid an duplicated swap cache lookup for SYNCHRONOUS_IO device Kairui Song
2023-11-21 17:15   ` Chris Li
2023-11-22 18:08     ` Kairui Song
2023-11-19 19:47 ` [PATCH 16/24] mm/swap: reduce scope of get_swap_device in swapin path Kairui Song
2023-11-19 21:12   ` Matthew Wilcox
2023-11-20 11:14     ` Kairui Song
2023-11-21 17:25   ` Chris Li
2023-11-22  0:36   ` Huang, Ying
2023-11-23 11:13     ` Kairui Song
2023-11-24  0:40       ` Huang, Ying
2023-11-19 19:47 ` [PATCH 17/24] mm/swap: fix false error when swapoff race with swapin Kairui Song
2023-11-19 19:47 ` [PATCH 18/24] mm/swap: introduce a helper non fault swapin Kairui Song
2023-11-20  1:07   ` kernel test robot
2023-11-22  4:40   ` Chris Li
2023-11-28 11:22     ` Kairui Song
2023-12-13  2:22       ` Chris Li
2023-11-19 19:47 ` [PATCH 19/24] shmem, swap: refactor error check on OOM or race Kairui Song
2023-11-20  7:04   ` Chris Li
2023-11-20 11:17     ` Kairui Song
2023-11-19 19:47 ` [PATCH 20/24] swap: simplify and make swap_find_cache static Kairui Song
2023-11-22  5:01   ` Chris Li
2023-11-19 19:47 ` [PATCH 21/24] swap: make swapin_readahead result checking argument mandatory Kairui Song
2023-11-22  5:15   ` Chris Li
2023-11-24  8:14     ` Kairui Song
2023-11-19 19:47 ` [PATCH 22/24] swap: make swap_cluster_readahead static Kairui Song
2023-11-22  5:20   ` Chris Li
2023-11-19 19:47 ` [PATCH 23/24] swap: fix multiple swap leak when after cgroup migrate Kairui Song
2023-11-20  7:35   ` Huang, Ying
2023-11-20 11:17     ` Kairui Song
2023-11-22  5:34       ` Chris Li
2023-11-19 19:47 ` [PATCH 24/24] mm/swap: change swapin_readahead to swapin_page_fault Kairui Song
2023-11-20 19:09 ` [PATCH 00/24] Swapin path refactor for optimization and bugfix Yosry Ahmed
2023-11-20 20:22   ` Chris Li
2023-11-22  6:46     ` Kairui Song
2023-11-22  6:43   ` Kairui Song

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=CAMgjq7Bax-EOiXXDRxpHFoyrQWRQdSWbZ0KN7-W8YsocV5C97w@mail.gmail.com \
    --to=ryncsn@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=chrisl@kernel.org \
    --cc=david@redhat.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=willy@infradead.org \
    --cc=ying.huang@intel.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