From: Vlastimil Babka <vbabka@suse.cz>
To: js1304@gmail.com, Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
kernel-team@lge.com, Christoph Hellwig <hch@infradead.org>,
Roman Gushchin <guro@fb.com>,
Mike Kravetz <mike.kravetz@oracle.com>,
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
Michal Hocko <mhocko@suse.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>
Subject: Re: [PATCH v4 05/11] mm/migrate: clear __GFP_RECLAIM for THP allocation for migration
Date: Tue, 7 Jul 2020 14:17:55 +0200 [thread overview]
Message-ID: <e72709dd-c7e2-ff26-f450-dbd2a719b2e4@suse.cz> (raw)
In-Reply-To: <1594107889-32228-6-git-send-email-iamjoonsoo.kim@lge.com>
On 7/7/20 9:44 AM, js1304@gmail.com wrote:
> From: Joonsoo Kim <iamjoonsoo.kim@lge.com>
>
> In mm/migrate.c, THP allocation for migration is called with the provided
> gfp_mask | GFP_TRANSHUGE. This gfp_mask contains __GFP_RECLAIM and it
> would be conflict with the intention of the GFP_TRANSHUGE.
>
> GFP_TRANSHUGE/GFP_TRANSHUGE_LIGHT is introduced to control the reclaim
> behaviour by well defined manner since overhead of THP allocation is
> quite large and the whole system could suffer from it. So, they deals
> with __GFP_RECLAIM mask deliberately. If gfp_mask contains __GFP_RECLAIM
> and uses gfp_mask | GFP_TRANSHUGE(_LIGHT) for THP allocation, it means
> that it breaks the purpose of the GFP_TRANSHUGE(_LIGHT).
>
> This patch fixes this situation by clearing __GFP_RECLAIM in provided
> gfp_mask. Note that there are some other THP allocations for migration
> and they just uses GFP_TRANSHUGE(_LIGHT) directly. This patch would make
> all THP allocation for migration consistent.
>
> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> ---
> mm/migrate.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 02b31fe..ecd7615 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -1547,6 +1547,11 @@ struct page *new_page_nodemask(struct page *page,
> }
>
> if (PageTransHuge(page)) {
> + /*
> + * clear __GFP_RECALIM since GFP_TRANSHUGE is the gfp_mask
> + * that chooses the reclaim masks deliberately.
> + */
> + gfp_mask &= ~__GFP_RECLAIM;
> gfp_mask |= GFP_TRANSHUGE;
In addition to what Michal said...
The mask is not passed to this function, so I would just redefine it, as is done
in the hugetlb case. We probably don't even need the __GFP_RETRY_MAYFAIL for the
THP case asi it's just there to prevent OOM kill (per commit 0f55685627d6d ) and
the costly order of THP is enough for that.
> order = HPAGE_PMD_ORDER;
> }
>
next prev parent reply other threads:[~2020-07-07 12:17 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-07 7:44 [PATCH v4 00/11] clean-up the migration target allocation functions js1304
2020-07-07 7:44 ` [PATCH v4 01/11] mm/page_isolation: prefer the node of the source page js1304
2020-07-07 7:44 ` [PATCH v4 02/11] mm/migrate: move migration helper from .h to .c js1304
2020-07-07 7:44 ` [PATCH v4 03/11] mm/hugetlb: unify migration callbacks js1304
2020-07-07 11:05 ` Vlastimil Babka
2020-07-07 11:19 ` Michal Hocko
2020-07-07 7:44 ` [PATCH v4 04/11] mm/hugetlb: make hugetlb migration callback CMA aware js1304
2020-07-07 11:22 ` Vlastimil Babka
2020-07-08 7:16 ` Joonsoo Kim
2020-07-08 7:41 ` Michal Hocko
2020-07-08 9:26 ` Vlastimil Babka
2020-07-08 10:57 ` Aneesh Kumar K.V
2020-07-08 11:32 ` Michal Hocko
2020-07-09 6:43 ` Michal Hocko
2020-07-09 7:03 ` Joonsoo Kim
2020-07-09 0:27 ` Mike Kravetz
2020-07-07 11:31 ` Michal Hocko
2020-07-08 6:48 ` Michal Hocko
2020-07-08 7:12 ` Joonsoo Kim
2020-07-07 7:44 ` [PATCH v4 05/11] mm/migrate: clear __GFP_RECLAIM for THP allocation for migration js1304
2020-07-07 11:40 ` Michal Hocko
2020-07-08 7:19 ` Joonsoo Kim
2020-07-08 7:48 ` Michal Hocko
2020-07-09 3:26 ` Joonsoo Kim
2020-07-07 12:17 ` Vlastimil Babka [this message]
2020-07-08 7:17 ` Joonsoo Kim
2020-07-09 7:17 ` Joonsoo Kim
2020-07-07 7:44 ` [PATCH v4 06/11] mm/migrate: make a standard migration target allocation function js1304
2020-07-07 11:43 ` Michal Hocko
2020-07-07 14:49 ` Vlastimil Babka
2020-07-07 19:00 ` Michal Hocko
2020-07-09 7:15 ` Joonsoo Kim
2020-07-09 10:28 ` Michal Hocko
2020-07-07 7:44 ` [PATCH v4 07/11] mm/gup: use a standard migration target allocation callback js1304
2020-07-07 11:46 ` Michal Hocko
2020-07-08 7:21 ` Joonsoo Kim
2020-07-07 7:44 ` [PATCH v4 08/11] mm/mempolicy: " js1304
2020-07-07 7:44 ` [PATCH v4 09/11] mm/page_alloc: remove a wrapper for alloc_migration_target() js1304
2020-07-07 7:44 ` [PATCH v4 10/11] mm/memory-failure: " js1304
2020-07-07 11:48 ` Michal Hocko
2020-07-07 15:03 ` Vlastimil Babka
2020-07-07 18:55 ` Michal Hocko
2020-07-07 15:00 ` Vlastimil Babka
2020-07-07 7:44 ` [PATCH v4 11/11] mm/memory_hotplug: " js1304
2020-07-07 11:52 ` Michal Hocko
2020-07-07 15:09 ` Vlastimil Babka
2020-07-09 3:25 ` Joonsoo Kim
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=e72709dd-c7e2-ff26-f450-dbd2a719b2e4@suse.cz \
--to=vbabka@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=guro@fb.com \
--cc=hch@infradead.org \
--cc=iamjoonsoo.kim@lge.com \
--cc=js1304@gmail.com \
--cc=kernel-team@lge.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=n-horiguchi@ah.jp.nec.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).