From: Michal Hocko <mhocko@suse.com>
To: Alistair Popple <apopple@nvidia.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, jhubbard@nvidia.com, ying.huang@intel.com,
osalvador@suse.de, baolin.wang@linux.alibaba.com, ziy@nvidia.com,
shy828301@gmail.com, ryan.roberts@arm.com
Subject: Re: [PATCH 1/2] mm/migrate.c: Fix return code when migration fails
Date: Mon, 7 Aug 2023 10:39:37 +0200 [thread overview]
Message-ID: <ZNCtyVuO0vCiO1Yu@dhcp22.suse.cz> (raw)
In-Reply-To: <20230807063945.911582-1-apopple@nvidia.com>
On Mon 07-08-23 16:39:44, Alistair Popple wrote:
> When a page fails to migrate move_pages() returns the error code in a
> per-page array of status values. The function call itself is also
> supposed to return a summary error code indicating that a failure
> occurred.
>
> This doesn't always happen. Instead success can be returned even
> though some pages failed to migrate. This is due to incorrectly
> returning the error code from store_status() rather than the code from
> add_page_for_migration. Fix this by only returning an error from
> store_status() if the store actually failed.
Error reporting by this syscall has been really far from
straightforward. Please read through a49bd4d71637 and the section "On a
side note".
Is there any specific reason you are trying to address this now or is
this motivated by the code inspection?
> Signed-off-by: Alistair Popple <apopple@nvidia.com>
> Fixes: a49bd4d71637 ("mm, numa: rework do_pages_move")
> ---
> mm/migrate.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 24baad2571e3..bb3a37245e13 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -2222,7 +2222,9 @@ static int do_pages_move(struct mm_struct *mm, nodemask_t task_nodes,
> * If the page is already on the target node (!err), store the
> * node, otherwise, store the err.
> */
> - err = store_status(status, i, err ? : current_node, 1);
> + err1 = store_status(status, i, err ? : current_node, 1);
> + if (err1)
> + err = err1;
> if (err)
> goto out_flush;
>
> --
> 2.39.2
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2023-08-07 8:39 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-07 6:39 Alistair Popple
2023-08-07 6:39 ` [PATCH 2/2] selftests/migration: Disable NUMA balancing and check migration status Alistair Popple
2023-08-07 9:15 ` Ryan Roberts
2023-08-07 12:41 ` Alistair Popple
2023-08-07 13:42 ` Ryan Roberts
2023-08-09 4:21 ` Alistair Popple
2023-08-09 9:34 ` David Hildenbrand
2023-08-09 13:39 ` Ryan Roberts
2023-08-10 8:23 ` Alistair Popple
2023-08-09 9:35 ` David Hildenbrand
2023-08-09 10:46 ` Alistair Popple
2023-08-07 8:39 ` Michal Hocko [this message]
2023-08-07 12:31 ` [PATCH 1/2] mm/migrate.c: Fix return code when migration fails Alistair Popple
2023-08-07 13:07 ` Michal Hocko
2023-08-09 4:10 ` Alistair Popple
2023-08-11 7:37 ` Huang, Ying
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=ZNCtyVuO0vCiO1Yu@dhcp22.suse.cz \
--to=mhocko@suse.com \
--cc=akpm@linux-foundation.org \
--cc=apopple@nvidia.com \
--cc=baolin.wang@linux.alibaba.com \
--cc=jhubbard@nvidia.com \
--cc=linux-mm@kvack.org \
--cc=osalvador@suse.de \
--cc=ryan.roberts@arm.com \
--cc=shy828301@gmail.com \
--cc=ying.huang@intel.com \
--cc=ziy@nvidia.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