Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1404424
| From | Vlastimil Babka <vbabka@suse.cz> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [patch] mm, migrate: increment fail count on ENOMEM |
| Date | 2016-05-20 15:20 +0200 |
| Message-ID | <rASvo-5Wx-17@gated-at.bofh.it> (permalink) |
| References | <rAEsp-5GL-5@gated-at.bofh.it> <rASlH-5T9-11@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On 05/20/2016 03:06 PM, Michal Hocko wrote:
> On Thu 19-05-16 15:11:23, David Rientjes wrote:
>> If page migration fails due to -ENOMEM, nr_failed should still be
>> incremented for proper statistics.
>>
>> This was encountered recently when all page migration vmstats showed 0,
>> and inferred that migrate_pages() was never called, although in reality
>> the first page migration failed because compaction_alloc() failed to find
>> a migration target.
>>
>> This patch increments nr_failed so the vmstat is properly accounted on
>> ENOMEM.
>>
>> Signed-off-by: David Rientjes <rientjes@google.com>
>
> Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
> One question though
>
>> ---
>> mm/migrate.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/mm/migrate.c b/mm/migrate.c
>> --- a/mm/migrate.c
>> +++ b/mm/migrate.c
>> @@ -1171,6 +1171,7 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
>>
>> switch(rc) {
>> case -ENOMEM:
>> + nr_failed++;
>> goto out;
>> case -EAGAIN:
>> retry++;
>
> Why don't we need also to count also retries?
We could, but not like you suggest.
> ---
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 53ab6398e7a2..ef9c5211ae3c 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -1190,9 +1190,9 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
> }
> }
> }
> +out:
> nr_failed += retry;
> rc = nr_failed;
This overwrites rc == -ENOMEM, which at least compaction needs to
recognize. But we could duplicate "nr_failed += retry" in the case -ENOMEM.
> -out:
> if (nr_succeeded)
> count_vm_events(PGMIGRATE_SUCCESS, nr_succeeded);
> if (nr_failed)
>
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[patch] mm, migrate: increment fail count on ENOMEM David Rientjes <rientjes@google.com> - 2016-05-20 00:20 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Michal Hocko <mhocko@kernel.org> - 2016-05-20 15:10 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Vlastimil Babka <vbabka@suse.cz> - 2016-05-20 15:20 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Michal Hocko <mhocko@kernel.org> - 2016-05-20 15:40 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Andrew Morton <akpm@linux-foundation.org> - 2016-05-24 00:10 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Hugh Dickins <hughd@google.com> - 2016-05-24 01:50 +0200
Re: [patch] mm, migrate: increment fail count on ENOMEM Michal Hocko <mhocko@kernel.org> - 2016-05-24 08:20 +0200
csiph-web