[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3933e4b1-e6cd-3e6b-d1a3-7a835767ab44@linux.alibaba.com>
Date: Tue, 16 Jul 2019 10:19:17 -0700
From: Yang Shi <yang.shi@...ux.alibaba.com>
To: Vlastimil Babka <vbabka@...e.cz>, mhocko@...nel.org,
mgorman@...hsingularity.net, akpm@...ux-foundation.org
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [v2 PATCH 1/2] mm: mempolicy: make the behavior consistent when
MPOL_MF_MOVE* and MPOL_MF_STRICT were specified
On 7/16/19 1:47 AM, Vlastimil Babka wrote:
> On 7/16/19 10:12 AM, Vlastimil Babka wrote:
>>> --- a/mm/mempolicy.c
>>> +++ b/mm/mempolicy.c
>>> @@ -429,11 +429,14 @@ static inline bool queue_pages_required(struct page *page,
>>> }
>>>
>>> /*
>>> - * queue_pages_pmd() has three possible return values:
>>> + * queue_pages_pmd() has four possible return values:
>>> + * 2 - there is unmovable page, and MPOL_MF_MOVE* & MPOL_MF_STRICT were
>>> + * specified.
>>> * 1 - pages are placed on the right node or queued successfully.
>>> * 0 - THP was split.
>> I think if you renumbered these, it would be more consistent with
>> queue_pages_pte_range() and simplify the code there.
>> 0 - pages on right node/queued
>> 1 - unmovable page with right flags specified
>> 2 - THP split
> Ah, alternatively you could add a boolean to struct queue_pages
> accessible from mm_walk, set true to indicate that unmovable page has
> been encountered, without propagating it back through special return values.
I will try both to see which one (renumbering return value or use flag)
is better.
Thanks,
Yang
Powered by blists - more mailing lists