[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87ikxvozla.fsf@yhuang6-desk2.ccr.corp.intel.com>
Date: Thu, 27 Jun 2024 11:04:49 +0800
From: "Huang, Ying" <ying.huang@...el.com>
To: Tvrtko Ursulin <tvrtko.ursulin@...lia.com>
Cc: Tvrtko Ursulin <tursulin@...lia.com>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, kernel-dev@...lia.com, Mel Gorman
<mgorman@...e.de>, Peter Zijlstra <peterz@...radead.org>, Ingo Molnar
<mingo@...hat.com>, Rik van Riel <riel@...riel.com>, Johannes Weiner
<hannes@...xchg.org>, "Matthew Wilcox (Oracle)" <willy@...radead.org>,
Dave Hansen <dave.hansen@...el.com>, Andi Kleen <ak@...ux.intel.com>,
Michal Hocko <mhocko@...e.com>, David Rientjes <rientjes@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] mm/numa_balancing: Teach mpol_to_str about the
balancing mode
Tvrtko Ursulin <tvrtko.ursulin@...lia.com> writes:
> On 26/06/2024 09:48, Huang, Ying wrote:
>> Tvrtko Ursulin <tursulin@...lia.com> writes:
>>
>>> From: Tvrtko Ursulin <tvrtko.ursulin@...lia.com>
>>>
>>> If a task has had MPOL_F_NUMA_BALANCING set it is useful to show that in
>> IIUC, MPOL_F_NUMA_BALANCING works for VMA area via mbind() too.
>
> Ah okay.. I think I forgot to actually check and went by what commit
> text of bda420b98505 said, which is probably outdated.
>
>>> procfs. Teach the mpol_to_str helper about its existance and while at it
>>> update the comment to account for "weighted interleave" when suggesting
>>> a recommended buffer size.
>> Otherwise LGTM, Thanks!
>> Reviewed-by: "Huang, Ying" <ying.huang@...el.com>
>
> Thank you! Would you have an idea of which tree this should go to aka
> which maintainer to ask to merge it?
You can use scripts/get_maintainer.pl in kernel source tree to find out
the maintainer. And IIUC, Andrew Morton is the maintainer for this.
> Second question - I also have a patch which enables choosing balancing
> for tmpfs (mpol_parse_str) but I am unsure of its value. It would make
> things symmetrical, but is there some other benefit I don't know. Any
> thoughts on this?
I am not familiar with tmpfs NUMA policy, after checking the source
code, my understanding is that the change may have user visible effect.
For example, you can mount tmpfs with numa balancing enabled/disabled
and check whether pages can be balanced among NUMA nodes.
[snip]
--
Best Regards,
Huang, Ying
Powered by blists - more mailing lists