[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <017450b3-21b1-a6a0-adea-b0b4fa1870d3@suse.cz>
Date: Wed, 7 Oct 2020 00:34:38 +0200
From: Vlastimil Babka <vbabka@...e.cz>
To: Michal Hocko <mhocko@...e.com>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Pavel Tatashin <pasha.tatashin@...een.com>,
David Hildenbrand <david@...hat.com>,
Oscar Salvador <osalvador@...e.de>,
Joonsoo Kim <iamjoonsoo.kim@....com>
Subject: Re: [PATCH 6/9] mm, page_alloc: cache pageset high and batch in
struct zone
On 10/5/20 3:28 PM, Michal Hocko wrote:
> On Tue 22-09-20 16:37:09, Vlastimil Babka wrote:
>> All per-cpu pagesets for a zone use the same high and batch values, that are
>> duplicated there just for performance (locality) reasons. This patch adds the
>> same variables also to struct zone as a shared copy.
>>
>> This will be useful later for making possible to disable pcplists temporarily
>> by setting high value to 0, while remembering the values for restoring them
>> later. But we can also immediately benefit from not updating pagesets of all
>> possible cpus in case the newly recalculated values (after sysctl change or
>> memory online/offline) are actually unchanged from the previous ones.
>
> Advantage of this patch is not really clear from it in isolation. Maybe
> merge it with the patch which uses the duplicated state.
I'm not sure that would help its reviewability? As the patch that uses it is the
last, largest one. And there is already a small advantage right away as
changelog explains.
Powered by blists - more mailing lists