Re: [RFC PATCH 3/5] mm/page_alloc: stop instantly reusing freed page

From	Vlastimil Babka <vbabka@suse.cz>
Newsgroups	linux.kernel
Subject	Re: [RFC PATCH 3/5] mm/page_alloc: stop instantly reusing freed page
Date	2016-10-13 13:20 +0200
Message-ID	<srMdj-2NV-1@gated-at.bofh.it> (permalink)
References	<srJfr-Vf-3@gated-at.bofh.it> <srJfs-Vf-21@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 10/13/2016 10:08 AM, js1304@gmail.com wrote:
> From: Joonsoo Kim <iamjoonsoo.kim@lge.com>
>
> Allocation/free pattern is usually sequantial. If they are freed to
> the buddy list, they can be coalesced. However, we first keep these freed
> pages at the pcp list and try to reuse them until threshold is reached
> so we don't have enough chance to get a high order freepage. This reusing
> would provide us some performance advantages since we don't need to
> get the zone lock and we don't pay the cost to check buddy merging.
> But, less fragmentation and more high order freepage would compensate
> this overhead in other ways. First, we would trigger less direct
> compaction which has high overhead. And, there are usecases that uses
> high order page to boost their performance.
>
> Instantly resuing freed page seems to provide us computational benefit
> but the other affects more precious things like as I/O performance and
> memory consumption so I think that it's a good idea to weight
> later advantage more.

Again, there's also cache hotness to consider. And whether the 
sequential pattern is still real on a system with higher uptime. Should 
be possible to evaluate with tracepoints?

Thread

[RFC PATCH 0/5] Reduce fragmentation js1304@gmail.com - 2016-10-13 10:10 +0200
  [RFC PATCH 4/5] mm/page_alloc: add fixed migratetype pageblock infrastructure js1304@gmail.com - 2016-10-13 10:10 +0200
  [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the tail of the buddy list js1304@gmail.com - 2016-10-13 10:10 +0200
    Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Vlastimil Babka <vbabka@suse.cz> - 2016-10-13 11:30 +0200
      Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Joonsoo Kim <iamjoonsoo.kim@lge.com> - 2016-10-14 03:10 +0200
    Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Xishi Qiu <qiuxishi@huawei.com> - 2016-10-17 11:30 +0200
      Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Joonsoo Kim <iamjoonsoo.kim@lge.com> - 2016-10-26 06:40 +0200
        Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Joonsoo Kim <iamjoonsoo.kim@lge.com> - 2016-10-26 08:00 +0200
          Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Xishi Qiu <qiuxishi@huawei.com> - 2016-10-26 08:10 +0200
        Re: [RFC PATCH 1/5] mm/page_alloc: always add freeing page at the  tail of the buddy list Xishi Qiu <qiuxishi@huawei.com> - 2016-10-26 08:00 +0200
  [RFC PATCH 3/5] mm/page_alloc: stop instantly reusing freed page js1304@gmail.com - 2016-10-13 10:10 +0200
    Re: [RFC PATCH 3/5] mm/page_alloc: stop instantly reusing freed page Vlastimil Babka <vbabka@suse.cz> - 2016-10-13 13:20 +0200
      Re: [RFC PATCH 3/5] mm/page_alloc: stop instantly reusing freed page Joonsoo Kim <iamjoonsoo.kim@lge.com> - 2016-10-14 03:30 +0200

csiph-web