Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

From	Rik van Riel <riel@redhat.com>
Newsgroups	linux.kernel
Subject	Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance
Date	2016-02-26 16:00 +0100
Message-ID	<r6s26-4LU-15@gated-at.bofh.it> (permalink)
References	(1 earlier) <q8JNn-1Ma-1@gated-at.bofh.it> <r5YGK-qv-9@gated-at.bofh.it> <r6cJI-27E-21@gated-at.bofh.it> <r6dFN-2OI-31@gated-at.bofh.it> <r6jUS-7t5-13@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

[Multipart message — attachments visible in raw view] - view raw

On Thu, 2016-02-25 at 22:17 -0800, Hugh Dickins wrote:
> On Fri, 26 Feb 2016, Ebru Akagunduz wrote:
> > in Thu, Feb 25, 2016 at 05:35:50PM -0500, Rik van Riel wrote:
> 
> > > Am I forgetting anything obvious?
> > > 
> > > Is this too aggressive?
> > > 
> > > Not aggressive enough?
> > > 
> > > Could PGPGOUT + PGSWPOUT be a useful
> > > in-between between just PGSWPOUT or
> > > PGSTEAL_*?
> 
> I've no idea offhand, would have to study what each of those
> actually means: I'm really not familiar with them myself.

There are a few levels of page reclaim activity:

PGSTEAL_* - any page was reclaimed, this could just
            be file pages for streaming file IO,etc

PGPGOUT   - the VM wrote pages back to disk to reclaim
            them, this could include file pages

PGSWPOUT  - the VM wrote something to swap to reclaim
            memory

I am not sure which level of aggressiveness khugepaged
should check against, but my gut instinct would probably
be the second or third.

-- 
All Rights Reversed.

Thread

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Hugh Dickins <hughd@google.com> - 2016-02-25 08:40 +0100
  Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Rik van Riel <riel@redhat.com> - 2016-02-25 23:40 +0100
    Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Ebru Akagunduz <ebru.akagunduz@gmail.com> - 2016-02-26 00:40 +0100
      Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Hugh Dickins <hughd@google.com> - 2016-02-26 07:20 +0100
        Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Rik van Riel <riel@redhat.com> - 2016-02-26 16:00 +0100
          Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Ebru Akagunduz <ebru.akagunduz@gmail.com> - 2016-03-03 23:10 +0100
  Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp  performance Ebru Akagunduz <ebru.akagunduz@gmail.com> - 2016-02-26 00:20 +0100

csiph-web