public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Jeff Garzik <jgarzik@pobox.com>,
	Jens Axboe <jens.axboe@oracle.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Ian Campbell <Ian.Campbell@eu.citrix.com>
Subject: Re: [PATCH 5 of 5] xen: Avoid allocations causing swap activity on the resume path
Date: Sat, 31 May 2008 02:59:49 -0700	[thread overview]
Message-ID: <20080531025949.7c8b17bc.akpm@linux-foundation.org> (raw)
In-Reply-To: <48411F70.5020100@goop.org>

On Sat, 31 May 2008 10:50:40 +0100 Jeremy Fitzhardinge <jeremy@goop.org> wrote:

> Does __GFP_HIGH necessarily mean that it won't try to 
> do IO to push pages out?

Nope.

__GFP_FS: may enter filesystems
__GFP_IO: may perform IO

__GFP_IO also means "may do swapout".  Even when swap is on a regular
file.  This is because we do all the fs-related operations up-front
during swapon.  So at alloc_pages()-time we can go direct-to-disk-blocks.

So I assume for this application you'll need GFP_NOIO.  That's still
heaps better than GFP_ATOMIC, because it can sleep and wait for kswapd
to do stuff, and it can reclaim clean pagecache and clean swapcache.

Whether you should also add __GFP_HIGH to cause the page allocation to
bite harder into the page reserves is unclear to me, sorry.


  reply	other threads:[~2008-05-31 10:00 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-31  0:01 [PATCH 0 of 5] xen device driver updates [resend] Jeremy Fitzhardinge
2008-05-31  0:01 ` [PATCH 1 of 5] xen/blkfront: Make sure we don't use bounce buffers, we don't need them Jeremy Fitzhardinge
2008-05-31  1:55   ` Jeff Garzik
2008-05-31  0:01 ` [PATCH 2 of 5] xn/blkfront: Add the CDROM_GET_CAPABILITY ioctl to blkfront Jeremy Fitzhardinge
2008-05-31  1:57   ` Jeff Garzik
2008-05-31  7:08     ` Jeremy Fitzhardinge
2008-05-31  0:01 ` [PATCH 3 of 5] xen/blkfront: Make sure that the device is fully ready before allowing release Jeremy Fitzhardinge
2008-05-31  0:01 ` [PATCH 4 of 5] xen/blkfront: add __exit to module_exit() handlers Jeremy Fitzhardinge
2008-05-31  1:57   ` Jeff Garzik
2008-05-31  0:01 ` [PATCH 5 of 5] xen: Avoid allocations causing swap activity on the resume path Jeremy Fitzhardinge
2008-05-31  1:58   ` Jeff Garzik
2008-05-31  9:50     ` Jeremy Fitzhardinge
2008-05-31  9:59       ` Andrew Morton [this message]
2008-05-31 10:10         ` Jeremy Fitzhardinge
2008-05-31 23:47           ` Andrew Morton
2008-06-01  0:38             ` Jeremy Fitzhardinge
2008-05-31  1:54 ` [PATCH 0 of 5] xen device driver updates [resend] Jeff Garzik
2008-05-31  6:31   ` Jeremy Fitzhardinge
  -- strict thread matches above, loose matches on Subject: below --
2008-05-30 23:50 [PATCH 0 of 5] xen device driver updates Jeremy Fitzhardinge
2008-05-30 23:50 ` [PATCH 5 of 5] xen: Avoid allocations causing swap activity on the resume path Jeremy Fitzhardinge

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080531025949.7c8b17bc.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=Ian.Campbell@eu.citrix.com \
    --cc=jens.axboe@oracle.com \
    --cc=jeremy@goop.org \
    --cc=jgarzik@pobox.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox