From: Christoph Lameter <clameter@sgi.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andi Kleen <andi@firstfloor.org>, Nick Piggin <npiggin@suse.de>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
akpm@linux-foundation.org, dkegel@google.com,
David Miller <davem@davemloft.net>,
Daniel Phillips <phillips@google.com>
Subject: Re: [RFC 0/3] Recursive reclaim (on __PF_MEMALLOC)
Date: Wed, 15 Aug 2007 13:32:03 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.64.0708151330180.7326@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <1187186120.6114.56.camel@twins>
On Wed, 15 Aug 2007, Peter Zijlstra wrote:
> The thing I strongly objected to was the 20%.
Well then set it to 10%. We have min_free_kbytes now and so we are used
to these limits.
> Also his approach misses the threshold - the extra condition needed to
> break out of the various network deadlocks. There is no point that says
> - ok, and now we're in trouble, drop anything non-critical. Without that
> you'll always run into a wall.
Networking?
> That is his second patch-set, and I do worry about the irq latency that
> that will introduce. It very much has the potential to ruin everything
> that cares about interactiveness or latency.
Where is the patchset introducing additional latencies? Most of the time
it only saves and restores flags. We already enable and disable interrupts
in the reclaim path but we assume that interupts are always enabled when
we enter reclaim.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2007-08-15 20:32 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-14 14:21 [RFC 0/3] Recursive reclaim (on __PF_MEMALLOC) Christoph Lameter
2007-08-14 14:21 ` [RFC 1/3] Allow reclaim via __GFP_NOMEMALLOC reclaim Christoph Lameter
2007-08-14 14:21 ` [RFC 2/3] Use NOMEMALLOC reclaim to allow reclaim if PF_MEMALLOC is set Christoph Lameter
2007-08-14 14:21 ` [RFC 3/3] Test code for PF_MEMALLOC reclaim Christoph Lameter
2007-08-14 14:36 ` [RFC 0/3] Recursive reclaim (on __PF_MEMALLOC) Peter Zijlstra
2007-08-14 15:29 ` Christoph Lameter
2007-08-14 19:32 ` Peter Zijlstra
2007-08-14 19:41 ` Christoph Lameter
2007-08-15 12:22 ` Nick Piggin
2007-08-15 13:12 ` Peter Zijlstra
2007-08-15 14:15 ` Andi Kleen
2007-08-15 13:55 ` Peter Zijlstra
2007-08-15 14:34 ` Andi Kleen
2007-08-15 20:32 ` Christoph Lameter [this message]
2007-08-15 20:29 ` Christoph Lameter
2007-08-16 3:29 ` Nick Piggin
2007-08-16 20:27 ` Christoph Lameter
2007-08-20 3:51 ` Peter Zijlstra
2007-08-20 19:15 ` Christoph Lameter
2007-08-21 0:32 ` Nick Piggin
2007-08-21 0:28 ` Nick Piggin
2007-08-21 15:29 ` Peter Zijlstra
2007-08-23 3:02 ` Nick Piggin
2007-09-12 22:39 ` Christoph Lameter
2007-09-05 9:20 ` Daniel Phillips
2007-09-05 10:42 ` Christoph Lameter
2007-09-05 11:42 ` Nick Piggin
2007-09-05 12:14 ` Christoph Lameter
2007-09-05 12:19 ` Nick Piggin
2007-09-10 19:29 ` Christoph Lameter
2007-09-10 19:37 ` Peter Zijlstra
2007-09-10 19:41 ` Christoph Lameter
2007-09-10 19:55 ` Peter Zijlstra
2007-09-10 20:17 ` Christoph Lameter
2007-09-10 20:48 ` Peter Zijlstra
2007-09-11 7:41 ` Nick Piggin
2007-09-12 10:52 ` Peter Zijlstra
2007-09-12 22:47 ` Christoph Lameter
2007-09-13 8:19 ` Peter Zijlstra
2007-09-13 18:32 ` Christoph Lameter
2007-09-13 19:24 ` Peter Zijlstra
2007-09-05 16:16 ` Daniel Phillips
2007-09-08 5:12 ` Mike Snitzer
2007-09-18 0:28 ` Daniel Phillips
2007-09-18 3:27 ` Mike Snitzer
[not found] ` <200709172211.26493.phillips@phunq.net>
2007-09-18 8:11 ` Wouter Verhelst
2007-09-18 9:58 ` Peter Zijlstra
2007-09-18 16:56 ` Daniel Phillips
2007-09-18 19:16 ` Peter Zijlstra
2007-09-18 9:30 ` Peter Zijlstra
2007-09-18 18:40 ` Daniel Phillips
2007-09-18 20:13 ` Mike Snitzer
2007-09-10 19:25 ` Christoph Lameter
2007-09-10 19:55 ` Peter Zijlstra
2007-09-10 20:22 ` Christoph Lameter
2007-09-10 20:48 ` Peter Zijlstra
2007-10-26 17:44 ` Pavel Machek
2007-10-26 17:55 ` Christoph Lameter
2007-10-27 22:58 ` Daniel Phillips
2007-10-27 23:08 ` Daniel Phillips
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=Pine.LNX.4.64.0708151330180.7326@schroedinger.engr.sgi.com \
--to=clameter@sgi.com \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=davem@davemloft.net \
--cc=dkegel@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=npiggin@suse.de \
--cc=phillips@google.com \
/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;
as well as URLs for NNTP newsgroup(s).