From: Andi Kleen <ak@muc.de>
To: "Martin J. Bligh" <mbligh@mbligh.org>
Cc: jschopp@austin.ibm.com, Mel Gorman <mel@csn.ul.ie>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org
Subject: Re: Avoiding external fragmentation with a placement policy Version 12
Date: Thu, 02 Jun 2005 20:28:26 +0200 [thread overview]
Message-ID: <m14qcgwr3p.fsf@muc.de> (raw)
In-Reply-To: <434510000.1117670555@flay> (Martin J. Bligh's message of "Wed, 01 Jun 2005 17:02:35 -0700")
"Martin J. Bligh" <mbligh@mbligh.org> writes:
> It gets very messy when CIFS requires a large buffer to write back
> to disk in order to free memory ...
How about just fixing CIFS to submit memory page by page? The network
stack below it supports that just fine and the VFS above it does anyways,
so it doesnt make much sense that CIFS sitting below them uses
larger buffers.
> There's one example ... we can probably work around it if we try hard
> enough. However, the fundamental question becomes "do we support higher
> order allocs, or not?". If not fine ... but we ought to quit pretending
> we do. If so, then we need to make them more reliable.
My understanding was that the deal was that order 1 is supposed
to work but somewhat slower, and bigger orders are supposed to work
at boot up time.
-Andi
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@muc.de>
To: "Martin J. Bligh" <mbligh@mbligh.org>
Cc: jschopp@austin.ibm.com, Mel Gorman <mel@csn.ul.ie>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org
Subject: Re: Avoiding external fragmentation with a placement policy Version 12
Date: Thu, 02 Jun 2005 20:28:26 +0200 [thread overview]
Message-ID: <m14qcgwr3p.fsf@muc.de> (raw)
In-Reply-To: <434510000.1117670555@flay> (Martin J. Bligh's message of "Wed, 01 Jun 2005 17:02:35 -0700")
"Martin J. Bligh" <mbligh@mbligh.org> writes:
> It gets very messy when CIFS requires a large buffer to write back
> to disk in order to free memory ...
How about just fixing CIFS to submit memory page by page? The network
stack below it supports that just fine and the VFS above it does anyways,
so it doesnt make much sense that CIFS sitting below them uses
larger buffers.
> There's one example ... we can probably work around it if we try hard
> enough. However, the fundamental question becomes "do we support higher
> order allocs, or not?". If not fine ... but we ought to quit pretending
> we do. If so, then we need to make them more reliable.
My understanding was that the deal was that order 1 is supposed
to work but somewhat slower, and bigger orders are supposed to work
at boot up time.
-Andi
--
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:"aart@kvack.org"> aart@kvack.org </a>
next prev parent reply other threads:[~2005-06-02 18:28 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-05-31 11:20 Avoiding external fragmentation with a placement policy Version 12 Mel Gorman
2005-05-31 11:20 ` Mel Gorman
2005-06-01 20:55 ` Joel Schopp
2005-06-01 20:55 ` Joel Schopp
2005-06-01 23:09 ` Nick Piggin
2005-06-01 23:09 ` Nick Piggin
2005-06-01 23:23 ` David S. Miller
2005-06-01 23:23 ` David S. Miller, Nick Piggin
2005-06-01 23:28 ` Martin J. Bligh
2005-06-01 23:28 ` Martin J. Bligh
2005-06-01 23:43 ` Nick Piggin
2005-06-01 23:43 ` Nick Piggin
2005-06-02 0:02 ` Martin J. Bligh
2005-06-02 0:02 ` Martin J. Bligh
2005-06-02 0:20 ` Nick Piggin
2005-06-02 0:20 ` Nick Piggin
2005-06-02 13:55 ` Mel Gorman
2005-06-02 13:55 ` Mel Gorman
2005-06-02 15:52 ` Joel Schopp
2005-06-02 15:52 ` Joel Schopp
2005-06-02 19:50 ` Ray Bryant
2005-06-02 19:50 ` Ray Bryant
2005-06-02 20:10 ` Joel Schopp
2005-06-02 20:10 ` Joel Schopp
2005-06-04 16:09 ` Marcelo Tosatti
2005-06-04 16:09 ` Marcelo Tosatti
2005-06-03 3:48 ` Nick Piggin
2005-06-03 3:48 ` Nick Piggin
2005-06-03 4:49 ` David S. Miller
2005-06-03 4:49 ` David S. Miller, Nick Piggin
2005-06-03 5:34 ` Martin J. Bligh
2005-06-03 5:34 ` Martin J. Bligh
2005-06-03 5:37 ` David S. Miller
2005-06-03 5:37 ` David S. Miller, Martin J. Bligh
2005-06-03 5:42 ` Martin J. Bligh
2005-06-03 5:42 ` Martin J. Bligh
2005-06-03 5:51 ` David S. Miller
2005-06-03 5:51 ` David S. Miller, Martin J. Bligh
2005-06-03 13:13 ` Mel Gorman
2005-06-03 13:13 ` Mel Gorman
2005-06-03 6:43 ` Nick Piggin
2005-06-03 6:43 ` Nick Piggin
2005-06-03 13:57 ` Martin J. Bligh
2005-06-03 13:57 ` Martin J. Bligh
2005-06-03 16:43 ` Dave Hansen
2005-06-03 16:43 ` Dave Hansen
2005-06-03 18:43 ` David S. Miller
2005-06-03 18:43 ` David S. Miller, Dave Hansen
2005-06-04 1:44 ` Herbert Xu
2005-06-04 1:44 ` Herbert Xu
2005-06-04 2:15 ` Nick Piggin
2005-06-04 2:15 ` Nick Piggin
2005-06-05 19:52 ` David S. Miller
2005-06-05 19:52 ` David S. Miller, Nick Piggin
2005-06-03 13:05 ` Mel Gorman
2005-06-03 13:05 ` Mel Gorman
2005-06-03 14:00 ` Martin J. Bligh
2005-06-03 14:00 ` Martin J. Bligh
2005-06-08 17:03 ` Mel Gorman
2005-06-08 17:03 ` Mel Gorman
2005-06-08 17:18 ` Martin J. Bligh
2005-06-08 17:18 ` Martin J. Bligh
2005-06-10 16:20 ` Christoph Lameter
2005-06-10 16:20 ` Christoph Lameter
2005-06-10 17:53 ` Steve Lord
2005-06-10 17:53 ` Steve Lord
2005-06-02 18:28 ` Andi Kleen [this message]
2005-06-02 18:28 ` Andi Kleen
2005-06-02 18:42 ` Martin J. Bligh
2005-06-02 18:42 ` Martin J. Bligh
2005-06-02 13:15 ` Mel Gorman
2005-06-02 13:15 ` Mel Gorman
2005-06-02 14:01 ` Martin J. Bligh
2005-06-02 14:01 ` Martin J. Bligh
[not found] ` <20050603174706.GA25663@localhost.localdomain>
2005-06-03 17:56 ` Martin J. Bligh
2005-06-03 17:56 ` Martin J. Bligh
2005-06-01 23:47 ` Mike Kravetz
2005-06-01 23:47 ` Mike Kravetz
2005-06-01 23:56 ` Nick Piggin
2005-06-01 23:56 ` Nick Piggin
2005-06-02 0:07 ` Mike Kravetz
2005-06-02 0:07 ` Mike Kravetz
2005-06-02 9:49 ` Mel Gorman
2005-06-02 9:49 ` Mel Gorman
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=m14qcgwr3p.fsf@muc.de \
--to=ak@muc.de \
--cc=akpm@osdl.org \
--cc=jschopp@austin.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mbligh@mbligh.org \
--cc=mel@csn.ul.ie \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.