linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: "Ray Bryant" <raybry@mpdtxmail.amd.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: Andi Kleen <ak@suse.de>, Adam Litke <agl@us.ibm.com>,
	linux-mm@kvack.org, christoph@lameter.com,
	kenneth.w.chen@intel.com, akpm@osdl.org
Subject: Re: [PATCH 0/4] Demand faunting for huge pages
Date: Thu, 18 Aug 2005 10:29:00 -0500	[thread overview]
Message-ID: <200508181029.01238.raybry@mpdtxmail.amd.com> (raw)
In-Reply-To: <20050818003302.GE7103@localhost.localdomain>

On Wednesday 17 August 2005 19:33, David Gibson wrote:

>
> Strict accounting leads to nicer behaviour in some cases - you'll tend
> to die early rather than late - but it seems an awful lot of work for
> a fairly small improvement in behaviour.
>

The last time we went around on this (April 2004?) Andrew thought that adding 
demand allocation for hugetlb pages without strict accounting was effectively 
an ABI change -- in the current approach the mmap() will fail if you ask for 
too many hugetlb pages whilst in the demand fault approach you will get 
SIGBUS at a later point in time.   At one time this was considered serious 
enough to fix.

Andy Whitcroft provided some code for the patch that Ken and I did back in
April 2004 time frame.   I can't find that one but the following patch from
Christoph Lameter appears to be the code.  The idea is that at mmap() time
a strict reservation is made that guarantees the necessary number of 
hugetlb pages is available. 

http://marc.theaimsgroup.com/?l=linux-kernel&m=109842250714489&w=2

-- 
Ray Bryant
AMD Performance Labs                   Austin, Tx
512-602-0038 (o)                 512-507-7807 (c)

--
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>

  parent reply	other threads:[~2005-08-18 15:29 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-08-17 18:56 Adam Litke
2005-08-17 19:03 ` [PATCH 1/4] x86-pte_huge Adam Litke
2005-08-17 19:18   ` Dave Hansen
2005-08-17 19:27     ` Adam Litke
2005-08-17 19:03 ` [PATCH 2/4] x86-move-stale-pgtable Adam Litke
2005-08-17 19:04 ` [PATCH 3/4] x86-walk-check Adam Litke
2005-08-17 19:41   ` Dave Hansen
2005-08-17 19:05 ` [PATCH 4/4] htlb-fault Adam Litke
2005-08-17 21:04 ` [PATCH 0/4] Demand faunting for huge pages Andi Kleen
2005-08-18  0:33   ` David Gibson
2005-08-18  0:35     ` Andi Kleen
2005-08-18 20:33       ` Adam Litke
2005-08-18 15:29     ` Ray Bryant [this message]
2005-08-18 20:29   ` Adam Litke

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=200508181029.01238.raybry@mpdtxmail.amd.com \
    --to=raybry@mpdtxmail.amd.com \
    --cc=agl@us.ibm.com \
    --cc=ak@suse.de \
    --cc=akpm@osdl.org \
    --cc=christoph@lameter.com \
    --cc=david@gibson.dropbear.id.au \
    --cc=kenneth.w.chen@intel.com \
    --cc=linux-mm@kvack.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