From: "Martin J. Bligh" <mbligh@aracnet.com>
To: Andrew Morton <akpm@osdl.org>, Andy Whitcroft <apw@shadowen.org>
Cc: anton@samba.org, sds@epoch.ncsc.mil, ak@suse.de, raybry@sgi.com,
lse-tech@lists.sourceforge.net, linux-ia64@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] [0/6] HUGETLB memory commitment
Date: Thu, 25 Mar 2004 16:18:52 -0800 [thread overview]
Message-ID: <33780000.1080260332@flay> (raw)
In-Reply-To: <20040325155117.60dbc0e1.akpm@osdl.org>
> I think it's simply:
>
> - Make normal overcommit logic skip hugepages completely
>
> - Teach the overcommit_memory=2 logic that hugepages are basically
> "pinned", so subtract them from the arithmetic.
>
> And that's it. The hugepages are semantically quite different from normal
> memory (prefaulted, preallocated, unswappable) and we've deliberately
> avoided pretending otherwise.
It would be nice (to fix some of the posted problems) if hugepages didn't
have to be prefaulted ... if they had their own overcommit pool (that we
used whether normal overcommit was on or not), that'd be unnecessary.
Specifically:
1) SGI found that requesting oodles of large pages took forever.
2) NUMA allocation API wants to be able to specify policies, which
means not prefaulting them.
I'd agree that fixing stopping hugepages from using the main overcommit
pool is the first priority, but it'd be nice to go one stage further.
M.
next prev parent reply other threads:[~2004-03-26 0:43 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-03-25 16:54 [PATCH] [0/6] HUGETLB memory commitment Andy Whitcroft
2004-03-25 16:58 ` [PATCH] [1/6] " Andy Whitcroft
2004-03-25 16:59 ` [PATCH] [2/6] " Andy Whitcroft
2004-03-25 17:00 ` [PATCH] [3/6] " Andy Whitcroft
2004-03-25 17:01 ` [PATCH] [4/6] " Andy Whitcroft
2004-03-25 17:02 ` [PATCH] [5/6] " Andy Whitcroft
2004-03-25 17:03 ` [PATCH] [6/6] " Andy Whitcroft
2004-03-25 21:04 ` [PATCH] [0/6] " Andrew Morton
2004-03-25 23:27 ` Andy Whitcroft
2004-03-25 23:51 ` Andrew Morton
2004-03-25 23:59 ` Andy Whitcroft
2004-03-26 0:10 ` Keith Owens
2004-03-26 0:22 ` Andrew Morton
2004-03-26 8:58 ` [Lse-tech] " Suparna Bhattacharya
2004-03-26 3:39 ` Keith Owens
2004-03-26 17:15 ` Suparna Bhattacharya
2004-03-26 2:01 ` Andy Whitcroft
2004-03-26 0:18 ` Martin J. Bligh [this message]
2004-03-28 18:02 ` Ray Bryant
2004-03-28 19:10 ` Martin J. Bligh
2004-03-28 21:32 ` [Lse-tech] " Ray Bryant
2004-03-29 16:50 ` Martin J. Bligh
2004-03-29 12:30 ` Andy Whitcroft
2004-03-29 20:45 ` Chen, Kenneth W
2004-03-29 20:49 ` Chen, Kenneth W
2004-03-30 12:57 ` Andy Whitcroft
2004-03-30 20:04 ` Chen, Kenneth W
2004-03-30 21:48 ` Andy Whitcroft
2004-03-31 1:48 ` Andy Whitcroft
2004-03-31 8:51 ` Chen, Kenneth W
2004-03-31 16:20 ` Andy Whitcroft
2004-04-01 21:15 ` Andy Whitcroft
2004-04-01 22:50 ` Andy Whitcroft
2004-04-01 23:09 ` Chen, Kenneth W
2004-04-03 3:57 ` [PATCH] " Ray Bryant
2004-04-04 3:31 ` Chen, Kenneth W
2004-04-04 22:15 ` Ray Bryant
2004-04-05 15:26 ` [Lse-tech] " Ray Bryant
2004-04-05 17:01 ` Chen, Kenneth W
2004-04-05 18:22 ` Ray Bryant
2004-04-05 23:18 ` Chen, Kenneth W
2004-04-06 1:05 ` Ray Bryant
2004-04-06 16:14 ` Andy Whitcroft
2004-04-06 17:40 ` Chen, Kenneth W
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=33780000.1080260332@flay \
--to=mbligh@aracnet.com \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=anton@samba.org \
--cc=apw@shadowen.org \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lse-tech@lists.sourceforge.net \
--cc=raybry@sgi.com \
--cc=sds@epoch.ncsc.mil \
/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