From: Andi Kleen <ak@suse.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: travis@sgi.com, Andrew Morton <akpm@linux-foundation.org>,
Christoph Lameter <clameter@sgi.com>,
Jack Steiner <steiner@sgi.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 00/10] x86: Reduce memory and intra-node effects with large count NR_CPUs
Date: Mon, 14 Jan 2008 11:04:18 +0100 [thread overview]
Message-ID: <200801141104.18789.ak@suse.de> (raw)
In-Reply-To: <20080114081418.GB18296@elte.hu>
> i.e. we've got ~22K bloat per CPU - which is not bad, but because it's a
> static component, it hurts smaller boxes. For distributors to enable
> CONFIG_NR_CPU=1024 by default i guess that bloat has to drop below 1-2K
> per CPU :-/ [that would still mean 1-2MB total bloat but that's much
> more acceptable than 23MB]
Even 1-2MB overhead would be too much for distributors I think. Ideally
there must be near zero overhead for possible CPUs (and I see no principle
reason why this is not possible) Worst case a low few hundred KBs, but even
that would be much.
There are the cpusets which get passed around, but these are only one bit per
possible CPU.
-Andi
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@suse.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: travis@sgi.com, Andrew Morton <akpm@linux-foundation.org>,
Christoph Lameter <clameter@sgi.com>,
Jack Steiner <steiner@sgi.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 00/10] x86: Reduce memory and intra-node effects with large count NR_CPUs
Date: Mon, 14 Jan 2008 11:04:18 +0100 [thread overview]
Message-ID: <200801141104.18789.ak@suse.de> (raw)
In-Reply-To: <20080114081418.GB18296@elte.hu>
> i.e. we've got ~22K bloat per CPU - which is not bad, but because it's a
> static component, it hurts smaller boxes. For distributors to enable
> CONFIG_NR_CPU=1024 by default i guess that bloat has to drop below 1-2K
> per CPU :-/ [that would still mean 1-2MB total bloat but that's much
> more acceptable than 23MB]
Even 1-2MB overhead would be too much for distributors I think. Ideally
there must be near zero overhead for possible CPUs (and I see no principle
reason why this is not possible) Worst case a low few hundred KBs, but even
that would be much.
There are the cpusets which get passed around, but these are only one bit per
possible CPU.
-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:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2008-01-14 10:04 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-13 18:34 [PATCH 00/10] x86: Reduce memory and intra-node effects with large count NR_CPUs travis
2008-01-13 18:34 ` travis
2008-01-13 18:34 ` [PATCH 01/10] x86: Change size of APICIDs from u8 to u16 travis
2008-01-13 18:34 ` travis
2008-01-14 12:23 ` Mel Gorman
2008-01-14 12:23 ` Mel Gorman
2008-01-14 18:13 ` Mike Travis
2008-01-14 18:13 ` Mike Travis
2008-01-14 19:26 ` Mike Travis
2008-01-14 19:26 ` Mike Travis
2008-01-14 18:10 ` Jan Engelhardt
2008-01-14 18:10 ` Jan Engelhardt
2008-01-14 18:22 ` Mike Travis
2008-01-14 18:22 ` Mike Travis
2008-01-14 18:32 ` Mike Travis
2008-01-14 18:32 ` Mike Travis
2008-01-14 19:16 ` Christoph Lameter
2008-01-14 19:16 ` Christoph Lameter
2008-01-13 18:34 ` [PATCH 02/10] x86: Change size of node ids " travis
2008-01-13 18:34 ` travis
2008-01-13 20:01 ` Eric Dumazet
2008-01-13 20:01 ` Eric Dumazet
2008-01-13 18:34 ` [PATCH 03/10] x86: Change NR_CPUS arrays in powernow-k8 travis
2008-01-13 18:34 ` travis
2008-01-13 18:34 ` [PATCH 04/10] x86: Change NR_CPUS arrays in intel_cacheinfo travis
2008-01-13 18:34 ` travis
2008-01-13 18:34 ` [PATCH 05/10] x86: Change NR_CPUS arrays in smpboot_64 travis
2008-01-13 18:34 ` travis
2008-01-13 18:34 ` [PATCH 06/10] x86: Change NR_CPUS arrays in topology travis
2008-01-13 18:34 ` travis
2008-01-14 18:25 ` Jan Engelhardt
2008-01-14 18:25 ` Jan Engelhardt
2008-01-14 19:08 ` Mike Travis
2008-01-14 19:08 ` Mike Travis
2008-01-13 18:35 ` [PATCH 07/10] x86: Cleanup x86_cpu_to_apicid references travis
2008-01-13 18:35 ` travis
2008-01-13 18:35 ` [PATCH 08/10] x86: Change NR_CPUS arrays in numa_64 travis
2008-01-13 18:35 ` travis
2008-01-14 11:14 ` Ingo Molnar
2008-01-14 11:14 ` Ingo Molnar
2008-01-14 17:17 ` Mike Travis
2008-01-14 17:17 ` Mike Travis
2008-01-14 18:14 ` Jan Engelhardt
2008-01-14 18:14 ` Jan Engelhardt
2008-01-13 18:35 ` [PATCH 09/10] x86: Change NR_CPUS arrays in acpi-cpufreq travis
2008-01-13 18:35 ` travis
2008-01-13 18:35 ` [PATCH 10/10] x86: Change bios_cpu_apicid to percpu data variable travis
2008-01-13 18:35 ` travis
2008-01-14 8:14 ` [PATCH 00/10] x86: Reduce memory and intra-node effects with large count NR_CPUs Ingo Molnar
2008-01-14 8:14 ` Ingo Molnar
2008-01-14 9:00 ` Ingo Molnar
2008-01-14 9:00 ` Ingo Molnar
2008-01-14 17:52 ` Mike Travis
2008-01-14 17:52 ` Mike Travis
2008-01-14 10:04 ` Andi Kleen [this message]
2008-01-14 10:04 ` Andi Kleen
2008-01-14 10:11 ` Ingo Molnar
2008-01-14 10:11 ` Ingo Molnar
2008-01-14 11:30 ` Andi Kleen
2008-01-14 11:30 ` Andi Kleen
2008-01-16 7:34 ` Nick Piggin
2008-01-16 7:34 ` Nick Piggin
2008-01-16 18:07 ` Christoph Lameter
2008-01-16 18:07 ` Christoph Lameter
2008-01-14 18:00 ` Mike Travis
2008-01-14 18:00 ` Mike Travis
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=200801141104.18789.ak@suse.de \
--to=ak@suse.de \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mingo@elte.hu \
--cc=steiner@sgi.com \
--cc=travis@sgi.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 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.