public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Ulrich Drepper <drepper@redhat.com>
To: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>,
	Jakub Jelinek <jakub@redhat.com>,
	Roland McGrath <roland@redhat.com>,
	Arjan van de Ven <arjan@infradead.org>,
	"Randy.Dunlap" <rdunlap@xenotime.net>,
	akpm@osdl.org, linux-kernel@vger.kernel.org,
	libc-alpha@sourceware.org
Subject: Re: [PATCH] Use uname not sysctl to get the kernel revision
Date: Wed, 12 Jul 2006 22:17:51 -0700	[thread overview]
Message-ID: <44B5D77F.60200@redhat.com> (raw)
In-Reply-To: <m1k66i8ql5.fsf@ebiederm.dsl.xmission.com>

[-- Attachment #1: Type: text/plain, Size: 1533 bytes --]

Eric W. Biederman wrote:
> Ulrich what would be interesting besides the possibility of having
> multiple cpus?

What is needed for various things like memory handling etc is all
topology information.  Somebody might remember the numa library proposal
I had in April 2004 which was cast aside because people were only
looking for a "quick fix".  Well, the problem still isn't solved.

IMO the vdso should export information about:

- processors and their relationship (hyperthreads, cores)

- the CPU caches and how they relate to the cores (e.g., dual core
  with shared L2)

- local main memory for each processor

- relative costs of the memory access of the various memory regions
  (for numa local memory to a node, intra-node costs)

- ideally, relative costs main memory and CPU caches


All this information can be steadily updated by the kernel as new
CPUs/memory get added/removed.  The vdso should have functions to access
this information.  It's easy enough to make this access race free.

I guess I should try to come up with a representation for this
knowledge.  Collecting the information (except the costs) should be
easy.  Determining the costs also shouldn't be that hard but it can be
very useful.  Some of this information could be determined at userlevel
but you really don't want every process to compute all this from
scratch.  And stored data in a file is stale if the system changes.

-- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 251 bytes --]

  parent reply	other threads:[~2006-07-13  5:17 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-07-10 22:39 [PATCH] sysctl: Document that sys_sysctl will be removed Eric W. Biederman
2006-07-10 22:50 ` Randy.Dunlap
2006-07-11  4:10   ` Eric W. Biederman
2006-07-11  7:07     ` Arjan van de Ven
2006-07-12 16:25       ` [PATCH] Use uname not sysctl to get the kernel revision Eric W. Biederman
2006-07-12 16:50         ` Ulrich Drepper
2006-07-12 17:42           ` Eric W. Biederman
2006-07-12 23:24             ` Theodore Tso
2006-07-12 23:31               ` Andi Kleen
2006-07-13  0:12                 ` Theodore Tso
2006-07-13  2:33                   ` Eric W. Biederman
2006-07-13 12:15                   ` Andi Kleen
2006-07-12 23:44               ` Steve Munroe
2006-07-14 18:49                 ` Benjamin Herrenschmidt
2006-07-14 19:09                   ` Andi Kleen
2006-07-13  0:19               ` Eric W. Biederman
2006-07-12 18:44           ` Roland McGrath
2006-07-12 19:33             ` Ulrich Drepper
2006-07-12 19:53               ` Jakub Jelinek
2006-07-12 20:09                 ` H. Peter Anvin
2006-07-12 21:23                   ` Eric W. Biederman
2006-07-12 21:29                     ` Arjan van de Ven
2006-07-12 21:56                       ` Eric W. Biederman
2006-07-12 22:01                         ` Arjan van de Ven
2006-07-12 22:02                           ` H. Peter Anvin
2006-07-12 22:26                             ` Eric W. Biederman
2006-07-12 22:31                               ` H. Peter Anvin
2006-07-12 23:07                               ` Alan Cox
2006-07-12 23:19                                 ` H. Peter Anvin
2006-07-13 11:15                                   ` Alan Cox
2006-07-14 18:45                                 ` Benjamin Herrenschmidt
2006-07-14 19:11                                   ` H. Peter Anvin
2006-07-12 21:29                     ` H. Peter Anvin
2006-07-12 21:33                     ` Michael Tokarev
2006-07-13  5:17                     ` Ulrich Drepper [this message]
2006-07-13  6:27                       ` Ian Wienand
2006-07-13 14:39                       ` Eric W. Biederman
2006-07-13 15:05                         ` Arjan van de Ven
  -- strict thread matches above, loose matches on Subject: below --
2006-07-13  5:00 Albert Cahalan
2006-07-13  5:42 ` H. Peter Anvin
2006-07-13  6:09   ` Albert Cahalan
2006-07-13  6:13     ` Albert Cahalan
2006-07-13  6:38 ` Eric W. Biederman
2006-07-13 16:15   ` Albert Cahalan
2006-07-13 16:53     ` Eric W. Biederman
2006-07-13 17:06       ` Albert Cahalan
2006-07-13 15:20 ` Eric W. Biederman

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=44B5D77F.60200@redhat.com \
    --to=drepper@redhat.com \
    --cc=akpm@osdl.org \
    --cc=arjan@infradead.org \
    --cc=ebiederm@xmission.com \
    --cc=hpa@zytor.com \
    --cc=jakub@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rdunlap@xenotime.net \
    --cc=roland@redhat.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