All of lore.kernel.org
 help / color / mirror / Atom feed
From: Arjan van de Ven <arjan@linux.intel.com>
To: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Cc: linux-kernel@vger.kernel.org, ak@suse.de, torvalds@osdl.org,
	akpm@osdl.org, mingo@elte.hu
Subject: Re: Patch to reorder functions in the vmlinux to a defined order
Date: Thu, 23 Feb 2006 18:59:45 +0100	[thread overview]
Message-ID: <43FDF811.3090309@linux.intel.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0602230933080.1579@montezuma.fsmlabs.com>

Zwane Mwaikambo wrote:
> On Thu, 23 Feb 2006, Arjan van de Ven wrote:
> 
>> This patch puts the infrastructure in place to allow for a reordering of
>> functions based inside the vmlinux. The general idea is that it is possible
>> to put all "common" functions into the first 2Mb of the code, so that they
>> are covered by one TLB entry. This as opposed to the current situation where
>> a typical vmlinux covers about 3.5Mb (on x86-64) and thus 2 TLB entries.
>> (This patch depends on the previous patch to pin head.S as first in the order)
> 
> Hello Arjan,
> 	Assuming that functions defined in an object file are related and 
> hence benefit from cache spatial locality, doesn't this affect this 
> greatly? It would seem that with regards to the kernel image on x86, (2MB) 
> TLB usage isn't as scarce a resource as icache.

this is probably a mixed blessing; right now frequent and non-frequent 
code are highly intermixed, so even in the current situation the icache 
isn't optimally used.

So what you sort of want is a thing better than pure "sort", a sort that 
keeps existing order *within the hot functions*, but move the "cold" 
ones out. That is thankfully independent of the infrastructure; I'll 
keep that in mind when making the next version of the order list, it's a 
good idea.


  reply	other threads:[~2006-02-23 18:00 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-02-23 13:19 Patch to make the head.S-must-be-first-in-vmlinux order explicit Arjan van de Ven
2006-02-23 13:42 ` Andi Kleen
2006-02-23 13:57   ` Arjan van de Ven
2006-02-23 14:14     ` Andi Kleen
2006-02-23 14:40       ` Arjan van de Ven
2006-02-24 13:49       ` Eric W. Biederman
2006-02-24 13:53         ` Arjan van de Ven
2006-02-24 14:17           ` Eric W. Biederman
2006-02-24 14:32           ` Andi Kleen
2006-02-23 15:09 ` Patch to reorder functions in the vmlinux to a defined order Arjan van de Ven
2006-02-23 16:00   ` Andi Kleen
2006-02-23 16:02     ` Arjan van de Ven
2006-02-23 16:43     ` Arjan van de Ven
2006-02-23 16:48       ` Andi Kleen
2006-02-23 16:49         ` Arjan van de Ven
2006-02-23 17:08       ` Linus Torvalds
2006-02-23 17:20         ` Andi Kleen
2006-02-23 17:34           ` Alan Cox
2006-02-23 18:13             ` Andi Kleen
2006-02-23 17:44           ` Linus Torvalds
2006-02-23 18:14             ` Andi Kleen
2006-02-23 19:03         ` linux-os (Dick Johnson)
2006-02-23 19:32           ` Linus Torvalds
2006-02-24 12:43             ` linux-os (Dick Johnson)
2006-02-23 19:23         ` Rene Herman
2006-02-23 19:31           ` Andi Kleen
2006-02-23 19:34           ` Alan Cox
2006-02-23 19:48             ` Rene Herman
2006-02-23 19:35           ` Linus Torvalds
2006-02-23 19:44             ` Arjan van de Ven
2006-02-23 20:13             ` Rene Herman
2006-02-23 20:26               ` Dave Jones
2006-02-24  2:44                 ` Fernando Luis Vazquez Cao
2006-02-24  2:47                   ` Andi Kleen
2006-02-24 14:44                   ` Theodore Ts'o
2006-02-23 23:19               ` Linus Torvalds
2006-02-23 23:53                 ` Rene Herman
2006-02-24 14:23                   ` Eric W. Biederman
2006-02-24 14:23                   ` Eric W. Biederman
2006-02-24 15:30                     ` Rene Herman
2006-02-24 15:55                       ` Eric W. Biederman
2006-02-24 16:48                         ` Andi Kleen
2006-02-24 20:38                           ` Eric W. Biederman
2006-02-25  2:11                             ` Folkert van Heusden
2006-02-25  8:32                               ` Arjan van de Ven
2006-02-24 17:26                         ` Linus Torvalds
2006-02-24 17:57                           ` Rene Herman
2006-02-24 18:11                             ` Linus Torvalds
2006-03-07 23:36                             ` H. Peter Anvin
2006-02-24 14:11                 ` Eric W. Biederman
2006-02-24  1:38             ` Martin Bligh
2006-02-23 20:07         ` Andy Whitcroft
2006-02-23 16:48   ` Linus Torvalds
2006-02-23 16:54     ` Benjamin LaHaise
2006-02-23 18:01     ` Arjan van de Ven
2006-02-23 16:57   ` linux-os (Dick Johnson)
2006-02-23 17:36   ` Zwane Mwaikambo
2006-02-23 17:59     ` Arjan van de Ven [this message]
2006-02-24 22:19   ` Tony Luck
2006-02-24 22:21     ` Arjan van de Ven

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=43FDF811.3090309@linux.intel.com \
    --to=arjan@linux.intel.com \
    --cc=ak@suse.de \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=torvalds@osdl.org \
    --cc=zwane@arm.linux.org.uk \
    /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.