All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>,
	Martin Bligh <mbligh@google.com>,
	Dave Hansen <hansendc@us.ibm.com>,
	linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
	linux-mm@kvack.org,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Subject: Re: [PATCH 1/4] x86_64: (SPARSE_VIRTUAL doubles sparsemem speed)
Date: Sun, 8 Apr 2007 00:18:44 +0200	[thread overview]
Message-ID: <200704080018.45084.ak@suse.de> (raw)
In-Reply-To: <Pine.LNX.4.64.0704071455060.31468@schroedinger.engr.sgi.com>

On Sunday 08 April 2007 00:06:13 Christoph Lameter wrote:

> Results:
> 
> x86_64 boot with virtual memmap
> 
> Format:               #events totaltime (min/avg/max)
> 
> kfree_virt_to_page       598430 5.6ms(3ns/9ns/322ns)
> 
> x86_64 boot regular sparsemem
> 
> kfree_virt_to_page       596360 10.5ms(4ns/18ns/28.7us)
> 
> 
> On average sparsemem virtual takes half the time than of sparsemem.

Nice.  But on what workloads? 

Anyways it looks promising. I hope we can just
replace old style sparsemem support with this for x86-64.

> Time is measured using the cycle counter (TSC on IA32, ITC on IA64) which has
> a very low latency.

Sorry that triggered my usual RDTSC rant...

Not on NetBurst (hundred of cycles) And on the others (C2,K8) it is a bit dangerous 
to measure short code blocks because RDTSC is not guaranteed ordered with the surrounding 
instructions.

-Andi
 

WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@suse.de>
To: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>,
	Martin Bligh <mbligh@google.com>,
	Dave Hansen <hansendc@us.ibm.com>,
	linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
	linux-mm@kvack.org,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Subject: Re: [PATCH 1/4] x86_64: (SPARSE_VIRTUAL doubles sparsemem speed)
Date: Sun, 8 Apr 2007 00:18:44 +0200	[thread overview]
Message-ID: <200704080018.45084.ak@suse.de> (raw)
In-Reply-To: <Pine.LNX.4.64.0704071455060.31468@schroedinger.engr.sgi.com>

On Sunday 08 April 2007 00:06:13 Christoph Lameter wrote:

> Results:
> 
> x86_64 boot with virtual memmap
> 
> Format:               #events totaltime (min/avg/max)
> 
> kfree_virt_to_page       598430 5.6ms(3ns/9ns/322ns)
> 
> x86_64 boot regular sparsemem
> 
> kfree_virt_to_page       596360 10.5ms(4ns/18ns/28.7us)
> 
> 
> On average sparsemem virtual takes half the time than of sparsemem.

Nice.  But on what workloads? 

Anyways it looks promising. I hope we can just
replace old style sparsemem support with this for x86-64.

> Time is measured using the cycle counter (TSC on IA32, ITC on IA64) which has
> a very low latency.

Sorry that triggered my usual RDTSC rant...

Not on NetBurst (hundred of cycles) And on the others (C2,K8) it is a bit dangerous 
to measure short code blocks because RDTSC is not guaranteed ordered with the surrounding 
instructions.

-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>

  reply	other threads:[~2007-04-07 22:19 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-04-01  7:10 [PATCH 1/2] Generic Virtual Memmap suport for SPARSEMEM Christoph Lameter
2007-04-01  7:10 ` Christoph Lameter
2007-04-01  7:10 ` [PATCH 1/4] x86_64: Switch to SPARSE_VIRTUAL Christoph Lameter
2007-04-01  7:10   ` Christoph Lameter
2007-04-01  7:11   ` Christoph Lameter
2007-04-01  7:11     ` Christoph Lameter
2007-04-01 10:46   ` Andi Kleen
2007-04-01 10:46     ` Andi Kleen
2007-04-02 15:37     ` Christoph Lameter
2007-04-02 15:37       ` Christoph Lameter
2007-04-02 15:44       ` Andi Kleen
2007-04-02 15:44         ` Andi Kleen
2007-04-02 15:51         ` Martin J. Bligh
2007-04-02 15:51           ` Martin J. Bligh
2007-04-02 15:54         ` Christoph Lameter
2007-04-02 15:54           ` Christoph Lameter
2007-04-02 17:14           ` Andi Kleen
2007-04-02 17:14             ` Andi Kleen
2007-04-02 17:34             ` Christoph Lameter
2007-04-02 17:34               ` Christoph Lameter
2007-04-02 19:54           ` Dave Hansen
2007-04-02 19:54             ` Dave Hansen
2007-04-02 20:11             ` Christoph Lameter
2007-04-02 20:11               ` Christoph Lameter
2007-04-02 20:13       ` Dave Hansen
2007-04-02 20:13         ` Dave Hansen
2007-04-02 20:30         ` Christoph Lameter
2007-04-02 20:30           ` Christoph Lameter
2007-04-02 20:38           ` Martin Bligh
2007-04-02 20:38             ` Martin Bligh
2007-04-02 20:51             ` Christoph Lameter
2007-04-02 20:51               ` Christoph Lameter
2007-04-05 11:50               ` Andy Whitcroft
2007-04-05 11:50                 ` Andy Whitcroft
2007-04-05 18:27                 ` Christoph Lameter
2007-04-05 18:27                   ` Christoph Lameter
2007-04-07 22:06                   ` [PATCH 1/4] x86_64: (SPARSE_VIRTUAL doubles sparsemem speed) Christoph Lameter
2007-04-07 22:06                     ` Christoph Lameter
2007-04-07 22:18                     ` Andi Kleen [this message]
2007-04-07 22:18                       ` Andi Kleen
2007-04-09 16:40                     ` William Lee Irwin III
2007-04-09 16:40                       ` William Lee Irwin III
2007-04-09 17:16                       ` Christoph Lameter
2007-04-09 17:16                         ` Christoph Lameter
2007-04-09 18:20                         ` William Lee Irwin III
2007-04-09 18:20                           ` William Lee Irwin III
2007-04-02 21:08           ` [PATCH 1/4] x86_64: Switch to SPARSE_VIRTUAL Dave Hansen
2007-04-02 21:08             ` Dave Hansen
2007-04-02 21:28             ` Christoph Lameter
2007-04-02 21:28               ` Christoph Lameter
2007-04-02 21:43               ` Martin Bligh
2007-04-02 21:43                 ` Martin Bligh
2007-04-02 21:56               ` Dave Hansen
2007-04-02 21:56                 ` Dave Hansen
2007-04-02 22:29                 ` Christoph Lameter
2007-04-02 22:29                   ` Christoph Lameter
2007-04-02 22:31                 ` Andi Kleen
2007-04-02 22:31                   ` Andi Kleen
2007-04-02 22:37                   ` Christoph Lameter
2007-04-02 22:37                     ` Christoph Lameter
2007-04-02 22:41                     ` Martin Bligh
2007-04-02 22:41                       ` Martin Bligh
2007-04-02 22:49                       ` Christoph Lameter
2007-04-02 22:49                         ` Christoph Lameter
2007-04-02 22:52                         ` Martin Bligh
2007-04-02 22:52                           ` Martin Bligh
2007-04-05 12:07                   ` Andy Whitcroft
2007-04-05 12:07                     ` Andy Whitcroft
2007-04-04 21:27               ` Bob Picco
2007-04-04 21:27                 ` Bob Picco
2007-04-04 22:38                 ` Christoph Lameter
2007-04-04 22:38                   ` Christoph Lameter
2007-04-02 23:03       ` Bjorn Helgaas
2007-04-02 23:03         ` Bjorn Helgaas
2007-04-02 15:44     ` Christoph Lameter
2007-04-02 15:44       ` Christoph Lameter
2007-04-02 16:18     ` Christoph Lameter
2007-04-02 16:18       ` Christoph Lameter
2007-04-02 20:50 ` [PATCH 1/2] Generic Virtual Memmap suport for SPARSEMEM Dave Hansen
2007-04-02 20:50   ` Dave Hansen
2007-04-02 21:00   ` Christoph Lameter
2007-04-02 21:00     ` Christoph Lameter
2007-04-02 21:22     ` Dave Hansen
2007-04-02 21:22       ` Dave Hansen
2007-04-02 21:31       ` Christoph Lameter
2007-04-02 21:31         ` Christoph Lameter
2007-04-02 21:42         ` Dave Hansen
2007-04-02 21:42           ` Dave Hansen
2007-04-02 21:53           ` Christoph Lameter
2007-04-02 21:53             ` Christoph Lameter
2007-04-02 22:05             ` Dave Hansen
2007-04-02 22:05               ` Dave Hansen

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=200704080018.45084.ak@suse.de \
    --to=ak@suse.de \
    --cc=apw@shadowen.org \
    --cc=clameter@sgi.com \
    --cc=hansendc@us.ibm.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mbligh@google.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.