Linux Archive

Linux Archive (http://www.linux-archive.org/)
-   Fedora Development (http://www.linux-archive.org/fedora-development/)
-   -   compilation architecture (http://www.linux-archive.org/fedora-development/35654-compilation-architecture.html)

"Jakub 'Livio' Rusinek" 01-12-2008 04:26 PM

compilation architecture
 
do we need to support legacy cpu's by i386 compilation?
i586 would make fedora faster even 3 times.
difference is noticeable.
--
Jakub 'Livio' Rusinek

http://liviopl.jogger.pl/
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

drago01 01-12-2008 04:51 PM

compilation architecture
 
2008/1/12 Jakub 'Livio' Rusinek <jakub.rusinek@gmail.com>:
> do we need to support legacy cpu's by i386 compilation?
> i586 would make fedora faster even 3 times.
> difference is noticeable.

.....
where are your benchmarks for the "3 times faster" claim?
the i386 packages are already optimized for newer cpus (mtune vs. march)
where it makes sense to have i686 versions there are some (kernel, glibc)

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

"Jakub 'Livio' Rusinek" 01-12-2008 05:06 PM

compilation architecture
 
I've used openSUSE (i586) with the same configuration and packages installed and eg. Firefox 3 was started in < 1 sec. I have no benchmarks but everything runs faster and is more responsible.

Our optimalization does nothing compared to > i386 compilation.


2008/1/12, drago01 <drago01@gmail.com>:
2008/1/12 Jakub 'Livio' Rusinek <jakub.rusinek@gmail.com>:
> do we need to support legacy cpu's by i386 compilation?
> i586 would make fedora faster even 3 times.

> difference is noticeable.

.....
where are your benchmarks for the "3 times faster" claim?
the i386 packages are already optimized for newer cpus (mtune vs. march)
where it makes sense to have i686 versions there are some (kernel, glibc)


--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list



--
Jakub 'Livio' Rusinek
http://liviopl.jogger.pl/
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Patrice Dumas 01-12-2008 05:12 PM

compilation architecture
 
On Sat, Jan 12, 2008 at 07:06:50PM +0100, Jakub 'Livio' Rusinek wrote:
> I've used openSUSE (i586) with the same configuration and packages installed
> and eg. Firefox 3 was started in < 1 sec. I have no benchmarks but
> everything runs faster and is more responsible.
>
> Our optimalization does nothing compared to > i386 compilation.

I can only guess that there are many differences besides the arch
between suse and fedora. Did you try a more meaningfull test, for
instance rebuilding the firefox rpm as i586 and seeing what the gain is?

--
Pat

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Denis Leroy 01-12-2008 05:13 PM

compilation architecture
 
Jakub 'Livio' Rusinek wrote:
I've used openSUSE (i586) with the same configuration and packages
installed and eg. Firefox 3 was started in < 1 sec. I have no benchmarks
but everything runs faster and is more responsible.


I'm afraid you're going to have to come up with hard scientific evidence
to convince anyone. Firefox 3 startup is probably dominated by disk I/O
anyways (all those dynamic libraries it has to load). Still, It would be
a fun experiment to run. Compute the cold (i.e. just after boot) startup
time of firefox, then recompiles all of its dependencies (all the way
down to glibc) with i586, reboot, and recomputes. To be honest, I'd be
surprised if you had even one tenth of a second of difference.


-denis

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

"Jakub 'Livio' Rusinek" 01-12-2008 05:27 PM

compilation architecture
 
Don't be silly. Try some > i386 distro, and look at the difference.
About Firefox, I wasn't using Firefox 3 from RPM, but Trunk snapshot from Mozilla's FTP.
--
Jakub 'Livio' Rusinek

http://liviopl.jogger.pl/
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

John Reiser 01-12-2008 05:50 PM

compilation architecture
 
> Don't be silly. Try some > i386 distro, and look at the difference.

The rule is: Anyone who proposes a change because of performance,
must supply *measurements* which support the change.
The measurements must be reproducible on other machines
by other members of the community, so the environment and
test procedure must be well-specified, too.

--

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Ignacio Vazquez-Abrams 01-12-2008 05:54 PM

compilation architecture
 
On Sat, 2008-01-12 at 19:27 +0100, Jakub 'Livio' Rusinek wrote:
> Don't be silly. Try some > i386 distro, and look at the difference.
> About Firefox, I wasn't using Firefox 3 from RPM, but Trunk snapshot
> from Mozilla's FTP.

You're suggesting trying a completely different architecture... with
completely different features... and completely different build options?

http://en.wikipedia.org/wiki/Scientific_method

--
Ignacio Vazquez-Abrams <ivazqueznet@gmail.com>

PLEASE don't CC me; I'm already subscribed

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

"Jakub 'Livio' Rusinek" 01-12-2008 06:18 PM

compilation architecture
 
2008/1/12, Ignacio Vazquez-Abrams <ivazqueznet@gmail.com>:
On Sat, 2008-01-12 at 19:27 +0100, Jakub 'Livio' Rusinek wrote:
> Don't be silly. Try some > i386 distro, and look at the difference.
> About Firefox, I wasn't using Firefox 3 from RPM, but Trunk snapshot

> from Mozilla's FTP.

You're suggesting trying a completely different architecture... with
completely different features... and completely different build options?


http://en.wikipedia.org/wiki/Scientific_method

--
Ignacio Vazquez-Abrams <ivazqueznet@gmail.com>

PLEASE don't CC me; I'm already subscribed


--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list


So I can say only one thing: "fedora does something in wrong, performance-loss way".
--
Jakub 'Livio' Rusinek

http://liviopl.jogger.pl/
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

Jesse Keating 01-12-2008 06:33 PM

compilation architecture
 
On Sat, 12 Jan 2008 20:18:34 +0100
"Jakub 'Livio' Rusinek" <jakub.rusinek@gmail.com> wrote:

> So I can say only one thing: "fedora does something in wrong,
> performance-loss way".

You could say that, but you'd be wrong.

--
Jesse Keating
Fedora -- All my bits are free, are yours?
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list


All times are GMT. The time now is 10:27 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.