I would be curious to see the results of compiling libreoffice with PORTAGE_TMPDIR on btrfs with compress=lzo.

On Feb 19, 2012 9:37 AM, "Mick" <michaelkintzios@gmail.com> wrote:
On Sunday 19 Feb 2012 12:57:51 Meik Frischke wrote:
> Am Sonntag, 19. Februar 2012, 12:26:57 schrieb Mick:
> > On Sunday 19 Feb 2012 11:45:25 Neil Bothwick wrote:
> > > On Sun, 19 Feb 2012 12:58:35 +0200, Thanasis wrote:
> > > > > [root@hactar ~ 0]% genlop -t libreoffice
> > > > >
> > > > >  * app-office/libreoffice
> > > > >
> > > > >      Wed Feb 15 10:23:58 2012 >>> app-office/libreoffice-3.5.0.3
> > > > >
> > > > >        merge time: 40 minutes and 52 seconds.
> > > >
> > > > Impressive. What's your hardware configuration, Neil?
> > >
> > > Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz with 16GB RAM.
> >
> > Hmm ... I've got a 1st generation i7 on my laptop and it is no-where near
> > that fast.  :-(
> >
> > Intel(R) Core(TM) i7 CPU       Q 720  @ 1.60GHz  with 4G RAM.
> >
> >      Sun Jan 29 21:47:46 2012 >>> app-office/libreoffice-3.4.5.2
> >
> >        merge time: 1 hour, 34 minutes and 43 seconds.
> >
> > Is it just down to the CPU or are you running /var/tmp/portage on some
> > turbo- charged fs?
>
> IO does really slow down the compilation progress a lot. Having
> PORTAGE_TMPDIR on tmpfs saves about half an hour (compared to reiserfs on
> a 7200rpm disk) for me :
> Intel(R) Core(TM) i7 CPU         920  @ 2.67GHz
>       Thu Feb 16 17:42:35 2012 >>> app-office/libreoffice-3.5.0.3
>          merge time: 50 minutes and 39 seconds.
>
> With 4GB Ram thats not an option for you though...

My /var partition is on ext4.

My CPU is admittedly slower, but even so I wouldn't think that it would take
more than _twice_ as long as Neil's.
--
Regards,
Mick