This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Memory partitioning considered harmful (!?)
- From: Richard Guenther <rguenther at suse dot de>
- To: Toon Moene <toon at moene dot indiv dot nluug dot nl>
- Cc: Diego Novillo <dnovillo at redhat dot com>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 31 Jan 2007 13:10:21 +0100 (CET)
- Subject: Re: Memory partitioning considered harmful (!?)
- References: <45BF88A8.2080700@moene.indiv.nluug.nl>
On Tue, 30 Jan 2007, Toon Moene wrote:
> [ Sorry for breaking the thread. This e-mail hasn't reached my system
> yet, so I reply from the archive ... ]
>
> I applied Richi's patch and it also affects my swapping problem: Less
> swapping, sped up throughput, with the same CPU time.
>
> I now get an wall-clock time of 1 hour 10 minutes for 3200 seconds of CPU
> time.
>
> It also shows in the amount of swap space used: with the standard code I use ~
> 450 Mbyte of RAM, ~ 470 Mbyte of SWAP, with Richi's patch the SWAP requirement
> goes down to ~ 310 Mbyte.
>
> Now if someone could *explain* it, that would be nice :-)
Interesting. Or rather, impossible ;) (I assume all the above numbers
are *runtime* of whatever benchmark you compiled, not numbers from
compiling it, right?)
> Are automatic arrays affected by this patch ? Our code uses lots of local
> arrays whose size is based on an integer argument to the routine the array is
> local to.
I doubt - but can you check?
Richard.
--
Richard Guenther <rguenther@suse.de>
Novell / SUSE Labs