Dear Ole,

we are currently facing the same issue, I guess we will start installing 
software in April. At the moment we have not made a decision about the 
toolchains, but we probably start with which ever version of the foss-toolchain 
is available by then. In principle, you could use a site-specific definition of 
the foss-toolchain and replace the GCC-version, but that might lead to 
complications later on. One of these complications could be the combination of 
foss and CUDA, because the recent CUDA 12.1 only support GCC up to version 12.2.

Best wishes
Stefan

--
Scientific Computing

Carl von Ossietzky University Oldenburg
School of Mathematics and Natural Sciences
26111 Oldenburg, Germany

Office: W03 1-139
Phone: +49-441-798 3147
E-Mail: [email protected]
www: http://www.uni-oldenburg.de/fk5/wr

-----Ursprüngliche Nachricht-----
Von: [email protected] <[email protected]> Im 
Auftrag von Ole Holm Nielsen
Gesendet: Dienstag, 14. März 2023 11:05
An: [email protected]
Betreff: Re: [easybuild] Which toolchain to use on AMD EPYC 9004 "Genoa"?

ACHTUNG! Diese E-Mail kommt von Extern! WARNING! This email originated 
off-campus.

On 2/15/23 15:30, Jure Pečar wrote:
> On Wed, 15 Feb 2023 15:26:26 +0100
> Ole Holm Nielsen <[email protected]> wrote:
>
>> Is "foss" also the preferred toolchain on AMD Rome and Genoa?
>
> For now, yes.
>
> There's some work going on to create a toolchain around amd compilers 
> but it's questionable how much you gain as all features these 
> compilers and libs bring eventually end up in upstream llvm and gcc.
>
> So for Genoa look for upcoming foss toolchain built around gcc 13. For 
> Rome and Milan existing ones with gcc 11 and 12 are already working fine.

Sorry for my late reply.  The latest GCC in EasyBuild is 12.2.0.  Is there any 
way to build GCC 13 with EasyBuild so that we can test AMD Genoa?

Obviously, a lot of other modules are needed as well, so when can a foss 
toolchain with GCC 13 be hoped for?

Thanks,
Ole

--
Ole Holm Nielsen
PhD, Senior HPC Officer
Department of Physics, Technical University of Denmark

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to