clang support for compiler:openmp

Matthieu Volat mazhe at alkumuna.eu
Mon Nov 7 22:09:15 UTC 2016


On Mon, 7 Nov 2016 18:59:26 -0300
"Nilton Jose Rizzo" <rizzo at i805.com.br> wrote:

> Em Mon, 7 Nov 2016 22:12:59 +0100, Matthieu Volat escreveu
> > On Mon, 7 Nov 2016 16:59:36 +0000
> > Brooks Davis <brooks at freebsd.org> wrote:
> > [...]

> 
>   I have one question about a llvm and clang ports.  it's possible
> port the llvm 3.9 and clang 4.0.0 with cuda support?
> 
> 

I don't think we'll see that anytime soon: even if there are now some compilers that produce cuda PTX bytecode, you neeed userland and kernel code to actually compile it to native GPU code/push it/do data transfert, and only the nvidia official driver do that for now.

There's some of that in the linux compatibility part of the nvidia driver, and some people managed to use it in the past, but right now, I'm not so sure.


-- Mazhe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 163 bytes
Desc: OpenPGP digital signature
URL: <http://lists.freebsd.org/pipermail/freebsd-ports/attachments/20161107/4d729ec7/attachment.sig>


More information about the freebsd-ports mailing list