Accéder au contenu.
Menu Sympa

starpu-devel - Re: [Starpu-devel] CUDA task and GCC plugin

Objet : Developers list for StarPU

Archives de la liste

Re: [Starpu-devel] CUDA task and GCC plugin


Chronologique Discussions 
  • From: Mehdi AMINI <mehdi.amini@silkan.com>
  • To: Ludovic Courtès <ludovic.courtes@inria.fr>
  • Cc: starpu-devel@lists.gforge.inria.fr
  • Subject: Re: [Starpu-devel] CUDA task and GCC plugin
  • Date: Mon, 11 Jun 2012 18:05:21 +0200
  • List-archive: <http://lists.gforge.inria.fr/pipermail/starpu-devel>
  • List-id: "Developers list. For discussion of new features, code changes, etc." <starpu-devel.lists.gforge.inria.fr>

Hi,


Currently I always end up with a linker error.

Oups, that’s a genuine bug (I’ve reproduced it.) I’ll let you know when
a fix is available.

Thanks ! :-)



I also have an issue related to implicit cpu implementations and -O3
gcc flag which seems to be broken. I'm testing a simple task call with
a CPU and a CUDA implementation. Using -O3 the CPU task is always
executed while without the -O3 flag the CUDA task is correctly
executed. Is it a known issue ?

I believe there’s nothing wrong here: the default scheduler (‘eager’)
makes non-deterministic scheduling decisions, and just happened to favor
CPU over CUDA for some reason.


It seems to be reproducible, I mean 100%
Moreover StarPU was compiled with --disable-cpu ; and finally I even reproduce it using STARPU_NCPUS=0 !

My guess is that -O3 may trigger an inlining of the original implicit implementation at call site so that the StarPU wrapper is never called and instead there directly the code of the CPU function.

Now, the CPU back-end can be disabled at run-time, forcing the scheduler
to use GPUs only:

$ STARPU_NCPUS=0 ./compile-test.sh

You missed the -O3 flag here, try again with :

STARPU_NCPUS=0 OPT_FLAG=-O3 ./compile-test.sh


Mehdi





Archives gérées par MHonArc 2.6.19+.

Haut de le page