This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: GCC OpenAcc executables problems
- From: Alexander Monakov <amonakov at ispras dot ru>
- To: Esteban Hernández <eshernan at gmail dot com>
- Cc: gcc-help at gcc dot gnu dot org, Nathan Sidwell <nathan at acm dot org>
- Date: Thu, 17 Dec 2015 20:26:56 +0300 (MSK)
- Subject: Re: GCC OpenAcc executables problems
- Authentication-results: sourceware.org; auth=none
- References: <CAMk=KywqapnxQzY4kVXNES=3bLLck=MwLR0mLak5dx2-=Ur+1w at mail dot gmail dot com>
On Thu, 17 Dec 2015, Esteban HernÃndez wrote:
> Dear,
>
> I compile gcc 6 r231616, for offloading and host,
>
>
> When i compile a simple example with -fopenacc -foffload=nvptx-none
>
> The compilation process finish o.k, but when i Run the code in a
> machine with cuda device, obtain the following problem
>
>
> libgomp: cuModuleLoadData error: device kernel image is invalid
Are you perhaps using new Maxwell-generation GPU for testing (Geforce 750,
950+)? There's odd code in libgomp plugin that insists on producing binary
code for Kepler-class hardware only.
Nathan, is that a mistake, or was there some reason to fix JIT target? We are
removing that code on gomp-nvptx branch:
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=1099ad42d145faf428053018d02b26678c47adce
Alexander