This is the mail archive of the
mailing list for the GCC project.
Re: The nvptx port [1/11+] indirect jumps
- From: Richard Biener <richard dot guenther at gmail dot com>
- To: Bernd Schmidt <bernds at codesourcery dot com>
- Cc: Jakub Jelinek <jakub at redhat dot com>, Jeff Law <law at redhat dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 22 Oct 2014 10:18:49 +0200
- Subject: Re: The nvptx port [1/11+] indirect jumps
- Authentication-results: sourceware.org; auth=none
- References: <54451994 dot 9070209 at codesourcery dot com> <544519D8 dot 70606 at codesourcery dot com> <5446A55A dot 6060807 at redhat dot com> <5446C973 dot 3040102 at codesourcery dot com> <20141021213025 dot GE10376 at tucnak dot redhat dot com> <5446D0D4 dot 9000706 at codesourcery dot com>
On Tue, Oct 21, 2014 at 11:32 PM, Bernd Schmidt <firstname.lastname@example.org> wrote:
> On 10/21/2014 11:30 PM, Jakub Jelinek wrote:
>> At least for OpenMP, the best would be if the #pragma omp target regions
>> and/or #pragma omp declare target functions contain anything a particular
>> offloading accelerator can't handle, instead of failing the whole
>> compilation perhaps just emit some at least by default non-fatal warning
>> and not emit anything for the particular offloading target, which would
>> either host fallback, or, if some other offloading target succeeded, just
>> that target.
> I guess a test could be added to mkoffload if gcc were to return a different
> value for a sorry vs. any other compilation failure. The tool could then
> choose not to produce offloading support for that target.
But that would be for the whole file instead of for the specific region?
So maybe we should produce one LTO offload object for each offload
function and make the symbols they are supposed to provide weak
so a fail doesn't end up failing to link the main program?
Looks like this gets somewhat awkward with the LTO setup.