[Openmp-dev] [RFC] Clarify the absence of API stability for the *device runtime* (aka. libomptarget-nvptx-sm_XX.bc)

[Ye Luo via Openmp-dev]

It seems that my experience that mixing an older version of libomp and a
newer version of libomptarget hangs the application indicates the stability
of libomp is limited at this point in time.
Ye
===================
Ye Luo, Ph.D.
Computational Science Division & Leadership Computing Facility
Argonne National Laboratory


On Fri, Jul 10, 2020 at 1:43 AM Jonas Hahnfeld via Openmp-dev <
openmp-dev at lists.llvm.org> wrote:

> Am Donnerstag, den 09.07.2020, 17:00 -0500 schrieb Michael Kruse:
> > Am Do., 9. Juli 2020 um 15:57 Uhr schrieb Jonas Hahnfeld <
> hahnjo at hahnjo.de>:
> > > I asked the maintainer to install libomptarget and the plugin because
> > > otherwise there are link errors when compiling an OpenMP application
> > > with target directives. That's worse from a users perspective.
> >
> > Isn't subtly miscompiling because a runtime function semantics changed
> > between versions worse than a link-time error?
>
> At that time, there was no "changed semantics" in the host runtime and
> there was no intent to change any interface functions, that's a rather
> recent idea of some. Given the stability guarantees for libomp, I
> assume(d) the same for libomptarget and I think there were even
> discussions on some review (could be wrong about that).
>
> Jonas
> _______________________________________________
> Openmp-dev mailing list
> Openmp-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/openmp-dev/attachments/20200710/1b40993e/attachment.html>