This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix omp declare simd cloning (PR tree-optimization/60823)
- From: Rainer Orth <ro at CeBiTec dot Uni-Bielefeld dot DE>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: Richard Biener <rguenther at suse dot de>, gcc-patches at gcc dot gnu dot org
- Date: Thu, 15 May 2014 14:47:19 +0200
- Subject: Re: [PATCH] Fix omp declare simd cloning (PR tree-optimization/60823)
- Authentication-results: sourceware.org; auth=none
- References: <20140418133047 dot GD1817 at tucnak dot redhat dot com> <ydd61la3vys dot fsf at lokon dot CeBiTec dot Uni-Bielefeld dot DE> <ydd4n0t1sur dot fsf at CeBiTec dot Uni-Bielefeld dot DE> <20140513175050 dot GB10386 at tucnak dot redhat dot com> <yddd2ffrzea dot fsf at lokon dot CeBiTec dot Uni-Bielefeld dot DE> <20140515124217 dot GM10386 at tucnak dot redhat dot com>
Jakub Jelinek <jakub@redhat.com> writes:
> On Thu, May 15, 2014 at 02:37:33PM +0200, Rainer Orth wrote:
>> > If OpenMP declare simd doesn't work on Solaris/x86 (due to the bogus hw cap
>> > stuff), then supposedly vect_simd_clones effective target should fail there.
>>
>> I don't think it's bogus: it guards against a similar kind of problems
>> as symbol versioning. There, programs that depend on a missing
>> interface don't start to run instead of crashing in the middle of
>> execution when a function is missing. With hwcap, programs depending on
>> insns not supported by the host don't even start running instead of
>> crashing later on.
>
> Runtime selection of code is very common though, which is why I think that
> whole idea of hw cap flags checking is bogus.
That may be changing, but what I observe far more often is code compiled
on one system with -mcpu/arch=native, later you try to execute it on a
different system and it crashes, giving no real indication why.
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University