search for: omp2012

...ky for me -- I'm employed by a CPU vendor (Intel), and we have very strict rules and long processes for publishing benchmark results. I simply can't run a benchmark and say: "hey! clang has this number and gcc has that number". The only thing I can share is that we do tested SPEC OMP2012 ( https://www.spec.org/omp2012/), which is the industry standard for OMP benchmarks, on a non-server class Darwin machine, and the results are quite good and comparable with other compilers. Speaking on Phoronix, two benchmarks where clang always lose due to lack of OpenMP are "John the Rippe...

...n to eventually contribute everything to the clang trunk (initial patches have already been committed). This implementation supports 3.1 version of OpenMP standard in full; it passes all OpenMP tests we tried with it so far (this includes OpenMP Validation Suite from OpenUH Research Compiler, SPEC OMP2012 and internal Intel test suites). Performance-wise, it demonstrates similar gains and scalability as other compilers with OpenMP support. (Sorry, I can’t be more specific here, as properly reporting performance results is a precise and laborious process. You are welcome to try clang compiler with Op...

...r (Intel), and we >> have very strict rules and long processes for publishing benchmark results. >> I simply can't run a benchmark and say: "hey! clang has this number and gcc >> has that number". >> >> The only thing I can share is that we do tested SPEC OMP2012 >> (https://www.spec.org/omp2012/), which is the industry standard for OMP >> benchmarks, on a non-server class Darwin machine, and the results are quite >> good and comparable with other compilers. >> >> Speaking on Phoronix, two benchmarks where clang always lose due...

All, I'd like to resurrect the discussion on replacing libgomp with libiomp as the default OpenMP runtime library linked with -fopenmp. For reference, the previous discussion is accessible there: http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20140217/thread.html#99461 We are very close to getting *full* OpenMP 3.1 specification supported in clang (only one (!) clause is not