Jesse Lu via llvm-dev
2020-May-22 18:27 UTC
[llvm-dev] __syncthreads() inrtinsic seems to be missing from __clang_cuda_device_functions.h
Hi! I see the `__syncthreads_{and|or|count}()` CUDA intrinsics listed in https://github.com/llvm/llvm-project/blob/master/clang/lib/Headers/__clang_cuda_device_functions.h but the vanilla `__syncthreads()` intrinsic seems to be missing. Is this an oversight, or is it omitted for a specific reason? Thank you! -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200522/4082015f/attachment.html>
Johannes Doerfert via llvm-dev
2020-May-22 23:42 UTC
[llvm-dev] __syncthreads() inrtinsic seems to be missing from __clang_cuda_device_functions.h
We list it as a builtin and it should be available if the target is nvptx. Do you have problems due to the missing function? Cheers, Johannes On 5/22/20 1:27 PM, Jesse Lu via llvm-dev wrote:> Hi! > > I see the `__syncthreads_{and|or|count}()` CUDA intrinsics listed in > https://github.com/llvm/llvm-project/blob/master/clang/lib/Headers/__clang_cuda_device_functions.h > but > the vanilla `__syncthreads()` intrinsic seems to be missing. Is this an > oversight, or is it omitted for a specific reason? > > Thank you! > > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200522/2cf5060f/attachment.html>
Jesse Lu via llvm-dev
2020-May-25 16:11 UTC
[llvm-dev] __syncthreads() inrtinsic seems to be missing from __clang_cuda_device_functions.h
https://github.com/ycm-core/ycmd seems to require this, as detailed in https://reviews.llvm.org/D80464. I'll let the issues be sorted out there, thank you! On Fri, May 22, 2020 at 4:43 PM Johannes Doerfert < johannesdoerfert at gmail.com> wrote:> We list it as a builtin and it should be available if the target is nvptx. > > Do you have problems due to the missing function? > > > Cheers, > > Johannes > > > On 5/22/20 1:27 PM, Jesse Lu via llvm-dev wrote: > > Hi! > > I see the `__syncthreads_{and|or|count}()` CUDA intrinsics listed inhttps://github.com/llvm/llvm-project/blob/master/clang/lib/Headers/__clang_cuda_device_functions.h > but > the vanilla `__syncthreads()` intrinsic seems to be missing. Is this an > oversight, or is it omitted for a specific reason? > > Thank you! > > > > _______________________________________________ > LLVM Developers mailing listllvm-dev at lists.llvm.orghttps://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200525/f4d3b789/attachment-0001.html>