[cfe-dev] Parallelism TS implementation and feasibility of GPU execution policies

Ronan KERYELL ronan-list at keryell.fr
Thu Apr 9 22:23:21 PDT 2015


>>>>> On Thu, 2 Apr 2015 08:58:42 -0400, Andrew Corrigan <andrew.corrigan at nrl.navy.mil> said:

    Andrew> Newcomer here, I hope this isn’t off-topic, but this seemed
    Andrew> to be the most appropriate place to ask: Are there plans to
    Andrew> implement Parallelism-TS in libc++/Clang?  If so, what
    Andrew> execution policies might be supported?

    [...]

    Andrew> Finally, are there plans, or would it even be feasible, to
    Andrew> target OpenCL/SYCL/SPIR(-V) via Parallelism-TS?  I am aware
    Andrew> of existing OpenCL-based parallel algorithms library but I
    Andrew> am really hoping for a Parallelism TS execution policy for
    Andrew> OpenCL devices, so that it is a single-source,
    Andrew> fully-integrated approach that one can pass C++ function
    Andrew> objects to directly, as opposed to being restricted to
    Andrew> passing strings containing OpenCL C99 syntax, or having to
    Andrew> pre-instantiatiate template functors with macro wrappers.

I know that Codeplay has a Parallelism-TS implementation based on OpenCL
SYCL.

I guess they will talk about this during http://www.iwocl.org/ next
month where there will be several sessions on SYCL.

They presented their Clang/LLVM-based SYCL implementation during the
Workshop on the LLVM Compiler Infrastructure in HPC at SuperComputing
2014
http://www.codeplay.com/public/uploaded/publications/SC2014_LLVM_HPC.pdf
but I just checked and it does not talk about Parallelism-TS yet.

-- 
  Ronan KERYELL




More information about the cfe-dev mailing list