[libc-commits] [PATCH] D152486: [libc] Begin implementing a 'libmgpu.a' for math on the GPU

[Jon Chesterfield via Phabricator via libc-commits]

JonChesterfield added a subscriber: Matt.
JonChesterfield added a comment.

@Matt good news on ocml, thanks. I think we should add a wave size intrinsic, unconditionally expand it somewhere in clang and the backend and replace the current magic variable in ocml with it. The main appeal to encoding it that way is we can also immediately kill dead branches on the value of the intrinsic which would solve various O0 related problems from invalid but not called code reaching the backend.

@Joseph I'd really like us to be able to get rid of the header files that currently do the remapping from libm calls to the vendor ones. Have you spoken to the hip / cuda / openmp people to see if any of them are in principle willing to part with the current header translation thing?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D152486/new/

https://reviews.llvm.org/D152486