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

[Matt Arsenault via Phabricator via libc-commits]

arsenm added a comment.

In D152486#4408675 <https://reviews.llvm.org/D152486#4408675>, @JonChesterfield wrote:

> Math optimisations were a bit of a mess a few years ago.

And they still are. The one I'd currently like to tackle is the fact that ocml treats select llvm generic math intrinsics like they're target specific fast operations. Worst offender is sqrt and log


Repository:
  rG LLVM Github Monorepo

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

https://reviews.llvm.org/D152486