[all-commits] [llvm/llvm-project] 3a6124: clang/AMDGPU: Assume denormals are enabled for the...

[Matt Arsenault via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 3a612450508b314b7a6f4db142d0c619031d760e
      https://github.com/llvm/llvm-project/commit/3a612450508b314b7a6f4db142d0c619031d760e
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2020-04-15 (Wed, 15 Apr 2020)

  Changed paths:
    M clang/lib/Driver/ToolChains/AMDGPU.cpp
    M clang/test/Driver/cl-denorms-are-zero.cl
    M clang/test/Driver/cuda-flush-denormals-to-zero.cu

  Log Message:
  -----------
  clang/AMDGPU: Assume denormals are enabled for the default target.

Since the default logic was based on having fast denormal/fma
features, and the default target has no features, we assumed flushing
by default. This fixes incorrectly assuming flushing in builds for
"generic" IR libraries.

The handling for no specified --cuda-gpu-arch in HIP is kind of
broken. Somewhere else forces a default target of gfx803, which does
not enable denormal handling by default. We don't see this default
switching here, so you'll end up with a different denormal mode
depending on whether you explicitly requested gfx803, or used it by
default.