[all-commits] [llvm/llvm-project] 4eaff6: libclc: Use elementwise exp for exp functions

[Matt Arsenault via All-commits]

  Branch: refs/heads/users/arsenm/libclc/use-exp-intrinsic
  Home:   https://github.com/llvm/llvm-project
  Commit: 4eaff61d6c2e130dc91ceb8d9d243d87b86e574a
      https://github.com/llvm/llvm-project/commit/4eaff61d6c2e130dc91ceb8d9d243d87b86e574a
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2026-03-10 (Tue, 10 Mar 2026)

  Changed paths:
    M libclc/clc/lib/amdgpu/CMakeLists.txt
    A libclc/clc/lib/amdgpu/math/clc_exp.cl
    A libclc/clc/lib/amdgpu/math/clc_exp10.cl
    A libclc/clc/lib/amdgpu/math/clc_exp2.cl

  Log Message:
  -----------
  libclc: Use elementwise exp for exp functions

For amdgpu use the exp intrinisc. Really, this should be
the default generic implementation. But we're stuck in a
mess where essentially nothing works. All of the exp
intrinsics work for AMDGPU, but aren't really implemented
for spirv or nvptx. Ideally the intrinsic and/or libm call
would be the default implementation.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications