[all-commits] [llvm/llvm-project] 285b41: [libclc] Move sqrt to CLC library (#128748)

[Fraser Cormack via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 285b411e4635e8db2526d653488ee54dad2bff34
      https://github.com/llvm/llvm-project/commit/285b411e4635e8db2526d653488ee54dad2bff34
  Author: Fraser Cormack <fraser at codeplay.com>
  Date:   2025-02-27 (Thu, 27 Feb 2025)

  Changed paths:
    M libclc/CMakeLists.txt
    M libclc/amdgpu/lib/SOURCES
    R libclc/amdgpu/lib/math/sqrt.cl
    M libclc/clc/include/clc/float/definitions.h
    A libclc/clc/include/clc/math/clc_sqrt.h
    A libclc/clc/lib/amdgpu/SOURCES
    A libclc/clc/lib/amdgpu/math/clc_sqrt_fp64.cl
    M libclc/clc/lib/generic/SOURCES
    A libclc/clc/lib/generic/math/clc_sqrt.cl
    A libclc/clc/lib/generic/math/clc_sqrt.inc
    R libclc/generic/include/math/clc_sqrt.h
    M libclc/generic/lib/SOURCES
    M libclc/generic/lib/math/clc_hypot.cl
    R libclc/generic/lib/math/clc_sqrt.cl
    R libclc/generic/lib/math/clc_sqrt_impl.inc
    M libclc/generic/lib/math/sqrt.cl

  Log Message:
  -----------
  [libclc] Move sqrt to CLC library (#128748)

This is fairly straightforward for most targets.

We use the element-wise sqrt builtin by default. We also remove a legacy
pre-filtering of the input argument, which the intrinsic now officially
handles.

AMDGPU provides its own implementation of sqrt for double types. This
commit moves this into the implementation of CLC sqrt. It uses weak
linkage on the 'default' CLC sqrt to allow AMDGPU to only override the
builtin for the types it cares about.



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