[all-commits] [llvm/llvm-project] b02586: [AMDGPU] - Add clang builtins for tied WMMA intrin...

[Jessica Del via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b025864af8897bcc5791d93f5004cc1acbb397de
      https://github.com/llvm/llvm-project/commit/b025864af8897bcc5791d93f5004cc1acbb397de
  Author: Jessica Del <50999226+OutOfCache at users.noreply.github.com>
  Date:   2023-11-13 (Mon, 13 Nov 2023)

  Changed paths:
    M clang/include/clang/Basic/BuiltinsAMDGPU.def
    M clang/lib/CodeGen/CGBuiltin.cpp
    A clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w32-gfx10-err.cl
    A clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w32-param-err.cl
    M clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w32.cl
    A clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w64-gfx10-err.cl
    A clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w64-param-err.cl
    M clang/test/CodeGenOpenCL/builtins-amdgcn-wmma-w64.cl

  Log Message:
  -----------
  [AMDGPU] - Add clang builtins for tied WMMA intrinsics (#70669)

Add clang builtins for the new tied wmma intrinsics. 
These variations tie the destination
accumulator matrix to the input
accumulator matrix.

See https://github.com/llvm/llvm-project/pull/69903 for context.