[all-commits] [llvm/llvm-project] bd860f: [NVPTX] Add intrinsics for redux.sync f32 instruct...

[Srinivasa Ravi via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: bd860f986406b6630e49b1836b3c208acd721d3e
      https://github.com/llvm/llvm-project/commit/bd860f986406b6630e49b1836b3c208acd721d3e
  Author: Srinivasa Ravi <srinivasar at nvidia.com>
  Date:   2025-02-14 (Fri, 14 Feb 2025)

  Changed paths:
    M clang/include/clang/Basic/BuiltinsNVPTX.td
    A clang/test/CodeGenCUDA/redux-f32-builtins.cu
    M llvm/include/llvm/IR/IntrinsicsNVVM.td
    M llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
    A llvm/test/CodeGen/NVPTX/redux-sync-f32.ll

  Log Message:
  -----------
  [NVPTX] Add intrinsics for redux.sync f32 instructions (#126664)

Adds NVVM intrinsics, NVPTX codegen and Clang builtins for `redux.sync`
f32 instructions introduced in ptx8.6 for sm_100a.
Tests added in `CodeGen/NVPTX/redux-sync.ll` and
`CodeGenCUDA/redux-builtins.cu` and verified through ptxas 12.8.0.

PTX Spec Reference:

https://docs.nvidia.com/cuda/parallel-thread-execution/#parallel-synchronization-and-communication-instructions-redux-sync



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