[all-commits] [llvm/llvm-project] a131fb: Reland "[NVPTX] deprecate nvvm.rotate.* intrinsics...

[Alex MacLean via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a131fbf1687a641265ced45b19dc97ee1c9e3484
      https://github.com/llvm/llvm-project/commit/a131fbf1687a641265ced45b19dc97ee1c9e3484
  Author: Alex MacLean <amaclean at nvidia.com>
  Date:   2024-09-27 (Fri, 27 Sep 2024)

  Changed paths:
    M llvm/include/llvm/IR/IntrinsicsNVVM.td
    M llvm/lib/IR/AutoUpgrade.cpp
    M llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp
    M llvm/lib/Target/NVPTX/NVPTXISelLowering.h
    M llvm/lib/Target/NVPTX/NVPTXInstrInfo.td
    M llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
    M llvm/test/Assembler/auto_upgrade_nvvm_intrinsics.ll
    M llvm/test/CodeGen/NVPTX/rotate.ll
    M llvm/test/CodeGen/NVPTX/rotate_64.ll

  Log Message:
  -----------
  Reland "[NVPTX] deprecate nvvm.rotate.* intrinsics, cleanup funnel-shift handling" (#110025)

This change deprecates the following intrinsics which can be trivially
converted to llvm funnel-shift intrinsics:

- @llvm.nvvm.rotate.b32
- @llvm.nvvm.rotate.right.b64
- @llvm.nvvm.rotate.b64

This fixes a bug in the previous version (#107655) which flipped the
order of the operands to the PTX funnel shift instruction. In LLVM IR
the high bits are the first arg and the low bits are the second arg,
while in PTX this is reversed.



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