[all-commits] [llvm/llvm-project] dfc512: [NVPTX] Consistently check fast-math flags when lo...

[Alex MacLean via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: dfc5125946ade289840fa119716957ebce2d31d2
      https://github.com/llvm/llvm-project/commit/dfc5125946ade289840fa119716957ebce2d31d2
  Author: Alex MacLean <amaclean at nvidia.com>
  Date:   2025-06-12 (Thu, 12 Jun 2025)

  Changed paths:
    M llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp
    M llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h
    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/CodeGen/NVPTX/fast-math.ll
    M llvm/test/CodeGen/NVPTX/sqrt-approx.ll

  Log Message:
  -----------
  [NVPTX] Consistently check fast-math flags when lowering fsqrt (#143776)

Ensure that we check the global, function-level, and instruction-level
flags when considering whether to use `sqrt.rn` or `sqrt.approx` to
lower either `@llvm.sqrt.f32` or `@llvm.nvvm.sqrt.f`



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