[all-commits] [llvm/llvm-project] e72fb6: [mlir][arith] Convert fastmath to LLVM dialect for...

[Slava Zakharin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e72fb692104ebab682d1ea5aeec39358b11ad407
      https://github.com/llvm/llvm-project/commit/e72fb692104ebab682d1ea5aeec39358b11ad407
  Author: Slava Zakharin <szakharin at nvidia.com>
  Date:   2022-11-07 (Mon, 07 Nov 2022)

  Changed paths:
    M mlir/lib/Conversion/ArithToLLVM/ArithToLLVM.cpp
    M mlir/test/Conversion/ArithToLLVM/arith-to-llvm.mlir

  Log Message:
  -----------
  [mlir][arith] Convert fastmath to LLVM dialect for some arith ops.

This is a follow-up on D126305 and D136225.
We can now preserve fastmath for arith::MaxFOp,MinFOp,RemFOp during
ArithToLLVM conversion.

Reviewed By: rriddle

Differential Revision: https://reviews.llvm.org/D137456