[Mlir-commits] [mlir] [mlir][arith] Add LUT-based lowering of arith.extf for FP8 to F32. (PR #196321)

[Jakub Kuderski]

================

----------------
kuhar wrote:

It depends on what your target cares about more, the caches, etc. If you had a vector of values to convert, I’d expect the final sign computation to be very cheap all things considered. (This is not a blocking comment, just something to consider.)

https://github.com/llvm/llvm-project/pull/196321