[Mlir-commits] [mlir] Revert "Support float8_e3m4 and float8_e4m3 in np_to_memref (#186453)" (PR #186677)

[Maksim Levental]

makslevental wrote:

Change was made here https://github.com/llvm/llvm-project/pull/122684 - looks like it's just conservative ie requires only free threading variant uses at least 0.5. we can just make that the "universal" minimum.

https://github.com/llvm/llvm-project/pull/186677