[Mlir-commits] [mlir] [mlir][linalg] convert arith ops to destination-passing-style. (PR #157854)

[Javed Absar]

javedabsar1 wrote:

> Why do we go to `linalg.generic` instead of `linalg.elementwise` (and potentially generalize those later)? Are you worried about support for fast math flags etc.?

We could just use `linalg-morph-ops=generic-to-named` etc.

https://github.com/llvm/llvm-project/pull/157854