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

Javed Absar llvmlistbot at llvm.org
Sat Sep 13 02:14:25 PDT 2025


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.?

answered above.

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


More information about the Mlir-commits mailing list