[Mlir-commits] [mlir] [mlir][x86] Add extra patterns for AMX nano-kernel lowering (acc with zero constant) (PR #222513)

[Arun Thangamani]

arun-thmn wrote:

@jopperm Updated the patch, to ignore the un-necessary loop with loads/stores for VNNI packed with cst.
Please check.

https://github.com/llvm/llvm-project/pull/222513