[Mlir-commits] [mlir] [MLIR][XeGPU] Enable WG-level mxfp GEMM via generalized shape_cast collapse inference (PR #201496)

[Sang Ik Lee]

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

----------------
silee2 wrote:

sg_data needs to be at least [32, 16] that is the minimum size to use load_nd with a f8E8M0FNU element type.

https://github.com/llvm/llvm-project/pull/201496