Quinn Dawkins 985f7ff632 [mlir][gpu] Add support for integer types in gpu.subgroup_mma ops ...

The signedness is carried by `!gpu.mma_matrix` types to most closely
match the Cooperative Matrix specification which determines signedness
with the type (and sometimes the operation).

See: https://htmlpreview.github.io/?https://github.com/KhronosGroup/SPIRV-Registry/blob/master/extensions/NV/SPV_NV_cooperative_matrix.html

To handle the lowering from vector to gpu, ops such as arith.extsi are
pattern matched next to `vector.transfer_read` and `vector.contract` to
determine the signedness of the matrix type.

Enables s8 and u8 WMMA types in NVVM for the GPUToNVVM conversion.

Reviewed By: ThomasRaoux

Differential Revision: https://reviews.llvm.org/D143223

2023-02-07 17:58:01 -05:00

..

gpu-to-nvvm.mlir

[mlir] Introduce a pattern to lower gpu.subgroup_reduce to nvvm.redux_op

2023-01-20 13:56:23 +01:00

wmma-ops-to-nvvm.mlir

[mlir][gpu] Add support for integer types in gpu.subgroup_mma ops

2023-02-07 17:58:01 -05:00