@ThomasRaoux Bump on the review :)

clarify helpers and cleanup

lint

Use invert helper instead of specialized function

rebase and use getPosition

Use getElementTypeOrSelf

lint

address comments

In D144113#4129557, @antiagainst wrote:

IMO, the accessor name could use some improvements---it's very confusing and error-prone just being getVectorType(); making it something like getSrcVectorType() would make it far easier to avoid pitfalls. At least for me, I need to read the impl to double check what it means by just getVectorType().

LGTM

I will let others review the transform dialect piece, but I see the movement of these patterns from IREE as a step in the right direction. Also, it would probably be good to pull the FileCheck tests from IREE as well, even if the intent is to later generalize the patterns as the transform won't change.

assert exactly one result

rebase and address comment

@ThomasRaoux I added changes to the GPUToNVVM lowering to support integer types. Would it be better to move those changes to a separate revision?

Add support for int8 types in lowering to nvvm and rebase

Remove check for empty uses

The lowering to SPIRV will just take the type from the mma_matrix type so as long as the types are set properly here then there should be no problems. The lowering otherwise doesn't look any different because the ops are the same. For the lowering to NVVM I realize I'm not as certain so I can either leave support for that to another PR or try and verify correctness now.

add back initial commit

Generalize pattern to arbitrary bit widths

This revision was erroneously opened.

rebase

Add description of transpose and relax check for transposed load.

Switch to OptionalAttr<UnitAttr> and update tests accordingly. Additionally changes the check for a transposed vector.transfer_read to directly check for a 2d transpose.