[Mlir-commits] [mlir] [mlir][xegpu] SIMT distribution patterns for XeGPU CreateNdTdesc, LoadNd, StoreNd and Dpas Ops. (PR #135271)

[Frank Schlimbach]

fschlimb wrote:

> Hi @fschlimb, this is Charitha from the IMEX team. We have the initial part of the XeGPU subgroup - SIMT distribution work ready for review. If you are interested and have the bandwidth, please have a look and give us feedback/approval. Thanks!

Thanks for getting in touch, interesting!

I added a few comments. This is not an area that I typically work in, so they are mostly on monkey level. 

While distribution is different in this context than in the context of distributed memory, there a commonalities (similar to similarities in tiling and sharding). It seems to me that tiling/vector-distribution are special cases of general sharding/spmdization. Unification might be worth considering.

Out of curiosity, in addition to the question about the forward propagation, I wonder how tensor shapes that do not evenly divide to lane-sizes would be treated.


https://github.com/llvm/llvm-project/pull/135271