[Mlir-commits] [mlir] [mlir][linalg] Fix inferConvolutionDimsImpl (depthwise convs) (PR #90057)

[Quinn Dawkins]

https://github.com/qedawkins commented:

Yeah simplifications to the code would be welcome. There is a `depth` dim classification for this though, maybe would it make more sense if we renamed it to `depthChannel`? Also I'm not necessarily opposed to a change in classification, I just wrote the current impl to be more in line with the `inferContractionDims` impl and I haven't worked much with depthwise convs.

https://github.com/llvm/llvm-project/pull/90057