[Mlir-commits] [mlir] [mlir][vector] Add vector.transpose with unit-dim to vector.shape_cast pattern (PR #72105)

[llvmlistbot at llvm.org]

MaheshRavishankar wrote:

Actually I tried one of the examples added with this PR. Using `convert-vector-to-spirv` does nothing. That pass is set up to apply partial conversion so it doesn't fail but does no conversion of the generated shape_cast itself. 
I am not sure how much this test would help. The solution here as Lei suggested is to move it out of general Vector lowering of transpose and use a separate populate method to use for different back ends. 

https://github.com/llvm/llvm-project/pull/72105