[Mlir-commits] [mlir] [mlir][tosa] Fix tosa-reduce-transposes to handle large constants better (PR #148755)

[Suraj Sudhir]

https://github.com/sjarus approved this pull request.

Thanks for this nice optimization! 

Regarding testing, since this is a performance optimization, we couldn't accomplish anything further than the existing LIT tests in LLVM/MLIR.

Could you also enable this as default within TorchToTosa in Torch-MLIR, assuming that's your frontend ? 

https://github.com/llvm/llvm-project/pull/148755