[Mlir-commits] [mlir] [mlir][ArmSME] Fix loop bounds of masked loads/stores (PR #78983)

[Andrzej Warzyński]

https://github.com/banach-space commented:

Based on the updates in [arm-sme-to-scf.mlir](https://github.com/llvm/llvm-project/pull/78983/files#diff-5b29cd224513275ab0dad25b89c92f3db789d4ba32708bf27c8a004fe6cb6bde), this should be fairly straightforward, but the diff is quite large.

It seems that a not-so-small refactor is combined with the actual fix. Would it make sense to split this into two patches:
* refactor
* fix
? WDYT?

https://github.com/llvm/llvm-project/pull/78983