[Mlir-commits] [mlir] [mlir][mesh] Handling changed halo region sizes during spmdization (PR #114238)

[Frank Schlimbach]

fschlimb wrote:

> Wouldn't it be more intuitive to have offsets starting from zero unless there is a specific reason not to follow this approach?

Yes, we could do that. It would require one the array to be of size N+1, though. I was undecided at the time; if you have a clear preference I will make the change.

https://github.com/llvm/llvm-project/pull/114238