[Mlir-commits] [mlir] [MLIR] Support interrupting AffineExpr walks (PR #74792)

[Uday Bondhugula]

bondhugula wrote:

> LG, but would be better with some tests showing the interruption: do we have unit-tests for AffineExpr?

It's being exercised by an affine utility - the change to it is included here, but it doesn't really prove it's working, but only that compilation (of the compiler) is succeeding. We'll need to think of a new unit test where the interrupt emits a diagnostic which is checked. Is this what you had in mind? 

https://github.com/llvm/llvm-project/pull/74792