[Mlir-commits] [mlir] [mlir][gpu] Warp execute terminator getter (PR #154729)

[Charitha Saumya]

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

LGTM. very useful change. 

Existing code uses `cast<gpu::YieldOp>(warpOp.getBodyRegion().getBlocks().begin()->getTerminator());` to get the terminator. But this approach uses `cast<gpu::YieldOp>(getBody()->getTerminator())`. Wondering if they are the same or could be different in come cases?

https://github.com/llvm/llvm-project/pull/154729