[Mlir-commits] [mlir] [mlir] [memref] Elevate `AllocOp`s to `GlobalOp`s pass (PR #211141)

[ioana ghiban]

ioghiban wrote:

> @ioghiban what do you think about writing a alloc_to_global transform which like the alloca_to_global transform would offload the complexity of performing checks to the user of the transform? This would likely work for our use case as well.

SGTM, I'd just add some comments making `MemRefAllocToGlobalOp`'s assumptions transparent to the user

https://github.com/llvm/llvm-project/pull/211141