[PATCH] D73620: Add 'gpu.terminator' operation.

[Alex Zinenko via Phabricator via llvm-commits]

ftynse added inline comments.


================
Comment at: mlir/test/mlir-cuda-runner/shuffle.mlir:24
     store %value, %kernel_dst[%tx] : memref<?xf32>
-    gpu.return
   }
----------------
herhut wrote:
> ftynse wrote:
> > Nit: we may want to skip this terminator while printing, but I won't insist.
> I was wondering the same. The only reason I left it printed was that I don't like a basic block without terminator. Confused my mental parser. So I only make them implicit for single-block operations. We could drop it if there is only a single block? Or is that too incoherent?
It sounds like too much code for sugaring the IR. Let's keep it visible all the time, having it consistent between single- and multiple-block regions is a good argument.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D73620/new/

https://reviews.llvm.org/D73620