[PATCH] D75391: [mlir][Linalg] Fix load/store operations generated while lower loops when output has zero rank.

[Nicolas Vasilache via Phabricator via llvm-commits]

nicolasvasilache accepted this revision.
nicolasvasilache added a comment.
This revision is now accepted and ready to land.

Let's push this forward to unblock progress, in the future the understanding is that linalg ops will not see 0-D memrefs anymore.



================
Comment at: mlir/lib/Dialect/Linalg/Transforms/LinalgToLoops.cpp:257
+      Value output = genericOp.getOutputBuffer(i);
+      if (!output.getType().cast<ShapedType>().getRank()) {
+        indexedValues[nInputs + i] = std_load(output);
----------------
Can we invert the conditional and make it `if (X > 0) {} else {}` ?
I find it more readable than the `not` test. 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75391/new/

https://reviews.llvm.org/D75391