[PATCH] D78226: [MLIR] Improve support for 0-dimensional Affine Maps.

[River Riddle via Phabricator via llvm-commits]

rriddle accepted this revision.
rriddle marked an inline comment as done.
rriddle added inline comments.
This revision is now accepted and ready to land.


================
Comment at: mlir/include/mlir/Dialect/Utils/StructuredOpsUtils.h:34
+  auto mapC = AffineMapAttr::get(AffineMap::get(3, 0, {m, n}, context));
+  auto maps = ArrayAttr::get({mapA, mapB, mapC}, context);
   return indexingMaps == maps;
----------------
Unrelated to this revision, but I don't see why this ArrayAttr needs to be generated.


================
Comment at: mlir/lib/IR/AffineMap.cpp:130
 inferFromExprList(ArrayRef<AffineExprContainer> exprsList) {
+  assert(exprsList.size() != 0);
+  assert(exprsList[0].size() != 0);
----------------
nit: prefer empty instead of size.


================
Comment at: mlir/lib/IR/Builders.cpp:300
 AffineMap Builder::getDimIdentityMap() {
-  return AffineMap::get(/*dimCount=*/1, /*symbolCount=*/0,
-                        {getAffineDimExpr(0)});
+  return AffineMap::get(/*dimCount=*/1, /*symbolCount=*/0, getAffineDimExpr(0));
 }
----------------
Is this wrapped at 80 characters?


================
Comment at: mlir/lib/Transforms/PipelineDataTransfer.cpp:102
+  auto modTwoMap =
+      AffineMap::get(/*dimCount=*/1, /*symbolCount=*/0, d0.floorDiv(step) % 2);
   auto ivModTwoOp = bInner.create<AffineApplyOp>(forOp.getLoc(), modTwoMap,
----------------
Same here, is this formatted?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78226/new/

https://reviews.llvm.org/D78226