[flang-commits] [flang] [flang] Canonicalize fir.array_coor by pulling in embox/rebox. (PR #92858)

[Razvan Lupusoru via flang-commits]

================
@@ -394,11 +394,17 @@ mlir::LogicalResult fir::ArrayCoorOp::verify() {
     } else {
       auto s = mlir::cast<fir::ShiftType>(shapeTy);
       shapeTyRank = s.getRank();
+      // TODO: it looks like PreCGRewrite and CodeGen can support
+      // fir.shift with plain array reference, so we may consider
+      // removing this check.
       if (!mlir::isa<fir::BaseBoxType>(getMemref().getType()))
         return emitOpError("shift can only be provided with fir.box memref");
     }
     if (arrDim && arrDim != shapeTyRank)
       return emitOpError("rank of dimension mismatched");
+    // TODO: support slicing with changing the numbder of dimensions,
+    // e.g. when array_coord represents an element access to array(:,1,:)
----------------
razvanlupusoru wrote:

array_coord -> array_coor

https://github.com/llvm/llvm-project/pull/92858