[Mlir-commits] [mlir] [mlir][sparse] complete migration to dim2lvl/lvl2dim in library (PR #69268)

Peiming Liu llvmlistbot at llvm.org
Mon Oct 16 19:34:04 PDT 2023


================
@@ -334,8 +301,7 @@ class SparseTensorStorage final : public SparseTensorStorageBase {
   /// Partially specialize lexicographical insertions based on template types.
   void lexInsert(const uint64_t *lvlCoords, V val) final {
     assert(lvlCoords);
-    // TODO: get rid of this! canonicalize all-dense "sparse" array into dense
-    // tensors.
+    // TODO: needed?
----------------
PeimingLiu wrote:

Depending on whether we still want to handle "all-dense sparse" tensor. If so, I think we can remove the TODO here.

https://github.com/llvm/llvm-project/pull/69268


More information about the Mlir-commits mailing list