[Mlir-commits] [mlir] [mlir][sparse] Expand LevelType to 64 bits and implement parsing n out of m (PR #79935)

[Yinying Li]

================
@@ -26,20 +26,20 @@ MLIR_DECLARE_CAPI_DIALECT_REGISTRATION(SparseTensor, sparse_tensor);
 /// If updating, keep them in sync and update the static_assert in the impl
 /// file.
 enum MlirSparseTensorLevelType {
----------------
yinying-lisa-li wrote:

Good point! I looked into this and I found that "The choice of type is implementation-defined, but shall be capable of representing the values of all the members of the enumeration." 
[https://stackoverflow.com/questions/17045661/are-there-any-guarantees-on-the-representation-of-large-enum-values](url)

https://github.com/llvm/llvm-project/pull/79935