On Dec 18, 2007, at 11:12 PM, Duncan Sands wrote: > PS: > >> TREE_CODE(type) == POINTER_TYPE || >> TREE_CODE(type) == REFERENCE_TYPE) && "not a sequential >> type!"); >> // This relies on gcc types with constant size mapping to LLVM >> types with the >> - // same size. >> - return !VOID_TYPE_P(TREE_TYPE(type)) && >> isInt64(TYPE_SIZE(TREE_TYPE(type)), true); >> + // same size. It is possible for the component type not to have >> a size: >> + // struct foo; extern foo bar[]; >> + return !VOID_TYPE_P(TREE_TYPE(type)) && >> + (!TYPE_SIZE(TREE_TYPE(type)) || >> + isInt64(TYPE_SIZE(TREE_TYPE(type)), true)); >> } > > does the !TYPE_SIZE test also catch !VOID_TYPE_P case?
Yes, it seems TYPE_SIZE(void) is currently null. It would be equally logical to represent it as size 0, but I guess the code would still work in that case. _______________________________________________ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
