Great questions! I think you're right this is a tooling issue. Haskell has valgrind-like heap/memory analysis built right in but not much for thunk-based memory

Great questions! I think you're right this is a tooling issue. Haskell has valgrind-like heap/memory analysis built right in but not much for thunk-based memory allocation. Neil's article is really about Haskell's thunk analysis specifically and I imagine we will see some valgrind-for-thunk tools come out of this discussion.