The Cost of a Closure in C, The Rest

Performance benchmarking reveals that C++ lambdas with perfect type information deliver the best closure performance, while GNU Nested Functions perform poorly enough to question their viability. Adding a single level of indirection (like int* instead of int) significantly degrades performance. Static and thread_local variables for passing context are slower than adding explicit function parameters. The analysis suggests C needs both type-preserving closures and thin type erasure (wide function pointers) for optimal performance, as existing extensions like Apple Blocks and GNU Nested Functions have fundamental design issues.