Andreas Rossberg

Generalising the exception proposal with resumption was the starting point for our work on effect handlers (which are more or less exceptions with resumption). That generalisation sounds attractive at first,...

> Unresumable exceptions are a special case of resumable exceptions (e.g. inhabitable resume type), so a coherent design would treat the two alike. Not once you start thinking about the...

> [Here](https://www.cs.cornell.edu/courses/cs4120/2011fa/lectures/lec39-fa11.pdf) is how you run generators on the same stack, a technique that is easily adapted to resumable exceptions (assuming you and I mean the same thing by that...

> You listed a bunch of requirements that always hold for the standard foreach/yield syntactic constructs of many languages with generators. Same holds true for the standard syntactic constructs for...

Finally can be expressed with catch and (re)throw, so isn't required as a primitive. Compilers tend to compile it away (e.g. LLVM, AFAIK).

@winksaville, not quite, Wasm is an abstraction defining a virtual ISA. Automatic interoperability between different compilers is explicitly a non-goal (they would need to collaboratively define a common ABI).

@bartwe, in Wasm, `return` is just a special case of a `br`, where the same applies. The same semantics exists in e.g. C++, yet, compilers like clang on LLVM compile...

Interesting points. To me, the former rather seems to suggest introducing a `iNN.nez` instruction instead of duplicating every bool-consuming instruction. In fact, I remember earlier complaints that Wasm does not...

It wouldn't be polymorphic (= work for all types of a certain shape), though, it would be overloaded (= work for a special finite set of types). Wasm has intentionally...

> Are you seriously arguing that we shouldn't extend support for 64-bit pointers because one day we may have 128-bit pointers?? I am arguing that an instruction set preferably is...