stack-switching

stack-switching copied to clipboard

It can be difficult to discuss different aspects of stack switching without some form of common terms for different aspects. This note is an attempt to support the design process...

fgmccabe

Stacks Design

1

As requested, here is a write-up of the design I mentioned in my presentation (WebAssembly/meetings#934). # Overview The idea of the design is to give wasm programs access to first-class...

RossTate

Motivation

13

At the moment, the overview gives these goals: > Motivation > * Support for Asynch/await programming pattern. > * Support for green threads. > * Support for yield-style generators. Of...

aykevl

Thank you for your contribution to this proposal. May I ask if there is a approximate time plan for this proposal? thx.

wangfakang

Consistent parameter order between cont.bind and switch tags

23

The explainer currently proposes to change the order in which `cont.bind` binds arguments. However, it needs to remain possible to use `cont.bind` to supply the (regular) arguments to a continuation...

rossberg

This came up as a point of ambiguity in a recent discussion, with the other possible design being trapping when the search finds a handler with the correct tag but...

tlively

[explainer] Remove unnecessary constraint on switch

4

Although it is necessary for the target continuation's return type to match the return type expected at the handler and communicated via the tag, it is not necessary for the...

tlively

WasmFX reference interpreter: Errorenous treatment of non-final types

2

There seems to be a validation problem when WasmFX instructions non-final types. Consider the following program: ```wat (module (type $ft1 (func (param i32))) (type $ct1 (sub (cont $ft1))) (type $ft0...

frank-emrich

WasmFX direct switching w/ dynamic lookup

34

The continuations explainer includes a section on direct (i.e. symmetric) switching: https://github.com/WebAssembly/stack-switching/blob/main/proposals/continuations/Explainer.md#direct-switching The second design in that section, reproduced below, gives typing rules for direct switching with dynamic handler lookup:...

tlively

Asymmetric coroutines without search

5

I'm definitely not proposing any changes to either proposed design, but I wanted to share this idea for discussion. I haven't talked with anyone about this yet. Our discussion in...

tlively