gc

gc copied to clipboard

Using this validation text from MVP.md: ``` array.new_canon_data $t $d : [i32 i32] -> [(ref $t)] iff expand($t) = array (mut t') and t' is numeric or packed numeric and...

eqrion

Pushing i31ref to post-MVP

20

Are any production languages going to require i31ref in the MVP? If not, I think we should consider pushing it to a separate proposal after the MVP. SpiderMonkey doesn't have...

eqrion

3-pronged subtyping hierarchy

7

This makes the necessary changes to the MVP doc for introducing the 3-pronged subtype hierarchy. For now, I haven’t done any renaming, so I went with types `any` and `data`...

rossberg

Using arrays for spatial locality

2

Forgive me if this is already addressed in the proposal, but I was wondering whether there's wider interest in supporting array allocation specifically as a means of achieving spatial locality...

ouillie

MVP `array.copy`

6

@manoskouk proposed `array.copy` in https://github.com/WebAssembly/gc/pull/278#discussion_r810944560, but it was not included in that PR because it deserved separate discussion and a separate PR. Would anyone object to adding `array.copy` to the...

tlively

1a implementation for toolchains: dummy initializers

10

I sort of lost track of the discussion on non-nullable locals. Really nice to see it has been settled and there is progress on 1a. After reading discussion on https://github.com/WebAssembly/function-references/issues/44...

carlopi

This could be a post-MVP feature, just posting for discussion. Currently, function types have declared subtyping like structs and arrays, and are limited to a single supertype. There isn't any...

titzer

.NET Notes

14

I was asked to add some notes based on the needs of Microsoft .NET implementations wrt GC in WebAssembly. 1. Object layout .NET objects in memory usually consist of a...

vargaz

Remove redundant typeidx on gc instructions

36

This removes the remaining redundant type indices on GC instructions, following the discussion in WebAssembly/function-references#27 and #238. It makes GC instructions consistent with what we already decided on for other...

rossberg

Field Accessors

29

This is a suggestion for a different way to design the instructions for getting/setting fields, which for lack of a better term I'll refer to as "field accessors". ## Design...

RossTate