Rust9x 1.84.0-beta 🎉 - "Tier 4" target for Windows 9x/NT
https://seri.tools/blog/rust9x-1-84/
[#]rust #rustlang
=> More informations about this toot | More toots from seri@hachyderm.io
omg i forgot to add the logo :c
=> More informations about this toot | More toots from seri@hachyderm.io
@seri Do you know the answer offhand to this? https://github.com/cr1901/ballast/blob/main/src/retrieval.rs#L75-L77
=> More informations about this toot | More toots from cr1901@mastodon.social
@cr1901 there are two distinct things here - does win95 have futexes? most definitely not, WaitOnAddress is like windows 8. however mpsc uses the thread parker from what you have linked, which you can implement differently
=> More informations about this toot | More toots from seri@hachyderm.io
@cr1901 rust9x has 3 different implementations: waitonaddress/futex for 8 and up, keyed events for xp and up, and a generic one for everything else, so mpsc should work on win95 in rust9x.
=> More informations about this toot | More toots from seri@hachyderm.io
@seri Yea, I'm not tied to futuxes, my question in the comment is more "I am wrapping an async 'receive with cancel' function in a executor that blocks. If I am to fake my receive with cancellation at all in Windows 95, even with a shitty "wake up every 100ms*" loop, I need mpsc. Can this be done?"
=> More informations about this toot | More toots from cr1901@mastodon.social
@cr1901 all basic functionality of the stdlib should work, that's the point of rust9x, so if you can do it with the stdlib alone, it should definitely work
=> More informations about this toot | More toots from seri@hachyderm.io
@seri I will also have to reread how smol::block_on and pollster are implemented. mpsc might indeed be all I need, but didn't commit the impl to memory.
=> More informations about this toot | More toots from cr1901@mastodon.social This content has been proxied by September (ba2dc).Proxy Information
text/gemini