search

esp-rs / esp-hal

no_std Hardware Abstraction Layers for ESP32 microcontrollers
https://docs.esp-rs.org/esp-hal/
Apache License 2.0
737 stars 207 forks source link

Decide on MSRV #788

Closed bjoernQ closed 1 year ago

bjoernQ commented 1 year ago

See https://github.com/esp-rs/esp-hal/pull/787

I like the CURRENT_VERSION - 1 idea and with https://github.com/esp-rs/esp-hal/pull/779/commits/7b45dca3c088183c9e6da86c362c4fb068ff8954 that should be easy to implement

jessebraham commented 1 year ago

Is there any reason not to just use 1.67.0, which is currently required for the async example, and remove the distinction there altogether? I don't think there are really any new language features we want to take advantage of (correct me if I'm wrong) so giving some headroom would be nice.

bjoernQ commented 1 year ago

Also fine for me. Don't have a strong opinion on this other than having to pin a (transitive) dependency because of the MSRV is not a good thing

MabezDev commented 1 year ago

I don't think we should blindly upgrade to CURRENT_VERSION - 1 whenever its available, but if there is a feature we want we can just go ahead and update provided its at least one version behind from current stable. We could also of course scrap that idea and just address it on a case-by-case basis.

bjoernQ commented 1 year ago

779 additionally requires toml_edit to be pinned in esp32c6-lp-procmacros

If we need to bump MSRV we should make sure it's at least CURRENT_VERSION - 1 or older. But we won't bump the MSRV just because we can

bjoernQ commented 1 year ago

The only thing left to do to close this issue is bump to MSRV 1.67 and unpin the toml_edit dependency