Karoy Lorentey
Karoy Lorentey
> Yeah, I did see that. I was wondering, what's the point of MinimalTypes? Those are reusable types with deliberately over-pedantic (and/or unusually configurable) conformances that can be used to...
> > Do you think we need to explicitly expose that as a core operation? (Isn't it generally equivalent to the proposed set[members: k...]?) > > I kinda think you...
This is WIP -- we will either add these files as part of the 1.1.0 release preparation work, or we will simply link to DocC-generated documentation.
Rescheduling to 1.2.
Hm, I think so -- it looks like this should use `Swift.max`, not `Swift.min`. The capacity in bytes also looks a little low to me -- we'll probably want to...
~The upcoming 1.1 release is scheduled to bump the minimum toolchain version to Swift 5.6, allowing double wide atomics to be unconditionally enabled. This will obsolete the ENABLE_DOUBLEWIDE_ATOMICS condition, resolving...
Correction: Apologies, I commented on the wrong bug -- unfortunately fixing this will likely need us to remove these two functions, which requires us to first fix https://github.com/apple/swift/issues/56105.
@ktoso would fixing this break these dependents now?
Getting rid of the C module with something along the lines of https://github.com/apple/swift-atomics/pull/74 would likely get rid of these issues. However, this will require some Swift compiler & stdlib work,...
CI is going to fail, but just for fun: @swift-ci test