Saleem Abdulrasool
Saleem Abdulrasool
I don't think that's a good idea. This is going to increase memory pressure on windows, which already has a slower build rate due to the fork exec heavy nature...
> Can we limit this to Windows only then? I'm okay with scoping this to Windows. I don't remember if POSIX guarantees that the mmap I/O backing store is modifiable...
> I would naively expect the latter, but it's quite possible we're holding the files open while the ASTContext is alive (and maybe we do actually need to). That shouldn't...
I think that this is ready to merge, going to merge to improve the windows state.
> @compnerd Is there a chance you could sneak that one line into a TBC Swift toolchain build, to see if that gets rid of the failing assertion? I would...
The C++ module maps are not compiler resources, but are part of the Swift SDK ("Platform Swift Modules"). They are auxiliary content that is overlaid on the Platform SDK. We...
@egorzhdan right, there's no ABI stability currently for that module. But the final place once that is achieved would be the SDK. The module maps however do not have any...
@rprichard if this is interesting, I can look at trying to setup some CI with GHA to get coverage for x86, x64, and ARM64.
This doesn’t se like an error of the generated executable. Paths are case insensitive on windows, and it seems that the drive is getting normalized to lower case. This is...
@swift-ci please smoke test macOS