Jakub Konka

This is by design. You are cross-compiling to a foreign arch from your host therefore it is up to you to point the compiler and linker at the sysroot, and...

> Thanks for the reply @kubkon! Unfortunately this did not work: > > ``` > $ zig build-exe --sysroot $(xcrun --show-sdk-path) -target x86_64-macos -lobjc hello-world.zig > MachO Flush... warning(link): library...

Overflow error is due to an error in consolidating instructions between parent, then, and else branches in `airCondBr` lowering. This usually can be one of two things: either a bug...

Rebasing on latest master makes the original panic go away which is quite worrying and I'll investigate locally why that is. In the meantime, in order not to block your...

Re the original error (overflow): it was caused because we tried building `libcompiler_rt.a` with `x86_64` backend which not advanced enough and an overflow in `airCondBr` is usually caused by an...

> Thank you so much @kubkon . I'll try to apply those changes as soon as possible. I'll also take a look if rebasing fixes #13231 and close it if...

This is awesome, thanks Andrew! I actually didn't experience your error which looks like we are feeding the linker archives in the wrong format - archives built for ELF have...

Some benchmark results: * `redis-server` ``` ❯ hyperfine ./zld ./lld ./ld64 --warmup 60 Benchmark 1: ./zld Time (mean ± σ): 35.6 ms ± 0.4 ms [User: 35.9 ms, System: 10.4...

Re this error > tried testing this on the self-hosted compiler to find out if it's faster, but I got this: ``` MachO Flush... error(link): incomplete symbol table: expected symbol...

I'm surprised by the linker errors. Are we perhaps missing custom re-exports with renames of globals in compiler-rt for some 32bit targets?