Alyssa Ross
Alyssa Ross
> Thank you for looking at this change - i'm not really sure this is particularly helpful. Labelling this method as unsafe doesn't help since there is no "safe" way...
> Okay - I agree that there are probably bugs in the code due to mistakes with assumptions but i'm not 100% sure that labelling a function that just populates...
No — they still need to be fixed so they're not marked as safe when they aren't.
I had to build with `-Wno-error=deprecated-declarations` to get past `QCheckBox::stateChanged(int)` being deprecated now, but with that as well, it worked with that branch. Thanks!
I have the same issue, but without anything in the kernel logs suggesting a problem with my TPM. The only lines containing "tpm" in `journalctl -k` are: ``` Aŭg 21...
I must be misunderstanding something here, I think. How does `__FILE__` end up being expanded in outputs? Are these preprocessed headers?
Thanks! Checking the x86_64-linux tests locally because they often time out on OfBorg. Looks like we also need 8.2.7 on 24.05. Are you interested in doing that?
The fix doesn't seem right — mesh/mesh-config-json.cc includes both mesh/missing.h and libgen.h, so it gets two conflicting definitions for basename, which is causes a compiler error: ``` > In file...
Assuming you meant to ask me. Yes it does. It's musl 1.2.5.
[config.log](https://github.com/user-attachments/files/17016832/config.log)