*Gcc packages does print errors for packages during build

[NobbZ]

When using emacsGgg or emacsPgtkGcc in my config, I get errors like the following during compilition of most, if not all, packages:

emacs-magit> Debugger entered--Lisp error: (native-compiler-error-empty-byte "/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-..." "/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-...")
emacs-magit>   signal(native-compiler-error-empty-byte ("/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-..." "/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-..."))
emacs-magit>   comp--native-compile("/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-...")
emacs-magit>   batch-native-compile()
emacs-magit>   command-line-1(("-f" "batch-native-compile" "/nix/store/f54kjchgzmf7dz6yikph9c0f65yagks6-emacs-..."))
emacs-magit>   command-line()
emacs-magit>   normal-top-level()

It seems as if the pathes are indeed different for each package.

Is there a way to make the packages compile correctly?

I'm observing this behaviour for a while now (approx half a year), though now I found the time to actually create an issue about it.

At least on a first glance, there are no additional runtime issues. (I was able to open some random files in emacs without having any errors or warning in the message buffer that haven't been there before)

[collares]

This is arguably not an error, even if it says "error" on the message. The Nix native compilation step asks emacs to native-compile every .el file, and all this message is saying is that an .el file didn't have anything interesting to compile (either it didn't contain any top-level forms or it explicitly set no-byte-compile to t). If you byte-compile such a file you'll also get no .elc file because there's nothing to byte-compile; the native-compiler is just emitting an overzealous warning. This does not stop compilation of a package, and everything that can be compiled is being compiled if you don't see any other types of errors.

Probably we should ask upstream to turn this error into a warning, but I already bother the gccemacs maintainer enough :)

[NobbZ]

Thanks for the explanation!

Yes, this should indeed be fixed upstream then. Though I am not used to the emacs processes to report a bug/issue with GccEmacs or Emacs at all…

[collares]

Reported this upstream at http://debbugs.gnu.org/cgi/bugreport.cgi?bug=47169

[collares]

@NobbZ Thanks for the bug report! The gccemacs maintainer fixed this upstream, so emacsGcc should not emit as many native-compiler-error-empty-byte errors, if any. Eventually this will get merged to emacsPgtkGcc.

[talyz]

The fix should be in all relevant branches by now.