View previous topic :: View next topic |
Author |
Message |
Dominique_71 Veteran
Joined: 17 Aug 2005 Posts: 1886 Location: Switzerland (Romandie)
|
Posted: Fri Apr 12, 2024 9:46 am Post subject: [solved]ABI: libgtk2_interface.so(64) func(-2) [BREAKING] |
|
|
When merging alsaplayer with its last git code, I get the following message:
Code: | >>> Installing (1 of 1) media-sound/alsaplayer-9999::test
* CMP: =media-sound/alsaplayer-9999 with media-sound/alsaplayer-9999/image
* ABI: libgtk2_interface.so(64) func(-2) [BREAKING]
* SIZE: 14.36MiB -> 1.40MiB, 51 -> 51 files
* ------> ABI(-2,>B<) SIZE(-90.21%) |
What does that mean? And what can I do about that?
Edit: The precedent build was a debug build for strace, and this message showed up when doing a standard build. If I do it again, it have disappeared:
Code: | >>> Merging media-sound/alsaplayer-9999 to /
* CMP: =media-sound/alsaplayer-9999 with media-sound/alsaplayer-9999/image
* ------> ABI(nodebug) |
_________________ "Confirm You are a robot." - the singularity |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2837
|
Posted: Fri Apr 12, 2024 9:28 pm Post subject: |
|
|
Debug symbols are needed to get an accurate abidiff report, removing debug symbols essentially just hides (most) abi changes thus why it says: "(nodebug)" instead
That aside, whether it's an issue or not and if need to do something about it varies. First it needs to be a library that other packages are using (aka not a private one), and next it'd need to be symbols that reverse dependencies are using (sometimes it's just private internals that were visible that got removed, not that it couldn't be breaking something still, it's a case-by-case thing and it's just there to warn about the possibility).
The message comes from iwdevtools anyhow (only when dev-util/libabigail is also installed), see the cdparanoia example in:
https://github.com/ionenwks/iwdevtools?tab=readme-ov-file#qa-cmp
A more recent event where it could've prevented temporary breakage if maintainer had used abidiff and verified, was with snappy-1.2.0: https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=4c7214ce7f45 (bug #929199) -- not the first time snappy's upstream pulled this without a soversion bump too. |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|