So, I haven't updated my laptop in a month or so and I go to give it a full update, and find that for some reason I cannot emerge gtk+ 2.12.5 or higher. I have an older version installed right now which I have not tried to recompile, but it was fine a while back. I've done a revdep-rebuild with no broken links or anything, and am kind of at a loss as to what I can do from here. I've tried the newer package that is currently keyworded also, it fails during the compile of the exact same file.
I have the tail of a build log here.
Someone in the IRC channel mentioned that it could be due to my -march CFLAG being set to k8 in a 32-bit environment, but I do not understand how this could be a problem as it is clearly compiling 32-bit code, and every source I can find indicates that this should not be an issue. Does anyone else know about this?
Edit: I should note that I also tried compiling it with 'march=i686' with no success, same issue. I haven't rebuild my whole system with that march yet though.
Issues compiling gtk+
Message
Ok, fixed my issue with 'emerge -e gtk+', a few packages were updated and like 250 were rebuilt, so it's hard to tell what was causing it but it was either a broken package or the ebuild isn't properly updated for a dependency, it might need a higher version of something than the .ebuild asked for, or I could've just had a broken package from something.
Either way, the op in #gentoo who told me that the 'amd64' march would compile 32-bit incompatible code even with an i686 CHOST was completely wrong, so I guess he banned me for no reason for not believing him.
Either way, the op in #gentoo who told me that the 'amd64' march would compile 32-bit incompatible code even with an i686 CHOST was completely wrong, so I guess he banned me for no reason for not believing him.
march k8 is not incompatible with an i686 chost, it is the chost which defines the word length. March tells your gcc the instruccion range that is available, but doesn't imply 32 or 64 bits code at all.Vash63 wrote: Either way, the op in #gentoo who told me that the 'amd64' march would compile 32-bit incompatible code even with an i686 CHOST was completely wrong, so I guess he banned me for no reason for not believing him.