Re: Build of gcc12 fails

On 25 Sep 2022, at 2:19 am, Lucier, Bradley J <[email protected]> wrote:
> 
> A linker snapshot was created at:
> 	/tmp/libstdc++.6.dylib-2022-09-24-173114.ld-snapshot
> ld: Assertion failed: (_file->_atomsArrayCount == computedAtomCount && "more atoms allocated than expected"), function parse, file macho_relocatable_file.cpp, line 2061.
> collect2: error: ld returned 1 exit status
> 
Confirmed for 12.2 and 11.3 as well.
Apparently this is a bug in Xcode 14, see 
https://stackoverflow.com/questions/73714336/xcode-update-to-version-2395-ld-compile-problem-occurs-computedatomcount-m

and the linked discussions. At the moment the only way around this seems downgrading back
to Command Line Tools 13.4 as described in
https://github.com/iains/gcc-12-branch/issues/6#issuecomment-1255285642

(as noted there, the first 14.1 beta does not yet include a fix; I also installed the full Xcode 14.1 beta 2,
but this comes with the same CLT version as the first one).

> I previously had problems with `fink self update`, it reported
> 
> [Bradleys-Mac-mini:/opt/sw] lucier% fink selfupdate
> 12.6 does not match the expected value of 12.5. Please run `fink selfupdate` to download a newer version of fink at /opt/sw/lib/perl5/Fink/Services.pm line 1380.
> 
> and I commented line 1380 of /opt/sw/lib/perl5/Fink/Services.pm to get things to run at all. (I’m running macOS Monterey version 12.6.)

I replaced the conditional above with
if (($sw_vers == 11.6 && $darwin_osx == 11.5) || ($sw_vers == 12.6 && $darwin_osx == 12.5)) {

would be good to know if such “fuzzy” matches will be common in further macOS releases...

_______________________________________________
Fink-users mailing list
[email protected]
List archive:
https://sourceforge.net/p/fink/mailman/fink-users/
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users