The update works, and the module errors are gone.

This fixes the restore problem

System boots and runs OK. I didn't verify the bug fix.

Works on my desktop and works for the GCC std::chrono impl.

Trying again with a new libphonenumber (and heaptrack made it in this time too)

@aleasto yeah lots of packages failed to build, that always happens with a large update like this. heaptrack failed with: Could NOT find Libunwind (missing: LIBUNWIND_HAS_UNW_BACKTRACE) I don't think that's not a problem with the new boost or tbb, so I'm not fixing it. A bug will get created for it if it still fails during the F40 mass rebuild.

@rineau I've built CGAL in the side tag now, since we're delaying the merge anyway.

Yeah, there's definitely a real problem: koji was completely unusable while I was trying to do this and so I couldn't do the rebuilds. I was advised to just merge this back to rawhide as-is and then keep doing the builds there, so that the mass rebuild isn't held up. I'm just trying to do the right thing, based on what I've been advised.

Looks like the tests failed here:

:: [ 21:19:46 ] :: [  BEGIN   ] :: Running 'cd /home/boost.HaBUzVqRpH/BUILD/boost*'
/usr/share/beakerlib/testing.sh: line 896: cd: too many arguments
:: [ 21:19:47 ] :: [   FAIL   ] :: Command 'cd /home/boost.HaBUzVqRpH/BUILD/boost*' (Expected 0, got 1)
:: [ 21:19:47 ] :: [  BEGIN   ] :: Running 'su -c './bootstrap.sh &>/home/boost.HaBUzVqRpH/bootstrap.log' bstbld'
:: [ 21:19:47 ] :: [   FAIL   ] :: Command 'su -c './bootstrap.sh &>/home/boost.HaBUzVqRpH/bootstrap.log' bstbld' (Expected 0, got 127)

The boost code hasn't actually changed since the last update

With this update I can run the GCC 13 PSTL tests on F37