Martin,
Well, from the comments in the PR, Geoff was protecting the
10.4 version of libgcc. I believe they are free to add that
symbol to 10.5's libgcc however. FYI, what I have been doing is
using gcc4 built with the current gcc trunk to build xplor-nih.
In doing so I noticed a c++ regression in the xplor-nih testsuite.
While Geoff was looking into that, I decided to regress out all of
the gcc4.patch and limit it to a varasm.c patch required for
gfortran. This is when I noticed massive c++ regressions when
gcc trunk was built under fink. These problems are due to the
MACOSX_DEPLOYMENT_TARGET setting of 10.4. I have now eliminated
those c++ regressions by not setting MACOSX_DEPLOYMENT_TARGET.
At this point, this still leaves me with the c++ regression
in xplor-nih built with gcc trunk. Geoff suspects this is due
to some change to libgcc which hasn't been ported from the
Apple branch to the trunk.
I'll post the gcc4.info and gcc4.patch I am using tonight
to the list. It would be really nice if someone with MacIntel
could try to determine exactly which patches are still missing
for MacIntel and we can yell at Geoff about that.
Jack
ps One other issue that still needs to be addressed is the odcctools
package. The current gcc4.info in 10.4 unstable uses --disable-multilib.
This is because the odcctools isn't the current cctools with the ld64
fixes. This one is in Xcode 2.3 and also exists in the odcctools cvs.
I have built a local odcctools package with a tarball rolled from the
odcctools cvs and will test that next. Currently I am building against
the cctools of Xcode 2.3.
_______________________________________________
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel
