Matthew Dillon dillon at apollo.backplane.com
Sun Apr 10 12:13:52 PDT 2005

:>    I don't think we want to rip up our dfports stuff until we actually
:>    have something to replace it with.
:That wasn't what I suggested.
:>    I'm not sure I understand what you mean about gcc 3.x not having any
:>    freebsd compatibility.  Binaries compiled with gcc 3.x work just fine
:>    on a FreeBSD-4.x box.  Maybe you are thinking of the FreeBSD-5 signal
:>    changes that broke FreeBSD 5's backwards compatibility with FreeBSD-4
:>    (and now also new system calls that are made unconditionally in FreeBSD-5
:>    which also break compatibility with FreeBSD-4).
:Nope.  Not what I meant as well.
:Current dfports compiles as freebsd-4.8, GCC 2.95.x is configured for the
:FreeBSD fall back.  GCC 3.x is not, we need to move forward into using our
:own uname for compiling ports if you want to rip out gcc 2.95.x like you
:said in your plans on the 1.2 release page.
:Jeroen Ruigrok van der Werven <asmodai(at)wxs.nl> / asmodai / kita no mono

    Ach, right.  You know I totally forgot we had those uname hacks in

    We still don't want to rip it out quite yet.  The GCC-3-as-default is
    not something we can do in a week, there's a lot of stuff that is going
    to go into the HEAD branch all at once and it will probably be at least
    two months before we slip the PREVIEW tag again.  So we have time to
    think about how to deal with the ports issue.
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>

More information about the Kernel mailing list