Re: FreeBSD 5.0 XFree86 4_2_1 source anyone?

nil wrote:

> Thanks for the datapoint.
> I did not use HEAD for either cvsup.
> I tried 5.0 HEAD for 2 days, each day I got different compile errors,
> leading me to conclude  it wasn't stable enough to do an initial make
> world.
> So I used RELENG_5_0, but the date was 12/30, close enough.
> Like you, I also  put the xf86 cvs source tree elsewhere.
> For xf86, I decided to use 4_2_1 because of what I saw on 5.0,
> but from what you're describing, HEAD is stable, so maybe I'll try that
> then...

Hi,

I just check the docs for 4.2.1 and it looks like i810e is supported.
In that case, I'd try install the XFree86-4 port first. I use the
head of cvs because my card is not supported in 4.2.1. But you
definitely should make your port tree up to date. Though the main
x11 port (x11/XFree86-4) has version 4.2.0, the server port
x11-servers/XFree86-4-Server does have version 4.2.1.

If you still have problems building or installing the ports, then
you should try [email protected] to see if it's FreeBSD issue.

Cheers,

yushun.

>
> Anyone has other suggestions?
> (I did find out that the ports does have 4.2.0 available, so that's
> another option.)
> I have i810e display, so I assume whichever version would work?
>
> Regards and thanks,
> Gilbert
>
> Yu-Shun Wang wrote:
>
> > Hi,
> >
> > I built and installed xfree86 from the head of cvs a while
> > ago under 5.0 because the one from port could not recognize
> > my ATI Radeon 9000. I did not have the problems you described
> > below, but keep in mind that my -CURRENT might be different
> > from your -CURRENT. :-)
> >
> > Here's a snapshot from my world.lod:
> >
> > ===========================================================================
> >
> > Building Release 6.6 of the X Window System.
> >
> > I hope you checked the configuration parameters in ./config/cf
> > to see if you need to pass BOOTSTRAPCFLAGS.
> >
> > Mon Dec 23 17:02:25 PST 2002
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > cd ./config/imake && make   -f Makefile.ini BOOTSTRAPCFLAGS="" CC="cc"
> > clean
> >  ...
> >
> >  ...
> > make -f xmakefile   VerifyOS
> >
> > Building on FreeBSD 5.0-CURRENT i386 [ELF] (5.0.0).
> >
> > GCC version: 3.2
> > ^^^^^^^^^^^^^^^^^
> >
> >
> >
> > ===========================================================================
> >
> >
> > Install went fine, too. Notice the date which is most likely the date
> > of the head of cvs for both FreeBSD CURRENT and XFree86.
> >
> > I did have to install the XFree86 CVS build into a separate directory
> > (/usr/local/X11R6) while keeping the one from ports (/usr/X11R6) to
> > satisfy tons of port dependencies (such as gnome, etc.).
> >
> > Just a data point for your reference because the version of gcc is
> > not the same as yours (3.2.1).
> >
> > Cheers,
> >
> > yushun.
> >
> >
> > [email protected] wrote:
> >
> >> I've cvsup RELENG_5_0 yesterday, make kernel, make world.
> >> Then I've cvsup xf-4_2_1 into another source directory, make World, make
> >> install.
> >> make World gave warnings of no big big integers, but completed.
> >> make install failed because the FreeBSD.cf supplied apparently called
> >> for
> >> ansic.inc version that's different.
> >> [all names above I recall from memory, maybe slightly off..]
> >>
> >> Question: anyone has include file or source or patch or instruction to
> >> make above combination work?
> >> I'm guessing it's because 5.0 uses gcc 3.2.1 and not 2.9 anymore, and
> >> include files changed...
> >>
> >> If above combination is not good, any suggestions?  I'm starting off
> >> with
> >> a new hard drive, though I'm sort of committed now using FreeBSD
> >> 5.0.  (I
> >> also have it dual boot to Windows and Linux.)  As came with the cvsup,
> >> the ports supplied is Xfree86 3.3.6...
> >>
> >> Regards,
> >> Gilbert
> >>
> >
> >
>
>
> _______________________________________________
> Xpert mailing list
> [email protected]
> http://XFree86.Org/mailman/listinfo/xpert



-- 
____________________________________________________________________________
Yu-Shun Wang <[email protected]>               Information Sciences Institute
                                            University of Southern California