Yet Another GCC 1.39 and Sys Vr4 problem.

Tom Armistead toma at swsrv1.cirr.com
Fri May 3 17:26:09 AEST 1991


In article <10990 at hub.ucsb.edu> jim at ferkel.ucsb.edu (Jim Lick) writes:
>*I posted this to gnu.gcc.help, but haven't received a response yet.  I'm*
>*reposting to comp.unix.sysv386, as I forgot to do this for the original!*
>
>Hello.  I have UHC SVR4.0 version 2.0.  I have been attempting to build gcc
>1.39 on this system.  From recent discussion, I was under the impression
>that this version would build with no problems on my system.  However, it
>appears to not be the case quite yet.  
>
>Here is what I am doing:
>
>1) Edit the Makefile to change INSTALL and CLIB as noted therein.
>
>2) config.gcc i386-sysv4
>
>3) make
>
   lots of stuff deleted...

I had ALL the same experiences that you speek of using DELL's Sys Vr4.?.

1.   The warning messages don't hurt anything.
1.5. It seems that that C compiler doesn't like '^L' in the source file?
     I just editted it out.
2.   The hard-parms.c error requires you to edit that file and take out the
     define for 'long double' types (just define them a plain double).
     hard-parms.c is rebuilt by the Makefile, so run make and edit the file
     after you get the error and then re-run make. (Seems like there were 2
     places in the file where you had to do this?)
3.   Now I was able to re-compile gcc with itself, but the '-g' option is not
     supported under Release 4 (at least not yet).  So no debugging - your code
     has to be perfect (like mine %^#^(*^&$&).

Tom
-- 
Tom Armistead - Software Services - 2918 Dukeswood Dr. - Garland, Tx  75040
===========================================================================
toma at swsrv1.cirr.com                {egsner,letni,ozdaltx,void}!swsrv1!toma



More information about the Comp.unix.sysv386 mailing list