[gnu.gcc] gcc compilation problems on sun4

roberto@cwi.nl (Rob ten Kroode) (03/23/89)

Hello,

(sun 4/110, SunOs 4.0)
I have some problems compiling gcc 1.34. I compiled the compiler without any 
problems, but when I try to compile the compiler with itself I get the following
messages:

stage1/gcc -g -O -Bstage1/  -sun4 -c  version.c
stage1/gcc -g -O -Bstage1/  -sun4 -c  obstack.c
stage1/gcc -g -O -Bstage1/  -o gccnew gcc.o version.o obstack.o
_edata: ld: user attempt to redefine loader-defined symbol
_end: user attemp to redefine loader-defined symbol
_text: /lib/libc.a(des_crypt.o): multiply defined
*** Error code 1
make: Fatal error: Command failed for target `gcc'

I have had no such problems on our sun3's running SunOs 3.5.

Any hints, sollutions?

-- 
                                  |
Rob ten Kroode (roberto@cwi.nl)   |         Don't read this !! 
                                  |
"When they said 'sit down' I stood up",  Growing up - Bruce Springsteen

schmidt@zola.ics.uci.edu (Doug Schmidt) (03/24/89)

Please post GNU C bug reports to bug-gcc@prep.ai.mit.edu.  The
info-gcc@prep.ai.mit.edu mailing list is intended for disseminating
periodic information about new compiler releases, etc. (and occasionally
it's used to vent moral indignation over petty attacks from ``software
hoarders'' ;-)).

thanks,

   Doug
--
schmidt@ics.uci.edu | On a clear day, under blue skies, one need not seek
office:             | And asking about Buddha 
(714) 856-4043      | Is like proclaiming innocence,
                    | With loot in your pocket.

peter@apexepa.UUCP (Peter Palij) (03/24/89)

In article <7972@boring.cwi.nl> roberto@cwi.nl () writes:
>Hello,
>
>(sun 4/110, SunOs 4.0)
>I have some problems compiling gcc 1.34. I compiled the compiler without any 
>problems, but when I try to compile the compiler with itself I get the 
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>following messages:
>
>*** Error code 1
>make: Fatal error: Command failed for target `gcc'
>

I just hit the same problem on a sun 3/50, no floating-point, diskless
client, OS4.01, except that the problem occured on the initial invocation
of stage1/gcc... .

Any advice/help would be appreciated.

Peter
-- 
-----------------
Peter Palij                                               uunet!apexepa!peter
Apex Software Corporation   peter@apexepa.uucp          Phone: (412) 681-4343

warsaw@cme.nbs.gov (Barry A. Warsaw) (03/25/89)

In article <7972@boring.cwi.nl> roberto@cwi.nl (Rob ten Kroode) writes:

> (sun 4/110, SunOs 4.0)
> I have some problems compiling gcc 1.34. I compiled the compiler without any 
> problems, but when I try to compile the compiler with itself I get the following
> messages:
 
> stage1/gcc -g -O -Bstage1/  -sun4 -c  version.c
> stage1/gcc -g -O -Bstage1/  -sun4 -c  obstack.c
> stage1/gcc -g -O -Bstage1/  -o gccnew gcc.o version.o obstack.o
> _edata: ld: user attempt to redefine loader-defined symbol
> _end: user attemp to redefine loader-defined symbol
> _text: /lib/libc.a(des_crypt.o): multiply defined
> *** Error code 1
> make: Fatal error: Command failed for target `gcc'
 
> I have had no such problems on our sun3's running SunOs 3.5.
 
> Any hints, sollutions?

Perhaps you need to upgrade to SunOS 4.0.1, if you haven't already.
I'm just in the process of building gcc for Sun3-OS3.5 and
Sun4-OS4.0.1 and both have succeeded in compiling correctly. At least
they compile themselves (stage1 and stage2) without any errors, and a
simple hello world program.  I'm using gcc 1.34 which I just snagged
last week.

However, I can't get gnu emacs to compile using gcc on the Sun3-OS3.5
(haven't yet tried the Sun4).  The make will spit out many warnings
(some function was declared implicitly 'extern' then later declared
'static') but these are not fatal.

Then I get warnings of garbage at the end of #ifndef's of the
following files when it compiles src/sunfns.c:

/usr/include/pixrect/pixrect_hs.h (line 11)
/usr/include/pixrect/traprop.h    (line 7)

Then the fatal error occurs:

/usr/include/sundev/vuid_event.h:39:unterminated character constant

which is a line inside a commented #ifdef.  The only funky word on
that line is "don't".  Could the "'" be screwing things up.
Admittedly I haven't hack on this much (it just happened as I was
reading the original article :-) but I figured, kill two birds with
one stone, so to speak.  Any ideas?

> Rob ten Kroode (roberto@cwi.nl)   |         Don't read this !! 

-Barry

warsaw@cme.nbs.gov (Barry A. Warsaw) (03/25/89)

Oops, sorry, I forgot my .signature.

NAME: Barry A. Warsaw                USMAIL: National Institute of Standards
TELE: (301) 975-3460                         and Technology (formerly NBS)
UUCP: {...}!uunet!cme-durer!warsaw           Rm. B-124, Bldg. 220
ARPA: warsaw@cme.nbs.gov                     Gaithersburg, MD 20899
  or: warsaw@cme-durer.arpa

wjj@SUN-VALLEY.STANFORD.EDU (03/27/89)

> However, I can't get gnu emacs to compile using gcc on the Sun3-OS3.5
> (haven't yet tried the Sun4).  The make will spit out many warnings
> (some function was declared implicitly 'extern' then later declared
> 'static') but these are not fatal.
>
> Then I get warnings of garbage at the end of #ifndef's of the
>following files when it compiles src/sunfns.c:
>
> /usr/include/pixrect/pixrect_hs.h (line 11)
> /usr/include/pixrect/traprop.h    (line 7)
>
> Then the fatal error occurs:
>
> /usr/include/sundev/vuid_event.h:39:unterminated character constant

Hi,
    I had the very same problem in intalling emacs 18.53 on Sun OS
3.5.  The problem is that that the two header files pixrect_hs.h and
traprop.h don't conform to the ANSI standard and cpp is unhappy.  To
correct this, copy these files to /usr/local/lib/gcc-include/pixrect
and change :
   
    TRAPROP.H  ==>   TRAPROP_H
    PIXRECT_HS.H ==> PIXRECT_HS_H

It is the .H in the #ifndef that in not ANSI standard.  

Also, in ~/dist-18.53/src/config.h, at the top, include

#ifdef __STDC__
#define CAT(a,b) a ## b
#endif

or use the -traditional option for gcc.

Good luck,

Warren J. Jasper
wjj@sun-valley.stanford.edu