🚀 go-pugleaf

This is a followup post to the one in comp.unix.ultrix, with more
information on what I am doing and what is going wrong.

I've been having a major major problem getting gcc-2.0 to compile on a
DECstation 3100, running Ultrix 4.2a with 60Meg of swap.  I have used
the following steps:

1.  untar the sucker
2a. configure --target=mips-dec-ultrix
2b. configure decstation
3.  make LANGUAGE=c

this is where the problems first crop up.  The compile works fine
until it gets to enquire.c and it blows up.  All the compiles before
this point work fine, they just take a while. Here is the output of the
make LANGUAGE=c:

  ...stuff deleted...
  cp ./enquire.c . > /dev/null 2>&1
  *** Error code 1 (ignored)
  ./gcc  -O1  -g -B./    -DNO_SC -DNO_MEM -DNO_STDDEF -O0 -I. -c enquire.c
  as1: Error: /usr/tmp/cca10919.s, line 1: Not enough core
  Fatal error in: /usr/lib/cmplrs/cc/as1 Segmentation fault - core dumped
  as1: Error: /usr/tmp/cca10919.s, line 1: Not enough core
  Fatal error in: /usr/lib/cmplrs/cc/as1 Segmentation fault - core dumped
  *** Error code 1

  Stop.

There are two sets of Errors from the assembler because it first tries
using as, then tries to use as2.1 which I assume differ in some
respect.

From looking at the output of the errors produced, it looks like the
assembler tries to optimize the code and starts using lots and lots of
memory until it runs out of swap and dies.  If I remove the -g, it
dies on line 766, if the -g is still there, it makes it down to around
line 1500+ before the error message.  

I then tried changing the -c to -S to produce assembler code output,
and then run the assembler manually to compile it.  Here is what I did:

  ./gcc -O1 -g -B./ -DNO_SC -DNO_MEM -DNO_STDDEF -O0 -I. -S enquire.c

  as2.1 -v -nocpp -O0 -o enquire enquire.s

  /usr/lib/cmplrs/cc2.1/as0 -v -G 8 -EL -g0 -O0 enquire.s -o
  /tmp/ctmaa10962 -t /tmp/ctmsta10962

  as0: overflow setmode farewell croak xmalloc maximum_int main
  eek_a_bug describe i_define u_define f_define floor_log ceil_log
  exponent f_rep bitpattern endian missing fmissing false type_of
  ftype_of typerr ftyperr promotions check_defines cprop basic sprop
  usprop fStore fSum fDiff fMul fDiv fSelf fCheck fprop efprop iprop
  uiprop dStore dSum dDiff dMul dDiv dSelf dCheck dprop edprop lprop
  ulprop ldStore ldSum ldDiff ldMul ldDiv ldSelf ldCheck ldprop eldprop 
  5.4u 1.0s 0:10 59% 86+439k 2+59io 19pf+0w

  /usr/lib/cmplrs/cc2.1/as1 -v -G 8 -p0 -EL -g0 -O0 /tmp/ctmaa10962 -o
  enquire -t /tmp/ctmsta10962
  as1: as1: Error: enquire.s, line 1: Not enough core
  19.1u 49.0s 7:57 14% 20+5081k 2+4148io 10270pf+1w
  Fatal error in: /usr/lib/cmplrs/cc2.1/as1 Segmentation fault - core dumped

From looking at this all, I think the problem might be either in the
way gcc creates the code it feeds the assembler, or in the assembler I
am using here.  Has anyone else run into this type of problem? 

I finally cheated big time by compiling enquire.c with cc.  This
worked and I was able to then "make stage1" and then do the "make
CC=stage1/gcc CFLAGS="-g -O -Bstage1/" where it promptly died.  I
assume this was because I cheated in compiling enquire.c

I would be really really really grateful for any help on this problem,
since it is driving me nuts!  I don't think our system is any
different from the standard DEC distribution and setup for the 3100
machines.  Is it Ultrix 4.2a?  Or am I just doing something really
stupid and dumb?

Thanks...

				John





--
Youth of today!  Join me in a mass rally for traditional mental attitudes!
-------------------------------------------------------------------------------
john@wpi.wpi.edu | BELA LUGOSI is my co-pilot..