🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Thread View: gnu.g++.lib.bug
3 messages
3 total messages Started by daniel Fri, 15 Dec 1989 01:39
libg++ 1.36.1 pros w/ g++ 1.36.1
#536
Author: daniel
Date: Fri, 15 Dec 1989 01:39
100 lines
4313 bytes

machine: sun 3/50
version: g++ 1.36/libg++ 1.36.1

problem: libg++ doesn't compile.  Problem with builtin.cc (and possibly
others)

Makefile output is:
(cd src; make GXX="g++"  GXXFLAGS=" -I/usr/local/src/languages/libg++/g++-includ
e -g -O -fstrength-reduce  -felide-constructors -fschedule-insns -fdelayed-branc
h -fsave-memoized  -Wall " GXXLDFLAGS="-L/usr/local/src/languages/libg++/src -lg
++ -lm " LIBDIR="/usr/local/lib/gcc" SRCIDIR="/usr/local/src/languages/libg++/g+
+-include" CC="gcc" CFLAGS=" -I/usr/local/lib/gcc-include -I/usr/include -I/usr/
local/src/languages/libg++/g++-include -g -O -fstrength-reduce -fdelayed-branch 
 -Wall " RANLIB="ranlib" LDXX="/usr/local/lib/gcc/gcc-ld" GXXCRT1="/usr/local/li
b/gcc/crt1+.o" MAKE="make" prefix="/usr/local" )
g++ -I/usr/local/src/languages/libg++/g++-include -g -O -fstrength-reduce  -feli
de-constructors -fschedule-insns -fdelayed-branch -fsave-memoized  -Wall  -c  Al
locQueue.cc
g++ -I/usr/local/src/languages/libg++/g++-include -g -O -fstrength-reduce  -feli
de-constructors -fschedule-insns -fdelayed-branch -fsave-memoized  -Wall  -c  bu
iltin.cc
builtin.cc: In function char *dtoa (double, char (=  103 ), int (=  0 ), int (= 
 6 )):
builtin.cc:499: warning: unused variable `carry'
builtin.cc:386: warning: unused variable `fworkend'
/usr/local/src/languages/libg++/g++-include/time.h: At top level:
In file included from /usr/local/src/languages/libg++/g++-include/osfcn.h:4, fro
m builtin.cc:763:
/usr/local/src/languages/libg++/g++-include/time.h:60: warning: type declaration
 of timezone shadows previous declaration
/usr/local/src/languages/libg++/g++-include/time.h:60: warning: `timezone' was d
eclared `extern' and later `static'
builtin.cc:848: warning: structure has no data members
as: error (/usr/tmp/cca24306.s:2409): Invalid operand
as: error (/usr/tmp/cca24306.s:2436): Invalid operand
*** Error code 1

Stop.
*** Error code 1

Stop.

Lines 2400-2450 of the assembler output at this stage is:
  2400          jbsr __free__7ObstackPv
  2401          .stabd 68,0,93
  2402  LBE52:
  2403  LBE50:
  2404          .stabs "builtin.cc",132,0,0,Ltext
  2405          .stabd 68,0,850
  2406          unlk a6
  2407          rts
  2408  .stabs "_GLOBAL_$D$builtin_cc:F15",36,0,0,__GLOBAL_$D$builtin_cc
  2409  .stabs "___DTOR_LIST__",22,0,0,__GLOBAL_$D$builtin_cc
  2410  .stabs "_init",10,0,0,0
  2411  .stabs "__GLOBAL_$I$builtin_cc",4,0,0,0
  2412          .even
  2413  .globl __GLOBAL_$I$builtin_cc
  2414  __GLOBAL_$I$builtin_cc:
  2415          .stabd 68,0,850
  2416          link a6,#0
  2417  LBB53:
  2418          .stabd 68,0,155
  2419          pea 4:w
  2420          pea 4092:w
  2421          pea __libgxx_io_ob
  2422          jbsr ___7Obstackii
  2423          .stabd 68,0,164
  2424          pea 20:w
  2425          pea __libgxx_fmtq
  2426          jbsr ___10AllocQueuei
  2427          .stabd 68,0,848
  2428  LBB54:
  2429          .stabd 68,0,846
  2430  LBE54:
  2431  LBE53:
  2432          .stabd 68,0,848
  2433          unlk a6
  2434          rts
  2435  .stabs "_GLOBAL_$I$builtin_cc:F15",36,0,0,__GLOBAL_$I$builtin_cc
  2436  .stabs "___CTOR_LIST__",22,0,0,__GLOBAL_$I$builtin_cc
  2437  .stabs "New_Time:S80",40,0,0,_New_Time
  2438  .lcomm _New_Time,72
  2439  .stabs "Old_Time:S80",40,0,0,_Old_Time
  2440  .lcomm _Old_Time,72
  2441  .stabs "chr_buf:S91=ar1;0;1;2",40,0,0,_chr_buf
  2442  .lcomm _chr_buf,2
  2443  .stabs "_libgxx_maxfmt:c=i20",0x80,0,0,0

The error is stated to be on two similar .stabs statements!  I
generated gcc claiming that I was a sun3 (or sun3os3) system, which I
am.

Any hints would be greatly appreciated.  I have failed to get g++ to
work on a Convex.  Failing on a sun3 is very dissapointing!

-- Daniel Pommert.
 pommert@uiuc.edu
-- 
Re: libg++ 1.36.1 pros w/ g++ 1.36.1
#537
Author: Ron Guilmette
Date: Fri, 15 Dec 1989 04:28
21 lines
1082 bytes
In article <1023100001@ux1.cso.uiuc.edu> daniel@ux1.cso.uiuc.edu writes:
>
>
>machine: sun 3/50
>version: g++ 1.36/libg++ 1.36.1
>
>problem: libg++ doesn't compile.  Problem with builtin.cc (and possibly
>others)
>
<... makefile output deleted...>

>as: error (/usr/tmp/cca24306.s:2409): Invalid operand
>as: error (/usr/tmp/cca24306.s:2436): Invalid operand

This question comes up about once a week.

If you get "Invalid operand" assembler errors from G++, it means that you
are not using GAS (as you should be).  I believe that the README file
or the INSTALL file says that you must use GAS for a SUN system.

// rfg
Re: libg++ 1.36.1 pros w/ g++ 1.36.1
#538
Author: daniel
Date: Fri, 15 Dec 1989 17:35
37 lines
1865 bytes
In reply by rfg@ics.uci.edu, it was stated that gas should be used:
>In article <1023100001@ux1.cso.uiuc.edu> daniel@ux1.cso.uiuc.edu writes:
>>machine: sun 3/50
>>version: g++ 1.36/libg++ 1.36.1
>>problem: libg++ doesn't compile.  Problem with builtin.cc (and possibly
>>others)
>>as: error (/usr/tmp/cca24306.s:2409): Invalid operand
>>as: error (/usr/tmp/cca24306.s:2436): Invalid operand
>This question comes up about once a week.
>If you get "Invalid operand" assembler errors from G++, it means that you
>are not using GAS (as you should be).  I believe that the README file
>or the INSTALL file says that you must use GAS for a SUN system.
>
>// rfg

However in gnu.g++.bugs,
tale                 GAS, GLD, GCC and G++       12:03 am  Dec 15, 1989
(at TURING.CS.RPI.EDU)                               (From News system)

Asks the OTHER weekly question:

How do I get gcc and g++ to use gas?  I have it installed in
/usr/local/bin/gas.

Patrick.Sanchez@irisa.fr mailed me this hint:
By the way, to use gas you have to put it in the directory where you
put gnu-ld gnu-cpp ... should be ..../lib/

He comments, however:
I managed to compile libg++ using gas (gnu-assembler)...
unfortunately I had to reverse to normal as for some files.

Where is the ``official'' answer documented?  What is it?

-- Daniel Pommert.
 pommert@uiuc.edu
Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads