Compiling TOM/FRODO on an IRIS Indigo

Compiling TOM/FRODO on an IRIS Indigo

Post by Carsten Men » Fri, 03 Sep 1993 17:21:30



Hi,
 I have got a little problem when I try to compile TOM on our IRIS Indigo:

1 33 MHZ IP12 Processor
FPU: MIPS R2010A/R3010 VLSI Floating Point Chip Revision: 4.0
CPU: MIPS R2000A/R3000 Processor Chip Revision: 3.0
On-board serial ports: 2
Data cache size: 32 Kbytes
Instruction cache size: 32 Kbytes
Main memory size: 32 Mbytes
Integral Ethernet: ec0, version 0
Disk drive: unit 1 on SCSI controller 0
Integral SCSI controller 0: Version WD33C93B, revision C
Iris Audio Processor: revision 3
Graphics board: GR2-XS24 with Z-buffer

OS: IRIX 4.0.5F

I get the following error message by uld when I try to link the opject code:

        /usr/bin/f77 -G 2 -w -col120 -DIRIS4D -O3 -kp extern -o tom abinit.o addall.o addmo.o addsc.o addtxt.o anchor.o assign.o  atid.o atomty.o back.o backobj.o bfdevi.o bfitnew.o bitgen.o  blob.o bndang.o bndbrk.o bond.o bonsub.o boxlim.o change.o  chat1.o chat2.o chnfix.o cntrl.o colcircles.o colsub.o  consrf.o contax.o contsym.o date.o defrng.o delete.o  dgloop.o dglsub.o dglx.o disabl.o disang.o disgen.o dougal.o  drawlines.o dummy.o empty.o enable.o enterid.o faslnk.o
fbrt.o fit.o  fitang.o fitarrays.o fitbet.o fitcg.o fitder.o fitdsn2.o  fitesum.o fithyd.o fitnow.o fitout.o fitrdf.o fitrot.o  fitsort.o fittom.o fitupd.o fitzed.o fitzon.o fixtre.o  frodbd.o getatd.o getput.o guide.o hbond.o helix.o help.o huetbl.o  huetbl0.o id.o idgen.o insert.o inter.o isitma.o ldigr.o  lista.o listr.o loada.o loddic.o lodtom.o lsqp.o makeat.o  makedi.o makems.o makepdb.o makespl.o makewh.o mapctr.o  matfit.o memps.o molobj.o move.o noda.o noid.o orig.o
ortho.o pdb.o pjoin.o print.o prnt.o probld.o psplot.o  problz.o prompt.o pseudo.o rama.o readan.o readat.o  readdi.o readmo.o readvg.o readwh.o redatd.o refi.o refine.o  rename.o renorm.o replce.o reset.o rigid.o ringio.o rock.o  rotmat.o rsrbit.o rtmr.o rtmrsub.o sam.o save.o seprat.o set.o  setaty.o setmar.o setmov.o spcgrpentry.o spcgrps.o spjoin.o  squeze.o ssbrid.o star.o stat.o subs.o surf.o swpbak.o symbig.o  symerr.o symgen.o symtry.o syscom.o tajps0.o tom.o tor.o  
torfix.o torgen.o torrot.o trans.o trebil.o update.o  updmol.o utilty.o vector.o verify.o vgout.o view.o water.o weight.o  deffnt.o gtime.o make.o scrsave.o tmpnam.o tomhard.o exec.o tajps1.u graphics.u reddials.u maps.u pen.u   -L -L/usr/lib  -lfgl -lgl_s  
uld:
assertion failed: pmext->aext.asym.value == pmext->usym_size at ext_tbl.c line 530
*** Error code 1

Stop.

Unfortunately I can not find ext_tbl.c which might be a part of a library nor
I am able to find the varibles pmext->aext.asym.value or pmext->usym_size.

Could anybody please give me any hints to solve this problem ?

Thank You very much

Carsten Menke

-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_

Universitaet Bielefeld       Phone [int+49] +521-106-6276
Fakultaet fuer Chemie        Fax   [int+49] +521-106-6146
Anorganische Chemie I
Postfach

D-33501 Bielefeld  (Germany)



-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_

 
 
 

Compiling TOM/FRODO on an IRIS Indigo

Post by Andras Aszo » Fri, 03 Sep 1993 21:35:19


Quote:>uld:
>assertion failed: pmext->aext.asym.value == pmext->usym_size at ext_tbl.c line 530
>*** Error code 1
>Stop.
>Unfortunately I can not find ext_tbl.c which might be a part of a library nor
>I am able to find the varibles pmext->aext.asym.value or pmext->usym_size.

I had the same problem. I thought it was due to the fact that -O2 and -O3 level
optimisations were both used and the final link was made between .o and .u files.
I edited the Makefiles and compiled every module with -O3, giving rise to
.u files only. In this case, the /tmp directory was filled up completely when performing
the global optimisation and linking. I could have set TMPDIR to /usr/tmp,
but gave up instead. I don't like packages which claim they can be compiled
straight away any more: I had too many counterexamples in the past. Thank God, they were
all free at least.

Of course there's no such program as ext_tbl.c. ;-)
Andras
===============================



   | |          Fax: ++44 81 913 8545
  / -+-^        
_/   L_/

 
 
 

Compiling TOM/FRODO on an IRIS Indigo

Post by Rodrigo Lop » Fri, 03 Sep 1993 22:10:40



|> Hi,
|>  I have got a little problem when I try to compile TOM on our IRIS Indigo:
|>
|> 1 33 MHZ IP12 Processor
|> FPU: MIPS R2010A/R3010 VLSI Floating Point Chip Revision: 4.0
|> CPU: MIPS R2000A/R3000 Processor Chip Revision: 3.0
|> On-board serial ports: 2
|> Data cache size: 32 Kbytes
|> Instruction cache size: 32 Kbytes
|> Main memory size: 32 Mbytes
|> Integral Ethernet: ec0, version 0
|> Disk drive: unit 1 on SCSI controller 0
|> Integral SCSI controller 0: Version WD33C93B, revision C
|> Iris Audio Processor: revision 3
|> Graphics board: GR2-XS24 with Z-buffer
|>
|> OS: IRIX 4.0.5F
|>
|>
|>
|> I get the following error message by uld when I try to link the opject code:
|>
|>
|>         /usr/bin/f77 -G 2 -w -col120 -DIRIS4D -O3 -kp extern -o tom abinit.o addall.o addmo.o addsc.o addtxt.o anchor.o assign.o  atid.o atomty.o back.o backobj.o bfdevi.o bfitnew.o bitgen.o  blob.o bndang.o bndbrk.o bond.o bonsub.o boxlim.o change.o  cha|> t1.o chat2.o chnfix.o cntrl.o colcircles.o colsub.o  consrf.o contax.o contsym.o date.o defrng.o delete.o  dgloop.o dglsub.o dglx.o disabl.o disang.o disgen.o dougal.o  drawlines.o dummy.o empty.o enable.o enterid.o faslnk.o
                                                 ^^^^
                                            set this to -O1

|> fbrt.o fit.o  fitang.o fitarrays.o fitbet.o fitcg.o fitder.o fitdsn2.o  fitesum.o fithyd.o fitnow.o fitout.o fitrdf.o fitrot.o  fitsort.o fittom.o fitupd.o fitzed.o fitzon.o fixtre.o  frodbd.o getatd.o getput.o guide.o hbond.o helix.o help.o huetbl.o  hue|> tbl0.o id.o idgen.o insert.o inter.o isitma.o ldigr.o  lista.o listr.o loada.o loddic.o lodtom.o lsqp.o makeat.o  makedi.o makems.o makepdb.o makespl.o makewh.o mapctr.o  matfit.o memps.o molobj.o move.o noda.o noid.o orig.o
|> ortho.o pdb.o pjoin.o print.o prnt.o probld.o psplot.o  problz.o prompt.o pseudo.o rama.o readan.o readat.o  readdi.o readmo.o readvg.o readwh.o redatd.o refi.o refine.o  rename.o renorm.o replce.o reset.o rigid.o ringio.o rock.o  rotmat.o rsrbit.o rtmr.o|>  rtmrsub.o sam.o save.o seprat.o set.o  setaty.o setmar.o setmov.o spcgrpentry.o spcgrps.o spjoin.o  squeze.o ssbrid.o star.o stat.o subs.o surf.o swpbak.o symbig.o  symerr.o symgen.o symtry.o syscom.o tajps0.o tom.o tor.o  
|> torfix.o torgen.o torrot.o trans.o trebil.o update.o  updmol.o utilty.o vector.o verify.o vgout.o view.o water.o weight.o  deffnt.o gtime.o make.o scrsave.o tmpnam.o tomhard.o exec.o tajps1.u graphics.u reddials.u maps.u pen.u   -L -L/usr/lib  -lfgl -lgl_|> s  
|> uld:
|> assertion failed: pmext->aext.asym.value == pmext->usym_size at ext_tbl.c line 530
|> *** Error code 1
|>
|> Stop.
|>
|> Unfortunately I can not find ext_tbl.c which might be a part of a library nor
|> I am able to find the varibles pmext->aext.asym.value or pmext->usym_size.
|>
|> Could anybody please give me any hints to solve this problem ?
|>
|> Thank You very much
|>
|>
|> Carsten Menke
|>
|>  
|> -_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
|>
|> Universitaet Bielefeld       Phone [int+49] +521-106-6276
|> Fakultaet fuer Chemie        Fax   [int+49] +521-106-6146
|> Anorganische Chemie I
|> Postfach
|>
|> D-33501 Bielefeld  (Germany)
|>


|>
|> -_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
|>
|>
|>

R:)

--
***************************************************************************

* Norwegian EMBnet node  Tel:xx-47-22958756 Fax:xx-47-22694130            *
***************************************************************************  

 
 
 

Compiling TOM/FRODO on an IRIS Indigo

Post by Vas » Fri, 03 Sep 1993 22:29:20


I am having problems print a postscript file from a Crimson.  The lab here has
a HP Laserjet 4 printer and we are using the serial port.  The printer flags
an error 22.  After hitting CONTINUE a few times on the printer.  It prints out
a part of the picture but the rest if a buch of dark lines.  I have tried all
combinations of setting on the printer.  Any suggetions to help me would be
appreciated.
Thank you,
Vasu
--
 
 
 

Compiling TOM/FRODO on an IRIS Indigo

Post by David B.Anders » Sat, 04 Sep 1993 01:21:08



Quote:>Hi,
> I have got a little problem when I try to compile TOM on our IRIS Indigo:
>OS: IRIX 4.0.5F
>I get the following error message by uld when I try to link the opject code:
>        /usr/bin/f77 -G 2 -w -col120 -DIRIS4D -O3 -kp extern -o tom abinit.o addall.o addmo.o addsc.o addtxt.o anchor.o assign.o  atid.o atomty.o back.o backobj.o bfdevi.o bfitnew.o bitgen.o  blob.o bndang.o bndbrk.o bond.o bonsub.o boxlim.o change.o  chat1.o chat2.o chnfix.o cntrl.o colcircles.o colsub.o  consrf.o contax.o contsym.o date.o defrng.o delete.o  dgloop.o dglsub.o dglx.o disabl.o disang.o disgen.o dougal.o  drawlines.o dummy.o empty.o enable.o enterid.o faslnk.o
[ ]
>torfix.o torgen.o torrot.o trans.o trebil.o update.o  updmol.o utilty.o vector.o verify.o vgout.o view.o water.o weight.o  deffnt.o gtime.o make.o scrsave.o tmpnam.o tomhard.o exec.o tajps1.u graphics.u reddials.u maps.u pen.u   -L -L/usr/lib  -lfgl -lgl_s  
>uld:
>assertion failed: pmext->aext.asym.value == pmext->usym_size at ext_tbl.c line 530
>*** Error code 1

ext_tbl.c is part of uld (which is just a link to ld).
The ext_tbl.c code is:
                if (pcurext->asym.sc == scCommon ||
                    pcurext->asym.sc == scSCommon) {
                    if (pcurext->asym.value > pmext->usym_size)
                        pmext->usym_size = pcurext->asym.value;
                    if (pcurext->asym.value > pmext->aext.asym.value)
                        pmext->aext.asym.value = pcurext->asym.value;
                    assert(pmext->aext.asym.value == pmext->usym_size);
               }
(quoting the ld code just for fun :-)

Apparently ld believes that some variable in 'common' has
internally-inconsistent data.  Unfortunately it is not telling us what
the name of the variable is...

As a workaround/experiment, I'd suggest using -O2 rather than O3 (thus
avoiding uld and this code).  That means recompiling anything compiled
-O3 and then relinking.

I presume you have the 2.40 or 3.10 compilers.  It is possible that
3.10.1 compilers fix this bug.  Note that this is a hope rather than a
promise...
        Check your compiler release with:
versions -Ivb maint_dev maint_ftn dev ftn
For reference, the 3.10.1 compiler versions output is:
   Name                 Date      Description
I  dev                  06/07/93  Development System, 4.1.1
I  ftn                  06/07/93  Fortran 77, 3.5.1

Hope this helps.

 
 
 

Compiling TOM/FRODO on an IRIS Indigo

Post by Gerard Kleijwe » Sun, 05 Sep 1993 02:35:45


Why on Earth are so many people still using FRODO in the
first place ?
There's a MUCH better successor, called O, by the same
author; it runs on SGIs & ESVs.  Since it comes as
an executable you don't need to compile/link anything.

--Gerard

******************************************************************
             Gerard J. Kleywegt              _____
Department of Molecular Biology              |   |   / \
              Biomedical Centre             / \ ---  | |
          University of Uppsala             | | | |  | |
                        Uppsala             | | | |  | |
                         SWEDEN             | | \ /  ---
                                            ---  |____|

******************************************************************
           "He's probably pining for the fiords ..."
******************************************************************
  The opinions in this mail/post are fictional.  Any similarity
   to actual opinions, living or dead, is purely coincidental.
******************************************************************

 
 
 

1. Iris INDIGO inthedoor

Just received 2 Indigos with 24bit color and some other stuff.
Will be using SGI-Explorer (Craig Upson) as front end for
production animation support tools.  Mostly compositing and output
device control.  

Got the C++ developers kit which came with 3 pubs from AT&T.
This is < 3.0 and does not support templates.  No word on 3.0 C++
from SGI yet.

Came with some great books about XWindows (O'Reilly) and Motif
programming.  Supports what is called mix-model programming for
using GL calls in X code (nifty).

Yes, there is a fractal on the cover of their graphics manuals.

The whole package is very nice, the CD distributions are much more
compact than tapes (ITIOTY).  Now if I can turn the thing on....

Andrew Rose
San Francisco Production Group

415 495 5595

BTW I was definitely not psyched about shelling $525 for registration at
SIGGRAPH and $250 rt from SFO and then figuring out where to stay and
stuff.  Someday maybe deep pocket will foot the bill, but not this year.

But I love graphics, and crowds, and the smell of highly polished salesforces,
and the Video Toaster Demo, and the papers and stuff so I guess its worth it.

2. Camera Tracking

3. X error problems on IRIS Indigo.

4. Mac .flc/.fli viewers

5. Change Video Resolution on iris/indigo

6. 3D Studio ver 2.0

7. Pixmode on Personal IRIS versus Indigo

8. Glut changes timings?

9. GL different on IRIS 4D/35 and Indigo ELAN

10. Iris Indigo WordPerfect Viewer File.

11. PHIGS on Silicon Graphics IRIS indigo

12. unable to compile Xemacs on SGI Indigo running IRIX 4.0.5H