[M3devel] segmentation fault upon ./install-cm3-compiler.sh upgrade

Jay jayk123 at hotmail.com
Thu Jan 24 23:32:08 CET 2008


 > Stable releases cannot necessarily be  > expected to run on newer operating systems when the ABI changes, as they  > don't adapt themselves ;-) 
Wow. No stable ABI? What's up with that?
I  understand that building with current headers/libs might not work on older systems, but building with older headers/libs should almost always work on newer systems, AT LEAST at the ABI level, if not at the exact behavioral level, that's much harder, if there is to be any change or any bug fixes. People get away with the craziest things, like double frees (ok, not in safe Modula-3), only to have newer better heaps be more strict and catch them...
 
 > Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". 
 Is that normal? 
 
 > Program received signal SIGSEGV, Segmentation fault.  > 0x9e841069 in ?? ()  > (gdb) bt  > #0 0x9e841069 in ?? ()  > Cannot access memory at address 0xf05fc929  > (gdb) If the above is normal..can you break on main? And..I don't know what the name is...a bisection search..for how far it gets before it crashes? You know, either systematically, or at least randomly, step over and step into function calls, if you go to far and it crashes, step into next time instead of step over.
> > If debugging symbols are missing, we've got to fix that.
 
Sorry, I was speculating, based mainly on "strip" appear in scripts/*.sh, didn't look if it is used.
> Well, that doesn't mean debugging symbols are missing, sometimes this > kind of thing happens anyway (right?).
Right.
 - Jay



> Date: Thu, 24 Jan 2008 23:11:08 +0100> From: neels at elego.de> To: wagner at elegosoft.com> CC: m3devel at elego.de; m3-support at elego.de> Subject: Re: [M3devel] segmentation fault upon ./install-cm3-compiler.sh upgrade> > > > Olaf Wagner wrote:> > Quoting Neels Janosch Hofmeyr <neels at elego.de>:> >> >> btw, I am using the "stable release" source tarballs with the 5.4.0> >> > Hi Neels,> >> > 5.4.0 is rather old now. I assume Ubuntu has made some incompatible> > changes to its kernel or C library interfaces that the old cm3 binary> > cannot cope with. What about trying one the more recent installation> > archives, like d.5.5.1? Once you have a working cm3 and cm3cg, compiling> > the rest should be no problem. Stable releases cannot necessarily be> > expected to run on newer operating systems when the ABI changes, as they> > don't adapt themselves ;-)> >> > Try this, for example:> >> > http://modula3.elegosoft.com/cm3/snaps/cm3-min-POSIX-LINUXLIBC6-d5.5.1-2008-01-24-03-35-01.tgz > >> I'll try it next.> btw, I am writing an instructions text for installing on ubuntu 7.10 > along the way...> >> > It should also be possible to get a backtrace of the cm3 crash> > within gdb. There should be debugging symbols in the distributed> > binaries. What does bt show?> I tried bt, it shows nothing:> > $ gdb /usr/local/cm3/bin/cm3> GNU gdb 6.6-debian> Copyright (C) 2006 Free Software Foundation, Inc.> GDB is free software, covered by the GNU General Public License, and you are> welcome to change it and/or distribute copies of it under certain > conditions.> Type "show copying" to see the conditions.> There is absolutely no warranty for GDB. Type "show warranty" for details.> This GDB was configured as "i486-linux-gnu"...> Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1".> (gdb) run -version> Starting program: /usr/local/cm3/bin/cm3 -version> > Program received signal SIGSEGV, Segmentation fault.> 0x9e841069 in ?? ()> (gdb) bt> #0 0x9e841069 in ?? ()> Cannot access memory at address 0xf05fc929> (gdb)> > >> > If debugging symbols are missing, we've got to fix that.> Well, that doesn't mean debugging symbols are missing, sometimes this > kind of thing happens anyway (right?).> >> > Olaf> >> >> Neels Janosch Hofmeyr wrote:> >>> Hi lists,> >>>> >>> I am trying to install cm3 5.4.0 on Ubuntu 7.10. I know that it > >>> worked on 7.4, and am doing the exact same steps. I have now tried > >>> it with cm3-min-...-5.4.0, just as I did last year.> >>>> >>> But now, I get this output:> >>>> >>> neels at oubantu:~/cm3-build/scripts> >>> $ ./install-cm3-compiler.sh upgrade> >>> cp /usr/local/cm3/bin/cm3 /usr/local/cm3/bin/cm3-5.4.0> >>> cp /usr/local/cm3/bin/cm3cg /usr/local/cm3/bin/cm3cg-5.4.0> >>> Segmentation fault (core dumped)> >>> cp /home/neels/cm3-build/m3-sys/cm3/LINUXLIBC6/cm3 > >>> /usr/local/cm3/bin/cm3-> >>> cp /home/neels/cm3-build/m3-sys/m3cc/LINUXLIBC6/cm3cg > >>> /usr/local/cm3/bin/cm3cg-> >>> cp /usr/local/cm3/bin/cm3- /usr/local/cm3/bin/cm3> >>> cp /usr/local/cm3/bin/cm3cg- /usr/local/cm3/bin/cm3cg> >>>> >>> I have done this a second time, making sure everything is cleaned > >>> out and monitored things. From adding a `set -x' in the > >>> install-cm3-compiler.sh, it becomes obvious that > >>> `/usr/local/cm3/bin/cm3 -version' exits with a Segmentation Fault.> >>>> >>> After doing ./cminstall (the minimal binary install), cm3 -version said> >>> $ /usr/local/cm3/bin/cm3 -version> >>> Critical Mass Modula-3 version 5.4.0> >>> last updated: 2006-10-11> >>> configuration: /usr/local/cm3/bin/cm3.cfg> >>>> >>> After doing ./do-cm3-core.sh buildship, it still said the same.> >>> So, somewhere in install-cm3-compiler.sh, a disfunctional cm3 > >>> binary gets installed in /usr/local/cm3/bin/, after which cm3 > >>> yields only segmentation faults.> >>>> >>> After install-cm3-compiler.sh, cm3 -version says> >>> $ /usr/local/cm3/bin/cm3 -version> >>>> >>>> >>> ***> >>> *** runtime error:> >>> *** Segmentation violation - possible attempt to dereference NIL> >>> *** pc = 0x9e841069> >>> ***> >>>> >>> Aborted (core dumped)> >>>> >>> Trying to backtrace in gdb apparently doesn't work -- I don't know > >>> how to compile debugging symbols into it. Giving up.> >>>> >>> I think now is the time to remove the statement "[cm3 is] > >>> easy-to-use [and] easy-to-install" from modula3.elegosoft.com. No > >>> piece of software I have ever encountered is as difficult to use > >>> and as impossible to install as critical mass modula3. I *am* > >>> following all the instructions!> >>>> >>> argh,> >>> Neels> >>>> >>> >> -- > >> Neels Janosch Hofmeyr> >> Software Developer> >>> >> neels at elego.de> >> Public Key: http://binarchy.net/neels/neels.hofmeyr.public.key.asc> >>> >> elego Software Solutions GmbH http://www.elegosoft.com> >> Gustav-Meyer-Allee 25, Gebäude 12 HRB 77719> >> 13355 Berlin, Germany Amtsgericht Charlottenburg> >> Tel.: +49 30 23 45 86 96 Sitz der Gesellschaft: Berlin> >> Fax: +49 30 23 45 86 95 Geschäftsführer: Olaf Wagner> >> >> >> > -- > Neels Janosch Hofmeyr> Software Developer> > neels at elego.de> Public Key: http://binarchy.net/neels/neels.hofmeyr.public.key.asc> > elego Software Solutions GmbH http://www.elegosoft.com> Gustav-Meyer-Allee 25, Gebäude 12 HRB 77719> 13355 Berlin, Germany Amtsgericht Charlottenburg> Tel.: +49 30 23 45 86 96 Sitz der Gesellschaft: Berlin> Fax: +49 30 23 45 86 95 Geschäftsführer: Olaf Wagner> > 
_________________________________________________________________
Need to know the score, the latest news, or you need your Hotmail®-get your "fix".
http://www.msnmobilefix.com/Default.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m3lists.elegosoft.com/pipermail/m3devel/attachments/20080124/9a5f337f/attachment-0002.html>


More information about the M3devel mailing list