Weber, Michael J. (US SSA)
2004-Apr-08 20:12 UTC
Building 3.7-p1 on i686-pc-linux-gnu for execution oni386-pc-linux-gnu
Peter, thanks for the idea! I did it exactly as you described... and I still get "Illegal instruction" on the 386 platform. I also tried # CFLAGS='-mcpu=i386 -march=i386' ./configure --prefix=/my/prefix/here and # CFLAGS="-mcpu=i386 -march=i386" ./configure --prefix=/my/prefix/here just for grins, and all had the same result. Still trying... Thanks, Mike> -----Original Message----- > From: > openssh-unix-dev-bounces+michael.weber=baesystems.com at mindrot. > org > [mailto:openssh-unix-dev-bounces+michael.weber=baesystems.com@ > mindrot.org] On Behalf Of Peter Stuge > Sent: Thursday, April 08, 2004 3:14 PM > To: openssh-unix-dev at mindrot.org > Subject: Re: Building 3.7-p1 on i686-pc-linux-gnu for > execution oni386-pc-linux-gnu > > > On Thu, Apr 08, 2004 at 12:24:19PM -0400, Weber, Michael J. > (US SSA) wrote: > > I have tried putting -march=i386 and -mcpu=i386 on the command line > > before running configure, but it doesn't work > > How did you do this, exactly? > > Try: > > $ CFLAGS=-mcpu=i386 ./configure > > > //Peter > > _______________________________________________ > openssh-unix-dev mailing list > openssh-unix-dev at mindrot.org > http://www.mindrot.org/mailman/listinfo/openssh-unix-dev >
Ben Lindstrom
2004-Apr-08 20:22 UTC
Building 3.7-p1 on i686-pc-linux-gnu for execution oni386-pc-linux-gnu
What may be more helpful would be if you had gdb and see what aspect of the code is failing with "Illegal instruction". That way one can know what the issue is. For all we know your build platform compiler could be be inserting code that it should not be from your environment. - Ben On Thu, 8 Apr 2004, Weber, Michael J. (US SSA) wrote:> Peter, thanks for the idea! I did it exactly as you described... and I > still get "Illegal instruction" on the 386 platform. > > I also tried > > # CFLAGS='-mcpu=i386 -march=i386' ./configure --prefix=/my/prefix/here > > and > > # CFLAGS="-mcpu=i386 -march=i386" ./configure --prefix=/my/prefix/here > > just for grins, and all had the same result. > > Still trying... > > Thanks, > Mike > > > -----Original Message----- > > From: > > openssh-unix-dev-bounces+michael.weber=baesystems.com at mindrot. > > org > > [mailto:openssh-unix-dev-bounces+michael.weber=baesystems.com@ > > mindrot.org] On Behalf Of Peter Stuge > > Sent: Thursday, April 08, 2004 3:14 PM > > To: openssh-unix-dev at mindrot.org > > Subject: Re: Building 3.7-p1 on i686-pc-linux-gnu for > > execution oni386-pc-linux-gnu > > > > > > On Thu, Apr 08, 2004 at 12:24:19PM -0400, Weber, Michael J. > > (US SSA) wrote: > > > I have tried putting -march=i386 and -mcpu=i386 on the command line > > > before running configure, but it doesn't work > > > > How did you do this, exactly? > > > > Try: > > > > $ CFLAGS=-mcpu=i386 ./configure > > > > > > //Peter > > > > _______________________________________________ > > openssh-unix-dev mailing list > > openssh-unix-dev at mindrot.org > > http://www.mindrot.org/mailman/listinfo/openssh-unix-dev > > > > _______________________________________________ > openssh-unix-dev mailing list > openssh-unix-dev at mindrot.org > http://www.mindrot.org/mailman/listinfo/openssh-unix-dev >
Weber, Michael J. (US SSA)
2004-Apr-08 22:12 UTC
Building 3.7-p1 on i686-pc-linux-gnu for execution oni386-pc-linux-gnu
Thanks for the tip Ben... after much poking and prodding, I managed to learn that the ssh tools were receiving SIGILL from the libcrytpo library, so it's back to OpenSSL... sorry for the false alarm. I'm still fighting all of the packages, but now that I have another tool in my box it should be easier to pinpoint the issues. Thanks! Mike> -----Original Message----- > From: Ben Lindstrom [mailto:mouring at etoh.eviladmin.org] > Sent: Thursday, April 08, 2004 4:22 PM > To: Weber, Michael J. (US SSA) > Cc: OpenSSH Development > Subject: RE: Building 3.7-p1 on i686-pc-linux-gnu for > execution oni386-pc-linux-gnu > > > > What may be more helpful would be if you had gdb and see what > aspect of the code is failing with "Illegal instruction". > That way one can know what the issue is. > > For all we know your build platform compiler could be be > inserting code that it should not be from your environment. > > - Ben
Seemingly Similar Threads
- Building 3.7-p1 on i686-pc-linux-gnu for execution on i386-pc-linux-gnu
- build fails dumping R code in 'methods' on i686-pc-linux-gnu (PR#1632)
- build fails dumping R code in 'methods' on i686-pc-linux-gnu (PR#1630)
- SEM model testing with identical goodness of fits
- [dspam-users] mysql fetch row error