tags 374978 + moreinfo thanks Hello,
Dr. Frank Dohrmann wrote: > Package: paw > Version: 2005.05.09.dfsg-7 > Severity: normal > > Hi Kevin, > > for the current testing version > > fun1 1 x 100 0 100 > > seg-violates, which it should not ... > ... x*x/x would work ... > ... paw++ is affected as well So if I understand correctly, there is no segfault if you enter fun1 2 x*x/x 100 0 100 into PAW, and this function is graphed correctly, is that right? Unfortunately I am not able to reproduce the bug in plotting f(x) = x, neither in testing nor in unstable on i386, nor in unstable on amd64. So there must be something system-dependent about it. Could you please answer the following questions for me? One of them might provide relevant information. - What is the output of the command "which pawX11" ? (Type it at the shell prompt without the surrounding quotes.) - ... of "ldd $(which pawX11)" ? - ... of "ls -l /usr/bin/paw*" ? - ... of "dpkg -l libc6-dev" ? - What are the contents of /etc/ld.so.conf ? - ... of /etc/apt/sources.list ? - Are you using an i386 computer as stated by reportbug in your report? (You can check this with the command "dpkg --print-architecture".) - Do you have any libraries installed in /usr/local/lib or somewhere else listed in /etc/ld.so.conf that might be interfering? - Did you compile any Cernlib-related packages on your own machine, or are all of them the binary debs provided by Debian? - If you run "gdb /usr/bin/pawX11" and try to run the same command in PAW, what is the output of gdb when the segfault occurs? (Of course you'll need to install the gdb package for this test.) Type "where" at the gdb prompt after the segfault to get a backtrace. - Do you have any config files for PAW in your home directory, or in the directory where you run the segfaulting command? If so, please provide them. (Such files would include .pawlogon.kumac in $HOME, and pawlogon.kumac, paw.metafile, last.kumac and higz_windows.dat in the current directory.) - Is PAW creating directories /tmp/comisXXXXXX when run (with XXXXXX replaced by random strings; it should be able to do so)? If you have $TMPDIR set, also look there, and let me know the value of $TMPDIR . - Do you have any non-Debian versions of Cernlib installed on this computer? - Anything else that might be at all unusual about this system? - Are you able to reproduce the segfault on any other computer(s) running Etch, if you have one available? - What about on computer(s) running Sarge or an up-to-date Sid? best regards, -- Kevin B. McCarty <[EMAIL PROTECTED]> Physics Department WWW: http://www.princeton.edu/~kmccarty/ Princeton University GPG: public key ID 4F83C751 Princeton, NJ 08544 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
