Hi all, thanks to all, who helped me to find the error in my Real-Time application.
To remember the problem: When using big matrices, I got a "unable to handle kernel paging request"-error when *unloading* the kernel module. The error occured when I tried to do a matrix inversion on a matrix bigger than 10x10. Steven Ivandich gave me the hint to allocate memory for the matrices when starting the program. So I included some kmalloc()'s in the init_module()-function (and the same number or kfree()'s in cleanup_module() ;-) ). After having done that, I wondered, why I now got the same faults like Stephen Bridges reported in his posting "Memory leak". Doing some hours on memory testing with "memtest86" didn't show any malfunctions, so I looked at my code again. After doing some work, I realized, that there were some pointer errors in it. After correcting this, I was able to do matrix inversion up to 100x100 without errors. Bigger matrices will also work, I suppose, but I haven't tested. Conclusion: If you get "kernel paging request"-errors, allocate more mem for your variables, but be careful, not to run into pointer errors. Greetings from Wuppertal, Germany Arne Linder Dipl.-Ing. Arne Linder Labor fuer elektrische Maschinen und Antriebe Fachbereich 13 Bergische Universitaet - Gesamthochschule Wuppertal D-42097 Wuppertal e-mail: [EMAIL PROTECTED] -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] -- For more information on Real-Time Linux see: http://www.rtlinux.org/
