[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-rt
Subject:    Re: [Rtl] Problems debugging a realtime module
From:       Jean-Sacha Melon <zxmzu23 () student ! uni-tuebingen ! de>
Date:       2003-11-17 13:18:39
[Download RAW message or body]

[Attachment #2 (multipart/signed)]


Hello,


> > Greetings, 
> > I have a little problem debugging my module: after starting gdb and 
> > doing 'target remote /dev/rtf10' I get the following message: 
> > 
> > (gdb) target remote /dev/rtf10 Remote debugging using /dev/rtf10 [New
> > Thread -159613952] [Switching to Thread -159613952] 0xf8a352f6 in ?? ()
> > warning: shared library handler failed to enable breakpoint
> > 
> > I don't understand why this warning occurs, the effect is that I can not 
> > debug my program. It halts because of the breakpoint() but thats all.
> > 
> > Some software versions:
> > gnu gdb 5.3
> > gcc 2.95.3
> > rtlinux-3.2-pre2
> > linux kernel 2.4.20
> > 
> can't reproduce this:
> 
> does the example in the debugger directory work ? 
> if the example works - can you send me the code causing the problem ?

Okay I was a little bit wrong: 
I tried to run an older rtlinux/comedi program on a new machine with a new comedi \
installation. In the older version the comedi_open(...) command (in the realtime \
environment!) used an integer to specify the device, in my case  comedi0 -> so I used \
the integer 0. In the new version of comedi this changed (what I did not know), and \
so I got a  null pointer exception (what I did not recognize at first because the \
kernel shows me that null pointer exception only  after I try to remove the module \
from kernel space). My breakpoint was never reached!

After removing this error the debugger is working but not properly: The above error \
message is remaining. Breakpoints  are sometimes ignored and sometimes the program is \
breaked but not at the exact position of the breakpoint. My solutions  to this is not \
to use breakpoint() but a division by zero - that is working properly for my \
debugging purposes.

May be now someone knows why this problem occurs? Its not important for me now, but \
interesting nevertheless...


Bye, Sacha.



-----------------------------------------------------
Sofienstraße 6
72070 Tübingen
Tel.: < not yet - thanks to german telekom >

Reply only in plain text! 
Only the finest schnickschnack!


[Attachment #5 (application/pgp-signature)]

_______________________________________________
Rtl mailing list
Rtl@rtlinux.org
http://www2.fsmlabs.com/mailman/listinfo/rtl


[prev in list] [next in list] [prev in thread] [next in thread]