|
Known problems
Stepping
The single-stepping currently used in gdb-cris is software
single-stepping performed on the host side. As a
consequence, stepping is not fool-proof. If you try to step
past an assembly instruction, and the program just continues
until the end of the program (or until next breakpoint) that's
a good sign of the single-stepping mechanism failing. If step
or next fails, try setting a breakpoint where you
want to stop and then continue to reach the desired
point in your program. (If you can isolate the place where
single-stepping fails and send us the code we'd be grateful.)
Backtrace in Threads
When backtracing in a thread, gdb is unable to determine
that the clone function terminates the call stack:
#0
thread_function (arg=0x0) at hello.c:7
#1 0x3556e06a
in pthread_start_thread (arg=0x0) at manager.c:262
#2 0x3556e0b2
in pthread_start_thread_event (arg=0x9f7ffc00) at
manager.c:285
#3 0x3560d244
in clone () from /usr/local/cris/r59/cris-axis-linux-gnu/lib/libc.so.6
#4 0x3560d244
in clone () from /usr/local/cris/r59/cris-axis-linux-gnu/lib/libc.so.6
Previous frame identical to this frame (corrupt stack?)
The stack
is not corrupt however. The reason is that clone lacks the
necessary debug information.
Reporting Bugs
Report bugs to the standard developer mailing list (dev-etrax@axis.com).
|
