You can reproduce the problem by stracing a system call which, for
instance, tries to load a module. i.e. open() of a device whose module
is not loaded.
I don't think user-space debuggers should need to know about the
internals of the kernel, and so there is no reason to keep this.
One possible fix would be to add a test in
arch/i386/kernel/ptrace.c:syscall_trace (and similarly for other arches)
to see if the syscall was made from kernel mode. If so, the parent
should not be notified.
If someone will suggest a good way to detect kernel mode, I can submit a
patch.
I'd appreciate it if replies could be cc'ed to me. Thanks,
--Nate Eldredge nate@cartsys.com
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/