Re: Debugging threads

Uwe Bonnes <bon-1JbLm1bU5j5ZIx36JBfelj3+ndqKAYMe9FMPySWZwLkb1SvskN2V4Q@public.gmane.org> Thu, 28 Jun 2018 20:19:02 +0200
Newsgroups gmane.comp.hardware.microcontrollers.ethernut
Message-ID <[email protected]>
>>>>> "Ole" == Ole Reinhardt <ole.reinhardt-L1vi/[email protected]> writes:

    Ole> Hi Uwe, I remember we had a similar discussion quite some time
    Ole> ago. If I remember correctly, someone just wrote Nut/OS for gdb
    Ole> some (long) time ago. But I'm very unsure...

    Ole> I think you'd need a tracing solution. If the debugger stops in the
    Ole> idle thread you'll quite likely be in the idle thread, which means
    Ole> that your code waits for something (for some data on a socket or
    Ole> uart, a mutex, a sleep, etc.) an no other thread is runnable.

    Ole> So you ideally need to trace back your code path into the past to
    Ole> see from where you changed into the idle thread.

Hello Ole,

nice to hear from you again!

In the meantime, I revived and fixed some issues in
 http://openocd.zylin.com/#/c/3881/

It works soemhow with on a nucle_l053 and I can see the threads. However
the hand does not happen on L053. On f3 discovery, where th thread hang
happens, OpenOcd has the old problems with halting the target and the thread
list does not make sense. I have to dig deeper. 

And maybe I have to activate and understand SWO...

Bye
-- 
Uwe Bonnes                bon-1JbLm1bU5j5ZIx36JBfelj3+ndqKAYMe9FMPySWZwLkb1SvskN2V4Q@public.gmane.org

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 1623569 ------- Fax. 06151 1623305 ---------
_______________________________________________
http://lists.egnite.de/mailman/listinfo/en-nut-discussion