Teresa Noviello wrote: > ... > [319540.247471] > [319540.247503] *** RTnet 0.9.2 - built on Mar 17 2006 19:18:22 *** > [319540.247505] > [319540.248950] RTnet: initialising real-time networking > [319540.998472] eepro100-rt.c:1.36-RTnet-0.7 2002-2005 Jan Kiszka < > [EMAIL PROTECTED]> > [319540.998507] eepro100-rt.c: based on eepro100.c 1.36 by D. Becker, A. V. > Savochkin and others > [319541.042267] rteth0: 0000:01:00.0, 00:00:5B:00:8C:2F, IRQ 11. > [319541.077999] Receiver lock-up workaround activated. > [319541.079381] RTnet: registered rteth0 > [319541.228624] initializing loopback... > [319541.231114] RTnet: registered rtlo > [319541.429589] RTcap: real-time capturing interface > [319543.168199] RTcfg: init real-time configuration distribution protocol > [319543.347488] RTmac: init realtime media access control > [319543.532678] RTmac/TDMA: init time division multiple access control > mechanism > [319543.751381] RTcfg: rtcfg_do_main_event() rtdev=1, > event=RTCFG_CMD_CLIENT, state=RTCFG_MAIN_OFF > [319543.751441] RTcfg: next main state=RTCFG_MAIN_CLIENT_0 > > (now i do 'rtnet stop')
To reproduce your shutdown procedure: you run "rtnet start" in one terminal, leave it running, and then issue "rtnet stop" on some other terminal in parallel? Or do you first interrupt "rtnet start" and then issue the stop? The latter is the "usual" scenario, the former may have received insufficient testing yet and /may/ trigger a bug (which would still need fixing). The trace you reported can be very helpful. It points at the cleanup code of tdma. I will have a closer look later. Jan
signature.asc
Description: OpenPGP digital signature
