Hello. On Tue, 2010-04-13 at 12:11, Hennerich, Michael wrote: > Stefan Schmidt wrote on 2010-04-13: > > The driver is ready as far it can be. However I'm not sure you noticed - the > entire mac802154 softmac is still in some work in progress state.
I learned that the hard way over the last days. :) I missed the fact that ieee802154 has already landed mainline but not mac802154. My bad. > And therefore not yet fully ready for 'Mainline' inclusion. > > A few examples: > 1) Try to assoc your wpan-phy with a coordinator running on the same phy. > You will notice that it doesn't work. > (which is probably the root cause of the problem you have reported in one of > your earlier mailing list posts). It is part of the problem. Testing this with fakelb works fine. Testing this with cc2420 on one device brings the problem you are describing. On the other hand I see the same symptoms when running the coordinator on one device and trying to associate from another. (Two imotes with cc2420 chip) The reason seems to be a different one here thought. Sniffing the air traffic shows me a malformated packet coming in when I'm trying to associate. Even the CRC failed. Going down this raod now to see what breaks the packet. > 2) Missing return value handling of xmit(). > There should be at least something like this: > RETVAL: > XMIT_SUCCESS > XMIT_SUCCESS_DATPEND > XMIT_FAILURE_CSMACA > XMIT_FAILURE_NOACK > > And upper layer should respond with appropriate actions. > > Since stuff is still in flux - I really prefer to keep debug options in the > driver for now. Agreed. I thought this driver was just one hop (Dmitry's tree) away from Linus and was wondering why it was applied so fast without much back and forth review. (Same for cc2420). No that I understand that mac802154 will need some more time to mature it makes more sense to me. > > The you should consider fixing this setup: Documentation/email- > > clients.txt > > There is not short term option for me right now. > I'll need to submit a patch as attachment. > Unfortunately this mailing list has a message size limit of 40kB. > So inline patch and attachment got rejected. 40kB is indeed a bit outdated. Dmitry, is this something you can fix? regards Stefan Schmidt ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Linux-zigbee-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-zigbee-devel
