1*a1fd9364Slogwang$Id: TODO,v 1.1 2002/11/24 19:47:06 max Exp $ 2*a1fd9364Slogwang$FreeBSD$ 3*a1fd9364Slogwang 4*a1fd9364SlogwangFIXME/TODO list 5*a1fd9364Slogwang 6*a1fd9364Slogwang0) Ping itself. Should L2CAP layer loopback data? 7*a1fd9364Slogwang 8*a1fd9364Slogwang1) Locking/SMP 9*a1fd9364Slogwang 10*a1fd9364Slogwang External code now uses ng_send_fn to inject data into Netgraph, so 11*a1fd9364Slogwang it should be fine as long as Netgraph is SMP safe. Just need to 12*a1fd9364Slogwang verify it. 13*a1fd9364Slogwang 14*a1fd9364Slogwang2) Understand and implement L2CAP QoS 15*a1fd9364Slogwang 16*a1fd9364Slogwang Will fix later. I only have CSR based hardware and it does not 17*a1fd9364Slogwang support QoS. 18*a1fd9364Slogwang 19*a1fd9364Slogwang3) Better functions to manage CIDs and command ident's. 20*a1fd9364Slogwang 21*a1fd9364Slogwang Resource manager is not good because it uses MTX_DEF mutexes, 22*a1fd9364Slogwang (i.e. could block/sleep) 23*a1fd9364Slogwang 24*a1fd9364Slogwang4) Implement group channels (multicast) 25*a1fd9364Slogwang 26*a1fd9364Slogwang Will fix later 27*a1fd9364Slogwang 28*a1fd9364Slogwang5) Add bytes/packets counters and commands to get/reset them 29*a1fd9364Slogwang 30*a1fd9364Slogwang Will fix later. What to count? 31*a1fd9364Slogwang 32*a1fd9364Slogwang6) Better way to get information about channels 33*a1fd9364Slogwang 34*a1fd9364Slogwang L2CAP can support about 65000 channels. Need define some good way 35*a1fd9364Slogwang to get data from kernel to user space. For example if we need to pass 36*a1fd9364Slogwang 1K of information for every channel, then worst case is that we need 37*a1fd9364Slogwang to pass 65Mbytes of data from kernel to user space. Not good. 38*a1fd9364Slogwang 39*a1fd9364Slogwang7) Deal properly with "shutdown"s and hook "disconnect"s 40*a1fd9364Slogwang 41*a1fd9364Slogwang For now we destroy all channels when upstream hook is disconnected. 42*a1fd9364Slogwang Is there a better way to handle this? 43*a1fd9364Slogwang 44