1114878Sjulian$Id: TODO,v 1.1 2002/11/24 19:47:06 max Exp $ 2107120Sjulian$FreeBSD$ 3107120Sjulian 4107120SjulianFIXME/TODO list 5107120Sjulian 6107120Sjulian0) Ping itself. Should L2CAP layer loopback data? 7107120Sjulian 8107120Sjulian1) Locking/SMP 9107120Sjulian 10107120Sjulian External code now uses ng_send_fn to inject data into Netgraph, so 11107120Sjulian it should be fine as long as Netgraph is SMP safe. Just need to 12107120Sjulian verify it. 13107120Sjulian 14107120Sjulian2) Understand and implement L2CAP QoS 15107120Sjulian 16107120Sjulian Will fix later. I only have CSR based hardware and it does not 17107120Sjulian support QoS. 18107120Sjulian 19107120Sjulian3) Better functions to manage CIDs and command ident's. 20107120Sjulian 21107120Sjulian Resource manager is not good because it uses MTX_DEF mutexes, 22107120Sjulian (i.e. could block/sleep) 23107120Sjulian 24107120Sjulian4) Implement group channels (multicast) 25107120Sjulian 26107120Sjulian Will fix later 27107120Sjulian 28107120Sjulian5) Add bytes/packets counters and commands to get/reset them 29107120Sjulian 30107120Sjulian Will fix later. What to count? 31107120Sjulian 32107120Sjulian6) Better way to get information about channels 33107120Sjulian 34107120Sjulian L2CAP can support about 65000 channels. Need define some good way 35107120Sjulian to get data from kernel to user space. For example if we need to pass 36107120Sjulian 1K of information for every channel, then worst case is that we need 37107120Sjulian to pass 65Mbytes of data from kernel to user space. Not good. 38107120Sjulian 39107120Sjulian7) Deal properly with "shutdown"s and hook "disconnect"s 40107120Sjulian 41107120Sjulian For now we destroy all channels when upstream hook is disconnected. 42107120Sjulian Is there a better way to handle this? 43107120Sjulian 44