#qi-hardware IRC log for Friday, 2015-11-27

--- Fri Nov 27 201500:00
larsceintopf: did you do anything with the adf7242?14:15
eintopfnot yet, sorry :(14:15
larscno problem14:16
eintopfI will cc you when I have patches14:16
larscwe were thinking about getting it ready for mainline14:16
eintopfcan you provide a link of current state of driver?14:16
larscits probably what you have14:17
larscjust wanted to make sure there are no conflicting changes on your side14:17
eintopfah 4.014:17
larscframework got completely rewritten in 4.1?14:18
eintopfnot completely14:18
eintopfwe have no xmit_async callback14:18
eintopfI am still not sure if this with running workqueues between ndo_xmit (netdev callback to transmit some buffer, which is irq context) and xmit_sync callback has races which we don't know14:20
eintopfbecause it's lost context information14:20
eintopfwith xmit_async you need to use spi_async for transmit messages14:20
eintopfand replace complete(&lp->tx_complete); with ieee802154_xmit_complete14:21
larscbecause it can't sleep?14:21
eintopfit can't do things like ifdown :-)14:21
eintopffor receive/transmit handling I would suggest to use spi_async handling, the other callbacks for 802.15.4 settings. might_sleep is allowed14:23
eintopfand we have now support for get phy capabilities14:24
eintopfto get "which transmit power does the transceiver support"14:24
eintopfthis might different on country and non ISM14:24
eintopfwe need for that something like regulator db in wireless14:25
eintopfbut we have not support for this right now14:25
larscok, thanks14:29
eintopfand example for providing tx power capabilities14:30
eintopfit's a simple array...14:31
eintopfin mbm values14:31
eintopf2.3 dbm == 230014:31
--- Sat Nov 28 201500:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!