BMC/Intel-BMC/linux.git/drivers/net/ieee802154, branch dev

at86rf230: cleanup start and stop callbacks

2015-06-16T16:55:11+00:00

This code cleanups the start and stop callbacks by removing hw->priv and using the already dereferenced variable lp which is the same. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

at86rf230: add recommended csma backoffs settings

2015-06-16T16:55:11+00:00

This patch adds support for a new random csma backoffs settings when going into sleep state. This is recommended according at86rf2xx datasheets. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

at86rf230: fix phy settings while sleeping

2015-06-16T16:55:11+00:00

While in sleep state then we can't access the at86rf2xx registers. This patch checks if the transceiver is in sleep state before sending spi messages via regmap. Regmap is used on every driver ops callback except for receive and xmit handling, but while receive and xmit handling the phy should not be inside the sleep state. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

fakelb: add xmit_async after stop testcase

2015-06-14T23:25:06+00:00

This patch adds a suspended testcase into the xmit_async functionality. In the hope that we can found race conditions when xmit_async is called after an ieee802154_ops stop. This should never happen. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

at86rf230: add support for sleep state

2015-06-14T23:22:19+00:00

This patch adds support for sleep state when between stop and start period. In this period the transceiver isn't used by the subsystem, in this time we disable the irq and going into the sleep state. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

at86rf230: use level high as fallback default

2015-06-14T23:22:19+00:00

This patch use high level interrupt type as fallback handling when no irq type is given. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann

ieee802154: cc2520: check for return values in cc2520_filter()

2015-06-09T12:05:22+00:00

neither ram nor register write return values have been checked here. Checking both now. Assign ret with 0 as all other assignments are inside if blocks and might not happen before we return ret. CID: 1230469 Signed-off-by: Stefan Schmidt Acked-by: Varka Bhadram Signed-off-by: Marcel Holtmann

ieee802154/mrf24j40: make sure we do not override return values

2015-06-09T08:55:36+00:00

If we run into an error during rx we set the the error code in ret, but override it afterwards. Using a different variable for the extra case avoids this situation. CID: 1226982, 1226983 Signed-off-by: Stefan Schmidt Cc: Alan Ott Signed-off-by: Marcel Holtmann

mac802154: remove aack hw flag

2015-06-07T07:13:32+00:00

This patch removes the hardware auto acknowdledge flag which indicates that the transceiver supports this handling. This flag is never evaluated inside mac802154 and all transceivers should support this handling by default per hardware. Suggested-by: Lennert Buytenhek Cc: Alan Ott Signed-off-by: Alexander Aring Reviewed-by: Varka Bhadram Acked-by: Stefan Schmidt Acked-by: Varka Bhadram Signed-off-by: Marcel Holtmann

mac802154: remove unneeded vif struct

2015-06-07T07:13:32+00:00

This patch removes the virtual interface structure from sub if data struct, because it isn't used anywhere. This structure could be useful for give per interface information at softmac driver layer. Nevertheless there exist no use case currently and it contains the interface type information currently. This information is also stored inside wpan dev which is now used to check on the wpan dev interface type. Signed-off-by: Alexander Aring Reviewed-by: Varka Bhadram Acked-by: Varka Bhadram Signed-off-by: Marcel Holtmann