- 28 Nov, 2011 27 commits
-
-
Franky Lin authored
Change parameter to device pointer for bus layer interface function brcmf_sdbrcm_bus_stop. This is part of the fullmac bus interface refactoring. Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
Franky Lin <frankyl@broadcom.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Franky Lin authored
brcmf_bus_get_device is no longer necessary. Use dongle device pointer saved in brcmf_pub directly. This is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
Change parameter to device pointer for bus layer interface function brcmf_sdbrcm_bus_init. This is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
Change parameter to device pointer for bus layer interface function brcmf_sdbrcm_bus_txdata. This is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
Change paramter to device pointer for bus layer interface function brcmf_sdbrcm_bus_txctl. This is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
brcmf_sdbrcm_bus_rxctl acts as an interface function of bus layer. Change parameter from struct brcmf_sdio to device pointer in order to provide a more compatible interface for different bus layers. This is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
structure brcmf_pub contains context for generic layer and should not be used in brcmf_sdbrcm_bus_watchdog. This patch is part of fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
busstate keeps track of the bus (USB/SDIO) status and is used by both generic layer and bus layer. Move it to brcmf_bus helps to clean up the interface. This patch is part of fullmac bus interface refactoring. Reviewed-by:
Alwin Beukers <alwin@broadcom.com> Reviewed-by:
-
Franky Lin authored
struct brcmf_bus will contain function porinter, bus specific private structure pointer and interface context of generic layer and bus layer. It will be the only shared structure between generic and bus layer. This patch is part of fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
Rename sdio bus structure brcmf_bus to brcmf_sdio for preparation of USB bus support. This patch is part of the fullmac bus interface refactoring. Reviewed-by:
-
Franky Lin authored
bus_no, slot, func and bustype are no longer needed by brcmf_sdbrcm_probe. This patch removes them. Reviewed-by:
-
Franky Lin authored
brcmf_c_init only init brcmf_msg_level used for debug. It's no longer needed as brcmf_msg_level doesn't cause trouble to multiple instances. Reviewed-by:
-
Franky Lin authored
This patch is part of the fullmac bus interface refactoring series. It moves the module init/exit code to bus layer. Reviewed-by:
-
Franky Lin authored
A new feature in the dongle firmware requires a handshake during firmware intialization. The request is sent in event packets which the host driver is not able to handle before any net device registered. Discard those packets as the context for handling it is missing. The initialization handler will be added as part of feature support code in the future. Signed-off-by:
-
Arend van Spriel authored
The wait queue request_packet_wait was used in request_chain function and for sake of consistency it has been renamed to request_chain_wait. Reviewed-by:
-
Arend van Spriel authored
The check for alignment is not valid anymore and can be removed. The function is collapsed with brcmf_sdioh_request_packet() as a consequence because it did not add much functionality any longer. Reviewed-by:
-
Alwin Beukers authored
Added support for handling FIF_PROMISC_IN_BSS, FIF_FCSFAIL, FIF_CONTROL, FIF_OTHER_BSS and FIF_PSPOLL. Reviewed-by:
-
Alwin Beukers authored
Reviewed-by:
-
Franky Lin authored
The nvram file contains info for firmware which varies with different hardware designs. Use more common firmware/nvram file names instead of those in Linux firmware repository to avoid misunderstanding. Signed-off-by:
-
Arend van Spriel authored
The function is only called with sk_buff parameter being non-zero so the prototype does not need to support passing a char buffer any longer. When the function is called with a NULL sk_buff parameter it returns -EINVAL now. Reviewed-by:
-
Arend van Spriel authored
The function brcmf_sdioh_request_buffer() was requiring a parameter in its prototype that was not used within the function. It has been removed consequently. Reviewed-by:
-
Arend van Spriel authored
In the receive path the buffer used to store the receive data from the device can be a chain of sk_buff. It has been separated to allow the use of skb queues. Reported-by:
Johannes Berg <johannes@sipsolutions.net> Reviewed-by:
-
Arend van Spriel authored
In two places the next pointer was used to process a sk_buff chain but it will always get a single sk_buff so this has been removed. Reviewed-by:
-
Arend van Spriel authored
In the receive path there was still code using the next pointer to access all packets in skb_queue. This patch fixes that. Reported-by:
-
Mohammed Shafi Shajakhan authored
Signed-off-by:
Mohammed Shafi Shajakhan <mohammed@qca.qualcomm.com> Signed-off-by:
-
Rajkumar Manoharan authored
The conformance test limits (CTL) for each regulatory domains (FCC/ETSI/MKK) are programmed for each runtime modes (11B,11G, HT20 and HT40) in EEPROM. The lowest ctledge power value of a particular running mode should not be used while computing ctledge power for a different running mode.(i.e 11G's min ctledge power should not be used while computing ctledge power for HT20). Currently, the code does not handle this properly which would result in incorrect txpowers in certain cases. So reset the twiceMaxEdgePower to the default while computing min ctlegepower for every mode. Cc: David Quan <dquan@qca.qualcomm.com> Signed-off-by:
Rajkumar Manoharan <rmanohar@qca.qualcomm.com> Signed-off-by:
-
Nikolay Martynov authored
When tx agg is being stopped TID is flushed using ath_tx_flush_tid. It is possible that ath_tx_flush_tid completelly flushes TID (if all packets in this TID have already been retried). If this happened ath_tx_aggr_stop would leave TID in cleanup state permanently. Fix this by making ath_tx_flush_tid remove AGGR_ADDBA_COMPLETE and AGGR_CLEANUP flags from TID status if TID is empty. Signed-off-by:
Nikolay Martynov <mar.kolya@gmail.com> Signed-off-by:
-
- 22 Nov, 2011 1 commit
-
-
John W. Linville authored
drivers/net/wireless/rtlwifi/usb.c: In function ‘_usb_read_sync’: drivers/net/wireless/rtlwifi/usb.c:102:6: warning: ‘status’ may be used uninitialized in this function drivers/net/wireless/rtlwifi/usb.c:102:6: note: ‘status’ was declared here My compiler is dumb, but better to eliminate the warning than to have anyone waste time evaluating this again... Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net>
-
- 21 Nov, 2011 12 commits
-
-
Christian Lamparter authored
In the past, it was fine to simply call dev_kfree_skb when it was impossible to transmit a skb. However, with the new tx status API: "mac80211: implement wifi TX status" Every loose skb needs to be handed back to mac80211. Signed-off-by:
Christian Lamparter <chunkeey@googlemail.com> Signed-off-by:
-
Christian Lamparter authored
In the past, it was fine to simply call dev_kfree_skb when it was impossible to transmit a skb. However, with the new tx status API: "mac80211: implement wifi TX status" Every loose skb needs to be handed back to mac80211. Signed-off-by:
-
Jussi Kivilinna authored
NET_TYPE_11FB actually has never been used. Signed-off-by:
Jussi Kivilinna <jussi.kivilinna@mbnet.fi> Signed-off-by:
-
Jussi Kivilinna authored
Convert old WEXT reporting to use new cfg80211_pmksa_candidate_notify(). Signed-off-by:
-
Jussi Kivilinna authored
Some structures were missing __packed. Signed-off-by:
-
Jussi Kivilinna authored
cfg80211_roamed() now has channel parameter so add passing current channel info. Signed-off-by:
-
Jussi Kivilinna authored
Split getting current channel channel from hardware to separate function as this function will be needed later in patch 'pass channel info to cfg80211_roamed()'. Signed-off-by:
-
Johannes Berg authored
Currently mac80211 implements these for all devices, but given restrictions of some devices that isn't really true, so prepare for being able to remove the capability for some mac80211 devices. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Acked-by:
Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by:
-
Joe Perches authored
Using dev_err instead of dev_printk(KERN_ERR uses fewer arguments and is a bit smaller. Deduplicating formats used by IWL_DEBUG_QUIET_RFKILL also makes the object a bit smaller. Neatened the macros, used ##__VA_ARGS__. $ size drivers/net/wireless/iwlwifi/built-in.o* text data bss dec hex filename 462652 8646 92576 563874 89aa2 drivers/net/wireless/iwlwifi/built-in.o.new 467557 8646 92592 568795 8addb drivers/net/wireless/iwlwifi/built-in.o.old Signed-off-by:
Joe Perches <joe@perches.com> Acked-by:
Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by:
-
Wey-Yi Guy authored
just use iwl_bus, remove the redundancy Signed-off-by:
-
Emmanuel Grumbach authored
Remove this redundancy. Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by:
-
Emmanuel Grumbach authored
Use the CPUed version of the variables when printing data from the BA notification. Signed-off-by:
-
