- 06 Jan, 2014 14 commits
-
-
Arend van Spriel authored
Instead of calling brcmf_cfg80211_detach() in brcmf_del_if() when deleting the primary interface, call it in brcmf_detach() after deleting all interfaces. Reviewed-by:
Hante Meuleman <meuleman@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
Arend van Spriel <arend@broadcom.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Arend van Spriel authored
The wiphy_unregister() call was done in brcmf_free_vif() when the last interface was being removed. This is not the obvious place to do that. This patch moves it to the brcmf_cfg80211_detach(). This removes the need to keep count of interfaces. Reviewed-by:
-
Arend van Spriel authored
Upon unload of the brcmfmac driver it gave a kernel warning because cfg80211 still believed to be connected to an AP. The brcmfmac had already transitioned to disconnected state during unload. This patch adds informing cfg80211 about this transition. This will get rid of warning from cfg80211 seen upon module unload: ------------[ cut here ]------------ WARNING: CPU: 3 PID: 24303 at net/wireless/core.c:952 cfg80211_netdev_notifier_call+0x193/0x640 [cfg80211]() Modules linked in: brcmfmac(O-) brcmutil(O) cfg80211(O) ... [last unloaded: bcma] CPU: 3 PID: 24303 Comm: rmmod Tainted: G W O 3.13.0-rc4-wl-testing-x64-00002-gb472b6d-dirty #1 Hardware name: Dell Inc. Latitude E6410/07XJP9, BIOS A07 02/15/2011 00000000000003b8 ffff8800b211faf8 ffffffff815a7fcd 0000000000000007 0000000000000000 ffff8800b211fb38 ffffffff8104819c ffff880000000000 ffff8800c889d008 ffff8800b2000220 ffff8800c889a000 ffff8800c889d018 Call Trace: [<ffffffff815a7fcd>] dump_stack+0x46/0x58 [<ffffffff8104819c>] warn_slowpath_common+0x8c/0xc0 [<ffffffff810481ea>] warn_slowpath_null+0x1a/0x20 [<ffffffffa173fd83>] cfg80211_netdev_notifier_call+0x193/0x640 [cfg80211] [<ffffffff81521ca8>] ? arp_ifdown+0x18/0x20 [<ffffffff8152d75a>] ? fib_disable_ip+0x3a/0x50 [<ffffffff815b143d>] notifier_call_chain+0x4d/0x70 [<ffffffff8106d6e6>] raw_notifier_call_chain+0x16/0x20 [<ffffffff814b9ae0>] call_netdevice_notifiers_info+0x40/0x70 [<ffffffff814b9b26>] call_netdevice_notifiers+0x16/0x20 [<ffffffff814bb59d>] rollback_registered_many+0x17d/0x280 [<ffffffff814bb74d>] rollback_registered+0x2d/0x40 [<ffffffff814bb7c8>] unregister_netdevice_queue+0x68/0xd0 [<ffffffff814bb9c0>] unregister_netdev+0x20/0x30 [<ffffffffa180069e>] brcmf_del_if+0xce/0x180 [brcmfmac] [<ffffffffa1800b3c>] brcmf_detach+0x6c/0xe0 [brcmfmac] Reviewed-by:
-
Arend van Spriel authored
The condition to disable the clock at the end of brcmf_sdio_bus_init() was wrong as the bus state is updated by the calling function. Hence, the clock was always disabled after brcmf_sdio_bus_init() which was not the intended behaviour. Reviewed-by:
Franky Lin <frankyl@broadcom.com> Reviewed-by:
-
Arend van Spriel authored
Change condition in brcmf_sdio_wd_timer() function to program watchdog only when in BRCMF_BUS_DATA state. This avoids watchdog being active during initialization. During initialization the SDIO save&restore capability is determined which affect the bus sleep mechanism used in watchdog thread. Reviewed-by:
-
Arend van Spriel authored
The detection of the save&restore capability in brcmf_sdio_sr_capable() is only valid for certain chipsets. This patch should cover it for all chipsets currently supported. Reviewed-by:
-
Arend van Spriel authored
The table for BCM4334 SDIO drive strength programming was missing from the driver. Adding it with this patch set. Reviewed-by:
-
Arend van Spriel authored
Using 'iw phy' only showed HT20 support in the HT capabilities info. This patch determines support for HT40 using a firmware query that is supposed to work for all supported devices. Reviewed-by:
-
Arend van Spriel authored
Moving code from helper functions to the calling function as it makes code easier to read. Reviewed-by:
-
Arend van Spriel authored
Two helper functions in the sdio remove path were very thin and only used once. So its code is moved to the calling function. Reviewed-by:
-
Hante Meuleman authored
An control request or set message length is restricted to ETH frame length for the buffer from host to device. This is limitation is imposed by the firmware. Reviewed-by:
-
Hante Meuleman authored
Reviewed-by:
-
Arend van Spriel authored
The logic in the SDIO register access functions was hard to read and contained a lot of conditional code path. This rework attempts to clean it up. Reviewed-by:
-
-
- 03 Jan, 2014 26 commits
-
-
Paul Gortmaker authored
None of these files are actually using any __init type directives and hence don't need to include <linux/init.h>. Most are just a left over from __devinit and __cpuinit removal, or simply due to code getting copied from one driver to the next. Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Acked-by:
Larry Finger <Larry.Finger@lwfinger.net> Acked-by:
Christian Lamparter <chunkeey@googlemail.com> Acked-by:
Gertjan van Wingerde <gwingerde@gmail.com> Signed-off-by:
-
Hauke Mehrtens authored
CHECK drivers/ssb/driver_chipcommon_sflash.c drivers/ssb/driver_chipcommon_sflash.c:40:11: warning: Using plain integer as NULL pointer drivers/ssb/driver_chipcommon_sflash.c:58:11: warning: Using plain integer as NULL pointer drivers/ssb/driver_chipcommon_sflash.c:69:11: warning: Using plain integer as NULL pointer Signed-off-by:
Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by:
-
Hauke Mehrtens authored
CHECK drivers/bcma/driver_chipcommon_sflash.c drivers/bcma/driver_chipcommon_sflash.c:41:11: warning: Using plain integer as NULL pointer drivers/bcma/driver_chipcommon_sflash.c:59:11: warning: Using plain integer as NULL pointer drivers/bcma/driver_chipcommon_sflash.c:70:11: warning: Using plain integer as NULL pointer Signed-off-by:
-
Sujith Manoharan authored
The maximum permissible values for noise floor in ETSI and Japan domains have been updated with new EU regulations for CCA detection. Adjust the values in the driver accordingly. Signed-off-by:
Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by:
-
Sujith Manoharan authored
AR955x has problems with RX sensitivity in 2G. This patch adds a routine to select range_osdac dynamically on a per-chain basis to address this issue. Signed-off-by:
-
Christian Engelmayer authored
Fix a memory leak in the cw1200_wow_suspend() error handling path. Signed-off-by:
Christian Engelmayer <cengelma@gmx.at> Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h, ieee80211_bar defined in include/linux/ieee80211.h and ath_common defined in drivers/net/wireless/ath/ath.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> Acked-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: at76_priv defined in drivers/net/wireless/at76c50x-usb.h and ieee80211_mgmt defined in include/linux/ieee80211.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structure involved is libipw_network defined in drivers/net/wireless/ipw2x00/libipw.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h ath_common defined in drivers/net/wireless/ath/ath.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h, il_priv defined in drivers/net/wireless/iwlegacy/common.h and il_rxon_cmd defined in drivers/net/wireless/iwlegacy/commands.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h and rtl_mac defined in drivers/net/wireless/rtlwifi/wifi.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h and mwl8k_priv defined in drivers/net/wireless/mwl8k.c This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_hdr defined in include/linux/ieee80211.h and p54_common defined in drivers/net/wireless/p54/p54.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_mgmt defined in include/linux/ieee80211.h and ath_common defined in drivers/net/wireless/ath/ath.h This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Julia Lawall authored
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and can be used when each argument is an array within a structure that contains at least two bytes of data beyond the array. The structures involved are: ieee80211_bar defined in include/linux/ieee80211.h and rt2x00_bar_list_entry defined in drivers/net/wireless/rt2x00/rt2x00.h. This was done using Coccinelle (http://coccinelle.lip6.fr/). Signed-off-by:
-
Sujith Manoharan authored
Initialize "sync_cause" to zero since commit "ath9k: move ath9k_debug_sync_cause out of ath9k_hw" fills it conditionally based on ISR status. Not doing this results in garbage values in debugfs. Cc: Felix Fietkau <nbd@openwrt.org> Signed-off-by:
-
Sujith Manoharan authored
Signed-off-by:
-
Sujith Manoharan authored
AR9100 requires a larger delay after waking up the RTC. Signed-off-by:
-
Sujith Manoharan authored
* Move initialization of config variables to ath9k_hw_init_config(). * Move initialization of defaults to ath9k_hw_init_defaults(). Signed-off-by:
-
Sujith Manoharan authored
There is no point in trying to bring up the chip when the MAC version is not present in the supported list. Signed-off-by:
-
