- 08 Feb, 2013 40 commits
-
-
Hante Meuleman authored
Use the stored role in wdev struct to determine AP/GO role. This simplifies code. Fix stop_ap for GO role. Reviewed-by:
Arend Van Spriel <arend@broadcom.com> Signed-off-by:
Hante Meuleman <meuleman@broadcom.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Hante Meuleman authored
Add module variable with which a p2p0 netdev can be created. This netdev can be used by wpa-supplicant to configure and set up the p2p client/GO. Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
-
Arend van Spriel authored
cfg80211 api change needs to be applied. this change will be squashed with introducing commit: commit 68751daec95b07a74b18aa22b2f9c196991dbe81 Author: Hante Meuleman <meuleman@broadcom.com> Date: Mon Nov 26 12:21:27 2012 +0100 brcmfmac: add support for P2P listen mode. Signed-off-by: -
Hante Meuleman authored
In some rare situations the wpa_supplicant can lock up on a remain on channel command. Use actual cookies for the remain on channel related commands and allow for additional remain on channel while still having one set. Reviewed-by:
-
Hante Meuleman authored
With this patch it is possible to send action frames. Reviewed-by:
-
Hante Meuleman authored
Once wpa_supplicant has registered for P2P action frames all received action frames for the device are passed up to cfg80211. Reviewed-by:
-
Hante Meuleman authored
With this update it is possible to create an P2P go. Handle the p2p go role in creation and the update beacon from cfg80211. Also store primary bss in global struct. Needed to map cfg device back to primary device. Reviewed-by:
-
Hante Meuleman authored
In preparation of p2p connections update the sta connect setup and handling routines. Use join iovar and bsscfg commands where applicable. Handle assoc ind per bsscfg (vif) and generalize vndr ie setup to support assoc IEs and support all setup types, e.g. sta, ap and p2p. Reviewed-by:
-
Hante Meuleman authored
Mask is OR-ed with value in case of deregister. Should be AND-ed. Reviewed-by:
-
Hante Meuleman authored
Add support for changing existing interface into p2p go interface. Reviewed-by:
-
Hante Meuleman authored
Improve exception handling. Store and removed created vif in cfg.p2p.bss array. Fix big endian bug. Fix msec jiffies bug. Reviewed-by:
-
Hante Meuleman authored
Add support for scanning on non primary netdev (p2p) and simplify the p2p scanning preparation. Reviewed-by:
-
Arend van Spriel authored
Deletion of the virtual interface was not fully supported so adding it now. Tested using p2p_connect command in wpa_supplicant. Reviewed-by:
-
Arend van Spriel authored
The driver determines the P2P interface address and device address from the primary mac address. However, the function used an invalid input. This caused a mac address mismatch between firmware and driver. This was hidden because the brcmf_if_add() function corrected the error upon receiving the IF_ADD event from firmware. Reviewed-by:
-
Arend van Spriel authored
On 64-bit target a warning was given when using %d to print sizeof() value: CC [M] drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.o brcm80211/brcmfmac/wl_cfg80211.c: In function 'brcmf_alloc_vif': brcm80211/brcmfmac/wl_cfg80211.c:3884:2: error: format '%d' expects argument of type 'int', but argument 3 has type 'long unsigned int' [-Werror=format] Fixed using %zu instead. Reviewed-by:
-
Hante Meuleman authored
With this patch a device can be put in p2p listen mode and becomes visible for other p2p devices (via p2p_find). Reviewed-by:
-
Arend van Spriel authored
This patch allow the creation of P2P client and group owner virtual interfaces in the driver. Reviewed-by:
Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by:
-
Arend van Spriel authored
First part for P2P support. It is a variation on the existing scanning functionality and is used to discover peer-to-peer group owners and/or peer-to-peer devices in listen state. Tested with wpa_cli/wpa_supplicant v2.0-devel. Reviewed-by:
-
Piotr Haber authored
Locking host access in trap info processing is not needed as bus access functions do it on their own. Reviewed-by:
Piotr Haber <phaber@broadcom.com> Signed-off-by:
-
Piotr Haber authored
Make sure backplane clocks are on while reading crash data. Reviewed-by:
-
Piotr Haber authored
fix wrong locking in crash info processing Reviewed-by:
-
Hante Meuleman authored
When unloading it is possible that drvr is not null, but iflist[0] is. So check iflist[0] pointer before using it. Reviewed-by:
-
Hante Meuleman authored
Upon receiving an IF event from the firmware the interface was created and stored on a list using the interface index. With upcoming P2P feature the firmware will send a IF event in which two interfaces have the same interface index. To uniquely locate them on the list the bss index is now used. Reviewed-by:
-
Joe Perches authored
alloc failures already get standardized OOM messages and a dump_stack. Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Tim Gardner authored
Dynamically allocate the probe response template which avoids potential stack corruption. Observed with smatch: drivers/net/wireless/brcm80211/brcmsmac/main.c:7412 brcms_c_bss_update_probe_resp() warn: 'prb_resp' puts 512 bytes on stack Cc: Brett Rudley <brudley@broadcom.com> Cc: Arend van Spriel <arend@broadcom.com> Cc: "Franky (Zhenhui) Lin" <frankyl@broadcom.com> Cc: Hante Meuleman <meuleman@broadcom.com> Cc: "John W. Linville" <linville@tuxdriver.com> Cc: Seth Forshee <seth.forshee@canonical.com> Cc: Pieter-Paul Giesberts <pieterpg@broadcom.com> Cc: Hauke Mehrtens <hauke@hauke-m.de> Cc: linux-wireless@vger.kernel.org Cc: brcm80211-dev-list@broadcom.com Cc: netdev@vger.kernel.org Signed-off-by:
Tim Gardner <tim.gardner@canonical.com> Acked-by:
-
Tim Gardner authored
DOT11_MIN_BEACON_PERIOD and DOT11_MAX_BEACON_PERIOD are superfluous. The only invalid beacon period is 0. Comparing a 16 bit quantity to 0xffff also causes a compile warning: drivers/net/wireless/brcm80211/brcmsmac/main.c:5560 brcms_c_set_beacon_period() warn: impossible condition '(period > 65535) => (0-65535 > 65535)' Observed from smatch analysis. Cc: Brett Rudley <brudley@broadcom.com> Cc: Arend van Spriel <arend@broadcom.com> Cc: "Franky (Zhenhui) Lin" <frankyl@broadcom.com> Cc: Hante Meuleman <meuleman@broadcom.com> Cc: "John W. Linville" <linville@tuxdriver.com> Cc: Pieter-Paul Giesberts <pieterpg@broadcom.com> Cc: Seth Forshee <seth.forshee@canonical.com> Cc: Hauke Mehrtens <hauke@hauke-m.de> Cc: linux-wireless@vger.kernel.org Cc: brcm80211-dev-list@broadcom.com Cc: netdev@vger.kernel.org Signed-off-by:
-
Jiri Slaby authored
It can, if invalid argument given, return a negative value. In that case we would access arrays out-of-bounds and such. Check the value and yell loudly if that happened as it would be a bug in the implementation. (Instead of silently corrupting memory.) Signed-off-by:
Jiri Slaby <jslaby@suse.cz> Cc: Nick Kossifidis <mickflemm@gmail.com> Cc: "Luis R. Rodriguez" <mcgrof@qca.qualcomm.com> Signed-off-by:
-
Larry Finger authored
When the new_id entry in /sysfs is used for a foreign USB device, rtlwifi BUGS with a NULL pointer dereference because the per-driver configuration data is not available. The probe function has been restructured as suggested by Ben Hutchings <bhutchings@solarflare.com>. Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Cc: Stable <stable@vger.kernel.org> Signed-off-by:
-
Arend van Spriel authored
The bcm43236 device showed having an issue connection to an AP without security when auth_type was set to automatic. For this particular device a workaround is made by selecting open-system. This problem does not occur when using wpa_supplicant as it will not use auth type automatic. Reviewed-by:
-
Arend van Spriel authored
Added function to obtain chip information for chip specific code. Reviewed-by:
-
Arend van Spriel authored
Extend the bus interface with chip identifier and revision. This will be used when certain devices need special handling in the driver. Reviewed-by:
-
Hante Meuleman authored
Simplify code by using single function for converting channel to chanspec. No need to set 40MHz flags, as they are not needed for correct working in FW. Reviewed-by:
-
Hante Meuleman authored
Clean up function, removing some unnecessary code and initialisation. Reviewed-by:
-
Hante Meuleman authored
When supporting multiple netdevs it is usefull to log idx in netdev related functions. This will be used for features like P2P. Reviewed-by:
-
Hante Meuleman authored
Statistics were tracked by bus driver while it is to be tracked per ifp/netdev. Reviewed-by:
-
Hante Meuleman authored
Reviewed-by:
-
Hante Meuleman authored
Pending 8021x frames were tracked per dongle. This should be done per ifp. Reviewed-by:
-
Hante Meuleman authored
Reviewed-by:
-
Hante Meuleman authored
At netdev up the mac address from drvr is copied to netdev mac address. This is incorrect and mac address was already set by brcmf_net_attach by copying it from ifp. Reviewed-by:
-
Hante Meuleman authored
The tracking of up/down status in bus interface is unnecessary. Reviewed-by:
-
