- 27 Nov, 2012 2 commits
-
-
Eliad Peller authored
for some reason, the wl12xx fw is not able to rx/tx on the first start_sta cmd. Workaround it by issuing a dummy start_sta + stop_sta before starting the sta for the final time. Signed-off-by:
Eliad Peller <eliad@wizery.com> Reviewed-by:
Arik Nemtsov <arik@wizery.com> Signed-off-by:
Luciano Coelho <coelho@ti.com>
-
Eliad Peller authored
Make the connection flow simpler by starting sta role on bssid change. Currently, we start dev role when going idle-off, and start the sta role only after association indication. This complicates the connection flow with some possible intermediate states. Make it simpler by starting sta role on bssid change, which now happens *before* auth req get sent. Update the handling of mac80211's notifications and change wl1271_join/unjoin accordingly - * Split wl1271_join() into wlcore_join (tuning on a channel/bssid) and wlcore_set_assoc (configure sta after association). * Rename wl1271_unjoin() to wlcore_unset_assoc(), as it is no longer the inversion of wl1271_join() (now it's only used to disconnect associated sta / joined ibss, without stopping the role). * Set ssid before starting station role (needed for start_role(sta) While on it, split wl1271_bss_info_changed_sta() into some sub-functions. since we no longer use dev role in the connection flow, we now always use the hlid of the sta role. Signed-off-by:
-
- 16 Nov, 2012 7 commits
-
-
Tushar Behera authored
No need to check whether unsigned variable is less than 0. CC: Luciano Coelho <coelho@ti.com> CC: linux-wireless@vger.kernel.org CC: netdev@vger.kernel.org Signed-off-by:
Tushar Behera <tushar.behera@linaro.org> Signed-off-by:
Luciano Coelho <luca@coelho.fi>
-
Chuansheng Liu authored
The return value of wait_for_completion_timeout() is always >= 0 with unsigned int type. So the condition "ret < 0" or "ret >= 0" is pointless. Signed-off-by:
liu chuansheng <chuansheng.liu@intel.com> Signed-off-by:
-
Janusz.Dziedzic@tieto.com authored
In corner case for wl12xx_spi_raw_write() when len == SPI_AGGR_BUFFER_SIZE we don't setup correctly spi transfer_list. Next we will have garbage and strange errors reported by SPI framework (eg. wrong speed_hz, failed to transfer one message from queue) when iterate transfer_list. Signed-off-by:
Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by:
-
Kees Cook authored
This config item has not carried much meaning for a while now and is almost always enabled by default. As agreed during the Linux kernel summit, remove it. CC: Luciano Coelho <coelho@ti.com> CC: "John W. Linville" <linville@tuxdriver.com> Signed-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
Wei Yongjun authored
platform_device_unregister() only calls platform_device_del() and platform_device_put(), thus use platform_device_unregister() to simplify the code. Also the documents in platform.c shows that platform_device_del and platform_device_put must _only_ be externally called in error cases. All other usage is a bug. dpatch engine is used to auto generate this patch. (https://github.com/weiyj/dpatch) Signed-off-by:
Wei Yongjun <yongjun_wei@trendmicro.com.cn> Signed-off-by:
-
Wei Yongjun authored
platform_device_unregister() only calls platform_device_del() and platform_device_put(), thus use platform_device_unregister() to simplify the code. Also the documents in platform.c shows that platform_device_del and platform_device_put must _only_ be externally called in error cases. All other usage is a bug. dpatch engine is used to auto generate this patch. (https://github.com/weiyj/dpatch) Signed-off-by:
-
Julia Lawall authored
The function wl12xx_set_power_on is only called twice, once in wl12xx_chip_wakeup and once in wl12xx_get_hw_info. On the failure of the call in wl12xx_chip_wakeup, the containing function just returns, but on the failure of the call in wl12xx_get_hw_info, the containing function calls wl1271_power_off. This does not seem necessary, because if wl12xx_set_power_on has set the power on and then fails, it has already turned the power off. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r@ identifier f,free,a; parameter list[n] ps; type T; expression e; @@ f(ps,T a,...) { ... when any when != a = e if(...) { ... free(a); ... return ...; } ... when any } @@ identifier r.f,r.free; expression x,a; expression list[r.n] xs; @@ * x = f(xs,a,...); if (...) { ... free(a); ... return ...; } // </smpl> Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by:
-
- 14 Nov, 2012 31 commits
-
-
Nishant Sarmukadam authored
Set tx packet timestamp to 0 in following scenarios:- - All packets in STA mode - Mgmt packets in AP mode - Eapol packets in AP mode In STA mode, this field is unused in the firmware. In AP mode, we should not be expiring mgmt and eapol frames. Setting timestamp to 0 will ensure that. Signed-off-by:
Nishant Sarmukadam <nishants@marvell.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Nishant Sarmukadam authored
This can cause issues when clients try to connect when the traffic is heavy Signed-off-by:
-
Nishant Sarmukadam authored
We should unmap the DMA address in the error path, else it causes resource leaks. Fix this. Signed-off-by:
-
Yogesh Ashok Powar authored
We have 6.5 Mbps is minimum rate of the link as the criterion for creation of BA. Although we check this before creating the BA stream, by the time amdpu_action is called from the workqueue, the link can get affected in the meantime. Hence, add an additional check in amdpu_action. Signed-off-by:
Yogesh Ashok Powar <yogeshp@marvell.com> Signed-off-by:
-
Yogesh Ashok Powar authored
AP mode support upto 8 interfaces. Defining it using iface_combinations Signed-off-by:
-
Bing Zhao authored
ksdioirqd has higher priority than kworker. Process RX packets in SDIO IRQ thread (ksdioirqd/mmcX) directly instead of deferring the work to kworker to avoid the extra latency. This improves TCP throughput 15~20% on an ARM platform with SDIO 2.0 controller. Signed-off-by:
Bing Zhao <bzhao@marvell.com> Signed-off-by:
-
Arend van Spriel authored
Firmware events are passed to wl_cfg80211 module associated with the primary net device. With virtual interface support events can be received for different interfaces, ie. struct brcmf_if instances. Pass it in the event to determine appropriate net device associated with the event. 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:
Franky Lin <frankyl@broadcom.com> Signed-off-by:
-
Hante Meuleman authored
Add USB log level and update and add log messages in usb module. Reviewed-by:
-
Hante Meuleman authored
brcmf_netdev_stop shall first check bus_if status before bringing down cfg80211. brcmf_detach shall first check if driver is allocated. Reviewed-by:
-
Hante Meuleman authored
brcmf_netdev_start_xmit and brcmf_fil_cmd_data are checking bus state for down. These functions should check for data state. Reviewed-by:
-
Hante Meuleman authored
when brcmf_cfg80211_del_key was called with out of range key index then firmware would return error. Checking was added to brcmf_cfg80211_del_key to immediately return error. Reviewed-by:
-
Arend van Spriel authored
This fixes a build regression for x86_64 target that showed up in 3.7-rc1 due to: commit 1a873342 Author: Hante Meuleman <meuleman@broadcom.com> Date: Thu Sep 27 14:17:54 2012 +0200 brcmfmac: add hostap supoort. Reported-by:
Yuanhan Liu <yuanhan.liu@intel.com> Reported-by:
Geert Uytterhoeven <geert@linux-m68k.org> Reported-by:
Randy Dunlap <rdunlap@infradead.org> Reviewed-by:
-
Franky Lin authored
Use int instead of bool as the return type of function brcmf_sdio_hdparser to explicitly describe error returns. Signed-off-by:
-
Franky Lin authored
brcmf_sdbrcm_wait_for_event is now a one line function and only used by brcmf_sdbrcm_bus_txctl. Intergrate the function call wait_event_interruptible_timeout into brcmf_sdbrcm_bus_txctl. Signed-off-by:
-
Franky Lin authored
Semaphore sdsem is used to protect consecutive memory access through SDIO bus. Same functionality is provided by sdio_claim_host/ sdio_release_host interface as well. Replace sdsem with sdio_claim_host. Signed-off-by:
-
Franky Lin authored
Semaphore sdsem which protects consecutive SDIO bus access is used to lock down unnecessary wide range. Decrease the locking range provides the capability of parallel processing. Signed-off-by:
-
Franky Lin authored
Rxbuf in SDIO interface is used in normal frame and control frame read. Use dynamically allocated buffer in control frame read path for post processing to avoid conflicts. Signed-off-by:
-
Arend van Spriel authored
The function brcmf_fil_bsscfg_data_get() calls brcmf_create_iovar() with data pointer set to NULL, which caused a NULL pointer access. As it should be possible to provide data in message towards the firmware, it should just pass the data buffer instead. Reviewed-by:
-
Hante Meuleman authored
the refactored firmware interface layer made this function obsolete. Reviewed-by:
-
Arend van Spriel authored
The function brcmf_cfg80211_start_ap() had some variables declared that were not used or not needed any longer. This patch removes those variables. Reviewed-by:
-
Arend van Spriel authored
The function brcmf_set_management_ie() operates on virtual interface data and brcmf_cfg80211_vif structure provides all information needed for interfacing with firmware. As this function will also be needed for interface without an associated net device it makes sense to provide the vif instead of deriving it from the net device. Reviewed-by:
-
Hante Meuleman authored
sizeof calculation in setting pkt_filter was incorrect. This patch fixes that and removes related defines which have become obsolete. Reviewed-by:
-
Hante Meuleman authored
brcmf_netdev_wait_pend8021x was polling to see if 8021x data was already sent. Code was replaced using wait_event_timeout. Reviewed-by:
-
Hante Meuleman authored
On exception during brcmf_bus_start the netdev should be freed, if already allocated. Reviewed-by:
-
Hante Meuleman authored
Change setting up multicast, mac address and offloading to use the refactored firmware interface layer. Remove obsolete brcmf_proto_dcmd function. Reviewed-by:
-
Hante Meuleman authored
e-scan has become the default scanning method. This patch removes the i-scan related code and cleans up e-scan related code to be always enabled. Reviewed-by:
-
Arend van Spriel authored
The vendor ie buffer passed to brcmf_set_management_ie() is not modified and the caller always provided a constant buffer, which needed a cast. Better making the buffer parameter of the function constant so the casts can be removed. Reviewed-by:
-
Arend van Spriel authored
The function brcmf_inform_bss() validates the version received from the device, before processing the individual bss entries. This error condition is only applicable when the list contains entries. A specific scan, ie. for a specific ssid, can result in a scan completion without finding any bss entries. No need to flag it as an error in the log when this happens. Reviewed-by:
-
Arend van Spriel authored
Code flow was: err = foo(); if (!err) return err; else goto exit; return 0; Changed it to just to exit label if err is non-zero. Reviewed-by:
-
Arend van Spriel authored
The data stored in ap_info structure is no longer used so remove it from the driver. Signed-off-by:
-
Pontus Fuchs authored
A missing else caused a potential NULL dereference. Reported-by:
Yuanhan Liu <yuanhan.liu@linux.intel.com> Signed-off-by:
Pontus Fuchs <pontus.fuchs@gmail.com> Signed-off-by:
-
