Increasing robustness of the code, although no problem has been reported
in the field. Several code paths were unshielded for multi thread access.
Several lock acquisitions have been added to wl_mac80211.c
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Reviewed-by: Brett Rudley <brudley@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
A spinlock was acquired prior to calling the Mac80211 functions
ieee80211_wake_queues() and ieee80211_stop_queues() and Cfg80211 functions
wiphy_rfkill_set_hw_state() and wiphy_rfkill_start_polling().
This is not required and could even lead to instability. Therefore the locks
were removed.
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Reviewed-by: Brett Rudley <brudley@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Code cleanup. Added lock related comments to wl_mac80211.c. Also removed a
redundant function definition.
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Reviewed-by: Arend van Spriel <arend@broadcom.com>
Reviewed-by: Brett Rudley <brudley@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In the struct wlc_bsscfg a couple of attribute were held under a
preprocessor definition, but these are not needed in the mac80211
driver context.
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When associated on 5G the driver receives a probe request for 2G with
a 2G rate specified. The driver asserts as the operating band is still
5G when the probe request packet is given. Root cause was that ioctl
function did fail upon setting the channel as requested by mac80211
when we are associated.
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
- move ltoh16_buf/htol16_buf util/bcmsrom.c
- replace ltoh16_buf in brcmsmac/wlc_mac80211.c with several
le16_to_cpu's
Signed-off-by: Stanislav Fomichev <kernel@fomichev.me>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
With heavy transmit traffic, once in a while (range 15mins-1hr)
a tx packet was added to a full transmit queue. Under certain
conditions an other packet in the queue gets bumped to make room
for the new packet. This is not considered an error condition, but
normal operation. Despite that, there was an ASSERT(0) that caused
the driver to oops. The ASSERT(0) has been removed. Driver was tested
afterwards.
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Both zram and zcache use xvmalloc allocator. If xvmalloc
is compiled separately for both of them, we will get linker
error if they are both selected as "built-in". We can also
get linker error regarding missing xvmalloc symbols if zram
is not built.
So, we now compile xvmalloc separately and export its symbols
which are then used by both of zram and zcache.
Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Cancel the pending scan operation once the interface is going down to
avoid warnings from the cfg80211 module. Once the interface is down,
cfg80211 checks for any pending scan requests and dumps a warning if it
finds one. It expects the driver to abort any ongoing scan operation
once the driver detects that the interface is going down.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
We should check that optTxFrmCmd.optIEDataLen isn't too large before we
copy it into the data buffer.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Vipin Mehta <vipin.mehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The original code was written in a funny way where every statement was
part of else if blocks. I broke them up into separate statements by
adding breaks on failure conditions.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Vipin Mehta <vipin.mehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This updates the TODO file to reflect new changes on development.
Cc: Joe Perches <joe@perches.com>
Cc: Naveen Singh <nsingh@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lots of little ones all through the driver, mostly
all in a cut-and-paste header comment.
Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This is an initial staging driver for the GMA500. It's been stripped out
of the PVR drivers and crunched together from various bits of code and
different kernels.
Currently it's unaccelerated but still pretty snappy even compositing with
the frame buffer X server.
Lots of work is needed to rework the ttm and bo interfaces from being
ripped out and then 2D acceleration wants putting back for framebuffer and
somehow eventually via DRM.
There is no support for the parts without open source userspace (video
accelerators, 3D) as per kernel policy.
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In a previous commit, 7a09876d, ASSERT was changed to WARN_ON, but
the condition wasn't updated. This patch fixed this error.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
These fields have been assigned in netvsc_drv_init() before calling
netvsc_initialize(), so there is no need to check them.
The ASSERTs were already commented out, and this patch removes
them.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When jiffies wrap-over, all the BSS in the cache is removed. Wrap-over of
jiffies is not handled in the correct way. This cause the scan list to go
empty during this time for a small duration
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixing the code to avoid overwriting of the first index in the probed ssid
list maintained by the hardware. This index is used to store broadcast SSID.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Adding state in driver to track the sme state. The connect/disconnect
events from the driver were messing up the state maintained within the
cfg80211 module.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixing the cached copy of the BSS filter set by user.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The patch fixes a mismatch in the allocation and free of scatter HIF bus
requests in the suspend/resume path.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Adding host side interface to configure the excessive TX retry threshold.
It is used by the target to determine disconnection triggers. Additionally,
some definitions have been added to header file wmi.h to bridge the gap
for the newly added command.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Virtual Scatter Gather Lists not getting freed during the HTCStop(). The
patch adds some clean up code in the code path.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Move all the wmi configuration commands done after wmi_ready to a single
function.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixing the driver initialization for manufacturing mode that involves
downloading a firmware binary meant for RF tests on the factory floor.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Firmware design requires a WMI_DISCONNECT_CMD for every WMI_CONNECT_CMD to
clear the firmware previous profile state. There is one case in linux host
driver where two WMI_CONNECT_CMD are given without a WMI_DISCONNECT_CMD.
This causes firmware state to mismatch causing an ASSERT. Use the driver
state variable arConnectPending to track whether a WMI_CONNECT_CMD is
issued to firmware.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Return correct scan complete status to the cfg80211 module based on
the value returned from the hardware.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
A bug was observed during the reconnection phase in the WPA/WPA2-PSK scenario
where the EAPOL frames were going encrypted during an auto reconnection
attempt. The initial 4-way handshake would go fine but then the driver
was getting a command to set default keys sometime later. Setting of an
incorrect flag (TX_USAGE) in the hadrware was causing the EAPOL frames during
the subsequent 4-way handshake attempts to go encrypted causing the AP to
reject the station.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The driver was dereferencing a NULL pointer because of the device instance
being registered via the set_wiphy_dev() function. The function
ar6000_avail_ev() was passing the argument as NULL instead of using the one
returned by the MMC stack through the probe callback.
Signed-off-by: Vipin Mehta <vmehta@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently the device is initialized when first write is done to the
device. Any read attempt before the first write would fail, including
"hidden" read the user may not know about (as for example if he tries
to write a partial block).
This patch initializes the device on first request, whether read or
write.
Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
