- 05 Oct, 2011 7 commits
-
-
Franky Lin authored
rtnl lock is used improperly in fullmac. This patch intends to clean them up. Reported-by:
Johannes Berg <johannes@sipsolutions.net> Reviewed-by:
Arend van Spriel <arend@broadcom.com> Reviewed-by:
Roland Vossen <rvossen@broadcom.com> Reviewed-by:
Sukesh Srikakula <sukeshs@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieterpg@broadcom.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Roland Vossen authored
Fix double include as detected by checkincludes script. Signed-off-by:
-
Arend van Spriel authored
The kernel provides bit operations to replace brcmu_bitcount so using those instead. Reported-by:
-
Arend van Spriel authored
The functions brcmu_bprintf and brcmu_binit could be replaced by snprintf kernel function. These are no longer used in both brcm80211 drivers so they have been removed. Reported-by:
-
Arend van Spriel authored
In the receive path of the brcmsmac driver probe request frames and frames with invalid mac address fields are dropped. This is only fine for STA mode. Anticipating coming AP and monitor mode functionality this is unwanted behavior. Reported-by:
-
Arend van Spriel authored
The ssn should be given a valid value by the driver or it should not override the value. Removed override to zero from the driver. Reported-by:
-
Roland Vossen authored
Several fields were unused or were only set once. They have been removed. Uncalled functions that relied on these constants have been removed as well. Reviewed-by:
Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by:
-
- 04 Oct, 2011 33 commits
-
-
Mark Einon authored
The mechanism by which the device is put into low power sleep is broken in that the device can never come back out of low power mode afterwards. Temorary fix to bring the device back out of sleep almost immediately, until a suitable wake trigger can be found. Signed-off-by:
Mark Einon <mark.einon@gmail.com> Signed-off-by:
-
Mark Einon authored
Adding copyright notices and adding myself as a module author. Signed-off-by:
-
Mark Einon authored
set_mac is implemented, and there have been lots of new ethtool_ops added, so removing the TODO to add more kernel parameters. Signed-off-by:
-
Mark Einon authored
The error_timer was only getting initialised and an initial jiffies value set following a probe. This could result in the timer needlessly expiring immediately after et131x_open is called. Now this is all done from the open call instead. Signed-off-by:
-
Mark Einon authored
Added basic suspend & resume functionality. Tested on an et1310 device, and putting Fedora15 host in and out of hibernation successfully. Signed-off-by:
-
Mark Einon authored
adapter->bmsr is no longer being updated, but is also used to check the link state in places. Remove bmsr from adapter, and replace link state checks with phydev->link check. Signed-off-by:
-
Mark Einon authored
In several places in the code, the pm_csr register is read and the PHY_SW_COMA bit checked. Move this check into its own small function to make the code more readable. Signed-off-by:
-
Mark Einon authored
Removing this as I'm pretty sure its not true, and alloc_etherdev isn't provided by this driver anyway. Alternatively, its a badly written comment and I don't understand it. This drivers use of alloc_etherdev() is within keeping with other net devices, so I'm happy. Signed-off-by:
-
David Kilroy authored
The message is in the wrong place. We definitely don't want to return at this point, as we may need to turn off encryption. Reported-by:
Henk de Groot <pe1dnn@amsat.org> Signed-off-by:
David Kiroy <kilroyd@googlemail.com> Signed-off-by:
-
David Kilroy authored
The function no longer compiles, so we just remove it. Reported-by:
-
David Kilroy authored
in DBG machinery so you don't have to declare DBG_FUNC at the start of all functions. This just makes it easy to add DBG conforming to existing code. The patch reformats the changed #defines to satisfy checkstyle.pl Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of unnecessary comments. Signed-off-by:
K. Y. Srinivasan <kys@microsoft.com> Signed-off-by:
Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by:
-
K. Y. Srinivasan authored
Now that the mouse driver is functional, enable the auto-loading of the driver. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup mousevsc_on_device_add(). Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of mousevsc_on_receive_input_report() by inlining the code. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup all camel cased names. Signed-off-by:
-
K. Y. Srinivasan authored
Now get rid of the machinery for managing the life-cycle of the mousevsc_dev as this is not needed. Signed-off-by:
-
K. Y. Srinivasan authored
Now, cleanup mousevsc_remove(). The mouse driver once initialized only receives data from the host. So, by closing the channel first in the unload path, we can properly deal with inflight packets. So, we don't need the machinery for managing the life-cycle of the mousevsc_dev object. Get rid of the unnecessary code. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of mousevsc_on_device_remove() by inlining code. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup mousevsc_connect_to_vsp(). There is no need to take reference on the mousevsc device object when we are setting up the device. As part of this cleanup get rid of get_input_device() as this function is only used here. Signed-off-by:
-
K. Y. Srinivasan authored
We don't need to handle the "send complete" callback - nothing needs to be done here; get rid of the code. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup alloc_input_device(); you can directly set the reference count. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of the unused state: num_outstanding_req in struct mousevsc_dev. Signed-off-by:
-
K. Y. Srinivasan authored
Free all allocated memory in free_input_device(). Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of unnecessary pr_* calls. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup error handling in this driver; use standard Linux error codes. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of unnecessary DPRINT calls. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup and properly implement reportdesc_callback(); properly allocate the hid_device and properly initialize the hid device structure. Signed-off-by:
-
K. Y. Srinivasan authored
Use completion primitive to synchronize. Signed-off-by:
-
K. Y. Srinivasan authored
Correctly initialize the header size. Signed-off-by:
-
K. Y. Srinivasan authored
Handle the case where we may get bogus report desc size from the host. Signed-off-by:
-
K. Y. Srinivasan authored
Change the allocation flags to reflect interrupt context. Signed-off-by:
-
K. Y. Srinivasan authored
Fix the bogus WARN_ON() calls. Signed-off-by:
-
