- 18 Jun, 2014 30 commits
-
-
Jes Sorensen authored
The driver shouldn't be dealing with public action management frames. Let these go directly to the cfg80211 layer which can defer them to userland if needed. Signed-off-by:
Jes Sorensen <Jes.Sorensen@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Jes Sorensen authored
The driver should not be issuing BSS Coexistence management action frames when scanning. This isn't required by the spec, and if needed, these should come from wpa_s instead. Recommended by Johannes Berg. Signed-off-by:
-
Jes Sorensen authored
This gets rid of the rtw_set_fixed_ie23a() usage here. Note that this abuses ieee80211_mgmt.u.action.u.chan_switch for now, since the correct defines are currently missing, and this one matches. Need to fix up the definitions in ieee80211.h and afterwards fix it up here. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
This reduces the dependency of rtw_set_fixed_ie23a() which is scheduled to go away. Signed-off-by:
-
Jes Sorensen authored
Be consistent in parsing it's value Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Match the return type of rtw_restructure_ht_ie23a() accordingly Signed-off-by:
-
Jes Sorensen authored
Only one is used, and only describes the size of an array defined in a local function. Use sizeof() in the local function to deobfuscate it. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
staging: rtl8723au: rtw_update_ht_cap23a(): Use correct mask when determining 40MHz channel location commit 4dc5f8ba Author: Jes Sorensen <Jes.Sorensen@redhat.com> Date: Tue May 27 13:43:12 2014 +0200 staging: rtl8723au: Eliminate struct HT_info_element Yet another duplicate version of struct ieee80211_ht_operation Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Only retrieve beacon_interval from frames where it is present. This avoids setting BeaconPeriod to garbage for other frame types. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
This sort of stuff should be handled by the bonding driver Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
These were just stored to check whether the beacon info had changed and print a debug message in this case. If we ever need this again, we can determine this change differently. Signed-off-by:
-
Jes Sorensen authored
Yet another duplicate version of struct ieee80211_ht_operation Signed-off-by:
-
Jes Sorensen authored
WLAN_EID_HT_CAPABILITY carries a struct ieee80211_ht_operation - use the right struct instead of a self invented one. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
The original RTL code defined the MCS information as a 16 byte array and blindly applied it's masking over the full array, ignoring the fact that only the first 10 bytes describe the rx_mask and the rest are different fields. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
- 17 Jun, 2014 10 commits
-
-
Wahib Faizi authored
Fix coding style issue "line over 80 characters" detected by checkpatch.pl in usbip_host_driver.c. Signed-off-by:
Wahib Faizi <wahibfaizi@gmail.com> Signed-off-by:
-
Wahib Faizi authored
Fix coding style issue "do not use assignment in if condition" detected by checkpatch.pl in usbip_host_driver.c. Signed-off-by:
-
Jes Sorensen authored
Add missing firmware references, and request the correct one. Reported-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Jes Sorensen authored
The filename requested by the driver didn't match what we had sitting in /lib/firmware/ Reported-by:
Ben Hutchings <ben@decadent.org.uk> Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Some of this is hackish, but so was the original code. This needs to be cleaned up eventually. Signed-off-by:
-
Jes Sorensen authored
Given that the IE len is stored in a u8, there is no point having a 512 buffer to store it, nor check if the size exceeds this size. Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
Jes Sorensen authored
Signed-off-by:
-
