firefly-linux-kernel-4.4.55.git
10 years agowil6210: writeable ITR registers
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:09 +0000 (19:37 +0300)]
wil6210: writeable ITR registers

Interrupt threshold registers may be written to.

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: debugfs interface to send raw WMI command
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:08 +0000 (19:37 +0300)]
wil6210: debugfs interface to send raw WMI command

Debug aid

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: BACK: track last dropped SSN
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:07 +0000 (19:37 +0300)]
wil6210: BACK: track last dropped SSN

Track and print on debugfs

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: Allow driver load if FW not ready
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:06 +0000 (19:37 +0300)]
wil6210: Allow driver load if FW not ready

Usable for debugging, to be able to obtain FW traces

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: Tx performance monitoring
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:05 +0000 (19:37 +0300)]
wil6210: Tx performance monitoring

For performance monitoring, trace time intervals when Tx vring
is idle/not idle. Use CPU cycle counter for this, because jiffies is
too rough, and other precise time measurement methods involve
overhead while get_cycles() should be fast.
This used to provide some estimation for percentage when Tx vring
was idle, i.e. when hardware is under-utilized.
Estimation is not precise because of many reasons - CPU frequency scaling,
grt_cycles() may be per core etc. But still, it is good estimation

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: more debug info for vring
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:04 +0000 (19:37 +0300)]
wil6210: more debug info for vring

print used/available counters on debugfs;
print to dmesg when Tx vring becomes empty

This aids with performance investigation

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: print more info about BSS found
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:03 +0000 (19:37 +0300)]
wil6210: print more info about BSS found

print essential info to dmesg

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: trace wil->status on debugfs
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:02 +0000 (19:37 +0300)]
wil6210: trace wil->status on debugfs

For debug purposes

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: print debug info when starting AP
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:01 +0000 (19:37 +0300)]
wil6210: print debug info when starting AP

In the wil_cfg80211_start_ap(), debug print selected data:
- beacon (before and after fix-up)
- crypto parameters
- mark start/stop AP invocation

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: indicate mgmt_tx status
Vladimir Kondratiev [Mon, 16 Jun 2014 16:37:00 +0000 (19:37 +0300)]
wil6210: indicate mgmt_tx status

Inform cfg80211 about Tx result

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agowil6210: Tx mgmt frame from debugfs
Vladimir Kondratiev [Mon, 16 Jun 2014 16:36:59 +0000 (19:36 +0300)]
wil6210: Tx mgmt frame from debugfs

Provide 2 files on the debugfs:
- "rxon": write channel (1..4) to open Rx on it, 0 to rxoff
- "tx_mgmt": write binary frame, starting from MAC header

one need to care about turning receiver on/off before/after tx_mgmt

Correct sequence is:
 echo $channel > rxon
 cat mfmt_frame > tx_mgmt
 echo 0 > rxon

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changes for 40MHz
Jahnavi Meher [Mon, 16 Jun 2014 14:17:03 +0000 (19:47 +0530)]
rsi: Changes for 40MHz

Added code required for 40MHz.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Adding support for host based bgscan.
Jahnavi Meher [Mon, 16 Jun 2014 14:16:48 +0000 (19:46 +0530)]
rsi: Adding support for host based bgscan.

Added support for host based bgscan. The h/w queues are blocked
while bgscan is being performed and after coming to the connected
channel, the queues are unblocked.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Adding support for 5GHz
Jahnavi Meher [Mon, 16 Jun 2014 14:16:31 +0000 (19:46 +0530)]
rsi: Adding support for 5GHz

Adding support for 5GHz.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Sending QoS null packet via the mgmt queue.
Jahnavi Meher [Mon, 16 Jun 2014 14:16:13 +0000 (19:46 +0530)]
rsi: Sending QoS null packet via the mgmt queue.

Send the QoS null packet via mgmt queue.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Added debug messages.
Jahnavi Meher [Mon, 16 Jun 2014 14:15:45 +0000 (19:45 +0530)]
rsi: Added debug messages.

Added some debug messages.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changed the logic of dequeuing packets from hal queues.
Jahnavi Meher [Mon, 16 Jun 2014 14:15:03 +0000 (19:45 +0530)]
rsi: Changed the logic of dequeuing packets from hal queues.

The number of packets being dequeued from s/w queues was fixed -
changed it to a dynamic calculation based on txop. There are also
some fixes to the dequeuing algorithm.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changed the SDIO interrupt variables and some clean up.
Jahnavi Meher [Mon, 16 Jun 2014 14:14:12 +0000 (19:44 +0530)]
rsi: Changed the SDIO interrupt variables and some clean up.

Changed the SDIO interrupt variables and some clean ups.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Use SGI if configured for fixed rate transmission.
Jahnavi Meher [Mon, 16 Jun 2014 14:13:54 +0000 (19:43 +0530)]
rsi: Use SGI if configured for fixed rate transmission.

Use SGI if configured while sending data packets at a fixed rate.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Lower level debug messages and changed handling of confirm received for rsi_prog...
Jahnavi Meher [Mon, 16 Jun 2014 14:13:35 +0000 (19:43 +0530)]
rsi: Lower level debug messages and changed handling of confirm received for rsi_program_bb_rf().

Lower level debug messages for some frames and changed confirm
received for rsi_program_bb_rf() to a valid case.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changed rate handling.
Jahnavi Meher [Mon, 16 Jun 2014 14:13:09 +0000 (19:43 +0530)]
rsi: Changed rate handling.

Changed rate handling.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changed the rsi_set_channel() and rsi_program_bb_rf().
Jahnavi Meher [Mon, 16 Jun 2014 14:11:58 +0000 (19:41 +0530)]
rsi: Changed the rsi_set_channel() and rsi_program_bb_rf().

Made required changes to rsi_set_channel() and rsi_program_bb_rf()
functions.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Changed the radio caps frame.
Jahnavi Meher [Mon, 16 Jun 2014 14:11:41 +0000 (19:41 +0530)]
rsi: Changed the radio caps frame.

Changed the radio caps frame and added the required fields.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Add macros for endpoints and set default value of endpoint.
Jahnavi Meher [Mon, 16 Jun 2014 14:11:22 +0000 (19:41 +0530)]
rsi: Add macros for endpoints and set default value of endpoint.

Added macros for the endpoints and set the default value of endpoint
to 2.4GHz and 20MHz.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Using band from rsi_common to fill in ieee80211_rx_status
Jahnavi Meher [Mon, 16 Jun 2014 14:10:55 +0000 (19:40 +0530)]
rsi: Using band from rsi_common to fill in ieee80211_rx_status

Filling in band from common->band.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Fixed the kernel doc
Jahnavi Meher [Mon, 16 Jun 2014 14:10:02 +0000 (19:40 +0530)]
rsi: Fixed the kernel doc

Changed the function header to match the function name.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Mapping the debugfs stats to the correct s/w queues.
Jahnavi Meher [Mon, 16 Jun 2014 14:09:40 +0000 (19:39 +0530)]
rsi: Mapping the debugfs stats to the correct s/w queues.

Changed the queue numbers to macros, and corrected the mappings.

Signed-off-by: Jahnavi Meher <jahnavi.meher@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: Use dma_zalloc_coherent
Joe Perches [Sun, 15 Jun 2014 20:37:43 +0000 (13:37 -0700)]
rt2x00: Use dma_zalloc_coherent

Use the zeroing function instead of dma_alloc_coherent & memset(,0,)

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Advertise multichannel support
Rajkumar Manoharan [Wed, 11 Jun 2014 10:48:16 +0000 (16:18 +0530)]
ath9k: Advertise multichannel support

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: use separate HW queue for each channel context
Rajkumar Manoharan [Wed, 11 Jun 2014 10:48:15 +0000 (16:18 +0530)]
ath9k: use separate HW queue for each channel context

Use seperate tx queue for each AC in each channel context and expose
these information to mac80211 to avoid stopping one channel context
leads to stopping the entire traffic for that AC even on other contexts.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add recovery mechanism for hw TSF timer
Felix Fietkau [Wed, 11 Jun 2014 10:48:14 +0000 (16:18 +0530)]
ath9k: Add recovery mechanism for hw TSF timer

Configure the TSF based hardware timer for a channel switch.
Also set up backup software timer, in case the gen timer fails.
This could be caused by a hardware reset.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Update channel switch timer
Felix Fietkau [Wed, 11 Jun 2014 10:48:13 +0000 (16:18 +0530)]
ath9k: Update channel switch timer

TSF time might have been updated by the incoming beacon,
need update the channel switch timer to reflect the change.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Handle beacon miss on multi channel context
Felix Fietkau [Wed, 11 Jun 2014 10:48:12 +0000 (16:18 +0530)]
ath9k: Handle beacon miss on multi channel context

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add periodic NoA support
Felix Fietkau [Wed, 11 Jun 2014 10:48:11 +0000 (16:18 +0530)]
ath9k: Add periodic NoA support

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add multi-channel scheduling support
Felix Fietkau [Wed, 11 Jun 2014 10:48:10 +0000 (16:18 +0530)]
ath9k: Add multi-channel scheduling support

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Implement mgd_prepare_tx
Felix Fietkau [Wed, 11 Jun 2014 10:48:09 +0000 (16:18 +0530)]
ath9k: Implement mgd_prepare_tx

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Adjust AP beacon tsf based on station context
Felix Fietkau [Wed, 11 Jun 2014 10:48:08 +0000 (16:18 +0530)]
ath9k: Adjust AP beacon tsf based on station context

In multi channel context (AP + STA case), adjust the TSF time of
the AP chanctx to keep its beacons at half beacon interval offset
relative to the STA chanctx.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: switch channel after sending beacon
Felix Fietkau [Wed, 11 Jun 2014 10:48:07 +0000 (16:18 +0530)]
ath9k: switch channel after sending beacon

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add p2p go NoA attribute
Felix Fietkau [Wed, 11 Jun 2014 10:48:06 +0000 (16:18 +0530)]
ath9k: Add p2p go NoA attribute

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Store current offchannel duration
Rajkumar Manoharan [Wed, 11 Jun 2014 10:48:05 +0000 (16:18 +0530)]
ath9k: Store current offchannel duration

Update current offchannel duration during scan or roc request.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: switch channel context for beaconing
Felix Fietkau [Wed, 11 Jun 2014 10:48:04 +0000 (16:18 +0530)]
ath9k: switch channel context for beaconing

Add a basic state machine for switch channel context
for beacon transmission.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Summarize hw state per channel context
Rajkumar Manoharan [Wed, 11 Jun 2014 10:48:03 +0000 (16:18 +0530)]
ath9k: Summarize hw state per channel context

Group and set hw state (opmode, primary_sta, beacon conf) per
channel context instead of whole list of vifs. This would allow
each channel context to run in different mode (STA/AP).

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Move beacon config to channel context
Rajkumar Manoharan [Wed, 11 Jun 2014 10:48:02 +0000 (16:18 +0530)]
ath9k: Move beacon config to channel context

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: save tsf in channel context
Felix Fietkau [Wed, 11 Jun 2014 10:48:01 +0000 (16:18 +0530)]
ath9k: save tsf in channel context

Save TSF in channel context for multiple operating channels.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add ATH_OP_MULTI_CHANNEL
Felix Fietkau [Wed, 11 Jun 2014 10:48:00 +0000 (16:18 +0530)]
ath9k: Add ATH_OP_MULTI_CHANNEL

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Move caldata into channel context
Felix Fietkau [Wed, 11 Jun 2014 10:47:59 +0000 (16:17 +0530)]
ath9k: Move caldata into channel context

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Fetch appropriate operating channel context
Rajkumar Manoharan [Wed, 11 Jun 2014 10:47:58 +0000 (16:17 +0530)]
ath9k: Fetch appropriate operating channel context

Retrieve appropriate operating channel context while switching
between operating and off channels.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Implement channel context ops
Felix Fietkau [Wed, 11 Jun 2014 10:47:57 +0000 (16:17 +0530)]
ath9k: Implement channel context ops

Add channel context operations (add, remove, change, assign and
unassign) to enable support for multiple channels.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Implement remain-on-channal support
Felix Fietkau [Wed, 11 Jun 2014 10:47:56 +0000 (16:17 +0530)]
ath9k: Implement remain-on-channal support

Add remain on channel support in order to enable multi-channel
concurrency.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Implement hw_scan support
Felix Fietkau [Wed, 11 Jun 2014 10:47:55 +0000 (16:17 +0530)]
ath9k: Implement hw_scan support

Implement hw_scan support for enabling multi-channel cuncurrency.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: send powersave frame on channel switch
Felix Fietkau [Wed, 11 Jun 2014 10:47:54 +0000 (16:17 +0530)]
ath9k: send powersave frame on channel switch

While leaving from or entering to active channel context, send out
nullfunc frame to inform to the AP about the presence of station.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: channel context based transmission
Felix Fietkau [Wed, 11 Jun 2014 10:47:53 +0000 (16:17 +0530)]
ath9k: channel context based transmission

Force queueing of all frames that belong to a virtual interface on
a different channel context, to ensure that they are sent on the
correct channel.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add channel context worker thread
Felix Fietkau [Wed, 11 Jun 2014 10:47:52 +0000 (16:17 +0530)]
ath9k: Add channel context worker thread

The channel context worker is used to switch to next requested
channel context.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Move acq to channel context
Felix Fietkau [Wed, 11 Jun 2014 10:47:51 +0000 (16:17 +0530)]
ath9k: Move acq to channel context

Add support to maintain per-channel ACs list.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Move txpower limit to channel context
Felix Fietkau [Wed, 11 Jun 2014 10:47:50 +0000 (16:17 +0530)]
ath9k: Move txpower limit to channel context

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoath9k: Add channel context structure
Felix Fietkau [Wed, 11 Jun 2014 10:47:49 +0000 (16:17 +0530)]
ath9k: Add channel context structure

The channel context structure is defined to enable
multi-channel concurrency support.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortlwifi: btcoexist: avoid format string in printk
Kees Cook [Tue, 10 Jun 2014 17:37:24 +0000 (10:37 -0700)]
rtlwifi: btcoexist: avoid format string in printk

Since CL_PRINTF only ever takes a single argument, make sure a format
string cannot leak into printk.

Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi_91x_sdio: add error handling into rsi_module_init()
Alexey Khoroshilov [Sat, 7 Jun 2014 03:18:31 +0000 (07:18 +0400)]
rsi_91x_sdio: add error handling into rsi_module_init()

Fix rsi_module_init() to propagate sdio_register_driver() errors.

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agorsi: Use module_usb_driver
Alexey Khoroshilov [Sat, 7 Jun 2014 03:18:30 +0000 (07:18 +0400)]
rsi: Use module_usb_driver

module_usb_driver eliminates the boilerplate and makes the code simpler,
in addition to the fact currently rsi_module_init() ignores usb_deregister() error.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agomwifiex: wowlan: do not disconnect on suspend
Bing Zhao [Sat, 7 Jun 2014 02:47:44 +0000 (19:47 -0700)]
mwifiex: wowlan: do not disconnect on suspend

For users who do not need wowlan, load mwifiex.ko with
disconnect_on_suspend = 1; or iw phy0 wowlan disable.

Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agomwifiex: support wowlan magic-packet encapsulated as UDP packet
Zhiyuan Yang [Sat, 7 Jun 2014 02:47:42 +0000 (19:47 -0700)]
mwifiex: support wowlan magic-packet encapsulated as UDP packet

When magic-packet is generated as a UDP packet the offset should
be 20+8 more bytes to cover IPv4 header and UDP header. So the
total offset become 56.
Add a new MEF entry to support both magic-packet patterns
generated by different tools.

Cc: Andreas Fenkart <afenkart@gmail.com>
Signed-off-by: Zhiyuan Yang <yangzy@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agobcma: gpio: register all 32 GPIOs on BCM53572
Rafał Miłecki [Thu, 5 Jun 2014 18:20:44 +0000 (20:20 +0200)]
bcma: gpio: register all 32 GPIOs on BCM53572

I've here a device detected as:
bcma: bus0: Found chip with id 0xD144, rev 0x01 and package 0x08
I couldn't find GPIO handling hw button until trying GPIO 20. It seems
BCM53572 also has 32 GPIOs.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: do not initialize BCN_OFFSET registers
Stanislaw Gruszka [Thu, 5 Jun 2014 11:52:27 +0000 (13:52 +0200)]
rt2x00: do not initialize BCN_OFFSET registers

We setup BCN_OFFSET{0,1} registers dynamically, don't have to
initialize them.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: change order when stop beaconing
Stanislaw Gruszka [Thu, 5 Jun 2014 11:52:26 +0000 (13:52 +0200)]
rt2x00: change order when stop beaconing

When no beaconing is needed, first stop beacon queue (disable beaconing
globally) to avoid possible sending of not prepared beacon on short
period after clearing beacon and before stop of BCN queue.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: change default MAC_BSSID_DW1_BSS_BCN_NUM
Stanislaw Gruszka [Thu, 5 Jun 2014 11:52:25 +0000 (13:52 +0200)]
rt2x00: change default MAC_BSSID_DW1_BSS_BCN_NUM

We setup MAC_BSSID_DW1_BSS_BCN_NUM dynamically when numbers of active
beacons increase. Change default to 0 to tell hardware that we want to
send only one beacon as default.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: change beaconing setup on RT2800
Stanislaw Gruszka [Thu, 5 Jun 2014 11:52:24 +0000 (13:52 +0200)]
rt2x00: change beaconing setup on RT2800

As reported by Matthias, on 5572 chip, even if we clear up TXWI
of corresponding beacon, hardware still try to send it or do other
action that increase power consumption peak up to 1A.

To avoid the issue, setup beaconing dynamically by configuring offsets
of currently active beacons and MAC_BSSID_DW1_BSS_BCN_NUM variable,
which limit number of beacons that hardware will try to send.

Reported-by: Matthias Fend <Matthias.Fend@wolfvision.net>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agort2x00: change beaconing locking
Stanislaw Gruszka [Thu, 5 Jun 2014 11:52:23 +0000 (13:52 +0200)]
rt2x00: change beaconing locking

This patch is needed for further changes to keep global variables
consistent when changing beaconing on diffrent vif's.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agomwifiex: Remove custom world regulatory domain
Markus Pargmann [Wed, 4 Jun 2014 13:32:32 +0000 (15:32 +0200)]
mwifiex: Remove custom world regulatory domain

A custom regulatory domain was introduced in this commit:
cc0ba0d mwifiex: support custom world regulatory domain

The commit description says that it was introduced because the world
regulatory domain does not include channels 52-64 and 100-140. These
channels are described in the world regulatory domain now, so we can
drop this custom regulatory domain.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Acked-by: Bing Zhao <bzhao@marvell.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agonet: wireless: rtlwifi: rtl8192de: phy.c: Cleaning up uninitialized variable
Rickard Strandqvist [Sun, 1 Jun 2014 20:14:38 +0000 (22:14 +0200)]
net: wireless: rtlwifi: rtl8192de: phy.c: Cleaning up uninitialized variable

There is a risk that the variables will be used without being initialized.
Have also moved variable to the part of the code where it is used.

This was largely found by using a static code analysis program called cppcheck.

Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agonet: wireless: rt2x00: rt2x00mac.c: Cleaning up uninitialized variables
Rickard Strandqvist [Sun, 1 Jun 2014 12:44:13 +0000 (14:44 +0200)]
net: wireless: rt2x00: rt2x00mac.c: Cleaning up uninitialized variables

There is a risk that the variable will be used without being initialized.

This was largely found by using a static code analysis program called cppcheck.

Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agonet: wireless: libertas: cmd.c: Cleaning up uninitialized variables
Rickard Strandqvist [Sun, 1 Jun 2014 12:39:20 +0000 (14:39 +0200)]
net: wireless: libertas: cmd.c: Cleaning up uninitialized variables

There is a risk that the variable will be used without being initialized.

This was largely found by using a static code analysis program called cppcheck.

Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: PHY: drop channel_type (we can get this info from chandef)
Rafał Miłecki [Sat, 31 May 2014 18:49:41 +0000 (20:49 +0200)]
b43: PHY: drop channel_type (we can get this info from chandef)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: PHY: drop is_40mhz (get width info from chandef)
Rafał Miłecki [Sat, 31 May 2014 18:49:40 +0000 (20:49 +0200)]
b43: PHY: drop is_40mhz (get width info from chandef)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: PHY: drop own channel_freq (get it from chandef when needed)
Rafał Miłecki [Sat, 31 May 2014 18:49:39 +0000 (20:49 +0200)]
b43: PHY: drop own channel_freq (get it from chandef when needed)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: store current channel using struct cfg80211_chan_def
Rafał Miłecki [Sat, 31 May 2014 18:49:38 +0000 (20:49 +0200)]
b43: store current channel using struct cfg80211_chan_def

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: b43_op_config: set channel info before switching band
Rafał Miłecki [Sat, 31 May 2014 18:49:37 +0000 (20:49 +0200)]
b43: b43_op_config: set channel info before switching band

Band switching code needs to know what channel we switch to.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: PHY: don't force default channel during init
Rafał Miłecki [Sat, 31 May 2014 18:49:36 +0000 (20:49 +0200)]
b43: PHY: don't force default channel during init

PHY may need to be re-initialized during runtime (e.g. on band switch).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: b43_op_config: use IEEE80211_CONF_CHANGE_CHANNEL
Rafał Miłecki [Sat, 31 May 2014 18:49:35 +0000 (20:49 +0200)]
b43: b43_op_config: use IEEE80211_CONF_CHANGE_CHANNEL

This is tiny optimization and grouping band/channel ops.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: b43_op_config: drop check for core change
Rafał Miłecki [Sat, 31 May 2014 18:49:34 +0000 (20:49 +0200)]
b43: b43_op_config: drop check for core change

There aren't devices with multiple 802.11 cores supported by b43.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agob43: drop B43_DEFAULT_CHANNEL
Rafał Miłecki [Sat, 31 May 2014 18:49:33 +0000 (20:49 +0200)]
b43: drop B43_DEFAULT_CHANNEL

It was never used, b43_switch_channel is always called with hw_value
(from mac80211) or whatever get_default_chan returns.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: fix possible RX descriptor invalid data read
Andrea Merello [Sat, 31 May 2014 16:30:26 +0000 (18:30 +0200)]
rtl818x_pci: fix possible RX descriptor invalid data read

RX descriptor data must be read only if the descriptor has been fully
updated by HW.
There is a "ownership" flag in the descriptor itself to test this.

The driver code contains a read for the "ownership" flag and, after
it, other read access for descriptor data.

This is in DMA coherent memory, that is _not_ guaranteed to be immune
to instruction reordering, thus it is possible that the descriptor
data is read _before_ the "ownership" flag.

This can theoretically lead to a DMA/CPU race that may end up with the
driver reading the data when it is still not valid, and the "ownership"
bit just after enough time that the HW make the whole descriptor valid.
The driver will in this case believe the data is valid, but it will use
the invalid data read earlier.

In order to avoid this, this patch adds a rmb() to force the "ownership"
bit read to be issued before other descriptor data reads are attempted.

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: provide dBm signal information for rtl8185
Andrea Merello [Sat, 31 May 2014 16:30:13 +0000 (18:30 +0200)]
rtl818x_pci: provide dBm signal information for rtl8185

This patch makes the driver report signal information for rtl8185
boards using dBm instead of unspecified unit.
Rtl8180 remains untouched.

I did some tests to confirm the correctness of the measure performed
by the board and it seems reasonably correct.

The test setup has been made by connecting an AP with coax and
RF attenuators to the card antenna port.

In order to get a reference measure I tried with several cards with
different chipset I own. I found that many gave different results, and
I finally selected two cards that gave me consistent results to use
as reference: AR9271 and Prism54-usb (isl3887 with Frisbee radio).

Using this references I compared the RSSI information with my rtl8185
and I repeated tests with three different attenuation values, increasing
attenuation by 10dB each step.
I made only relative measures, making NO assumption about source power.

CCK measures seem very close to my references, OFDM are a little bit
less precise but, considering that these cards are not measuring
instrumentation, IMHO this is still fairly good.

CCK measures (1Mbps beacons)

 ATTENUATOR 1
p54usb:    -58dBm
ath9k_htc: -59dBm
rtl8185:   -59dBm

 ATTENUATOR 2
p54usb:    -67dBm
ath9k_htc: -68dBm
rtl8185:   -70dBm

 ATTENUATOR 3
p54usb:    -78dBm
ath9k_htc: -79dBm
rtl8185:   -79dBm

OFDM measures (54Mbps ping)

 ATTENUATOR 1
p54usb:    -58dBm
ath9k_htc: -57dBm
rtl8185:   -62dBm

 ATTENUATOR 2
p54usb:    -68dBm
rtl8185:   -71dBm

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agortl818x_pci: make RSSI code more readable
Andrea Merello [Sat, 31 May 2014 16:29:46 +0000 (18:29 +0200)]
rtl818x_pci: make RSSI code more readable

remove the if-else chains and use switch-case to make code more
readable and avoiding long lines that broke in several lines

Signed-off-by: Andrea Merello <andrea.merello@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agocw1200: Remove useless return variables
Peter Senna Tschudin [Sat, 31 May 2014 16:12:26 +0000 (13:12 -0300)]
cw1200: Remove useless return variables

This patch remove variables that are initialized with a constant,
are never updated, and are only used as parameter of return.
Return the constant instead of using a variable.

Verified by compilation only.

The coccinelle script that find and fixes this issue is:
// <smpl>
@@
type T;
constant C;
identifier ret;
@@
- T ret = C;
... when != ret
    when strict
return
- ret
+ C
;
// </smpl>

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agonet: wireless: Remove useless return variables
Peter Senna Tschudin [Sat, 31 May 2014 13:14:06 +0000 (10:14 -0300)]
net: wireless: Remove useless return variables

This patch remove variables that are initialized with a constant,
are never updated, and are only used as parameter of return.
Return the constant instead of using a variable.

wl_cfg80211.c verified by compilation only.
phy/phy_cmn.c unverified.

The coccinelle script that find and fixes this issue is:
// <smpl>
@@
type T;
constant C;
identifier ret;
@@
- T ret = C;
... when != ret
    when strict
return
- ret
+ C
;
// </smpl>

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoLinux 3.16-rc1
Linus Torvalds [Mon, 16 Jun 2014 03:45:28 +0000 (17:45 -1000)]
Linux 3.16-rc1

10 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Mon, 16 Jun 2014 02:37:03 +0000 (16:37 -1000)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Fix checksumming regressions, from Tom Herbert.

 2) Undo unintentional permissions changes for SCTP rto_alpha and
    rto_beta sysfs knobs, from Denial Borkmann.

 3) VXLAN, like other IP tunnels, should advertize it's encapsulation
    size using dev->needed_headroom instead of dev->hard_header_len.
    From Cong Wang.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  net: sctp: fix permissions for rto_alpha and rto_beta knobs
  vxlan: Checksum fixes
  net: add skb_pop_rcv_encapsulation
  udp: call __skb_checksum_complete when doing full checksum
  net: Fix save software checksum complete
  net: Fix GSO constants to match NETIF flags
  udp: ipv4: do not waste time in __udp4_lib_mcast_demux_lookup
  vxlan: use dev->needed_headroom instead of dev->hard_header_len
  MAINTAINERS: update cxgb4 maintainer

10 years agoMerge tag 'clk-for-linus-3.16-part2' of git://git.linaro.org/people/mike.turquette...
Linus Torvalds [Mon, 16 Jun 2014 02:02:20 +0000 (16:02 -1000)]
Merge tag 'clk-for-linus-3.16-part2' of git://git.linaro.org/people/mike.turquette/linux

Pull more clock framework updates from Mike Turquette:
 "This contains the second half the of the clk changes for 3.16.

  They are simply fixes and code refactoring for the OMAP clock drivers.
  The sunxi clock driver changes include splitting out the one
  mega-driver into several smaller pieces and adding support for the A31
  SoC clocks"

* tag 'clk-for-linus-3.16-part2' of git://git.linaro.org/people/mike.turquette/linux: (25 commits)
  clk: sunxi: document PRCM clock compatible strings
  clk: sunxi: add PRCM (Power/Reset/Clock Management) clks support
  clk: sun6i: Protect SDRAM gating bit
  clk: sun6i: Protect CPU clock
  clk: sunxi: Rework clock protection code
  clk: sunxi: Move the GMAC clock to a file of its own
  clk: sunxi: Move the 24M oscillator to a file of its own
  clk: sunxi: Remove calls to clk_put
  clk: sunxi: document new A31 USB clock compatible
  clk: sunxi: Implement A31 USB clock
  ARM: dts: OMAP5/DRA7: use omap5-mpu-dpll-clock capable of dealing with higher frequencies
  CLK: TI: dpll: support OMAP5 MPU DPLL that need special handling for higher frequencies
  ARM: OMAP5+: dpll: support Duty Cycle Correction(DCC)
  CLK: TI: clk-54xx: Set the rate for dpll_abe_m2x2_ck
  CLK: TI: Driver for DRA7 ATL (Audio Tracking Logic)
  dt:/bindings: DRA7 ATL (Audio Tracking Logic) clock bindings
  ARM: dts: dra7xx-clocks: Correct name for atl clkin3 clock
  CLK: TI: gate: add composite interface clock to OMAP2 only build
  ARM: OMAP2: clock: add DT boot support for cpufreq_ck
  CLK: TI: OMAP2: add clock init support
  ...

10 years agoMerge git://git.infradead.org/users/willy/linux-nvme
Linus Torvalds [Mon, 16 Jun 2014 01:58:03 +0000 (15:58 -1000)]
Merge git://git.infradead.org/users/willy/linux-nvme

Pull NVMe update from Matthew Wilcox:
 "Mostly bugfixes again for the NVMe driver.  I'd like to call out the
  exported tracepoint in the block layer; I believe Keith has cleared
  this with Jens.

  We've had a few reports from people who're really pounding on NVMe
  devices at scale, hence the timeout changes (and new module
  parameters), hotplug cpu deadlock, tracepoints, and minor performance
  tweaks"

[ Jens hadn't seen that tracepoint thing, but is ok with it - it will
  end up going away when mq conversion happens ]

* git://git.infradead.org/users/willy/linux-nvme: (22 commits)
  NVMe: Fix START_STOP_UNIT Scsi->NVMe translation.
  NVMe: Use Log Page constants in SCSI emulation
  NVMe: Define Log Page constants
  NVMe: Fix hot cpu notification dead lock
  NVMe: Rename io_timeout to nvme_io_timeout
  NVMe: Use last bytes of f/w rev SCSI Inquiry
  NVMe: Adhere to request queue block accounting enable/disable
  NVMe: Fix nvme get/put queue semantics
  NVMe: Delete NVME_GET_FEAT_TEMP_THRESH
  NVMe: Make admin timeout a module parameter
  NVMe: Make iod bio timeout a parameter
  NVMe: Prevent possible NULL pointer dereference
  NVMe: Fix the buffer size passed in GetLogPage(CDW10.NUMD)
  NVMe: Update data structures for NVMe 1.2
  NVMe: Enable BUILD_BUG_ON checks
  NVMe: Update namespace and controller identify structures to the 1.1a spec
  NVMe: Flush with data support
  NVMe: Configure support for block flush
  NVMe: Add tracepoints
  NVMe: Protect against badly formatted CQEs
  ...

10 years agonet: sctp: fix permissions for rto_alpha and rto_beta knobs
Daniel Borkmann [Sat, 14 Jun 2014 22:59:14 +0000 (00:59 +0200)]
net: sctp: fix permissions for rto_alpha and rto_beta knobs

Commit 3fd091e73b81 ("[SCTP]: Remove multiple levels of msecs
to jiffies conversions.") has silently changed permissions for
rto_alpha and rto_beta knobs from 0644 to 0444. The purpose of
this was to discourage users from tweaking rto_alpha and
rto_beta knobs in production environments since they are key
to correctly compute rtt/srtt.

RFC4960 under section 6.3.1. RTO Calculation says regarding
rto_alpha and rto_beta under rule C3 and C4:

  [...]
  C3)  When a new RTT measurement R' is made, set

       RTTVAR <- (1 - RTO.Beta) * RTTVAR + RTO.Beta * |SRTT - R'|

       and

       SRTT <- (1 - RTO.Alpha) * SRTT + RTO.Alpha * R'

       Note: The value of SRTT used in the update to RTTVAR
       is its value before updating SRTT itself using the
       second assignment. After the computation, update
       RTO <- SRTT + 4 * RTTVAR.

  C4)  When data is in flight and when allowed by rule C5
       below, a new RTT measurement MUST be made each round
       trip. Furthermore, new RTT measurements SHOULD be
       made no more than once per round trip for a given
       destination transport address. There are two reasons
       for this recommendation: First, it appears that
       measuring more frequently often does not in practice
       yield any significant benefit [ALLMAN99]; second,
       if measurements are made more often, then the values
       of RTO.Alpha and RTO.Beta in rule C3 above should be
       adjusted so that SRTT and RTTVAR still adjust to
       changes at roughly the same rate (in terms of how many
       round trips it takes them to reflect new values) as
       they would if making only one measurement per
       round-trip and using RTO.Alpha and RTO.Beta as given
       in rule C3. However, the exact nature of these
       adjustments remains a research issue.
  [...]

While it is discouraged to adjust rto_alpha and rto_beta
and not further specified how to adjust them, the RFC also
doesn't explicitly forbid it, but rather gives a RECOMMENDED
default value (rto_alpha=3, rto_beta=2). We have a couple
of users relying on the old permissions before they got
changed. That said, if someone really has the urge to adjust
them, we could allow it with a warning in the log.

Fixes: 3fd091e73b81 ("[SCTP]: Remove multiple levels of msecs to jiffies conversions.")
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge branch 'csum_fixes'
David S. Miller [Sun, 15 Jun 2014 08:00:56 +0000 (01:00 -0700)]
Merge branch 'csum_fixes'

Tom Herbert says:

====================
Fixes related to some recent checksum modifications.

- Fix GSO constants to match NETIF flags
- Fix logic in saving checksum complete in __skb_checksum_complete
- Call __skb_checksum_complete from UDP if we are checksumming over
  whole packet in order to save checksum.
- Fixes to VXLAN to work correctly with checksum complete
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agovxlan: Checksum fixes
Tom Herbert [Sun, 15 Jun 2014 06:24:36 +0000 (23:24 -0700)]
vxlan: Checksum fixes

Call skb_pop_rcv_encapsulation and postpull_rcsum for the Ethernet
header to work properly with checksum complete.

Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: add skb_pop_rcv_encapsulation
Tom Herbert [Sun, 15 Jun 2014 06:24:28 +0000 (23:24 -0700)]
net: add skb_pop_rcv_encapsulation

This function is used by UDP encapsulation protocols in RX when
crossing encapsulation boundary. If ip_summed is set to
CHECKSUM_UNNECESSARY and encapsulation is not set, change to
CHECKSUM_NONE since the checksum has not been validated within the
encapsulation. Clears csum_valid by the same rationale.

Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoudp: call __skb_checksum_complete when doing full checksum
Tom Herbert [Sun, 15 Jun 2014 06:24:20 +0000 (23:24 -0700)]
udp: call __skb_checksum_complete when doing full checksum

In __udp_lib_checksum_complete check if checksum is being done over all
the data (len is equal to skb->len) and if it is call
__skb_checksum_complete instead of __skb_checksum_complete_head. This
allows checksum to be saved in checksum complete.

Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: Fix save software checksum complete
Tom Herbert [Sun, 15 Jun 2014 06:24:03 +0000 (23:24 -0700)]
net: Fix save software checksum complete

Geert reported issues regarding checksum complete and UDP.
The logic introduced in commit 7e3cead5172927732f51fde
("net: Save software checksum complete") is not correct.

This patch:
1) Restores code in __skb_checksum_complete_header except for setting
   CHECKSUM_UNNECESSARY. This function may be calculating checksum on
   something less than skb->len.
2) Adds saving checksum to __skb_checksum_complete. The full packet
   checksum 0..skb->len is calculated without adding in pseudo header.
   This value is saved in skb->csum and then the pseudo header is added
   to that to derive the checksum for validation.
3) In both __skb_checksum_complete_header and __skb_checksum_complete,
   set skb->csum_valid to whether checksum of zero was computed. This
   allows skb_csum_unnecessary to return true without changing to
   CHECKSUM_UNNECESSARY which was done previously.
4) Copy new csum related bits in __copy_skb_header.

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agonet: Fix GSO constants to match NETIF flags
Tom Herbert [Sun, 15 Jun 2014 06:23:52 +0000 (23:23 -0700)]
net: Fix GSO constants to match NETIF flags

Joseph Gasparakis reported that VXLAN GSO offload stopped working with
i40e device after recent UDP changes. The problem is that the
SKB_GSO_* bits are out of sync with the corresponding NETIF flags. This
patch fixes that. Also, we add BUILD_BUG_ONs in net_gso_ok for several
GSO constants that were missing to avoid the problem in the future.

Reported-by: Joseph Gasparakis <joseph.gasparakis@intel.com>
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoMerge tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb...
Linus Torvalds [Sun, 15 Jun 2014 00:49:48 +0000 (19:49 -0500)]
Merge tag 'scsi-for-linus' of git://git./linux/kernel/git/jejb/scsi

Pull more SCSI updates from James Bottomley:
 "This is just a couple of drivers (hpsa and lpfc) that got left out for
  further testing in linux-next.  We also have one fix to a prior
  submission (qla2xxx sparse)"

* tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (36 commits)
  qla2xxx: fix sparse warnings introduced by previous target mode t10-dif patch
  lpfc: Update lpfc version to driver version 10.2.8001.0
  lpfc: Fix ExpressLane priority setup
  lpfc: mark old devices as obsolete
  lpfc: Fix for initializing RRQ bitmap
  lpfc: Fix for cleaning up stale ring flag and sp_queue_event entries
  lpfc: Update lpfc version to driver version 10.2.8000.0
  lpfc: Update Copyright on changed files from 8.3.45 patches
  lpfc: Update Copyright on changed files
  lpfc: Fixed locking for scsi task management commands
  lpfc: Convert runtime references to old xlane cfg param to fof cfg param
  lpfc: Fix FW dump using sysfs
  lpfc: Fix SLI4 s abort loop to process all FCP rings and under ring_lock
  lpfc: Fixed kernel panic in lpfc_abort_handler
  lpfc: Fix locking for postbufq when freeing
  lpfc: Fix locking for lpfc_hba_down_post
  lpfc: Fix dynamic transitions of FirstBurst from on to off
  hpsa: fix handling of hpsa_volume_offline return value
  hpsa: return -ENOMEM not -1 on kzalloc failure in hpsa_get_device_id
  hpsa: remove messages about volume status VPD inquiry page not supported
  ...

10 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
Linus Torvalds [Sun, 15 Jun 2014 00:48:43 +0000 (19:48 -0500)]
Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Pull more btrfs updates from Chris Mason:
 "This has a few fixes since our last pull and a new ioctl for doing
  btree searches from userland.  It's very similar to the existing
  ioctl, but lets us return larger items back down to the app"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  btrfs: fix error handling in create_pending_snapshot
  btrfs: fix use of uninit "ret" in end_extent_writepage()
  btrfs: free ulist in qgroup_shared_accounting() error path
  Btrfs: fix qgroups sanity test crash or hang
  btrfs: prevent RCU warning when dereferencing radix tree slot
  Btrfs: fix unfinished readahead thread for raid5/6 degraded mounting
  btrfs: new ioctl TREE_SEARCH_V2
  btrfs: tree_search, search_ioctl: direct copy to userspace
  btrfs: new function read_extent_buffer_to_user
  btrfs: tree_search, copy_to_sk: return needed size on EOVERFLOW
  btrfs: tree_search, copy_to_sk: return EOVERFLOW for too small buffer
  btrfs: tree_search, search_ioctl: accept varying buffer
  btrfs: tree_search: eliminate redundant nr_items check

10 years agoMerge git://git.kvack.org/~bcrl/aio-next
Linus Torvalds [Sun, 15 Jun 2014 00:43:27 +0000 (19:43 -0500)]
Merge git://git.kvack.org/~bcrl/aio-next

Pull aio fix and cleanups from Ben LaHaise:
 "This consists of a couple of code cleanups plus a minor bug fix"

* git://git.kvack.org/~bcrl/aio-next:
  aio: cleanup: flatten kill_ioctx()
  aio: report error from io_destroy() when threads race in io_destroy()
  fs/aio.c: Remove ctx parameter in kiocb_cancel

10 years agofix __swap_writepage() compile failure on old gcc versions
Al Viro [Sat, 14 Jun 2014 06:12:41 +0000 (07:12 +0100)]
fix __swap_writepage() compile failure on old gcc versions

Tetsuo Handa wrote:
 "Commit 62a8067a7f35 ("bio_vec-backed iov_iter") introduced an unnamed
  union inside a struct which gcc-4.4.7 cannot handle.  Name the unnamed
   union as u in order to fix build failure"

Let's do this instead: there is only one place in the entire tree that
steps into this breakage.  Anon structs and unions work in older gcc
versions; as the matter of fact, we have those in the tree - see e.g.
struct ieee80211_tx_info in include/net/mac80211.h

What doesn't work is handling their initializers:

struct {
int a;
union {
int b;
char c;
};
} x[2] = {{.a = 1, .c = 'a'}, {.a = 0, .b = 1}};

is the obvious syntax for initializer, perfectly fine for C11 and
handled correctly by gcc-4.7 or later.

Earlier versions, though, break on it - declaration is fine and so's
access to fields (i.e.  x[0].c = 'a'; would produce the right code), but
members of the anon structs and unions are not inserted into the right
namespace.  Tellingly, those older versions will not barf on struct {int
a; struct {int a;};}; - looks like they just have it hacked up somewhere
around the handling of .  and -> instead of doing the right thing.

The easiest way to deal with that crap is to turn initialization of
those fields (in the only place where we have such initializer of
iov_iter) into plain assignment.

Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reported-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>