aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/net/wireless/ath
AgeCommit message (Collapse)AuthorFilesLines
2010-06-04ath9k: fix dma sync in rx pathMing Lei1-3/+7
If buffer is to be accessed by cpu after dma is over, but between dma mapping and dma unmapping, we should use dma_sync_single_for_cpu to sync the buffer between cpu with device. And dma_sync_single_for_device is used to let device gain the buffer again. v2: Felix pointed out dma_sync_single_for_device is needed to return buffer to device if an unsuccessful status bit check is found. Signed-off-by: Ming Lei <tom.leiming@gmail.com> Acked-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k: fix queue stop/start based on the number of pending framesFelix Fietkau3-37/+25
Because there is a limited number of tx buffers available, once the queue has been filled to a certain point, ath9k needs to stop accepting new frames from mac80211. In order to prevent a full WMM queue from stopping another queue with fewer frames, this patch limits the number of queued frames to a quarter of the total available tx buffers, minus some reserved frames to be used for other purposes (e.g. beacons). Because tx buffers are reserved for frames when they're staged in software queues as well, the actual queue depth cannot be used for this, so this patch stores a reference to the tx queue in the ath_buf struct and keeps track of the total number of pending frames. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Fix fair beacon distributionSujith3-1/+31
This patch fixes beacon distribution in IBSS mode by configuring the hardware beacon queue properly. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Handle host RX disableSujith2-0/+11
The MIB counters used by ANI have to be disabled on the host because the FW doesn't do it on the target side. Also, flush the receive buffers before initializing RX on the target. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Handle monitor interface removalSujith1-6/+6
The monitor interface instance on the target has to be removed before setting it to FULLSLEEP. Handle this properly. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Fix locking for ps_idleSujith1-5/+13
ps_idle is protected by the htc_pm_lock mutex. Use it to protect the variable. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Remove useless cancel_work_syncSujith1-1/+0
There is no need to cancel the PS work when disassociation happens. The work handlers are cancelled in the stop() callback. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Fix bug in handling CONF_IDLESujith1-1/+1
Disable the radio only when mac80211 indicates it, through the IEEE80211_CONF_CHANGE_IDLE flag. Not handling this properly will result in multiple calls to radio_disable() even though the radio is already idle. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Add PS wrappersSujith1-5/+18
The HW has to be awake when registers are accessed. Ensure this is so by using the PS wrappers at appropriate places. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_htc: Revamp CONF_IDLE handlingSujith1-36/+93
This patch revamps IDLE power save handling in the driver. Two separate functions (radio enable/disable) are introduced, because the semantics of radio handling is just not the same as the start()/stop() callbacks. For example, the HW must not be disabled, instead, the PHY has to be disabled in radio_disable(). Also, the HW has to be reset properly in radio enable/disable and certain registers have to be programmed only once, in the start() callback. The radio_enable() routine doesn't need the PS wrappers since we set the HW power mode to AWAKE anyway before calling it. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: Fix async fifo for AR9287Sujith5-44/+55
Async fifo is now enabled only for versions 1.3 and above. Enable it in the appropriate place, in the reset routine, instead of process_ini(). Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: Update the PCI WAR registerSujith2-28/+60
This patch updates the PCI power save handling code, fixing ASPM hangs and handling device state D3 properly. The WAR register is programmed with the correct values now. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: Cleanup TX power calculation for AR9287Sujith3-140/+167
* Add a few comments, and move the updation of max_power_level to a helper routine. This is also done by non-4K based chipsets, this will be fixed in a separate patch. * Remove two WARs which are required for old AR5416 chipsets, and are not needed for AR9287. * Fix indentation and make things readable. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: Optimize ath9k_hw_ar9287_set_board_valuesSujith3-36/+40
Rather than doing a series of RMWs, calculate the value to be written to the register in question and do a single REGWRITE. This improves bringup time. This depends on the analog_shiftreg configuration option, which is currently buggy. For AP mode, a delay of 100us has to be the default. For station mode, this knob has to be enabled on a per-case basis, though it is a little unclear on when to enable a delay. This can be fixed later though. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: Cleanup eeprom_9287.cSujith1-137/+136
* Fix whitespace damage. * Remove unused debug messages. * Introduce a macro NUM_EEP_WORDS. * Convert AR9287 to lowercase in function names. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k: use the key handling code from ath9k_common instead of duplicating itFelix Fietkau1-230/+2
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_common: use allocated key cache entries for multi BSS crypto supportFelix Fietkau1-10/+27
This patch replaces the buggy 'ath9k: Group Key fix for VAPs' change. For AP mode group keys, use the BSSID as lookup mac address, with the multicast keysearch bit set. For IBSS mode, use the peer's MAC address with multicast keysearch. For STA mode, keep using the group key slots. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-03ath9k_hw: add multicast key search supportFelix Fietkau1-1/+12
If a MAC address for a key table entry is flagged with the multicast bit (0x01), indicate to the hardware that multicast lookup instead of unicast lookup should be used. The multicast bit itself never makes it to the actual keytable entry register, as it is shifted out. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: fix build error when ATH9K_HTC_DEBUGFS not enabledJohn W. Linville1-0/+2
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: cleanup: remove unneeded null checkDan Carpenter1-3/+0
We dereference "wmi" on the line before and also when we initialize "ah". This check has always been after a dereference since the first commit a couple months ago. Looking through the code, it looks like "wmi" can't actually be null here so I just removed the check. Signed-off-by: Dan Carpenter <error27@gmail.com> Acked-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Fix bug in validating received data length for edmaVasanthakumar Thiagarajan1-1/+2
The rx status length should also be taken into account while validating the length of a received frame. Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Fix bug in accessing skb->data of rx frame for edmaVasanthakumar Thiagarajan1-1/+2
Skip the rx status portion in skb->data before accessing ieee80211 frame header. Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Clean up few function parameters in recv.cVasanthakumar Thiagarajan1-15/+8
ath9k_rx_skb_preprocess() needs only ieee80211 frame header, pass only frame headers instead of skb to that function. Also remove ineffective frame dump in ath9k_process_rate(). Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Move driver specific structuresSujith2-63/+63
A bunch of data structures are present in the common module, which are internal to ath9k. Move them to ath9k.h Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Move ath9k specific RX code to driverSujith3-281/+263
This patch relocates RX processing code from the common module to ath9k. This reduces the size of the common module which is also used by ath9k_htc. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_hw: Enable TX IQ calibration on AR9003Luis R. Rodriguez3-11/+6
To enable it we now disable and re-enable the PHY chips after TX IQ calibration. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: print error message if ANI levels are out of rangeBruno Randolf1-6/+5
Since we have sysfs to manually set the ANI levels, we should print errors to the kernel log if the values are out of bounds. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: always calculate ANI listen timeBruno Randolf1-4/+5
Calculate 'listen' time also when automatic ANI is off, since this and the "busy" time is useful information also in manual mode. Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: add sysfs files for ANI parametersBruno Randolf5-1/+123
/sys/class/ieee80211/phy0/device/ani/ani_mode /sys/class/ieee80211/phy0/device/ani/noise_immunity_level /sys/class/ieee80211/phy0/device/ani/spur_level /sys/class/ieee80211/phy0/device/ani/firstep_level /sys/class/ieee80211/phy0/device/ani/ofdm_weak_signal_detection /sys/class/ieee80211/phy0/device/ani/cck_weak_signal_detection /sys/class/ieee80211/phy0/device/ani/noise_immunity_level_max /sys/class/ieee80211/phy0/device/ani/spur_level_max /sys/class/ieee80211/phy0/device/ani/firstep_level_max sysfs has a lot of symlinks, so you can find the files also in other locations, like (by PCI ID) /sys/devices/pci0000:00/0000:00:11.0/ani and others. Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: use ath5k_softc as driver dataBruno Randolf1-9/+6
It's our "private driver data"... It's used more often and hw is the mac80211 part. This makes more sense with the next (sysfs) patch. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: clarify logic when to enable spur mitigation filterBruno Randolf1-10/+5
The old code logically did not make sense and seems to have been confused by the fact that we could have newer EEPROMs on older hardware. In any case the spur mitigation filter was set if the srev was >= AR5K_SREV_AR5424. Spur info is available only from EEPROM versions bigger than 5.3 but but the EEPOM routines will use static values for older versions, so that should be o.k. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: remove ATH_TRACE macroBruno Randolf12-99/+0
Now that we have ftrace, it is not needed any more. Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: run NF calibration only every 60 secondsBruno Randolf2-7/+13
Since NF calibration interferes with TX and RX and also has been the cause of other problems (when it's run concurrently with ath5k_reset) we want to run it less often - every 60 seconds for now. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: Stop queues only for NF calibrationBruno Randolf1-5/+6
As far as we know, only NF calibration interferes with RX/TX so we can leave the queues enabled for the other calibrations. BTW: Stopping the queues is not enough for avoiding transmissions, since there might be packets in the queue + beacons are also sent regularly! But i leave it like this until we have a better solution (stopping TX DMA?). Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: move noise floor calibration into taskletBruno Randolf3-23/+12
Seperate noise floor calibration from other PHY calibration and move it to the tasklet. This is the first step to more separation of different calibrations. Also move out ath5k_hw_request_rfgain_probe(ah) so we have one clean function for I/Q calibration on 5111x parts. Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath5k: initialize calibration timersBruno Randolf1-0/+2
Initialize calibration timers on reset, since otherwise they might be in the future and the calibration tasklet might not be scheduled for a long time. Signed-off-by: Bruno Randolf <br1@einfach.org> Acked-by: Nick Kossifidis <mickflemm@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_hw: Enable auto sleep for ar9003Vasanthakumar Thiagarajan1-1/+1
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Fix power save with auto sleepingVasanthakumar Thiagarajan1-2/+3
Rx should not be disabed/disabled when hw supports auto sleeping. Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: Make sure null func frame is acked before going into PS for ar9003Vasanthakumar Thiagarajan1-0/+11
Add missing code to handle nullfunc frame completion in ath_tx_edma_tasklet(). Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02wireless: fix several minor description typosWalter Goldens2-2/+2
Signed-off-by: Walter Goldens <goldenstranger@yahoo.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Enable RX STBC for AR9271Sujith1-0/+2
Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Enable SGI in HT20 for AR9271Sujith3-3/+9
Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Update HT configuration properlySujith1-0/+30
Use BSS_CHANGED_HT to handle HT parameter changes. The rate information on the target has to be updated to handle changes in HT configuration. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Cleanup rate initializationSujith2-83/+32
This patch removes a large chunk of code dealing with rate management within the driver and simplifying things by removing the hacky method of calculating HT changes. A subsequent patch would fix this by just using BSS_CHANGED_HT from mac80211. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Initvals update for AR9271Sujith1-3/+3
Update from internal systems engineering team. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02drivers/net/wireless/ath/ath9k: Use kmemdupJulia Lawall1-3/+1
Use kmemdup when some other buffer is immediately copied into the allocated region. A simplified version of the semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression from,to,size,flag; statement S; @@ - to = \(kmalloc\|kzalloc\)(size,flag); + to = kmemdup(from,size,flag); if (to==NULL || ...) S - memcpy(to, from, size); // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k/debug: fixup the return codesDan Carpenter1-9/+9
Changed -EINVAL to -EFAULT if copy_to_user() failed. Changed 0 to -ENOMEM if allocations failed. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k/debug: improve the snprintf() handlingDan Carpenter1-8/+25
The snprintf() function returns the number of bytes that *would* have been written (not counting the NULL terminator) and that can potentally be more than the size of the buffer. In this patch if there were one liners where string clearly fits into the buffer, then I changed snprintf to sprintf(). It's confusing to use the return value of snprintf() as a limitter without verifying that it's smaller than size. This is what initially caught my attention here. If we use the return value of sprintf() instead future code auditors will assume we've verified that it fits already. Also I did find some places where it made sense to use the return value after we've verified that it is smaller than the buffer size. Finally the read_file_rcstat() function added an explicit NULL terminator before calling snprintf(). That's unnecessary because snprintf() will add the null terminator automatically. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k: fix dma direction for map/unmap in ath_rx_taskletMing Lei1-2/+2
For edma, we should use DMA_BIDIRECTIONAL, or else use DMA_FROM_DEVICE. Signed-off-by: Ming Lei <tom.leiming@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-02ath9k_htc: Increase credit sizeSujith1-1/+1
This is the maximum supported by the firmware. Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.