summaryrefslogtreecommitdiffstats
path: root/drivers (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Merge remote-tracking branch 'wireless-next/master' into mac80211-nextJohannes Berg2014-11-182100-40273/+106888
|\ | | | | | | | | | | | | This brings in some mwifiex changes that further patches will need to work on top to not cause merge conflicts. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
| * ath9k: Move roc completion to the offchannel timerSujith Manoharan2014-11-171-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, when a roc period expires, the offchannel timer calls ieee80211_remain_on_channel_expired(), but the roc state is cleared only when the queued work to switch to the operating channel gets a chance to run. This race is a problem because mac80211 can issue a new roc request in this window. To avoid this, handle roc completion in the offchannel timer itself. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Adjust tbtt delta properlySujith Manoharan2014-11-172-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In a GO/STA setup, when we switch to the STA context, the channel context timer is scheduled with a period of half the beacon interval. If a beacon is received in this duration, the timer is adjusted to accommodate TSF sync done by the HW. But, if the actual channel switch is delayed for some reason, we end up rearming the timer every time a new beacon is received. Avoid this by doing the adjustment only once. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Handle failure to send NoASujith Manoharan2014-11-171-1/+8
| | | | | | | | | | | | | | | | | | | | If for some reason a beacon with a new NoA is not sent out, then reset the mgd_prepare_tx flag. Not doing this will result in a situation where a GO will send a new NoA when it shouldn't. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Cancel pending offchannel operationsSujith Manoharan2014-11-171-13/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch makes sure that pending roc/scan operations are cancelled properly when a new context is assigned/unassigned. The flush_work() call to flush out any scheduled channel context work is removed, instead, sc->mutex is dropped to allow any pending work to get a chance to complete by the channel scheduler. Also, increase the timeout to allow a switch to an active GO. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Clear offchannel state properlySujith Manoharan2014-11-171-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | When a pending roc or scan operation is cancelled, the offchannel operation is cleared, but the offchannel state in the main scheduler is not cleared. This causes problems since an active GO will try to process a stale offchannel request that was deferred earlier. Fix this by clearing the state when there is no pending offchannel (roc/scan) operation. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Fix thermometer programmingSujith Manoharan2014-11-171-4/+5
| | | | | | | | | | | | | | | | | | | | | | The registers that control the on-chip thermometer need to be programmed based on the chainmask that the solution supports, not the chainmask that is present in the eeprom. Cc: Miaoqing Pan <miaoqing@qca.qualcomm.com> Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Store the chip chainmask in HW capabilitiesSujith Manoharan2014-11-172-10/+11
| | | | | | | | | | | | Cc: Miaoqing Pan <miaoqing@qca.qualcomm.com> Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Enable TSF2 for generic HW timersSujith Manoharan2014-11-173-1/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The base TSF is used for HW timers 0..7, but chips in the AR9003 family and above can support more generic timers. To use them, however, a second HW TSF needs to be enabled. This patch allows usage of the extra timers by starting the second TSF properly. The extra set of HW timers is apparently also present in AR9287, but we enable it only for the AR9003 family. Cc: Kobi Cohen-Arazi <kobic@qti.qualcomm.com> Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Fix LED configurationSujith Manoharan2014-11-176-5/+19
| | | | | | | | | | | | | | | | | | | | On some x86 platforms, the LED gpio is active high instead of active low. Identify such cards and modify the GPIO usage to make sure LED works properly. Cc: Russell Hu <rhu@qca.qualcomm.com> Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Use new QCA953x initvalsMiaoqing Pan2014-11-171-10/+41
| | | | | | | | | | | | | | | | This patch updates the initvals for QCA953x v1.1 and v2.0 Signed-off-by: Miaoqing Pan <miaoqing@qca.qualcomm.com> Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: Update QCA953x initvalsSujith Manoharan2014-11-171-30/+468
| | | | | | | | | | | | | | | | | | | | * Duplicates have been marked. * New initvals for 1.1 and 2.0 versions. * xPA support. * Fix for low power issue. Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k_htc: Enable software crypto for mgmt frame in Tx for USB devicesChun-Yeow Yeoh2014-11-171-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In secured mesh, the unicast mgmt frame is encrypted using the same key that used for encrypting the unicast data frame. This patch "ath9k_htc_firmware: fix the offset of CCMP header for mesh data frame" applied to open-ath9k-htc-firmware allows the ath9k_htc to be loaded without "nohwcrypt=1". Unfortunately, this is not working and we still need CCMP encryption of transmitted management frames to be done in software. So this patch allows the software encryption for transmitted management frame to be done in software but remain the hardware decryption for received management frame. This patch is tested with the following hardwares: - TP-Link TL-WN821N v3 802.11n [Atheros AR7010+AR9287] - AR9271 802.11n and managed to work with peer mesh STA equipped with ath9k. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k|ath9k_htc: Seperate the software crypto flag for Tx and RxChun-Yeow Yeoh2014-11-175-7/+11
| | | | | | | | | | | | | | | | | | | | Use the sw_mgmt_crypto_tx flag to trigger the CCMP encryption for transmitted management frames to be done in software while the sw_mgmt_crypto_rx flag is used to trigger the CCMP decryption for received management frames to be done in software. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: enable auto TDLS support for SD8887Avinash Patil2014-11-173-1/+9
| | | | | | | | | | | | | | | | | | Auto TDLS support is enabled per device. As of now add this feature only for SD8887. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: support for parsing TDLS discovery framesAvinash Patil2014-11-171-0/+38
| | | | | | | | | | | | | | | | | | This patch adds support for parsing TDLS discovery frames. After parsing, we update peer RSSI information. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: add auto TDLS supportAvinash Patil2014-11-178-2/+295
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds auto TDLS support to mwifiex. Auto TDLS functionality works as follows: 1. Whenever userspace application has triggered TDLS connection with any peer, driver would store this peer mac address details in its database. 2. After this driver whenever driver receives packet on direct link, it would store rssi and timestamp in peer information. 3. Whenever a packet is to be transmitted to non-AP peer in station mode, driver would check if TDLS link can be established by looking at peer RSSI information. Driver would initiate TDLS setup in such cases. 4. Periodic timer is used for updating peer information. 5. Auto TDLS peer list & timer are cleared during disconnection or driver unload. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: wmm support for TDLS linkAvinash Patil2014-11-171-0/+55
| | | | | | | | | | | | | | | | | | | | This patch adds WMM support for TDLS link. Patch add WMM info IE for TDLS setup request/response frames while WMM parameter for TDLS confirm frame. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: do not explicitly disable TDLS link during teardownAvinash Patil2014-11-171-3/+0
| | | | | | | | | | | | | | | | | | | | When Teardown event from FW is indicated to userspace, userspace would trigger tdls_oper handler to disable TDLS link. We need not do this explicitly here. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: update rx packet descriptor structure to match FWAvinash Patil2014-11-171-0/+1
| | | | | | | | | | | | | | | | | | RX packet descriptor structure has recently changed in FW. This patch updates rxpd accordingly. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: fix the assignment of hw queues for mesh interfaceChun-Yeow Yeoh2014-11-171-1/+2
| | | | | | | | | | | | | | | | | | We need to assign the hw queues for mesh interface. Otherwise, we are not able to bring up the mesh interface due to the IEEE80211_INVAL_HW_QUEUE error. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: common-spectral: don't depend from ATH9K_DEBUGFSOleksij Rempel2014-11-171-10/+0
| | | | | | | | | | | | | | | | | | | | | | | | we can have here two variants. Add ATH9K_CMN_DEBUGFS y if ATH9K_CMN_DEBUGFS || ATH9K_HTC_DEBUGFS wich will add more configurations and testcases. Or remove ATH9K_HTC_DEBUGFS which need more time to be done. So, make common-spectral ignore ATH9K_DEBUGFS option for now. Signed-off-by: Oleksij Rempel <linux@rempel-privat.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k_htc: ath9k_htc_op_ps_wakeup() can be statickbuild test robot2014-11-171-3/+3
| | | | | | | | | | | | | | | | | | drivers/net/wireless/ath/ath9k/htc_drv_init.c:56:6: sparse: symbol 'ath9k_htc_op_ps_wakeup' was not declared. Should it be static? drivers/net/wireless/ath/ath9k/htc_drv_init.c:61:6: sparse: symbol 'ath9k_htc_op_ps_restore' was not declared. Should it be static? drivers/net/wireless/ath/ath9k/htc_drv_init.c:66:19: sparse: symbol 'ath9k_htc_ps_ops' was not declared. Should it be static? Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: ath9k_op_ps_wakeup() can be statickbuild test robot2014-11-171-3/+3
| | | | | | | | | | | | | | | | | | drivers/net/wireless/ath/ath9k/init.c:91:6: sparse: symbol 'ath9k_op_ps_wakeup' was not declared. Should it be static? drivers/net/wireless/ath/ath9k/init.c:96:6: sparse: symbol 'ath9k_op_ps_restore' was not declared. Should it be static? drivers/net/wireless/ath/ath9k/init.c:101:19: sparse: symbol 'ath9k_ps_ops' was not declared. Should it be static? Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * mwifiex: module load parameter for interface creationAvinash Patil2014-11-172-0/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds module load parameter driver_mode for mwifiex which would enable driver to create AP or P2P client interface while loading module. driver_mode is bitmap of interface modes for station, AP and P2P client. Station interface is created by default and is unaffected by driver_mode parameter. Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * ath9k: fix misc debugfs when not using chan contextBen Greear2014-11-171-3/+4
| | | | | | | | | | | | | | | | | | | | | | When channel-context is not enabled, all vifs belong to the first context, but it is not configured as 'assigned'. Fix misc debugfs file to print out info for non-assigned contexts, and also print whether ctx is assigned or not. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * Merge branch 'for-linville' of git://github.com/kvalo/athJohn W. Linville2014-11-1713-295/+388
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Kalle Valo <kvalo@qca.qualcomm.com> says: "One ath6kl patch and rest for ath10k, but nothing really major which stands out. Most notable: o fix resume (Bartosz) o firmware restart is now faster and more reliable (Michal) o it's now possible to test hardware restart functionality without crashing the firmware using hw-restart parameter with simulate_fw_crash debugfs file (Michal)" Signed-off-by: John W. Linville <linville@tuxdriver.com>
| | * ath10k: fix mismatched wmi api callPeter Oh2014-11-171-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | Fix to use v10.2 wmi call for firmware v10.2. It turned out that peer association function was using v10.1 wmi call for v10.2 firmware during code review. Signed-off-by: Peter Oh <poh@qca.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: advertise support for AP mode channel width changesRajkumar Manoharan2014-11-171-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | This will enable AP mode to change channel width dynamically based on 20/40 intolerance report sent by associated client. Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: add modpram 'skip_otp' to ignore empty otp error during BMIRajkumar Manoharan2014-11-171-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | This patch would help bring up wifi interface with default board data in case of failures in otp download. It is useful for initial calibration. Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: handle ieee80211 header and payload tracing separatelyRajkumar Manoharan2014-11-174-16/+78
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For packet log, the transmitted frame 802.11 header alone is sufficient. Recording entire packet is also consuming lot of disk space. To optimize this, tx and rx data tracepoints are splitted into header and payload tracepoints. To record tx ieee80211 headers trace-cmd record -e ath10k_tx_hdr To record complete packets trace-cmd record -e ath10k_tx_hdr -e ath10k_tx_payload Cc: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: fix pm resume after suspendBartosz Markowski2014-11-041-4/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Firmware was crashing when we were trying to warm reset it after suspend. This was due to the fact that target registeres can be accessed only if the hardware is awaken. This patch makes sure to awake the device also on the hif up, not only in case of probe call. Signed-off-by: Bartosz Markowski <bartosz.markowski@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath6kl: remove incorrect reset_resume handlerAlexey Khoroshilov2014-11-031-9/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Existing implementation of reset_resume handler just calls ath6kl_usb_remove() that deallocates all resources. It can lead to double free, etc. on disconnect. The patch removes reset_resume handler, so usb core could conservatively reset the driver. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: speed up hw recoveryMichal Kazior2014-10-317-5/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In some cases hw recovery was taking an absurdly long time due to ath10k waiting for things that would never really complete. Instead of waiting for inevitable timeouts poke all completions and wakequeues and check if it's still worth waiting. Reading/writing ar->state requires conf_mutex. Since waiters might be holding it introduce a new flag CRASH_FLUSH so it's possible to tell waiters to abort whatever they were waiting for. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: expose hw restart via debugfsMichal Kazior2014-10-311-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Until now it was possible to simulate soft and hard fw crashes but it wasn't possible to trigger an immediately hw restart itself (without the fw crash). This can be useful when stress testing hw restarting stability, e.g. during heavy tx/rx traffic. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: fix possible bmi crashMichal Kazior2014-10-312-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While testing other things I've found that CE items aren't cleared properly. This could lead to null dereferences in BMI. To prevent that make sure CE revoking clears the nbytes value (which is used as a buffer completion indication) and memset the entire CE ring data shared between host and target when (re)initializing. Also make sure to check BMI xfer pointer and print a splat instead of crashing the kernel. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: don't reset chip on power_downMichal Kazior2014-10-311-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | Currently hif_power_up performs effectively a reset and hif_stop resets the chip as well so there's no point in resetting here. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: split reset logic from power upMichal Kazior2014-10-311-71/+80
| | | | | | | | | | | | | | | | | | | | | | | | The power up procedure was overly complex due to warm/cold reset workarounds and issues. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: make warm reset a bit safer and fasterMichal Kazior2014-10-311-62/+48
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | One of the problems with warm reset I've found is that it must be guaranteed that copy engine registers are not being accessed while being reset. Otherwise in worst case scenario the host may lock up. Instead of using sleeps and hoping the device is operational in some arbitrary timeframes use firmware indication register. As a side effect this makes driver boot/stop/recovery faster. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: change ce ring cleanup logicMichal Kazior2014-10-312-42/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make ath10k_pci_init_pipes() effectively only alter shared target-host data. The per_transfer_context is a host-only thing. It is necessary to preserve it's contents for a more robust ring cleanup. This is required for future warm reset fixes. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: avoid possible deadlock with scan timeoutMichal Kazior2014-10-311-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This should prevent deadlock predicted by the following splat: ====================================================== [ INFO: possible circular locking dependency detected ] 3.17.0-wl-ath+ #67 Not tainted ------------------------------------------------------- kworker/u32:1/7230 is trying to acquire lock: (&ar->conf_mutex){+.+.+.}, at: [<ffffffffa040a57d>] ath10k_scan_timeout_work+0x2d/0x50 [ath10k_core] but task is already holding lock: ((&(&ar->scan.timeout)->work)){+.+...}, at: [<ffffffff8106dae1>] process_one_work+0x151/0x470 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 ((&(&ar->scan.timeout)->work)){+.+...}: [<ffffffff810a12e5>] lock_acquire+0x85/0x100 [<ffffffff8106cb4d>] flush_work+0x3d/0x270 [<ffffffff8106e49d>] __cancel_work_timer+0x7d/0x110 [<ffffffff8106e543>] cancel_delayed_work_sync+0x13/0x20 [<ffffffffa0409f16>] ath10k_cancel_remain_on_channel+0x36/0x60 [ath10k_core] [<ffffffffa028c75c>] ieee80211_cancel_roc+0x1cc/0x2f0 [mac80211] [<ffffffffa028c8a2>] ieee80211_mgmt_tx_cancel_wait+0x22/0x30 [mac80211] [<ffffffffa0132288>] nl80211_tx_mgmt_cancel_wait+0xa8/0x130 [cfg80211] [<ffffffff816654a5>] genl_family_rcv_msg+0x1a5/0x3c0 [<ffffffff81665749>] genl_rcv_msg+0x89/0xc0 [<ffffffff81664e91>] netlink_rcv_skb+0xb1/0xc0 [<ffffffff816650bc>] genl_rcv+0x2c/0x40 [<ffffffff8166474d>] netlink_unicast+0x18d/0x200 [<ffffffff81664add>] netlink_sendmsg+0x31d/0x430 [<ffffffff8161a9ac>] sock_sendmsg+0x9c/0xd0 [<ffffffff8161b469>] ___sys_sendmsg+0x389/0x3a0 [<ffffffff8161bed9>] __sys_sendmsg+0x49/0x90 [<ffffffff8161bf32>] SyS_sendmsg+0x12/0x20 [<ffffffff8174c456>] system_call_fastpath+0x1a/0x1f -> #0 (&ar->conf_mutex){+.+.+.}: [<ffffffff810a0bde>] __lock_acquire+0x1b6e/0x1ce0 [<ffffffff810a12e5>] lock_acquire+0x85/0x100 [<ffffffff817491eb>] mutex_lock_nested+0x4b/0x370 [<ffffffffa040a57d>] ath10k_scan_timeout_work+0x2d/0x50 [ath10k_core] [<ffffffff8106db41>] process_one_work+0x1b1/0x470 [<ffffffff8106df63>] worker_thread+0x123/0x460 [<ffffffff81073f34>] kthread+0xe4/0x100 [<ffffffff8174c3ac>] ret_from_fork+0x7c/0xb0 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock((&(&ar->scan.timeout)->work)); lock(&ar->conf_mutex); lock((&(&ar->scan.timeout)->work)); lock(&ar->conf_mutex); *** DEADLOCK *** Reported-by: Marek Puzyniak <marek.puzyniak@tieto.com> Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: call correct function for frag thresholdMichal Kazior2014-10-241-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rts threshold was being configured instead of fragmentation threshold. Keep in mind available firmware binaries don't seem to support fragmentation anyway so this doesn't fix fragmentation threshold per se. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: add SURVEY_INFO_IN_USE for current channel on surveyFelix Fietkau2014-10-241-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When user space requests survey info, it is useful to know which of the survey data refers to the channel that is currently actively being used. One of the use cases is getting the current channel noise for status output. Without this flag you would have to look up the channel separately and then compare it against the frequency in the survey output in user space. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: re-use trace classMichal Kazior2014-10-241-27/+5
| | | | | | | | | | | | | | | | | | | | | | | | Instead of defining a completely new tracepoint use an existing tracepoint class. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: remove tsf argument from rx_desc tracingMichal Kazior2014-10-242-9/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fundamentally this was wrong. Tsf is only valid in last MPDU of a PPDU. This means tsf value was wrong most of the time during heavy traffic. Also I don't see much point in exposing a redundant (and broken) tsf value. Userspace can already read it from the dumped rx descriptor buffer. Cc: Rajkumar Manoharan <rmanohar@qti.qualcomm.com> Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: clear htt->rx_confused on loadMichal Kazior2014-10-241-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Once driver entered the rx_confused state it would refuse to rx even after firmware is restarted. Make sure to clear it so that rx works after, e.g. hw restart or after all interfaces are stopped. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: don't forget to replenish after fragmented RxMichal Kazior2014-10-241-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In theory it was possible to drain entire HTT Rx ring via fragmented Rx leading to Rx lockup. In practice non-data traffic would always trigger replenishment via the regular Rx handler. For correctness sake make sure to replenish the ring on fragmented Rx. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: add extra sanity check when popping amsduMichal Kazior2014-10-241-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | The netbuf pop can return NULL. Make sure to check for that. It shouldn't happen but better safe than sorry. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: don't drop frames aggressivelyMichal Kazior2014-10-241-15/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There's little point in dropping, e.g. frames with FCS error early in ath10k. This simplifies amsdu_allowed() and gets rid of htt_rx_mpdu_status usage finally. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
| | * ath10k: deduplicate htt rx dma unmappingMichal Kazior2014-10-241-21/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | Treat non-chained and chained popping the same way. Also this makes netbuf pop fully symmetrical to (re)filling. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-29 03:16:51 +0100