- 04 Feb, 2011 25 commits
-
-
Felix Fietkau authored
I can't think of a valid use case for this aside from debugging (which can also be done with a real monitor interface), and dropping these frames saves some precious CPU cycles. Signed-off-by:
Felix Fietkau <nbd@openwrt.org> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Jouni Malinen authored
All register writes to the key cache have to be done in pairs. However, the clearing of a separate MIC entry with hardware revisions that use combined MIC key layout did not do that with one of the registers. Add the matching register write to the following register to make the KEY4 register write actually complete. This is mostly a fix for a theoretical issue since the incorrect entry that could potentially be left behind in the key cache would not match with received frames. Anyway, better make this code clean the entry correctly using paired register writes. Signed-off-by:
Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by:
-
Rajkumar Manoharan authored
While leaving oper channel, STA informs sleep state to AP to stop sending data. Till sending ack for the nullfunc, AP continues to send the data to STA which restarts ps_timer that is causing unnecessary nullfunc exchange on timer expiry when the STA was already moved to offchannel. So don't restart ps_timer on data reception during scan. This issue was identified by the following warning. WARNING: at net/mac80211/tx.c:661 invoke_tx_handlers+0xf07/0x1330 [mac80211] wlan0: Dropped data frame as no usable bitrate found while scanning and associated. Target station: 00:03:7f:0b:a6:1b on 5 GHz band Call Trace: [<ffffffffa0413ba7>] invoke_tx_handlers+0xf07/0x1330 [mac80211] [<ffffffffa0414056>] ieee80211_tx+0x86/0x2c0 [mac80211] [<ffffffffa0414345>] ieee80211_xmit+0xb5/0x1d0 [mac80211] [<ffffffffa04037e0>] ieee80211_dynamic_ps_enable_work+0x0/0xb0 [mac80211] [<ffffffffa04158cf>] ieee80211_tx_skb+0x4f/0x60 [mac80211] [<ffffffffa04026e6>] ieee80211_send_nullfunc+0x46/0x60 [mac80211] [<ffffffffa0403885>] ieee80211_dynamic_ps_enable_work+0xa5/0xb0 [mac80211] Reviewed-by:
Johannes Berg <johannes@sipsolutions.net> Signed-off-by:
Rajkumar Manoharan <rmanoharan@atheros.com> Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
Jussi Kivilinna <jussi.kivilinna@mbnet.fi> Signed-off-by:
-
Jussi Kivilinna authored
It should be safe to enable AP-mode now. Signed-off-by:
-
Jussi Kivilinna authored
RX appears to freeze while idle. Resetting rx-urbs appears to be enough to fix this. Do reset 30 seconds after last rx. Signed-off-by:
-
Jussi Kivilinna authored
When driver fails to acquire device semaphore lock, device usually freezes soon afterwards. So failing to acquire lock indicates us that not everything is going right in device/fw. So reset device when this happens. Signed-off-by:
-
Jussi Kivilinna authored
When doing transfers at high speed for long time, tx queue can freeze. So add tx watchdog. TX-watchdog checks for locked tx-urbs and reset hardware when such is detected. Merely unlinking urb was not enough, device have to be reseted. Hw settings are restored so that any open link will stay on after reset. Signed-off-by:
-
Jussi Kivilinna authored
We need HW hard reset later in patchset to reset device after TX-stall. Collect all settings that we have set to driver for later reset and add restore function. Signed-off-by:
-
Jussi Kivilinna authored
Device command timeouts are set up very high (1 sec) and this causes AP beacon to lock up for long for example. Checking timeouts on device it's easy to see that 1 sec timeout is not needed, when device fails to response longer timeout doesn't help: [ 473.074419] zd1211rw 1-1:1.0: print_times() Read times: [ 473.175163] zd1211rw 1-1:1.0: print_time() 0 - 10 msec: 1506 [ 473.176429] zd1211rw 1-1:1.0: print_time() 11 - 50 msec: 0 [ 473.177955] zd1211rw 1-1:1.0: print_time() 51 - 100 msec: 0 [ 473.180703] zd1211rw 1-1:1.0: print_time() 101 - 250 msec: 0 [ 473.182101] zd1211rw 1-1:1.0: print_time() 251 - 500 msec: 0 [ 473.183221] zd1211rw 1-1:1.0: print_time() 500 - 1000 msec: 20 [ 473.184381] zd1211rw 1-1:1.0: print_time() 1000 - ... msec: 18 Also vendor driver doesn't use this long timeout. Therefore change timeout to 50msec. Signed-off-by:
-
Jussi Kivilinna authored
As might lower beacon update CPU usage. Signed-off-by:
-
Jussi Kivilinna authored
Use stack for allocing small < 64 byte arrays in zd_chip.c and preallocated buffer in zd_usb.c. This might lower CPU usage for beacon setup. v2: - Do not use stack buffers in zd_usb.c as they would be used for urb transfer_buffer. Signed-off-by:
-
Jussi Kivilinna authored
Beacon config function writes beacon to hw one write per byte. This is very slow (usually taking more than 100ms to finish) and causes high CPU usage when in AP-mode (kworker at ~50% on Intel Atom N270). By batching commands together zd_mac_config_beacon() runtime can be lowered to 1/5th and lower CPU usage to saner levels (<10% on Atom). Signed-off-by:
-
Jussi Kivilinna authored
When doing tx/rx at high packet rate (for example simply using ping -f), device starts to fail to respond to control messages. On non-AP modes this only causes problems for LED updating code but when we are running in AP-mode we are writing new beacon to HW usually every 100ms. Now if control message fails in HW beacon setup, device lock is kept locked and beacon data partially written. This can and usually does cause: 1. HW beacon setup fail now on, as driver cannot acquire device lock. 2. Beacon-done interrupt stop working as device has incomplete beacon. Therefore make zd_mac_config_beacon() always try to release device lock and add beacon watchdog to restart beaconing when stall is detected. Also fix zd_mac_config_beacon() try acquiring device lock for max 500ms, as what old code appeared to be trying to do using loop and msleep(1). Signed-off-by:
-
Christian Lamparter authored
On review of 'zd1211rw: implement beacon fetching and handling ieee80211_get_buffered_bc()', Christian Lamparter noted that [1]: Since zd_beacon_done also uploads the next beacon so long in advance, there could be an equally long race between the outdated state of the next beacon's DTIM broadcast traffic indicator (802.11-2007 7.3.2.6) which -in your case- was uploaded almost a beacon interval ago and the xmit of ieee80211_get_buffered_bc *now*. The dtim bc/mc bit might be not set, when a mc/bc arrived after the beacon was uploaded, but before the "beacon done event" from the hardware. So, dozing stations don't expect the broadcast traffic and of course, they might miss it completely. It's probably better to fix this in mac80211 (see the attached hack). [1] http://marc.info/?l=linux-wireless&m=129435041117256&w=2 CC: Christian Lamparter <chunkeey@googlemail.com> Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
-
Jussi Kivilinna authored
Add support for AP-mode beacon. Also disable beacon when interface is set down as otherwise hw will keep flooding NEXT_BCN interrupts. Signed-off-by:
-
Jussi Kivilinna authored
For reasons not very clear yet to me, filter_ack leaves matching tx-packet pending with 'ack_pending'. This causes tx-packet to be passed back to upper layer after next packet has been transfered and tx-packets might end up coming come out of monitor interface in wrong order vs. rx. Because of this when enable AP-mode, hostapd monitor interface would get packets in wrong order causing problems in WPA association. So don't use mac->ack_pending when in AP-mode. Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
-
Jussi Kivilinna authored
As bss_info_changed may sleep, we can as well set RTS_CTS register right away. Keep mac->short_preamble for later use (hw reset). Signed-off-by:
-
Jussi Kivilinna authored
Workers not needed anymore since configure_filter may sleep. Keep mac->multicast_hash for later use (hw reset). Signed-off-by:
Jussi Kivilinna <jussi.kivilina@mbnet.fi> Signed-off-by:
-
Jussi Kivilinna authored
Vendor driver uses CR_BNC_INTERVAL at various places, one is HW_EnableBeacon() that combinies beacon interval with BSS-type flag and DTIM value in upper 16bit of u32. The other one is HW_UpdateBcnInterval() that set_aw_pt_bi() appears to be based on. HW_UpdateBcnInterval() takes interval argument as u16 and uses that for calculations, set_aw_pt_bi() uses u32 value that has flags and dtim in upper part. This clearly seems wrong. Also HW_UpdateBcnInterval() updates only lower 16bit part of CR_BNC_INTERVAL. So make set_aw_pt_bi() do calculations on only lower u16 part of s->beacon_interval. Also set 32bit beacon interval register before reading values from device, as HW_EnableBeacon() on vendor driver does. This is required to make beacon work on AP-mode, simply reading and then writing updated values is not enough at least with zd1211b. Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
-
Jussi Kivilinna authored
OOPS if worker is running and disconnect() is called (triggered by unpluging device). Much harder to trigger at this stage but later when we have AP beacon work in process_intr it happens very easy. Signed-off-by:
-
Jussi Kivilinna authored
When stress testing AP-mode I hit OOPS when unpluging or rmmodding driver. It appears that when tx-queue is disabled, tx-urbs might be left pending. These can cause ehci to call non-existing tx_urb_complete() (after rmmod) or uninitialized/reseted private structure (after disconnect()). Add skb queue for submitted packets and unlink pending urbs on zd_usb_disable_tx(). Part of the problem seems to be usb->free_urb_list that isn't always working as it should, causing machine freeze when trying to free the list in zd_usb_disable_tx(). Caching free urbs isn't what other drivers seem to be doing (usbnet for example) so strip free_usb_list. Patch makes tx-urb handling saner with use of urb anchors. Signed-off-by:
-
- 03 Feb, 2011 15 commits
-
-
Jouni Malinen authored
TKIP countermeasures depend on devices being able to detect Michael MIC failures on received frames and for stations to report errors to the AP. In order to test that behavior, it is useful to be able to send out TKIP frames with incorrect Michael MIC. This testing behavior has minimal effect on the TX path, so it can be added to mac80211 for convenient use. The interface for using this functionality is a file in mac80211 netdev debugfs (tkip_mic_test). Writing a MAC address to the file makes mac80211 generate a dummy data frame that will be sent out using invalid Michael MIC value. In AP mode, the address needs to be for one of the associated stations or ff:ff:ff:ff:ff:ff to use a broadcast frame. In station mode, the address can be anything, e.g., the current BSSID. It should be noted that this functionality works correctly only when associated and using TKIP. Signed-off-by:
-
Jouni Malinen authored
The TKIP implementation was originally prepared to be a bit more flexible in the way Michael MIC TX/RX keys are configured. However, we are now taking care of the TX/RX MIC key swapping in user space, so this code will not be needed. Similarly, there were some remaining WPA testing code that won't be used in their current form. Remove the unneeded extra complexity. Signed-off-by:
-
Justin P. Mattock authored
The below patch fixes a typo comamnd to command. Signed-off-by:
Justin P. Mattock <justinmattock@gmail.com> Signed-off-by:
-
Rajkumar Manoharan authored
The beaconing vif addition is based on max beacon slot available. So it is better to reserve a beacon slot on interface addition and let it be configured properly on bss_info change. Signed-off-by:
-
Johannes Berg authored
There's no point in disallowing scanning for a GO interface when it's not beaconing yet. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
Arik Nemtsov authored
Beacons from external BSSes are required for updating overlapping BSS info (i.e. ERP protection). Pass them up unconditionally. Signed-off-by:
Arik Nemtsov <arik@wizery.com> Signed-off-by:
-
Arik Nemtsov authored
When operating in AP mode the wl1271 hardware filters out null-data packets as well as management packets. This makes it impossible for mac80211 to monitor the PS mode by using the PM bit of incoming frames. Implement a HW flag to indicate that mac80211 should ignore the PM bit. In addition, expose ieee80211_sta_ps_transition() to make low-level drivers capable of controlling PS-mode. Signed-off-by:
-
Arik Nemtsov authored
When rate-control is performed in HW, we cannot calculate frame duration as we do not have the skb transmission rate in SW. ieee80211_tx_h_calculate_duration() should only be called when ieee80211_tx_h_rate_ctrl() has been called before to initialize data in skb->cb. This doesn't happen for drivers with HW rate-control. Fixes the following warning when operating in AP-mode in a driver with HW rate-control. WARNING: at net/mac80211/tx.c:57 ieee80211_duration+0x54/0x1d8 [mac80211]() Modules linked in: wl1271_sdio wl1271 firmware_class crc7 mac80211 cfg80211 [<c0046090>] (unwind_backtrace+0x0/0x124) from [<c0064c10>] (warn_slowpath_common+0x4c/0x64) [<c0064c10>] (warn_slowpath_common+0x4c/0x64) from [<c0064c40>] (warn_slowpath_null+0x18/0x1c) [<c0064c40>] (warn_slowpath_null+0x18/0x1c) from [<bf040e34>] (ieee80211_duration+0x54/0x1d8 [mac80211]) [<bf040e34>] (ieee80211_duration+0x54/0x1d8 [mac80211]) from [<bf04200c>] (invoke_tx_handlers+0xfa0/0x1088 [mac80211]) [<bf04200c>] (invoke_tx_handlers+0xfa0/0x1088 [mac80211]) from [<bf042178>] (ieee80211_tx+0x84/0x248 [mac80211]) [<bf042178>] (ieee80211_tx+0x84/0x248 [mac80211]) from [<bf042f44>] (ieee80211_tx_pending+0x12c/0x278 [mac80211]) [<bf042f44>] (ieee80211_tx_pending+0x12c/0x278 [mac80211]) from [<c0069a9c>] (tasklet_action+0x68/0xbc) [<c0069a9c>] (tasklet_action+0x68/0xbc) from [<c006a044>] (__do_softirq+0x84/0x114) [<c006a044>] (__do_softirq+0x84/0x114) from [<c006a1b8>] (do_softirq+0x48/0x54) [<c006a1b8>] (do_softirq+0x48/0x54) from [<c006a4f8>] (local_bh_enable+0x98/0xcc) [<c006a4f8>] (local_bh_enable+0x98/0xcc) from [<bf074e60>] (wl1271_rx+0x2e8/0x3a4 [wl1271]) [<bf074e60>] (wl1271_rx+0x2e8/0x3a4 [wl1271]) from [<bf071ae4>] (wl1271_irq_work+0x230/0x310 [wl1271]) [<bf071ae4>] (wl1271_irq_work+0x230/0x310 [wl1271]) from [<c0076864>] (process_one_work+0x208/0x350) [<c0076864>] (process_one_work+0x208/0x350) from [<c0076e14>] (worker_thread+0x1cc/0x300) [<c0076e14>] (worker_thread+0x1cc/0x300) from [<c007bb88>] (kthread+0x84/0x8c) [<c007bb88>] (kthread+0x84/0x8c) from [<c0041494>] (kernel_thread_exit+0x0/0x8) Signed-off-by:
-
Ben Greear authored
When a vif goes away, it could cause the super-chan to be recalculated differently, so do that calculation on iface removal. Signed-off-by:
Ben Greear <greearb@candelatech.com> Acked-by:
-
Ben Greear authored
Signed-off-by:
-
Rajkumar Manoharan authored
The opmode recalculation is accessing hw registers. When it is called from remove interface callback and if there are no vifs present then hw is moved to FULL SLEEP by radio disable. So use power save wrappers before accessing hw registers in calculating opmode state. Signed-off-by:
-
Rajkumar Manoharan authored
Signed-off-by:
-
Rajkumar Manoharan authored
move ath_update_txpow to common to remove code duplication in both ath9k & ath9k_htc. Signed-off-by:
-
Rajkumar Manoharan authored
Signed-off-by:
-
Rajkumar Manoharan authored
ani work is cancelled in dissaoctiation. But in some cases during suspend, deauthention never be called. So we failed to stop ani work which was identified by the following warning. Call Trace: [<ffffffffa0454a1d>] ieee80211_can_queue_work.clone.17+0x2d/0x40 [mac80211] [<ffffffffa0454a60>] ieee80211_queue_delayed_work+0x30/0x60 [mac80211] [<ffffffffa0567f82>] ath9k_ani_work+0x142/0x250 [ath9k_htc] [<ffffffff81073c70>] async_run_entry_fn+0x0/0x180 [<ffffffffa0567e40>] ath9k_ani_work+0x0/0x250 [ath9k_htc] Signed-off-by:
-
