- 04 Jan, 2011 10 commits
-
-
Christian Lamparter authored
By removing two "safety" msleeps (and an echo nop), the channel change delay is effectively halved. Previously, the delay could be as long as 260 ms and the device could not go off-channel without risking to miss the next DTIM beacon [interval ~307 ms]. Signed-off-by:
Christian Lamparter <chunkeey@googlemail.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Christian Lamparter authored
This patch adds a forgotten bail-out path. Signed-off-by:
-
Bob Copeland authored
ath5k_reset is called from process context and takes the cc_lock with plain spin_lock(), but cc_lock can also be taken from tasklets in softirq context. Thus we need to at least use spin_lock_bh. This fixes the following lockdep warning: [ 19.967874] sky2 0000:01:00.0: eth0: enabling interface [ 19.982761] ieee80211 phy0: device now idle [ 20.904809] NET: Registered protocol family 17 [ 21.243857] ieee80211 phy0: device no longer idle - scanning [ 21.404343] [ 21.404346] ================================= [ 21.404450] [ INFO: inconsistent lock state ] [ 21.404518] 2.6.37-rc7-wl+ #242 [ 21.404582] --------------------------------- [ 21.404650] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. [ 21.404721] kworker/u:4/982 [HC0[0]:SC1[3]:HE1:SE0] takes: [ 21.404792] (&(&common->cc_lock)->rlock){+.?...}, at: [<f8115780>] ath5k_ani_calibration+0x29/0x5d9 [ath5k] [ 21.405011] {SOFTIRQ-ON-W} state was registered at: [ 21.405011] [<c105cadd>] __lock_acquire+0x62f/0x13c1 [ 21.405011] [<c105d944>] lock_acquire+0xd5/0xf1 [ 21.405011] [<c12c978d>] _raw_spin_lock+0x45/0x72 [ 21.405011] [<f8111533>] ath5k_reset+0x2c0/0x349 [ath5k] [ 21.405011] [<f8111a7a>] ath5k_start+0xb8/0x139 [ath5k] [ 21.405011] [<f849c714>] ieee80211_do_open+0x13f/0x819 [mac80211] [ 21.405011] [<f849ce51>] ieee80211_open+0x63/0x66 [mac80211] [ 21.405011] [<c1258b2e>] __dev_open+0x8d/0xb6 [ 21.405011] [<c1255c64>] __dev_change_flags+0x9d/0x114 [ 21.405011] [<c1258a75>] dev_change_flags+0x18/0x44 [ 21.405011] [<c1262990>] do_setlink+0x23f/0x521 [ 21.405011] [<c1262d58>] rtnl_setlink+0xe6/0xea [ 21.405011] [<c126347c>] rtnetlink_rcv_msg+0x18a/0x1a0 [ 21.405011] [<c126d5f0>] netlink_rcv_skb+0x35/0x7b [ 21.405011] [<c12632eb>] rtnetlink_rcv+0x20/0x27 [ 21.405011] [<c126d370>] netlink_unicast+0x1bb/0x21e [ 21.405011] [<c126db21>] netlink_sendmsg+0x23b/0x288 [ 21.405011] [<c124823c>] sock_sendmsg+0xac/0xc4 [ 21.405011] [<c1248680>] sys_sendmsg+0x152/0x1a2 [ 21.405011] [<c1249b0d>] sys_socketcall+0x214/0x275 [ 21.405011] [<c10029d0>] sysenter_do_call+0x12/0x36 [ 21.405011] irq event stamp: 138032 [ 21.405011] hardirqs last enabled at (138032): [<c12ca252>] _raw_spin_unlock_irqrestore+0x3b/0x5e [ 21.405011] hardirqs last disabled at (138031): [<c12c98cc>] _raw_spin_lock_irqsave+0x18/0x7e [ 21.405011] softirqs last enabled at (138024): [<f84a570e>] ieee80211_tx_skb+0x47/0x49 [mac80211] [ 21.405011] softirqs last disabled at (138027): [<c100452b>] do_softirq+0x63/0xb4 [ 21.405011] [ 21.405011] other info that might help us debug this: [ 21.405011] 3 locks held by kworker/u:4/982: [ 21.405011] #0: (name){+.+.+.}, at: [<c1046158>] process_one_work+0x1b8/0x41b [ 21.405011] #1: ((&(&local->scan_work)->work)){+.+.+.}, at: [<c1046158>] process_one_work+0x1b8/0x41b [ 21.405011] #2: (&local->mtx){+.+.+.}, at: [<f84920fb>] ieee80211_scan_work+0x32/0x4a4 [mac80211] [ 21.405011] [ 21.405011] stack backtrace: [ 21.405011] Pid: 982, comm: kworker/u:4 Not tainted 2.6.37-rc7-wl+ #242 [ 21.405011] Call Trace: [ 21.405011] [<c12c6e68>] ? printk+0x1d/0x25 [ 21.405011] [<c105a742>] print_usage_bug+0x181/0x18b [ 21.405011] [<c105b196>] ? check_usage_forwards+0x0/0xb6 [ 21.405011] [<c105a9ec>] mark_lock+0x2a0/0x4aa [ 21.405011] [<c1059f6d>] ? trace_hardirqs_off+0xb/0xd [ 21.405011] [<c105ca68>] __lock_acquire+0x5ba/0x13c1 [ 21.405011] [<c1059eed>] ? trace_hardirqs_off_caller+0x18/0x8d [ 21.405011] [<c1059f6d>] ? trace_hardirqs_off+0xb/0xd [ 21.405011] [<c1050411>] ? local_clock+0x2c/0x4f [ 21.405011] [<c1059e00>] ? save_trace+0x2/0xa0 [ 21.405011] [<c105ac39>] ? mark_held_locks+0x43/0x5b [ 21.405011] [<c12ca252>] ? _raw_spin_unlock_irqrestore+0x3b/0x5e [ 21.405011] [<f8115780>] ? ath5k_ani_calibration+0x29/0x5d9 [ath5k] [ 21.405011] [<c105d944>] lock_acquire+0xd5/0xf1 [ 21.405011] [<f8115780>] ? ath5k_ani_calibration+0x29/0x5d9 [ath5k] [ 21.405011] [<c12c9b1a>] _raw_spin_lock_bh+0x4a/0x77 [ 21.405011] [<f8115780>] ? ath5k_ani_calibration+0x29/0x5d9 [ath5k] [ 21.405011] [<f8115780>] ath5k_ani_calibration+0x29/0x5d9 [ath5k] [ 21.405011] [<c105ac39>] ? mark_held_locks+0x43/0x5b [ 21.405011] [<f8113496>] ath5k_tasklet_ani+0x1d/0x27 [ath5k] [ 21.405011] [<c1037304>] tasklet_action+0x96/0x137 [ 21.405011] [<c10379b5>] __do_softirq+0xde/0x1c3 [ 21.405011] [<c10b0cef>] ? arch_get_unmapped_area_topdown+0x3b/0x127 [ 21.405011] [<c10378d7>] ? __do_softirq+0x0/0x1c3 [ 21.405011] <IRQ> [<c1036dee>] ? irq_exit+0x3d/0x49 [ 21.405011] [<c1003b4f>] ? do_IRQ+0x98/0xac [ 21.405011] [<c1002eee>] ? common_interrupt+0x2e/0x34 [ 21.405011] [<c103007b>] ? sys_unshare+0x57/0x226 [ 21.405011] [<c1047fee>] ? queue_delayed_work+0x1/0x27 [ 21.405011] [<f84a83a0>] ? ieee80211_queue_delayed_work+0x2e/0x33 [mac80211] [ 21.405011] [<f8492528>] ? ieee80211_scan_work+0x45f/0x4a4 [mac80211] [ 21.405011] [<c104620e>] ? process_one_work+0x26e/0x41b [ 21.405011] [<c1046158>] ? process_one_work+0x1b8/0x41b [ 21.405011] [<f84920c9>] ? ieee80211_scan_work+0x0/0x4a4 [mac80211] [ 21.405011] [<c10466b6>] ? worker_thread+0x18a/0x2a5 [ 21.405011] [<c12ca25e>] ? _raw_spin_unlock_irqrestore+0x47/0x5e [ 21.405011] [<c104652c>] ? worker_thread+0x0/0x2a5 [ 21.405011] [<c104abe7>] ? kthread+0x67/0x6c [ 21.405011] [<c104ab80>] ? kthread+0x0/0x6c [ 21.405011] [<c1002efa>] ? kernel_thread_helper+0x6/0x10 Signed-off-by:Bob Copeland <me@bobcopeland.com> Acked-by:
Bruno Randolf <br1@einfach.org> Signed-off-by:
-
Brian Prodoehl authored
For the AR9002, the spur frequency read from the EEPROM is mangled before being compared against AR_NO_SPUR. This results in the driver trying to set up the spur mitigation for bogus spurs, rather than cleanly breaking out. Signed-off-by:
Brian Prodoehl <bprodoehl@nomadio.net> Signed-off-by:
-
Akinobu Mita authored
Simplify write file operation for /proc files by using simple_write_to_buffer(). Signed-off-by:
Akinobu Mita <akinobu.mita@gmail.com> Cc: "John W. Linville" <linville@tuxdriver.com> Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by:
-
Senthil Balasubramanian authored
Disabling BH is not required while running from a tasklet context and so replace spin_lock_bh with just spin_lock. Signed-off-by:
Senthil Balasubramanian <senthilkumar@atheros.com> Signed-off-by:
-
Stanislaw Gruszka authored
Looks that we do not set correctly antennas when scanning on 5Ghz band and when bluetooth is enabled, because priv->cfg->scan_tx_antennas[band] is only defined for IEEE80211_BAND_2GHZ. To fix we check band before limiting antennas to first one. This allow to remove hard coded cfg->scan_tx_antennas[band]. Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Acked-by:
Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by:
-
Bruno Randolf authored
Move mac80211 functions into new file mac80211-ops.c to have a better separation and to make base.c smaller. Signed-off-by:
-
John W. Linville authored
Merge branch 'wireless-next-2.6' of git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-2.6
-
John W. Linville authored
-
- 26 Dec, 2010 3 commits
-
-
Wey-Yi Guy authored
The correct name should be used for the newer devices, remove reference to Gen2 Signed-off-by: -
Johannes Berg authored
# iw wlan0 interface add moni0 type monitor flags control # ip link set moni0 up causes a continuous spew of FH_ERROR from the device. Fix this by not setting the CTL2HOST filter by itself -- CTL + promisc works fine. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
Wey-Yi Guy authored
For "led_mode" module parameters, string "led_mode" is duplicated twice, remove one. Signed-off-by:
-
- 23 Dec, 2010 7 commits
-
-
Gustavo F. Padovan authored
Does not allow any channel different of HCI_CHANNEL_RAW and HCI_CHANNEL_CONTROL to bind. Signed-off-by:Gustavo F. Padovan <padovan@profusion.mobi>
-
Johan Hedberg authored
The initialization function used by hci_open_dev (hci_init_req) sends many different HCI commands. The __hci_request function should only return when all of these commands have completed (or a timeout occurs). Several of these commands cause hci_req_complete to be called which causes __hci_request to return prematurely. This patch fixes the issue by adding a new hdev->req_last_cmd variable which is set during the initialization procedure. The hci_req_complete function will no longer mark the request as complete until the command matching hdev->req_last_cmd completes. Signed-off-by:
Johan Hedberg <johan.hedberg@nokia.com> Acked-by:
Marcel Holtmann <marcel@holtmann.org> Signed-off-by:
-
Johan Hedberg authored
This patch adds Bluetooth Management interface events for controller addition and removal. The events correspond to the existing HCI_DEV_REG and HCI_DEV_UNREG stack internal events. Signed-off-by:
-
Johan Hedberg authored
This patch implements the read_info command which is used to fetch basic info about an adapter. Signed-off-by:
-
Johan Hedberg authored
This patch implements the read_index_list command through which userspace can get a list of current adapter indices. Signed-off-by:
-
Johan Hedberg authored
This patch implements the initial read_version command that userspace will use before any other management interface operations. Signed-off-by:
-
Johan Hedberg authored
The command handlers for bluetooth management messaging should be able to report errors (such as memory allocation failures) to the higher levels in the call stack. Signed-off-by:
-
- 22 Dec, 2010 20 commits
-
-
Chaoming Li authored
An RX buffer is set to 9100 bytes to receive 8K AMSDU; however, an skb of this size fails in the kernel. Signed-off-by:
Chaoming Li <chaoming_li@realsil.com.cn> Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Johannes Berg authored
There's a redundant rcu_read_lock/unlock pair, a redundant variable, and a few redundant accesses to the 1d_to_ac array. Fix this to make the code neater and easier to follow. Signed-off-by:
-
Mohammed Shafi Shajakhan authored
It looks like some hardware registers are left into undefined state after suspend/resume. At minimum, this can cause odd issues related to key cache and hardware trying to encrypt/decrypt frames unexpectedly. This seems to happen even when there is no keys configured, i.e., hardware can end up touching TX frames just based of invalid key cache context even if the driver is not asking a specific entry to be used. In addition, RX can likely be affected. This patch fixes this issue. Signed-off-by:
Jouni Malinen <Jouni.Malinen@Atheros.com> Signed-off-by:
Mohammed Shafi Shajakhan <mshajakhan@atheros.com> Signed-off-by:
-
Larry Finger authored
If the radio enable switch is off when the driver is loaded, it is not possible to get radio output until the driver is unloaded and reloaded with the switch on. Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
Jussi Kivilinna <jussi.kivilinna@mbnet.fi> Signed-off-by:
-
Jussi Kivilinna authored
Device poller already reads current RSSI, so add support for set_cqm_rssi_config there. Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
-
Jussi Kivilinna authored
Signed-off-by:
-
Jussi Kivilinna authored
Radio should be off when interface is down. Signed-off-by:
-
Jussi Kivilinna authored
rndis_set_default_key did call add_wep_key to set default key on device, even if key is WPA. This caused rndis_wlan not work with wpa_supplicant in nl80211 mode (causing disconnect from AP). Signed-off-by:
-
Jussi Kivilinna authored
Sometimes device returns wrong number of items in bssid-list. Appears that some specific beacons trigger this problem and leads to very poor scanning results. Workaround by ignoring num_items received from device and walkthrough full bssid-list buffer. v2: Fix buffer range checks and reading next item length. Old code read behind buffer on last item but didn't use those values as 'count' would also reach zero. Also fix resizing of buffer if device has larger buffer, old code assumed that BSSID-list OID would return same buffer size when it really can return yet another new larger length. Tested-by:Luís Picciochi <Pitxyoki@gmail.com> Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
Rafał Miłecki <zajec5@gmail.com> Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Rafał Miłecki authored
Signed-off-by:
-
Vasanthakumar Thiagarajan authored
The size of the eeprom data is 1088 bytes for AR9485. But a sanity check is done against 4K which would result in a 'potential read past the end of the buffer' smatch complaint. Reported-by:
Dan Carpenter <error27@gmail.com> Signed-off-by:
Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by:
-
Bruno Randolf authored
Introduce a helper function to get the EEPROM mode from channel and remove multiple similar switch statements. Also since it's now easy to get the EEPROM mode from the channel, use them inside the functions which need it, instead of passing a redundant ee_mode parameter. Signed-off-by:
Nick Kossifidis <mickflemm@gmail.com> Signed-off-by:
-
