OpenCloudOS-Kernel/drivers/net/wireless/ath/ath9k/main.c

2208 lines
55 KiB
C
Raw Normal View History

/*
* Copyright (c) 2008-2009 Atheros Communications Inc.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#include <linux/nl80211.h>
#include <linux/delay.h>
#include "ath9k.h"
#include "btcoex.h"
static u8 parse_mpdudensity(u8 mpdudensity)
{
/*
* 802.11n D2.0 defined values for "Minimum MPDU Start Spacing":
* 0 for no restriction
* 1 for 1/4 us
* 2 for 1/2 us
* 3 for 1 us
* 4 for 2 us
* 5 for 4 us
* 6 for 8 us
* 7 for 16 us
*/
switch (mpdudensity) {
case 0:
return 0;
case 1:
case 2:
case 3:
/* Our lower layer calculations limit our precision to
1 microsecond */
return 1;
case 4:
return 2;
case 5:
return 4;
case 6:
return 8;
case 7:
return 16;
default:
return 0;
}
}
static bool ath9k_has_pending_frames(struct ath_softc *sc, struct ath_txq *txq)
{
bool pending = false;
spin_lock_bh(&txq->axq_lock);
if (txq->axq_depth || !list_empty(&txq->axq_acq))
pending = true;
else if (sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)
pending = !list_empty(&txq->txq_fifo_pending);
spin_unlock_bh(&txq->axq_lock);
return pending;
}
bool ath9k_setpower(struct ath_softc *sc, enum ath9k_power_mode mode)
{
unsigned long flags;
bool ret;
spin_lock_irqsave(&sc->sc_pm_lock, flags);
ret = ath9k_hw_setpower(sc->sc_ah, mode);
spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
return ret;
}
void ath9k_ps_wakeup(struct ath_softc *sc)
{
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
unsigned long flags;
enum ath9k_power_mode power_mode;
spin_lock_irqsave(&sc->sc_pm_lock, flags);
if (++sc->ps_usecount != 1)
goto unlock;
power_mode = sc->sc_ah->power_mode;
ath9k_hw_setpower(sc->sc_ah, ATH9K_PM_AWAKE);
/*
* While the hardware is asleep, the cycle counters contain no
* useful data. Better clear them now so that they don't mess up
* survey data results.
*/
if (power_mode != ATH9K_PM_AWAKE) {
spin_lock(&common->cc_lock);
ath_hw_cycle_counters_update(common);
memset(&common->cc_survey, 0, sizeof(common->cc_survey));
spin_unlock(&common->cc_lock);
}
unlock:
spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
}
void ath9k_ps_restore(struct ath_softc *sc)
{
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
unsigned long flags;
spin_lock_irqsave(&sc->sc_pm_lock, flags);
if (--sc->ps_usecount != 0)
goto unlock;
spin_lock(&common->cc_lock);
ath_hw_cycle_counters_update(common);
spin_unlock(&common->cc_lock);
if (sc->ps_idle)
ath9k_hw_setpower(sc->sc_ah, ATH9K_PM_FULL_SLEEP);
else if (sc->ps_enabled &&
!(sc->ps_flags & (PS_WAIT_FOR_BEACON |
PS_WAIT_FOR_CAB |
PS_WAIT_FOR_PSPOLL_DATA |
PS_WAIT_FOR_TX_ACK)))
ath9k_hw_setpower(sc->sc_ah, ATH9K_PM_NETWORK_SLEEP);
unlock:
spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
}
static void ath_start_ani(struct ath_common *common)
{
struct ath_hw *ah = common->ah;
unsigned long timestamp = jiffies_to_msecs(jiffies);
struct ath_softc *sc = (struct ath_softc *) common->priv;
if (!(sc->sc_flags & SC_OP_ANI_RUN))
return;
if (sc->sc_flags & SC_OP_OFFCHANNEL)
return;
common->ani.longcal_timer = timestamp;
common->ani.shortcal_timer = timestamp;
common->ani.checkani_timer = timestamp;
mod_timer(&common->ani.timer,
jiffies +
msecs_to_jiffies((u32)ah->config.ani_poll_interval));
}
static void ath_update_survey_nf(struct ath_softc *sc, int channel)
{
struct ath_hw *ah = sc->sc_ah;
struct ath9k_channel *chan = &ah->channels[channel];
struct survey_info *survey = &sc->survey[channel];
if (chan->noisefloor) {
survey->filled |= SURVEY_INFO_NOISE_DBM;
survey->noise = chan->noisefloor;
}
}
/*
* Updates the survey statistics and returns the busy time since last
* update in %, if the measurement duration was long enough for the
* result to be useful, -1 otherwise.
*/
static int ath_update_survey_stats(struct ath_softc *sc)
{
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
int pos = ah->curchan - &ah->channels[0];
struct survey_info *survey = &sc->survey[pos];
struct ath_cycle_counters *cc = &common->cc_survey;
unsigned int div = common->clockrate * 1000;
int ret = 0;
if (!ah->curchan)
return -1;
if (ah->power_mode == ATH9K_PM_AWAKE)
ath_hw_cycle_counters_update(common);
if (cc->cycles > 0) {
survey->filled |= SURVEY_INFO_CHANNEL_TIME |
SURVEY_INFO_CHANNEL_TIME_BUSY |
SURVEY_INFO_CHANNEL_TIME_RX |
SURVEY_INFO_CHANNEL_TIME_TX;
survey->channel_time += cc->cycles / div;
survey->channel_time_busy += cc->rx_busy / div;
survey->channel_time_rx += cc->rx_frame / div;
survey->channel_time_tx += cc->tx_frame / div;
}
if (cc->cycles < div)
return -1;
if (cc->cycles > 0)
ret = cc->rx_busy * 100 / cc->cycles;
memset(cc, 0, sizeof(*cc));
ath_update_survey_nf(sc, pos);
return ret;
}
/*
* Set/change channels. If the channel is really being changed, it's done
* by reseting the chip. To accomplish this we must first cleanup any pending
* DMA, then restart stuff.
*/
int ath_set_channel(struct ath_softc *sc, struct ieee80211_hw *hw,
struct ath9k_channel *hchan)
{
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ieee80211_conf *conf = &common->hw->conf;
bool fastcc = true, stopped;
struct ieee80211_channel *channel = hw->conf.channel;
struct ath9k_hw_cal_data *caldata = NULL;
int r;
if (sc->sc_flags & SC_OP_INVALID)
return -EIO;
sc->hw_busy_count = 0;
del_timer_sync(&common->ani.timer);
cancel_work_sync(&sc->paprd_work);
cancel_work_sync(&sc->hw_check_work);
cancel_delayed_work_sync(&sc->tx_complete_work);
cancel_delayed_work_sync(&sc->hw_pll_work);
ath9k_ps_wakeup(sc);
spin_lock_bh(&sc->sc_pcu_lock);
/*
* This is only performed if the channel settings have
* actually changed.
*
* To switch channels clear any pending DMA operations;
* wait long enough for the RX fifo to drain, reset the
* hardware at the new frequency, and then re-enable
* the relevant bits of the h/w.
*/
ath9k_hw_disable_interrupts(ah);
stopped = ath_drain_all_txq(sc, false);
if (!ath_stoprecv(sc))
stopped = false;
if (!ath9k_hw_check_alive(ah))
stopped = false;
/* XXX: do not flush receive queue here. We don't want
* to flush data frames already in queue because of
* changing channel. */
if (!stopped || !(sc->sc_flags & SC_OP_OFFCHANNEL))
fastcc = false;
if (!(sc->sc_flags & SC_OP_OFFCHANNEL))
caldata = &sc->caldata;
ath_dbg(common, ATH_DBG_CONFIG,
"(%u MHz) -> (%u MHz), conf_is_ht40: %d fastcc: %d\n",
sc->sc_ah->curchan->channel,
channel->center_freq, conf_is_ht40(conf),
fastcc);
r = ath9k_hw_reset(ah, hchan, caldata, fastcc);
if (r) {
ath_err(common,
"Unable to reset channel (%u MHz), reset status %d\n",
channel->center_freq, r);
goto ps_restore;
}
if (ath_startrecv(sc) != 0) {
ath_err(common, "Unable to restart recv logic\n");
r = -EIO;
goto ps_restore;
}
ath9k_cmn_update_txpow(ah, sc->curtxpow,
sc->config.txpowlimit, &sc->curtxpow);
ath9k_hw_set_interrupts(ah, ah->imask);
if (!(sc->sc_flags & (SC_OP_OFFCHANNEL))) {
if (sc->sc_flags & SC_OP_BEACONS)
ath_beacon_config(sc, NULL);
ieee80211_queue_delayed_work(sc->hw, &sc->tx_complete_work, 0);
ieee80211_queue_delayed_work(sc->hw, &sc->hw_pll_work, HZ/2);
ath_start_ani(common);
}
ps_restore:
ieee80211_wake_queues(hw);
spin_unlock_bh(&sc->sc_pcu_lock);
ath9k_ps_restore(sc);
return r;
}
static void ath_paprd_activate(struct ath_softc *sc)
{
struct ath_hw *ah = sc->sc_ah;
struct ath9k_hw_cal_data *caldata = ah->caldata;
struct ath_common *common = ath9k_hw_common(ah);
int chain;
if (!caldata || !caldata->paprd_done)
return;
ath9k_ps_wakeup(sc);
ar9003_paprd_enable(ah, false);
for (chain = 0; chain < AR9300_MAX_CHAINS; chain++) {
if (!(common->tx_chainmask & BIT(chain)))
continue;
ar9003_paprd_populate_single_table(ah, caldata, chain);
}
ar9003_paprd_enable(ah, true);
ath9k_ps_restore(sc);
}
static bool ath_paprd_send_frame(struct ath_softc *sc, struct sk_buff *skb, int chain)
{
struct ieee80211_hw *hw = sc->hw;
struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ath_tx_control txctl;
int time_left;
memset(&txctl, 0, sizeof(txctl));
txctl.txq = sc->tx.txq_map[WME_AC_BE];
memset(tx_info, 0, sizeof(*tx_info));
tx_info->band = hw->conf.channel->band;
tx_info->flags |= IEEE80211_TX_CTL_NO_ACK;
tx_info->control.rates[0].idx = 0;
tx_info->control.rates[0].count = 1;
tx_info->control.rates[0].flags = IEEE80211_TX_RC_MCS;
tx_info->control.rates[1].idx = -1;
init_completion(&sc->paprd_complete);
txctl.paprd = BIT(chain);
if (ath_tx_start(hw, skb, &txctl) != 0) {
ath_dbg(common, ATH_DBG_XMIT, "PAPRD TX failed\n");
dev_kfree_skb_any(skb);
return false;
}
time_left = wait_for_completion_timeout(&sc->paprd_complete,
msecs_to_jiffies(ATH_PAPRD_TIMEOUT));
if (!time_left)
ath_dbg(ath9k_hw_common(sc->sc_ah), ATH_DBG_CALIBRATE,
"Timeout waiting for paprd training on TX chain %d\n",
chain);
return !!time_left;
}
void ath_paprd_calibrate(struct work_struct *work)
{
struct ath_softc *sc = container_of(work, struct ath_softc, paprd_work);
struct ieee80211_hw *hw = sc->hw;
struct ath_hw *ah = sc->sc_ah;
struct ieee80211_hdr *hdr;
struct sk_buff *skb = NULL;
struct ath9k_hw_cal_data *caldata = ah->caldata;
struct ath_common *common = ath9k_hw_common(ah);
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
int ftype;
int chain_ok = 0;
int chain;
int len = 1800;
if (!caldata)
return;
if (ar9003_paprd_init_table(ah) < 0)
return;
skb = alloc_skb(len, GFP_KERNEL);
if (!skb)
return;
skb_put(skb, len);
memset(skb->data, 0, len);
hdr = (struct ieee80211_hdr *)skb->data;
ftype = IEEE80211_FTYPE_DATA | IEEE80211_STYPE_NULLFUNC;
hdr->frame_control = cpu_to_le16(ftype);
hdr->duration_id = cpu_to_le16(10);
memcpy(hdr->addr1, hw->wiphy->perm_addr, ETH_ALEN);
memcpy(hdr->addr2, hw->wiphy->perm_addr, ETH_ALEN);
memcpy(hdr->addr3, hw->wiphy->perm_addr, ETH_ALEN);
ath9k_ps_wakeup(sc);
for (chain = 0; chain < AR9300_MAX_CHAINS; chain++) {
if (!(common->tx_chainmask & BIT(chain)))
continue;
chain_ok = 0;
ath_dbg(common, ATH_DBG_CALIBRATE,
"Sending PAPRD frame for thermal measurement "
"on chain %d\n", chain);
if (!ath_paprd_send_frame(sc, skb, chain))
goto fail_paprd;
ar9003_paprd_setup_gain_table(ah, chain);
ath_dbg(common, ATH_DBG_CALIBRATE,
"Sending PAPRD training frame on chain %d\n", chain);
if (!ath_paprd_send_frame(sc, skb, chain))
goto fail_paprd;
if (!ar9003_paprd_is_done(ah))
break;
if (ar9003_paprd_create_curve(ah, caldata, chain) != 0)
break;
chain_ok = 1;
}
kfree_skb(skb);
if (chain_ok) {
caldata->paprd_done = true;
ath_paprd_activate(sc);
}
fail_paprd:
ath9k_ps_restore(sc);
}
/*
* This routine performs the periodic noise floor calibration function
* that is used to adjust and optimize the chip performance. This
* takes environmental changes (location, temperature) into account.
* When the task is complete, it reschedules itself depending on the
* appropriate interval that was calculated.
*/
void ath_ani_calibrate(unsigned long data)
{
struct ath_softc *sc = (struct ath_softc *)data;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
bool longcal = false;
bool shortcal = false;
bool aniflag = false;
unsigned int timestamp = jiffies_to_msecs(jiffies);
u32 cal_interval, short_cal_interval, long_cal_interval;
unsigned long flags;
if (ah->caldata && ah->caldata->nfcal_interference)
long_cal_interval = ATH_LONG_CALINTERVAL_INT;
else
long_cal_interval = ATH_LONG_CALINTERVAL;
short_cal_interval = (ah->opmode == NL80211_IFTYPE_AP) ?
ATH_AP_SHORT_CALINTERVAL : ATH_STA_SHORT_CALINTERVAL;
/* Only calibrate if awake */
if (sc->sc_ah->power_mode != ATH9K_PM_AWAKE)
goto set_timer;
ath9k_ps_wakeup(sc);
/* Long calibration runs independently of short calibration. */
if ((timestamp - common->ani.longcal_timer) >= long_cal_interval) {
longcal = true;
ath_dbg(common, ATH_DBG_ANI, "longcal @%lu\n", jiffies);
common->ani.longcal_timer = timestamp;
}
/* Short calibration applies only while caldone is false */
if (!common->ani.caldone) {
if ((timestamp - common->ani.shortcal_timer) >= short_cal_interval) {
shortcal = true;
ath_dbg(common, ATH_DBG_ANI,
"shortcal @%lu\n", jiffies);
common->ani.shortcal_timer = timestamp;
common->ani.resetcal_timer = timestamp;
}
} else {
if ((timestamp - common->ani.resetcal_timer) >=
ATH_RESTART_CALINTERVAL) {
common->ani.caldone = ath9k_hw_reset_calvalid(ah);
if (common->ani.caldone)
common->ani.resetcal_timer = timestamp;
}
}
/* Verify whether we must check ANI */
ath9k: add new ANI implementation for AR9003 This adds support for ANI for AR9003. The implementation for ANI for AR9003 is slightly different than the one used for the older chipset families. It can technically be used for the older families as well but this is not yet fully tested so we only enable the new ANI for the AR5008, AR9001 and AR9002 families with a module parameter, force_new_ani. The old ANI implementation is left intact. Details of the new ANI implemention: * ANI adjustment logic is now table driven so that each ANI level setting is parameterized. This makes adjustments much more deterministic than the old procedure based logic and allows adjustments to be made incrementally to several parameters per level. * ANI register settings are now relative to INI values; so ANI param zero level == INI value. Appropriate floor and ceiling values are obeyed when adjustments are combined with INI values. * ANI processing is done once per second rather that every 100ms. The poll interval is now a set upon hardware initialization and can be picked up by the core driver. * OFDM error and CCK error processing are made in a round robin fashion rather than allowing all OFDM adjustments to be made before CCK adjustments. * ANI adjusts MRC CCK off in the presence of high CCK errors * When adjusting spur immunity (SI) and OFDM weak signal detection, ANI now sets register values for the extension channel too * When adjusting FIR step (ST), ANI now sets register for FIR step low too * FIR step adjustments now allow for an extra level of immunity for extremely noisy environments * The old Noise immunity setting (NI), which changes coarse low, size desired, etc have been removed. Changing these settings could affect up RIFS RX as well. * CCK weak signal adjustment is no longer used * ANI no longer enables phy error interrupts; in all cases phy hw counting registers are used instead * The phy error count (overflow) interrupts are also no longer used for ANI adjustments. All ANI adjustments are made via the polling routine and no adjustments are possible in the ISR context anymore * A history settings buffer is now correctly used for each channel; channel settings are initialized with the defaults but later changes are restored when returning back to that channel * When scanning, ANI is disabled settings are returned to (INI) defaults. * OFDM phy error thresholds are now 400 & 1000 (errors/second units) for low/high water marks, providing increased stability/hysteresis when changing levels. * Similarly CCK phy error thresholds are now 300 & 600 (errors/second) Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-12 12:33:45 +08:00
if ((timestamp - common->ani.checkani_timer) >=
ah->config.ani_poll_interval) {
aniflag = true;
common->ani.checkani_timer = timestamp;
}
/* Skip all processing if there's nothing to do. */
if (longcal || shortcal || aniflag) {
/* Call ANI routine if necessary */
if (aniflag) {
spin_lock_irqsave(&common->cc_lock, flags);
ath9k_hw_ani_monitor(ah, ah->curchan);
ath_update_survey_stats(sc);
spin_unlock_irqrestore(&common->cc_lock, flags);
}
/* Perform calibration if necessary */
if (longcal || shortcal) {
common->ani.caldone =
ath9k_hw_calibrate(ah,
ah->curchan,
common->rx_chainmask,
longcal);
}
}
ath9k_ps_restore(sc);
set_timer:
/*
* Set timer interval based on previous results.
* The interval must be the shortest necessary to satisfy ANI,
* short calibration and long calibration.
*/
cal_interval = ATH_LONG_CALINTERVAL;
if (sc->sc_ah->config.enable_ani)
ath9k: add new ANI implementation for AR9003 This adds support for ANI for AR9003. The implementation for ANI for AR9003 is slightly different than the one used for the older chipset families. It can technically be used for the older families as well but this is not yet fully tested so we only enable the new ANI for the AR5008, AR9001 and AR9002 families with a module parameter, force_new_ani. The old ANI implementation is left intact. Details of the new ANI implemention: * ANI adjustment logic is now table driven so that each ANI level setting is parameterized. This makes adjustments much more deterministic than the old procedure based logic and allows adjustments to be made incrementally to several parameters per level. * ANI register settings are now relative to INI values; so ANI param zero level == INI value. Appropriate floor and ceiling values are obeyed when adjustments are combined with INI values. * ANI processing is done once per second rather that every 100ms. The poll interval is now a set upon hardware initialization and can be picked up by the core driver. * OFDM error and CCK error processing are made in a round robin fashion rather than allowing all OFDM adjustments to be made before CCK adjustments. * ANI adjusts MRC CCK off in the presence of high CCK errors * When adjusting spur immunity (SI) and OFDM weak signal detection, ANI now sets register values for the extension channel too * When adjusting FIR step (ST), ANI now sets register for FIR step low too * FIR step adjustments now allow for an extra level of immunity for extremely noisy environments * The old Noise immunity setting (NI), which changes coarse low, size desired, etc have been removed. Changing these settings could affect up RIFS RX as well. * CCK weak signal adjustment is no longer used * ANI no longer enables phy error interrupts; in all cases phy hw counting registers are used instead * The phy error count (overflow) interrupts are also no longer used for ANI adjustments. All ANI adjustments are made via the polling routine and no adjustments are possible in the ISR context anymore * A history settings buffer is now correctly used for each channel; channel settings are initialized with the defaults but later changes are restored when returning back to that channel * When scanning, ANI is disabled settings are returned to (INI) defaults. * OFDM phy error thresholds are now 400 & 1000 (errors/second units) for low/high water marks, providing increased stability/hysteresis when changing levels. * Similarly CCK phy error thresholds are now 300 & 600 (errors/second) Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-12 12:33:45 +08:00
cal_interval = min(cal_interval,
(u32)ah->config.ani_poll_interval);
if (!common->ani.caldone)
cal_interval = min(cal_interval, (u32)short_cal_interval);
mod_timer(&common->ani.timer, jiffies + msecs_to_jiffies(cal_interval));
if ((sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_PAPRD) && ah->caldata) {
if (!ah->caldata->paprd_done)
ieee80211_queue_work(sc->hw, &sc->paprd_work);
else if (!ah->paprd_table_write_done)
ath_paprd_activate(sc);
}
}
static void ath_node_attach(struct ath_softc *sc, struct ieee80211_sta *sta)
{
struct ath_node *an;
struct ath_hw *ah = sc->sc_ah;
an = (struct ath_node *)sta->drv_priv;
#ifdef CONFIG_ATH9K_DEBUGFS
spin_lock(&sc->nodes_lock);
list_add(&an->list, &sc->nodes);
spin_unlock(&sc->nodes_lock);
an->sta = sta;
#endif
if ((ah->caps.hw_caps) & ATH9K_HW_CAP_APM)
sc->sc_flags |= SC_OP_ENABLE_APM;
if (sc->sc_flags & SC_OP_TXAGGR) {
ath_tx_node_init(sc, an);
an->maxampdu = 1 << (IEEE80211_HT_MAX_AMPDU_FACTOR +
sta->ht_cap.ampdu_factor);
an->mpdudensity = parse_mpdudensity(sta->ht_cap.ampdu_density);
}
}
static void ath_node_detach(struct ath_softc *sc, struct ieee80211_sta *sta)
{
struct ath_node *an = (struct ath_node *)sta->drv_priv;
#ifdef CONFIG_ATH9K_DEBUGFS
spin_lock(&sc->nodes_lock);
list_del(&an->list);
spin_unlock(&sc->nodes_lock);
an->sta = NULL;
#endif
if (sc->sc_flags & SC_OP_TXAGGR)
ath_tx_node_cleanup(sc, an);
}
void ath_hw_check(struct work_struct *work)
{
struct ath_softc *sc = container_of(work, struct ath_softc, hw_check_work);
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
unsigned long flags;
int busy;
ath9k_ps_wakeup(sc);
if (ath9k_hw_check_alive(sc->sc_ah))
goto out;
spin_lock_irqsave(&common->cc_lock, flags);
busy = ath_update_survey_stats(sc);
spin_unlock_irqrestore(&common->cc_lock, flags);
ath_dbg(common, ATH_DBG_RESET, "Possible baseband hang, "
"busy=%d (try %d)\n", busy, sc->hw_busy_count + 1);
if (busy >= 99) {
if (++sc->hw_busy_count >= 3)
ath_reset(sc, true);
} else if (busy >= 0)
sc->hw_busy_count = 0;
out:
ath9k_ps_restore(sc);
}
void ath9k_tasklet(unsigned long data)
{
struct ath_softc *sc = (struct ath_softc *)data;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
u32 status = sc->intrstatus;
u32 rxmask;
if (status & ATH9K_INT_FATAL) {
ath_reset(sc, true);
return;
}
ath9k_ps_wakeup(sc);
spin_lock(&sc->sc_pcu_lock);
/*
* Only run the baseband hang check if beacons stop working in AP or
* IBSS mode, because it has a high false positive rate. For station
* mode it should not be necessary, since the upper layers will detect
* this through a beacon miss automatically and the following channel
* change will trigger a hardware reset anyway
*/
if (ath9k_hw_numtxpending(ah, sc->beacon.beaconq) != 0 &&
!ath9k_hw_check_alive(ah))
ieee80211_queue_work(sc->hw, &sc->hw_check_work);
if (ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)
rxmask = (ATH9K_INT_RXHP | ATH9K_INT_RXLP | ATH9K_INT_RXEOL |
ATH9K_INT_RXORN);
else
rxmask = (ATH9K_INT_RX | ATH9K_INT_RXEOL | ATH9K_INT_RXORN);
if (status & rxmask) {
/* Check for high priority Rx first */
if ((ah->caps.hw_caps & ATH9K_HW_CAP_EDMA) &&
(status & ATH9K_INT_RXHP))
ath_rx_tasklet(sc, 0, true);
ath_rx_tasklet(sc, 0, false);
}
if (status & ATH9K_INT_TX) {
if (ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)
ath_tx_edma_tasklet(sc);
else
ath_tx_tasklet(sc);
}
if ((status & ATH9K_INT_TSFOOR) && sc->ps_enabled) {
/*
* TSF sync does not look correct; remain awake to sync with
* the next Beacon.
*/
ath_dbg(common, ATH_DBG_PS,
"TSFOOR - Sync with next Beacon\n");
sc->ps_flags |= PS_WAIT_FOR_BEACON | PS_BEACON_SYNC;
}
if (ah->btcoex_hw.scheme == ATH_BTCOEX_CFG_3WIRE)
if (status & ATH9K_INT_GENTIMER)
ath_gen_timer_isr(sc->sc_ah);
/* re-enable hardware interrupt */
ath9k_hw_enable_interrupts(ah);
spin_unlock(&sc->sc_pcu_lock);
ath9k_ps_restore(sc);
}
irqreturn_t ath_isr(int irq, void *dev)
{
#define SCHED_INTR ( \
ATH9K_INT_FATAL | \
ATH9K_INT_RXORN | \
ATH9K_INT_RXEOL | \
ATH9K_INT_RX | \
ATH9K_INT_RXLP | \
ATH9K_INT_RXHP | \
ATH9K_INT_TX | \
ATH9K_INT_BMISS | \
ATH9K_INT_CST | \
ATH9K_INT_TSFOOR | \
ATH9K_INT_GENTIMER)
struct ath_softc *sc = dev;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
enum ath9k_int status;
bool sched = false;
/*
* The hardware is not ready/present, don't
* touch anything. Note this can happen early
* on if the IRQ is shared.
*/
if (sc->sc_flags & SC_OP_INVALID)
return IRQ_NONE;
/* shared irq, not for us */
if (!ath9k_hw_intrpend(ah))
return IRQ_NONE;
/*
* Figure out the reason(s) for the interrupt. Note
* that the hal returns a pseudo-ISR that may include
* bits we haven't explicitly enabled so we mask the
* value to insure we only process bits we requested.
*/
ath9k_hw_getisr(ah, &status); /* NB: clears ISR too */
status &= ah->imask; /* discard unasked-for bits */
/*
* If there are no status bits set, then this interrupt was not
* for me (should have been caught above).
*/
if (!status)
return IRQ_NONE;
/* Cache the status */
sc->intrstatus = status;
if (status & SCHED_INTR)
sched = true;
/*
* If a FATAL or RXORN interrupt is received, we have to reset the
* chip immediately.
*/
if ((status & ATH9K_INT_FATAL) || ((status & ATH9K_INT_RXORN) &&
!(ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)))
goto chip_reset;
if ((ah->caps.hw_caps & ATH9K_HW_CAP_EDMA) &&
(status & ATH9K_INT_BB_WATCHDOG)) {
spin_lock(&common->cc_lock);
ath_hw_cycle_counters_update(common);
ar9003_hw_bb_watchdog_dbg_info(ah);
spin_unlock(&common->cc_lock);
goto chip_reset;
}
if (status & ATH9K_INT_SWBA)
tasklet_schedule(&sc->bcon_tasklet);
if (status & ATH9K_INT_TXURN)
ath9k_hw_updatetxtriglevel(ah, true);
if (ah->caps.hw_caps & ATH9K_HW_CAP_EDMA) {
if (status & ATH9K_INT_RXEOL) {
ah->imask &= ~(ATH9K_INT_RXEOL | ATH9K_INT_RXORN);
ath9k_hw_set_interrupts(ah, ah->imask);
}
}
if (status & ATH9K_INT_MIB) {
/*
* Disable interrupts until we service the MIB
* interrupt; otherwise it will continue to
* fire.
*/
ath9k_hw_disable_interrupts(ah);
/*
* Let the hal handle the event. We assume
* it will clear whatever condition caused
* the interrupt.
*/
spin_lock(&common->cc_lock);
ath9k_hw_proc_mib_event(ah);
spin_unlock(&common->cc_lock);
ath9k_hw_enable_interrupts(ah);
}
if (!(ah->caps.hw_caps & ATH9K_HW_CAP_AUTOSLEEP))
if (status & ATH9K_INT_TIM_TIMER) {
if (ATH_DBG_WARN_ON_ONCE(sc->ps_idle))
goto chip_reset;
/* Clear RxAbort bit so that we can
* receive frames */
ath9k_setpower(sc, ATH9K_PM_AWAKE);
ath9k_hw_setrxabort(sc->sc_ah, 0);
sc->ps_flags |= PS_WAIT_FOR_BEACON;
}
chip_reset:
ath_debug_stat_interrupt(sc, status);
if (sched) {
/* turn off every interrupt */
ath9k_hw_disable_interrupts(ah);
tasklet_schedule(&sc->intr_tq);
}
return IRQ_HANDLED;
#undef SCHED_INTR
}
static void ath9k_bss_assoc_info(struct ath_softc *sc,
struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
struct ieee80211_bss_conf *bss_conf)
{
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
if (bss_conf->assoc) {
ath_dbg(common, ATH_DBG_CONFIG,
"Bss Info ASSOC %d, bssid: %pM\n",
bss_conf->aid, common->curbssid);
/* New association, store aid */
common->curaid = bss_conf->aid;
ath9k_hw_write_associd(ah);
/*
* Request a re-configuration of Beacon related timers
* on the receipt of the first Beacon frame (i.e.,
* after time sync with the AP).
*/
sc->ps_flags |= PS_BEACON_SYNC;
/* Configure the beacon */
ath_beacon_config(sc, vif);
/* Reset rssi stats */
sc->last_rssi = ATH_RSSI_DUMMY_MARKER;
sc->sc_ah->stats.avgbrssi = ATH_RSSI_DUMMY_MARKER;
ath9k: Fix bug in starting ani There are few places where ANI is started without checking if it is right to start. This might lead to a case where ani timer would be left undeleted and cause improper memory acccess during module unload. This bug is clearly exposed with paprd support where the driver detects tx hang and does a chip reset. During this reset ani is (re)started without checking if it needs to be started. This would leave a timer scheduled even after all the resources are freed and cause a panic. This patch introduces a bit in sc_flags to indicate if ani needs to be started in sw_scan_start() and ath_reset(). This would fix the following panic. This issue is easily seen with ar9003 + paprd. BUG: unable to handle kernel paging request at 0000000000003f38 [<ffffffff81075391>] ? __queue_work+0x41/0x50 [<ffffffff8106afaa>] run_timer_softirq+0x17a/0x370 [<ffffffff81088be8>] ? tick_dev_program_event+0x48/0x110 [<ffffffff81061f69>] __do_softirq+0xb9/0x1f0 [<ffffffff810ba060>] ? handle_IRQ_event+0x50/0x160 [<ffffffff8100af5c>] call_softirq+0x1c/0x30 [<ffffffff8100c9f5>] do_softirq+0x65/0xa0 [<ffffffff81061e25>] irq_exit+0x85/0x90 [<ffffffff8155e095>] do_IRQ+0x75/0xf0 [<ffffffff815570d3>] ret_from_intr+0x0/0x11 <EOI> [<ffffffff812fd67b>] ? acpi_idle_enter_simple+0xe4/0x119 [<ffffffff812fd674>] ? acpi_idle_enter_simple+0xdd/0x119 [<ffffffff81441c87>] cpuidle_idle_call+0xa7/0x140 [<ffffffff81008da3>] cpu_idle+0xb3/0x110 [<ffffffff81550722>] start_secondary+0x1ee/0x1f5 Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 21:49:21 +08:00
sc->sc_flags |= SC_OP_ANI_RUN;
ath_start_ani(common);
} else {
ath_dbg(common, ATH_DBG_CONFIG, "Bss Info DISASSOC\n");
common->curaid = 0;
/* Stop ANI */
ath9k: Fix bug in starting ani There are few places where ANI is started without checking if it is right to start. This might lead to a case where ani timer would be left undeleted and cause improper memory acccess during module unload. This bug is clearly exposed with paprd support where the driver detects tx hang and does a chip reset. During this reset ani is (re)started without checking if it needs to be started. This would leave a timer scheduled even after all the resources are freed and cause a panic. This patch introduces a bit in sc_flags to indicate if ani needs to be started in sw_scan_start() and ath_reset(). This would fix the following panic. This issue is easily seen with ar9003 + paprd. BUG: unable to handle kernel paging request at 0000000000003f38 [<ffffffff81075391>] ? __queue_work+0x41/0x50 [<ffffffff8106afaa>] run_timer_softirq+0x17a/0x370 [<ffffffff81088be8>] ? tick_dev_program_event+0x48/0x110 [<ffffffff81061f69>] __do_softirq+0xb9/0x1f0 [<ffffffff810ba060>] ? handle_IRQ_event+0x50/0x160 [<ffffffff8100af5c>] call_softirq+0x1c/0x30 [<ffffffff8100c9f5>] do_softirq+0x65/0xa0 [<ffffffff81061e25>] irq_exit+0x85/0x90 [<ffffffff8155e095>] do_IRQ+0x75/0xf0 [<ffffffff815570d3>] ret_from_intr+0x0/0x11 <EOI> [<ffffffff812fd67b>] ? acpi_idle_enter_simple+0xe4/0x119 [<ffffffff812fd674>] ? acpi_idle_enter_simple+0xdd/0x119 [<ffffffff81441c87>] cpuidle_idle_call+0xa7/0x140 [<ffffffff81008da3>] cpu_idle+0xb3/0x110 [<ffffffff81550722>] start_secondary+0x1ee/0x1f5 Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 21:49:21 +08:00
sc->sc_flags &= ~SC_OP_ANI_RUN;
del_timer_sync(&common->ani.timer);
}
}
void ath_radio_enable(struct ath_softc *sc, struct ieee80211_hw *hw)
{
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ieee80211_channel *channel = hw->conf.channel;
int r;
ath9k_ps_wakeup(sc);
spin_lock_bh(&sc->sc_pcu_lock);
ath9k_hw_configpcipowersave(ah, 0, 0);
if (!ah->curchan)
ah->curchan = ath9k_cmn_get_curchannel(sc->hw, ah);
r = ath9k_hw_reset(ah, ah->curchan, ah->caldata, false);
if (r) {
ath_err(common,
"Unable to reset channel (%u MHz), reset status %d\n",
channel->center_freq, r);
}
ath9k_cmn_update_txpow(ah, sc->curtxpow,
sc->config.txpowlimit, &sc->curtxpow);
if (ath_startrecv(sc) != 0) {
ath_err(common, "Unable to restart recv logic\n");
goto out;
}
if (sc->sc_flags & SC_OP_BEACONS)
ath_beacon_config(sc, NULL); /* restart beacons */
/* Re-Enable interrupts */
ath9k_hw_set_interrupts(ah, ah->imask);
/* Enable LED */
ath9k_hw_cfg_output(ah, ah->led_pin,
AR_GPIO_OUTPUT_MUX_AS_OUTPUT);
ath9k_hw_set_gpio(ah, ah->led_pin, 0);
ieee80211_wake_queues(hw);
ieee80211_queue_delayed_work(hw, &sc->hw_pll_work, HZ/2);
out:
spin_unlock_bh(&sc->sc_pcu_lock);
ath9k_ps_restore(sc);
}
void ath_radio_disable(struct ath_softc *sc, struct ieee80211_hw *hw)
{
struct ath_hw *ah = sc->sc_ah;
struct ieee80211_channel *channel = hw->conf.channel;
int r;
ath9k_ps_wakeup(sc);
cancel_delayed_work_sync(&sc->hw_pll_work);
spin_lock_bh(&sc->sc_pcu_lock);
ieee80211_stop_queues(hw);
/*
* Keep the LED on when the radio is disabled
* during idle unassociated state.
*/
if (!sc->ps_idle) {
ath9k_hw_set_gpio(ah, ah->led_pin, 1);
ath9k_hw_cfg_gpio_input(ah, ah->led_pin);
}
/* Disable interrupts */
ath9k_hw_disable_interrupts(ah);
ath_drain_all_txq(sc, false); /* clear pending tx frames */
ath_stoprecv(sc); /* turn off frame recv */
ath_flushrecv(sc); /* flush recv queue */
if (!ah->curchan)
ah->curchan = ath9k_cmn_get_curchannel(hw, ah);
r = ath9k_hw_reset(ah, ah->curchan, ah->caldata, false);
if (r) {
ath_err(ath9k_hw_common(sc->sc_ah),
"Unable to reset channel (%u MHz), reset status %d\n",
channel->center_freq, r);
}
ath9k_hw_phy_disable(ah);
ath9k_hw_configpcipowersave(ah, 1, 1);
spin_unlock_bh(&sc->sc_pcu_lock);
ath9k_ps_restore(sc);
}
int ath_reset(struct ath_softc *sc, bool retry_tx)
{
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ieee80211_hw *hw = sc->hw;
int r;
sc->hw_busy_count = 0;
/* Stop ANI */
del_timer_sync(&common->ani.timer);
ath9k_ps_wakeup(sc);
spin_lock_bh(&sc->sc_pcu_lock);
ieee80211_stop_queues(hw);
ath9k_hw_disable_interrupts(ah);
ath_drain_all_txq(sc, retry_tx);
ath_stoprecv(sc);
ath_flushrecv(sc);
r = ath9k_hw_reset(ah, sc->sc_ah->curchan, ah->caldata, false);
if (r)
ath_err(common,
"Unable to reset hardware; reset status %d\n", r);
if (ath_startrecv(sc) != 0)
ath_err(common, "Unable to start recv logic\n");
/*
* We may be doing a reset in response to a request
* that changes the channel so update any state that
* might change as a result.
*/
ath9k_cmn_update_txpow(ah, sc->curtxpow,
sc->config.txpowlimit, &sc->curtxpow);
ath9k: fix regression on beacon loss after bgscan When we return to the home channel we were never reseting our beacon timers, this was casued by the fact that the scanning flag was still on even after we returned to our home channel. There are also other reasons why we would get a reset and if we are not off channel we always need to resynch our beacon timers, because a reset will clear them. This bug is a regression introduced on 2.6.36. The order of the changes are as follows: 5ee08656 - Sat Jul 31 - ath9k: prevent calibration during off-channel activity a0daa0e7 - Tue Jul 27 - Revert "mac80211: fix sw scan bracketing" 543708be - Fri Jun 18 - mac80211: fix sw scan bracketing mcgrof@tux ~/linux-2.6-allstable (git::master)$ git describe \ --contains 5ee0865615f65f84e6ee9174771a6716c29e08e1 v2.6.36-rc1~43^2~34^2~22 mcgrof@tux ~/linux-2.6-allstable (git::master)$ git describe \ --contains a0daa0e7592ada797d6835f11529097aabc27ad2 v2.6.36-rc1~571^2~64^2~13 mcgrof@tux ~/linux-2.6-allstable (git::master)$ git describe \ --contains 543708be320d7df692d24b349ca01a947b340764 v2.6.36-rc1~571^2~107^2~187 So 5ee08656 would have worked if a0daa0e7 was not committed but it was so this means 5ee08656 was broken since it assumed that when we were in the channel change routine the scan flag would be lifted. As it turns out the scan flag will be set when we are already on the home channel. For more details refer to: http://code.google.com/p/chromium-os/issues/detail?id=5715 These issues will need to be considered for our solution on reshifting the scan complete callback location on mac80211 on current development kernel work. This patch has stable fixes which apply down to [2.6.36+] Cc: stable@kernel.org Cc: Paul Stewart <pstew@google.com> Cc: Amod Bodas <amod.bodas@atheros.com> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-17 03:12:27 +08:00
if ((sc->sc_flags & SC_OP_BEACONS) || !(sc->sc_flags & (SC_OP_OFFCHANNEL)))
ath_beacon_config(sc, NULL); /* restart beacons */
ath9k_hw_set_interrupts(ah, ah->imask);
if (retry_tx) {
int i;
for (i = 0; i < ATH9K_NUM_TX_QUEUES; i++) {
if (ATH_TXQ_SETUP(sc, i)) {
spin_lock_bh(&sc->tx.txq[i].axq_lock);
ath_txq_schedule(sc, &sc->tx.txq[i]);
spin_unlock_bh(&sc->tx.txq[i].axq_lock);
}
}
}
ieee80211_wake_queues(hw);
spin_unlock_bh(&sc->sc_pcu_lock);
/* Start ANI */
ath_start_ani(common);
ath9k_ps_restore(sc);
return r;
}
/**********************/
/* mac80211 callbacks */
/**********************/
static int ath9k_start(struct ieee80211_hw *hw)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ieee80211_channel *curchan = hw->conf.channel;
struct ath9k_channel *init_channel;
int r;
ath_dbg(common, ATH_DBG_CONFIG,
"Starting driver with initial channel: %d MHz\n",
curchan->center_freq);
mutex_lock(&sc->mutex);
/* setup initial channel */
sc->chan_idx = curchan->hw_value;
init_channel = ath9k_cmn_get_curchannel(hw, ah);
/* Reset SERDES registers */
ath9k_hw_configpcipowersave(ah, 0, 0);
/*
* The basic interface to setting the hardware in a good
* state is ``reset''. On return the hardware is known to
* be powered up and with interrupts disabled. This must
* be followed by initialization of the appropriate bits
* and then setup of the interrupt mask.
*/
spin_lock_bh(&sc->sc_pcu_lock);
r = ath9k_hw_reset(ah, init_channel, ah->caldata, false);
if (r) {
ath_err(common,
"Unable to reset hardware; reset status %d (freq %u MHz)\n",
r, curchan->center_freq);
spin_unlock_bh(&sc->sc_pcu_lock);
goto mutex_unlock;
}
/*
* This is needed only to setup initial state
* but it's best done after a reset.
*/
ath9k_cmn_update_txpow(ah, sc->curtxpow,
sc->config.txpowlimit, &sc->curtxpow);
/*
* Setup the hardware after reset:
* The receive engine is set going.
* Frame transmit is handled entirely
* in the frame output path; there's nothing to do
* here except setup the interrupt mask.
*/
if (ath_startrecv(sc) != 0) {
ath_err(common, "Unable to start recv logic\n");
r = -EIO;
spin_unlock_bh(&sc->sc_pcu_lock);
goto mutex_unlock;
}
spin_unlock_bh(&sc->sc_pcu_lock);
/* Setup our intr mask. */
ah->imask = ATH9K_INT_TX | ATH9K_INT_RXEOL |
ATH9K_INT_RXORN | ATH9K_INT_FATAL |
ATH9K_INT_GLOBAL;
if (ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)
ah->imask |= ATH9K_INT_RXHP |
ATH9K_INT_RXLP |
ATH9K_INT_BB_WATCHDOG;
else
ah->imask |= ATH9K_INT_RX;
ah->imask |= ATH9K_INT_GTT;
if (ah->caps.hw_caps & ATH9K_HW_CAP_HT)
ah->imask |= ATH9K_INT_CST;
sc->sc_flags &= ~SC_OP_INVALID;
sc->sc_ah->is_monitoring = false;
/* Disable BMISS interrupt when we're not associated */
ah->imask &= ~(ATH9K_INT_SWBA | ATH9K_INT_BMISS);
ath9k_hw_set_interrupts(ah, ah->imask);
ieee80211_wake_queues(hw);
ieee80211_queue_delayed_work(sc->hw, &sc->tx_complete_work, 0);
if ((ah->btcoex_hw.scheme != ATH_BTCOEX_CFG_NONE) &&
!ah->btcoex_hw.enabled) {
ath9k_hw_btcoex_set_weight(ah, AR_BT_COEX_WGHT,
AR_STOMP_LOW_WLAN_WGHT);
ath9k_hw_btcoex_enable(ah);
if (common->bus_ops->bt_coex_prep)
common->bus_ops->bt_coex_prep(common);
if (ah->btcoex_hw.scheme == ATH_BTCOEX_CFG_3WIRE)
ath9k_btcoex_timer_resume(sc);
}
if (ah->caps.pcie_lcr_extsync_en && common->bus_ops->extn_synch_en)
common->bus_ops->extn_synch_en(common);
mutex_unlock:
mutex_unlock(&sc->mutex);
return r;
}
static void ath9k_tx(struct ieee80211_hw *hw, struct sk_buff *skb)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
struct ath_tx_control txctl;
struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
if (sc->ps_enabled) {
/*
* mac80211 does not set PM field for normal data frames, so we
* need to update that based on the current PS mode.
*/
if (ieee80211_is_data(hdr->frame_control) &&
!ieee80211_is_nullfunc(hdr->frame_control) &&
!ieee80211_has_pm(hdr->frame_control)) {
ath_dbg(common, ATH_DBG_PS,
"Add PM=1 for a TX frame while in PS mode\n");
hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM);
}
}
if (unlikely(sc->sc_ah->power_mode != ATH9K_PM_AWAKE)) {
/*
* We are using PS-Poll and mac80211 can request TX while in
* power save mode. Need to wake up hardware for the TX to be
* completed and if needed, also for RX of buffered frames.
*/
ath9k_ps_wakeup(sc);
if (!(sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_AUTOSLEEP))
ath9k_hw_setrxabort(sc->sc_ah, 0);
if (ieee80211_is_pspoll(hdr->frame_control)) {
ath_dbg(common, ATH_DBG_PS,
"Sending PS-Poll to pick a buffered frame\n");
sc->ps_flags |= PS_WAIT_FOR_PSPOLL_DATA;
} else {
ath_dbg(common, ATH_DBG_PS,
"Wake up to complete TX\n");
sc->ps_flags |= PS_WAIT_FOR_TX_ACK;
}
/*
* The actual restore operation will happen only after
* the sc_flags bit is cleared. We are just dropping
* the ps_usecount here.
*/
ath9k_ps_restore(sc);
}
memset(&txctl, 0, sizeof(struct ath_tx_control));
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
txctl.txq = sc->tx.txq_map[skb_get_queue_mapping(skb)];
ath_dbg(common, ATH_DBG_XMIT, "transmitting packet, skb: %p\n", skb);
if (ath_tx_start(hw, skb, &txctl) != 0) {
ath_dbg(common, ATH_DBG_XMIT, "TX failed\n");
goto exit;
}
return;
exit:
dev_kfree_skb_any(skb);
}
static void ath9k_stop(struct ieee80211_hw *hw)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
mutex_lock(&sc->mutex);
cancel_delayed_work_sync(&sc->tx_complete_work);
cancel_delayed_work_sync(&sc->hw_pll_work);
cancel_work_sync(&sc->paprd_work);
cancel_work_sync(&sc->hw_check_work);
if (sc->sc_flags & SC_OP_INVALID) {
ath_dbg(common, ATH_DBG_ANY, "Device not present\n");
mutex_unlock(&sc->mutex);
return;
}
/* Ensure HW is awake when we try to shut it down. */
ath9k_ps_wakeup(sc);
if (ah->btcoex_hw.enabled) {
ath9k_hw_btcoex_disable(ah);
if (ah->btcoex_hw.scheme == ATH_BTCOEX_CFG_3WIRE)
ath9k_btcoex_timer_pause(sc);
}
spin_lock_bh(&sc->sc_pcu_lock);
/* prevent tasklets to enable interrupts once we disable them */
ah->imask &= ~ATH9K_INT_GLOBAL;
/* make sure h/w will not generate any interrupt
* before setting the invalid flag. */
ath9k_hw_disable_interrupts(ah);
if (!(sc->sc_flags & SC_OP_INVALID)) {
ath_drain_all_txq(sc, false);
ath_stoprecv(sc);
ath9k_hw_phy_disable(ah);
} else
sc->rx.rxlink = NULL;
if (sc->rx.frag) {
dev_kfree_skb_any(sc->rx.frag);
sc->rx.frag = NULL;
}
/* disable HAL and put h/w to sleep */
ath9k_hw_disable(ah);
ath9k_hw_configpcipowersave(ah, 1, 1);
spin_unlock_bh(&sc->sc_pcu_lock);
/* we can now sync irq and kill any running tasklets, since we already
* disabled interrupts and not holding a spin lock */
synchronize_irq(sc->irq);
tasklet_kill(&sc->intr_tq);
tasklet_kill(&sc->bcon_tasklet);
ath9k_ps_restore(sc);
sc->ps_idle = true;
ath_radio_disable(sc, hw);
sc->sc_flags |= SC_OP_INVALID;
mutex_unlock(&sc->mutex);
ath_dbg(common, ATH_DBG_CONFIG, "Driver halt\n");
}
bool ath9k_uses_beacons(int type)
{
switch (type) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_MESH_POINT:
return true;
default:
return false;
}
}
static void ath9k_reclaim_beacon(struct ath_softc *sc,
struct ieee80211_vif *vif)
{
struct ath_vif *avp = (void *)vif->drv_priv;
ath9k_set_beaconing_status(sc, false);
ath_beacon_return(sc, avp);
ath9k_set_beaconing_status(sc, true);
sc->sc_flags &= ~SC_OP_BEACONS;
}
static void ath9k_vif_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
{
struct ath9k_vif_iter_data *iter_data = data;
int i;
if (iter_data->hw_macaddr)
for (i = 0; i < ETH_ALEN; i++)
iter_data->mask[i] &=
~(iter_data->hw_macaddr[i] ^ mac[i]);
switch (vif->type) {
case NL80211_IFTYPE_AP:
iter_data->naps++;
break;
case NL80211_IFTYPE_STATION:
iter_data->nstations++;
break;
case NL80211_IFTYPE_ADHOC:
iter_data->nadhocs++;
break;
case NL80211_IFTYPE_MESH_POINT:
iter_data->nmeshes++;
break;
case NL80211_IFTYPE_WDS:
iter_data->nwds++;
break;
default:
iter_data->nothers++;
break;
}
}
/* Called with sc->mutex held. */
void ath9k_calculate_iter_data(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
struct ath9k_vif_iter_data *iter_data)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
/*
* Use the hardware MAC address as reference, the hardware uses it
* together with the BSSID mask when matching addresses.
*/
memset(iter_data, 0, sizeof(*iter_data));
iter_data->hw_macaddr = common->macaddr;
memset(&iter_data->mask, 0xff, ETH_ALEN);
if (vif)
ath9k_vif_iter(iter_data, vif->addr, vif);
/* Get list of all active MAC addresses */
ieee80211_iterate_active_interfaces_atomic(sc->hw, ath9k_vif_iter,
iter_data);
}
/* Called with sc->mutex held. */
static void ath9k_calculate_summary_state(struct ieee80211_hw *hw,
struct ieee80211_vif *vif)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ath9k_vif_iter_data iter_data;
ath9k_calculate_iter_data(hw, vif, &iter_data);
ath9k_ps_wakeup(sc);
/* Set BSSID mask. */
memcpy(common->bssidmask, iter_data.mask, ETH_ALEN);
ath_hw_setbssidmask(common);
/* Set op-mode & TSF */
if (iter_data.naps > 0) {
ath9k_hw_set_tsfadjust(ah, 1);
sc->sc_flags |= SC_OP_TSF_RESET;
ah->opmode = NL80211_IFTYPE_AP;
} else {
ath9k_hw_set_tsfadjust(ah, 0);
sc->sc_flags &= ~SC_OP_TSF_RESET;
if (iter_data.nwds + iter_data.nmeshes)
ah->opmode = NL80211_IFTYPE_AP;
else if (iter_data.nadhocs)
ah->opmode = NL80211_IFTYPE_ADHOC;
else
ah->opmode = NL80211_IFTYPE_STATION;
}
/*
* Enable MIB interrupts when there are hardware phy counters.
*/
if ((iter_data.nstations + iter_data.nadhocs + iter_data.nmeshes) > 0) {
if (ah->config.enable_ani)
ah->imask |= ATH9K_INT_MIB;
ah->imask |= ATH9K_INT_TSFOOR;
} else {
ah->imask &= ~ATH9K_INT_MIB;
ah->imask &= ~ATH9K_INT_TSFOOR;
}
ath9k_hw_set_interrupts(ah, ah->imask);
ath9k_ps_restore(sc);
/* Set up ANI */
if ((iter_data.naps + iter_data.nadhocs) > 0) {
ath9k: Fix bug in starting ani There are few places where ANI is started without checking if it is right to start. This might lead to a case where ani timer would be left undeleted and cause improper memory acccess during module unload. This bug is clearly exposed with paprd support where the driver detects tx hang and does a chip reset. During this reset ani is (re)started without checking if it needs to be started. This would leave a timer scheduled even after all the resources are freed and cause a panic. This patch introduces a bit in sc_flags to indicate if ani needs to be started in sw_scan_start() and ath_reset(). This would fix the following panic. This issue is easily seen with ar9003 + paprd. BUG: unable to handle kernel paging request at 0000000000003f38 [<ffffffff81075391>] ? __queue_work+0x41/0x50 [<ffffffff8106afaa>] run_timer_softirq+0x17a/0x370 [<ffffffff81088be8>] ? tick_dev_program_event+0x48/0x110 [<ffffffff81061f69>] __do_softirq+0xb9/0x1f0 [<ffffffff810ba060>] ? handle_IRQ_event+0x50/0x160 [<ffffffff8100af5c>] call_softirq+0x1c/0x30 [<ffffffff8100c9f5>] do_softirq+0x65/0xa0 [<ffffffff81061e25>] irq_exit+0x85/0x90 [<ffffffff8155e095>] do_IRQ+0x75/0xf0 [<ffffffff815570d3>] ret_from_intr+0x0/0x11 <EOI> [<ffffffff812fd67b>] ? acpi_idle_enter_simple+0xe4/0x119 [<ffffffff812fd674>] ? acpi_idle_enter_simple+0xdd/0x119 [<ffffffff81441c87>] cpuidle_idle_call+0xa7/0x140 [<ffffffff81008da3>] cpu_idle+0xb3/0x110 [<ffffffff81550722>] start_secondary+0x1ee/0x1f5 Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 21:49:21 +08:00
sc->sc_flags |= SC_OP_ANI_RUN;
ath_start_ani(common);
} else {
sc->sc_flags &= ~SC_OP_ANI_RUN;
del_timer_sync(&common->ani.timer);
ath9k: Fix bug in starting ani There are few places where ANI is started without checking if it is right to start. This might lead to a case where ani timer would be left undeleted and cause improper memory acccess during module unload. This bug is clearly exposed with paprd support where the driver detects tx hang and does a chip reset. During this reset ani is (re)started without checking if it needs to be started. This would leave a timer scheduled even after all the resources are freed and cause a panic. This patch introduces a bit in sc_flags to indicate if ani needs to be started in sw_scan_start() and ath_reset(). This would fix the following panic. This issue is easily seen with ar9003 + paprd. BUG: unable to handle kernel paging request at 0000000000003f38 [<ffffffff81075391>] ? __queue_work+0x41/0x50 [<ffffffff8106afaa>] run_timer_softirq+0x17a/0x370 [<ffffffff81088be8>] ? tick_dev_program_event+0x48/0x110 [<ffffffff81061f69>] __do_softirq+0xb9/0x1f0 [<ffffffff810ba060>] ? handle_IRQ_event+0x50/0x160 [<ffffffff8100af5c>] call_softirq+0x1c/0x30 [<ffffffff8100c9f5>] do_softirq+0x65/0xa0 [<ffffffff81061e25>] irq_exit+0x85/0x90 [<ffffffff8155e095>] do_IRQ+0x75/0xf0 [<ffffffff815570d3>] ret_from_intr+0x0/0x11 <EOI> [<ffffffff812fd67b>] ? acpi_idle_enter_simple+0xe4/0x119 [<ffffffff812fd674>] ? acpi_idle_enter_simple+0xdd/0x119 [<ffffffff81441c87>] cpuidle_idle_call+0xa7/0x140 [<ffffffff81008da3>] cpu_idle+0xb3/0x110 [<ffffffff81550722>] start_secondary+0x1ee/0x1f5 Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 21:49:21 +08:00
}
}
/* Called with sc->mutex held, vif counts set up properly. */
static void ath9k_do_vif_add_setup(struct ieee80211_hw *hw,
struct ieee80211_vif *vif)
{
struct ath_softc *sc = hw->priv;
ath9k_calculate_summary_state(hw, vif);
if (ath9k_uses_beacons(vif->type)) {
int error;
/* This may fail because upper levels do not have beacons
* properly configured yet. That's OK, we assume it
* will be properly configured and then we will be notified
* in the info_changed method and set up beacons properly
* there.
*/
ath9k_set_beaconing_status(sc, false);
error = ath_beacon_alloc(sc, vif);
if (!error)
ath_beacon_config(sc, vif);
ath9k_set_beaconing_status(sc, true);
}
}
static int ath9k_add_interface(struct ieee80211_hw *hw,
struct ieee80211_vif *vif)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ath_vif *avp = (void *)vif->drv_priv;
int ret = 0;
mutex_lock(&sc->mutex);
switch (vif->type) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_WDS:
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_MESH_POINT:
break;
default:
ath_err(common, "Interface type %d not yet supported\n",
vif->type);
ret = -EOPNOTSUPP;
goto out;
}
if (ath9k_uses_beacons(vif->type)) {
if (sc->nbcnvifs >= ATH_BCBUF) {
ath_err(common, "Not enough beacon buffers when adding"
" new interface of type: %i\n",
vif->type);
ret = -ENOBUFS;
goto out;
}
}
if ((vif->type == NL80211_IFTYPE_ADHOC) &&
sc->nvifs > 0) {
ath_err(common, "Cannot create ADHOC interface when other"
" interfaces already exist.\n");
ret = -EINVAL;
goto out;
}
ath_dbg(common, ATH_DBG_CONFIG,
"Attach a VIF of type: %d\n", vif->type);
/* Set the VIF opmode */
avp->av_opmode = vif->type;
avp->av_bslot = -1;
sc->nvifs++;
ath9k_do_vif_add_setup(hw, vif);
out:
mutex_unlock(&sc->mutex);
return ret;
}
static int ath9k_change_interface(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
enum nl80211_iftype new_type,
bool p2p)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
int ret = 0;
ath_dbg(common, ATH_DBG_CONFIG, "Change Interface\n");
mutex_lock(&sc->mutex);
/* See if new interface type is valid. */
if ((new_type == NL80211_IFTYPE_ADHOC) &&
(sc->nvifs > 1)) {
ath_err(common, "When using ADHOC, it must be the only"
" interface.\n");
ret = -EINVAL;
goto out;
}
if (ath9k_uses_beacons(new_type) &&
!ath9k_uses_beacons(vif->type)) {
if (sc->nbcnvifs >= ATH_BCBUF) {
ath_err(common, "No beacon slot available\n");
ret = -ENOBUFS;
goto out;
}
}
/* Clean up old vif stuff */
if (ath9k_uses_beacons(vif->type))
ath9k_reclaim_beacon(sc, vif);
/* Add new settings */
vif->type = new_type;
vif->p2p = p2p;
ath9k_do_vif_add_setup(hw, vif);
out:
mutex_unlock(&sc->mutex);
return ret;
}
static void ath9k_remove_interface(struct ieee80211_hw *hw,
struct ieee80211_vif *vif)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
ath_dbg(common, ATH_DBG_CONFIG, "Detach Interface\n");
mutex_lock(&sc->mutex);
sc->nvifs--;
/* Reclaim beacon resources */
if (ath9k_uses_beacons(vif->type))
ath9k_reclaim_beacon(sc, vif);
ath9k_calculate_summary_state(hw, NULL);
mutex_unlock(&sc->mutex);
}
static void ath9k_enable_ps(struct ath_softc *sc)
{
struct ath_hw *ah = sc->sc_ah;
sc->ps_enabled = true;
if (!(ah->caps.hw_caps & ATH9K_HW_CAP_AUTOSLEEP)) {
if ((ah->imask & ATH9K_INT_TIM_TIMER) == 0) {
ah->imask |= ATH9K_INT_TIM_TIMER;
ath9k_hw_set_interrupts(ah, ah->imask);
}
ath9k_hw_setrxabort(ah, 1);
}
}
static void ath9k_disable_ps(struct ath_softc *sc)
{
struct ath_hw *ah = sc->sc_ah;
sc->ps_enabled = false;
ath9k_hw_setpower(ah, ATH9K_PM_AWAKE);
if (!(ah->caps.hw_caps & ATH9K_HW_CAP_AUTOSLEEP)) {
ath9k_hw_setrxabort(ah, 0);
sc->ps_flags &= ~(PS_WAIT_FOR_BEACON |
PS_WAIT_FOR_CAB |
PS_WAIT_FOR_PSPOLL_DATA |
PS_WAIT_FOR_TX_ACK);
if (ah->imask & ATH9K_INT_TIM_TIMER) {
ah->imask &= ~ATH9K_INT_TIM_TIMER;
ath9k_hw_set_interrupts(ah, ah->imask);
}
}
}
static int ath9k_config(struct ieee80211_hw *hw, u32 changed)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ieee80211_conf *conf = &hw->conf;
bool disable_radio = false;
mutex_lock(&sc->mutex);
/*
* Leave this as the first check because we need to turn on the
* radio if it was disabled before prior to processing the rest
* of the changes. Likewise we must only disable the radio towards
* the end.
*/
if (changed & IEEE80211_CONF_CHANGE_IDLE) {
sc->ps_idle = !!(conf->flags & IEEE80211_CONF_IDLE);
if (!sc->ps_idle) {
ath_radio_enable(sc, hw);
ath_dbg(common, ATH_DBG_CONFIG,
"not-idle: enabling radio\n");
} else {
disable_radio = true;
}
}
/*
* We just prepare to enable PS. We have to wait until our AP has
* ACK'd our null data frame to disable RX otherwise we'll ignore
* those ACKs and end up retransmitting the same null data frames.
* IEEE80211_CONF_CHANGE_PS is only passed by mac80211 for STA mode.
*/
if (changed & IEEE80211_CONF_CHANGE_PS) {
ath9k: fix power save race conditions ath9k has a race on putting the chip into network sleep and having registers read from hardware. The race occurs because although ath9k_ps_restore() locks its own callers it makes use of some variables which get altered in the driver at different code paths. The variables are the ps_enabled and ps_flags. This is easily reprodicible in large network environments when roaming with the wpa_supplicant simple bgscan. You'd get some 0xdeadbeef read out on certain registers such as: ath: timeout (100000 us) on reg 0x806c: 0xdeadbeef & 0x01f00000 != 0x00000000 ath: RX failed to go idle in 10 ms RXSM=0xdeadbeef ath: timeout (100000 us) on reg 0x7000: 0xdeadbeef & 0x00000003 != 0x00000000 ath: Chip reset failed The fix is to protect the ath9k_config(hw, IEEE80211_CONF_CHANGE_PS) calls with a spin_lock_irqsave() which will disable contendors for these variables from interrupt context, timers, re-entry from mac80211 on the same callback, and most importantly from ath9k_ps_restore() which is the only call which will put the device into network sleep. There are quite a few threads and bug reports on these a few of them are: https://bugs.launchpad.net/ubuntu/karmic/+source/linux/+bug/407040 http://code.google.com/p/chromium-os/issues/detail?id=5709 http://code.google.com/p/chromium-os/issues/detail?id=5943 Stable fixes apply to [2.6.32+] Cc: stable@kernel.org Cc: Paul Stewart <pstew@google.com> Cc: Amod Bodas <amod.bodas@atheros.com> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-17 03:12:26 +08:00
unsigned long flags;
spin_lock_irqsave(&sc->sc_pm_lock, flags);
if (conf->flags & IEEE80211_CONF_PS)
ath9k_enable_ps(sc);
else
ath9k_disable_ps(sc);
ath9k: fix power save race conditions ath9k has a race on putting the chip into network sleep and having registers read from hardware. The race occurs because although ath9k_ps_restore() locks its own callers it makes use of some variables which get altered in the driver at different code paths. The variables are the ps_enabled and ps_flags. This is easily reprodicible in large network environments when roaming with the wpa_supplicant simple bgscan. You'd get some 0xdeadbeef read out on certain registers such as: ath: timeout (100000 us) on reg 0x806c: 0xdeadbeef & 0x01f00000 != 0x00000000 ath: RX failed to go idle in 10 ms RXSM=0xdeadbeef ath: timeout (100000 us) on reg 0x7000: 0xdeadbeef & 0x00000003 != 0x00000000 ath: Chip reset failed The fix is to protect the ath9k_config(hw, IEEE80211_CONF_CHANGE_PS) calls with a spin_lock_irqsave() which will disable contendors for these variables from interrupt context, timers, re-entry from mac80211 on the same callback, and most importantly from ath9k_ps_restore() which is the only call which will put the device into network sleep. There are quite a few threads and bug reports on these a few of them are: https://bugs.launchpad.net/ubuntu/karmic/+source/linux/+bug/407040 http://code.google.com/p/chromium-os/issues/detail?id=5709 http://code.google.com/p/chromium-os/issues/detail?id=5943 Stable fixes apply to [2.6.32+] Cc: stable@kernel.org Cc: Paul Stewart <pstew@google.com> Cc: Amod Bodas <amod.bodas@atheros.com> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-17 03:12:26 +08:00
spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
}
if (changed & IEEE80211_CONF_CHANGE_MONITOR) {
if (conf->flags & IEEE80211_CONF_MONITOR) {
ath_dbg(common, ATH_DBG_CONFIG,
"Monitor mode is enabled\n");
sc->sc_ah->is_monitoring = true;
} else {
ath_dbg(common, ATH_DBG_CONFIG,
"Monitor mode is disabled\n");
sc->sc_ah->is_monitoring = false;
}
}
if (changed & IEEE80211_CONF_CHANGE_CHANNEL) {
struct ieee80211_channel *curchan = hw->conf.channel;
int pos = curchan->hw_value;
int old_pos = -1;
unsigned long flags;
if (ah->curchan)
old_pos = ah->curchan - &ah->channels[0];
if (hw->conf.flags & IEEE80211_CONF_OFFCHANNEL)
sc->sc_flags |= SC_OP_OFFCHANNEL;
else
sc->sc_flags &= ~SC_OP_OFFCHANNEL;
ath_dbg(common, ATH_DBG_CONFIG,
"Set channel: %d MHz type: %d\n",
curchan->center_freq, conf->channel_type);
ath9k_cmn_update_ichannel(&sc->sc_ah->channels[pos],
curchan, conf->channel_type);
/* update survey stats for the old channel before switching */
spin_lock_irqsave(&common->cc_lock, flags);
ath_update_survey_stats(sc);
spin_unlock_irqrestore(&common->cc_lock, flags);
/*
* If the operating channel changes, change the survey in-use flags
* along with it.
* Reset the survey data for the new channel, unless we're switching
* back to the operating channel from an off-channel operation.
*/
if (!(hw->conf.flags & IEEE80211_CONF_OFFCHANNEL) &&
sc->cur_survey != &sc->survey[pos]) {
if (sc->cur_survey)
sc->cur_survey->filled &= ~SURVEY_INFO_IN_USE;
sc->cur_survey = &sc->survey[pos];
memset(sc->cur_survey, 0, sizeof(struct survey_info));
sc->cur_survey->filled |= SURVEY_INFO_IN_USE;
} else if (!(sc->survey[pos].filled & SURVEY_INFO_IN_USE)) {
memset(&sc->survey[pos], 0, sizeof(struct survey_info));
}
if (ath_set_channel(sc, hw, &sc->sc_ah->channels[pos]) < 0) {
ath_err(common, "Unable to set channel\n");
mutex_unlock(&sc->mutex);
return -EINVAL;
}
/*
* The most recent snapshot of channel->noisefloor for the old
* channel is only available after the hardware reset. Copy it to
* the survey stats now.
*/
if (old_pos >= 0)
ath_update_survey_nf(sc, old_pos);
}
if (changed & IEEE80211_CONF_CHANGE_POWER) {
ath_dbg(common, ATH_DBG_CONFIG,
"Set power: %d\n", conf->power_level);
sc->config.txpowlimit = 2 * conf->power_level;
ath9k_ps_wakeup(sc);
ath9k_cmn_update_txpow(ah, sc->curtxpow,
sc->config.txpowlimit, &sc->curtxpow);
ath9k_ps_restore(sc);
}
if (disable_radio) {
ath_dbg(common, ATH_DBG_CONFIG, "idle: disabling radio\n");
ath_radio_disable(sc, hw);
}
mutex_unlock(&sc->mutex);
return 0;
}
#define SUPPORTED_FILTERS \
(FIF_PROMISC_IN_BSS | \
FIF_ALLMULTI | \
FIF_CONTROL | \
FIF_PSPOLL | \
FIF_OTHER_BSS | \
FIF_BCN_PRBRESP_PROMISC | \
FIF_PROBE_REQ | \
FIF_FCSFAIL)
/* FIXME: sc->sc_full_reset ? */
static void ath9k_configure_filter(struct ieee80211_hw *hw,
unsigned int changed_flags,
unsigned int *total_flags,
u64 multicast)
{
struct ath_softc *sc = hw->priv;
u32 rfilt;
changed_flags &= SUPPORTED_FILTERS;
*total_flags &= SUPPORTED_FILTERS;
sc->rx.rxfilter = *total_flags;
ath9k_ps_wakeup(sc);
rfilt = ath_calcrxfilter(sc);
ath9k_hw_setrxfilter(sc->sc_ah, rfilt);
ath9k_ps_restore(sc);
ath_dbg(ath9k_hw_common(sc->sc_ah), ATH_DBG_CONFIG,
"Set HW RX filter: 0x%x\n", rfilt);
}
static int ath9k_sta_add(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
struct ieee80211_sta *sta)
{
struct ath_softc *sc = hw->priv;
ath_node_attach(sc, sta);
return 0;
}
static int ath9k_sta_remove(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
struct ieee80211_sta *sta)
{
struct ath_softc *sc = hw->priv;
ath_node_detach(sc, sta);
return 0;
}
static int ath9k_conf_tx(struct ieee80211_hw *hw, u16 queue,
const struct ieee80211_tx_queue_params *params)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
struct ath_txq *txq;
struct ath9k_tx_queue_info qi;
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
int ret = 0;
if (queue >= WME_NUM_AC)
return 0;
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
txq = sc->tx.txq_map[queue];
mutex_lock(&sc->mutex);
memset(&qi, 0, sizeof(struct ath9k_tx_queue_info));
qi.tqi_aifs = params->aifs;
qi.tqi_cwmin = params->cw_min;
qi.tqi_cwmax = params->cw_max;
qi.tqi_burstTime = params->txop;
ath_dbg(common, ATH_DBG_CONFIG,
"Configure tx [queue/halq] [%d/%d], aifs: %d, cw_min: %d, cw_max: %d, txop: %d\n",
queue, txq->axq_qnum, params->aifs, params->cw_min,
params->cw_max, params->txop);
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
ret = ath_txq_update(sc, txq->axq_qnum, &qi);
if (ret)
ath_err(common, "TXQ Update failed\n");
if (sc->sc_ah->opmode == NL80211_IFTYPE_ADHOC)
ath9k: rework tx queue selection and fix queue stopping/waking The current ath9k tx queue handling code showed a few issues that could lead to locking issues, tx stalls due to stopped queues, and maybe even DMA issues. The main source of these issues is that in some places the queue is selected via skb queue mapping in places where this mapping may no longer be valid. One such place is when data frames are transmitted via the CAB queue (for powersave buffered frames). This is made even worse by a lookup WMM AC values from the assigned tx queue (which is undefined for the CAB queue). This messed up the pending frame counting, which in turn caused issues with queues getting stopped, but not woken again. To fix these issues, this patch removes an unnecessary abstraction separating a driver internal queue number from the skb queue number (not to be confused with the hardware queue number). It seems that this abstraction may have been necessary because of tx queue preinitialization from the initvals. This patch avoids breakage here by pushing the software <-> hardware queue mapping to the function that assigns the tx queues and redefining the WMM AC definitions to match the numbers used by mac80211 (also affects ath9k_htc). To ensure consistency wrt. pending frame count tracking, these counters are moved to the ath_txq struct, updated with the txq lock held, but only where the tx queue selected by the skb queue map actually matches the tx queue used by the driver for the frame. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Reported-by: Björn Smedman <bjorn.smedman@venatech.se> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-07 21:59:39 +08:00
if (queue == WME_AC_BE && !ret)
ath_beaconq_config(sc);
mutex_unlock(&sc->mutex);
return ret;
}
static int ath9k_set_key(struct ieee80211_hw *hw,
enum set_key_cmd cmd,
struct ieee80211_vif *vif,
struct ieee80211_sta *sta,
struct ieee80211_key_conf *key)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
int ret = 0;
if (ath9k_modparam_nohwcrypt)
return -ENOSPC;
if (vif->type == NL80211_IFTYPE_ADHOC &&
(key->cipher == WLAN_CIPHER_SUITE_TKIP ||
key->cipher == WLAN_CIPHER_SUITE_CCMP) &&
!(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {
/*
* For now, disable hw crypto for the RSN IBSS group keys. This
* could be optimized in the future to use a modified key cache
* design to support per-STA RX GTK, but until that gets
* implemented, use of software crypto for group addressed
* frames is a acceptable to allow RSN IBSS to be used.
*/
return -EOPNOTSUPP;
}
mutex_lock(&sc->mutex);
ath9k_ps_wakeup(sc);
ath_dbg(common, ATH_DBG_CONFIG, "Set HW Key\n");
switch (cmd) {
case SET_KEY:
ret = ath_key_config(common, vif, sta, key);
if (ret >= 0) {
key->hw_key_idx = ret;
/* push IV and Michael MIC generation to stack */
key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV;
if (key->cipher == WLAN_CIPHER_SUITE_TKIP)
key->flags |= IEEE80211_KEY_FLAG_GENERATE_MMIC;
if (sc->sc_ah->sw_mgmt_crypto &&
key->cipher == WLAN_CIPHER_SUITE_CCMP)
key->flags |= IEEE80211_KEY_FLAG_SW_MGMT;
ret = 0;
}
break;
case DISABLE_KEY:
ath_key_delete(common, key);
break;
default:
ret = -EINVAL;
}
ath9k_ps_restore(sc);
mutex_unlock(&sc->mutex);
return ret;
}
static void ath9k_bss_info_changed(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
struct ieee80211_bss_conf *bss_conf,
u32 changed)
{
struct ath_softc *sc = hw->priv;
struct ath_beacon_config *cur_conf = &sc->cur_beacon_conf;
struct ath_hw *ah = sc->sc_ah;
struct ath_common *common = ath9k_hw_common(ah);
struct ath_vif *avp = (void *)vif->drv_priv;
int slottime;
int error;
mutex_lock(&sc->mutex);
if (changed & BSS_CHANGED_BSSID) {
/* Set BSSID */
memcpy(common->curbssid, bss_conf->bssid, ETH_ALEN);
memcpy(avp->bssid, bss_conf->bssid, ETH_ALEN);
common->curaid = 0;
ath9k_hw_write_associd(ah);
/* Set aggregation protection mode parameters */
sc->config.ath_aggr_prot = 0;
ath_dbg(common, ATH_DBG_CONFIG, "BSSID: %pM aid: 0x%x\n",
common->curbssid, common->curaid);
/* need to reconfigure the beacon */
sc->sc_flags &= ~SC_OP_BEACONS ;
}
/* Enable transmission of beacons (AP, IBSS, MESH) */
if ((changed & BSS_CHANGED_BEACON) ||
((changed & BSS_CHANGED_BEACON_ENABLED) && bss_conf->enable_beacon)) {
ath9k_set_beaconing_status(sc, false);
error = ath_beacon_alloc(sc, vif);
if (!error)
ath_beacon_config(sc, vif);
ath9k_set_beaconing_status(sc, true);
}
if (changed & BSS_CHANGED_ERP_SLOT) {
if (bss_conf->use_short_slot)
slottime = 9;
else
slottime = 20;
if (vif->type == NL80211_IFTYPE_AP) {
/*
* Defer update, so that connected stations can adjust
* their settings at the same time.
* See beacon.c for more details
*/
sc->beacon.slottime = slottime;
sc->beacon.updateslot = UPDATE;
} else {
ah->slottime = slottime;
ath9k_hw_init_global_settings(ah);
}
}
/* Disable transmission of beacons */
if ((changed & BSS_CHANGED_BEACON_ENABLED) &&
!bss_conf->enable_beacon) {
ath9k_set_beaconing_status(sc, false);
avp->is_bslot_active = false;
ath9k_set_beaconing_status(sc, true);
}
if (changed & BSS_CHANGED_BEACON_INT) {
cur_conf->beacon_interval = bss_conf->beacon_int;
/*
* In case of AP mode, the HW TSF has to be reset
* when the beacon interval changes.
*/
if (vif->type == NL80211_IFTYPE_AP) {
sc->sc_flags |= SC_OP_TSF_RESET;
ath9k_set_beaconing_status(sc, false);
error = ath_beacon_alloc(sc, vif);
if (!error)
ath_beacon_config(sc, vif);
ath9k_set_beaconing_status(sc, true);
} else {
ath_beacon_config(sc, vif);
}
}
if (changed & BSS_CHANGED_ERP_PREAMBLE) {
ath_dbg(common, ATH_DBG_CONFIG, "BSS Changed PREAMBLE %d\n",
bss_conf->use_short_preamble);
if (bss_conf->use_short_preamble)
sc->sc_flags |= SC_OP_PREAMBLE_SHORT;
else
sc->sc_flags &= ~SC_OP_PREAMBLE_SHORT;
}
if (changed & BSS_CHANGED_ERP_CTS_PROT) {
ath_dbg(common, ATH_DBG_CONFIG, "BSS Changed CTS PROT %d\n",
bss_conf->use_cts_prot);
if (bss_conf->use_cts_prot &&
hw->conf.channel->band != IEEE80211_BAND_5GHZ)
sc->sc_flags |= SC_OP_PROTECT_ENABLE;
else
sc->sc_flags &= ~SC_OP_PROTECT_ENABLE;
}
if (changed & BSS_CHANGED_ASSOC) {
ath_dbg(common, ATH_DBG_CONFIG, "BSS Changed ASSOC %d\n",
bss_conf->assoc);
ath9k_bss_assoc_info(sc, hw, vif, bss_conf);
}
mutex_unlock(&sc->mutex);
}
static u64 ath9k_get_tsf(struct ieee80211_hw *hw)
{
struct ath_softc *sc = hw->priv;
u64 tsf;
mutex_lock(&sc->mutex);
ath9k_ps_wakeup(sc);
tsf = ath9k_hw_gettsf64(sc->sc_ah);
ath9k_ps_restore(sc);
mutex_unlock(&sc->mutex);
return tsf;
}
static void ath9k_set_tsf(struct ieee80211_hw *hw, u64 tsf)
{
struct ath_softc *sc = hw->priv;
mutex_lock(&sc->mutex);
ath9k_ps_wakeup(sc);
ath9k_hw_settsf64(sc->sc_ah, tsf);
ath9k_ps_restore(sc);
mutex_unlock(&sc->mutex);
}
static void ath9k_reset_tsf(struct ieee80211_hw *hw)
{
struct ath_softc *sc = hw->priv;
mutex_lock(&sc->mutex);
ath9k_ps_wakeup(sc);
ath9k_hw_reset_tsf(sc->sc_ah);
ath9k_ps_restore(sc);
mutex_unlock(&sc->mutex);
}
static int ath9k_ampdu_action(struct ieee80211_hw *hw,
struct ieee80211_vif *vif,
enum ieee80211_ampdu_mlme_action action,
struct ieee80211_sta *sta,
u16 tid, u16 *ssn, u8 buf_size)
{
struct ath_softc *sc = hw->priv;
int ret = 0;
local_bh_disable();
switch (action) {
case IEEE80211_AMPDU_RX_START:
if (!(sc->sc_flags & SC_OP_RXAGGR))
ret = -ENOTSUPP;
break;
case IEEE80211_AMPDU_RX_STOP:
break;
case IEEE80211_AMPDU_TX_START:
if (!(sc->sc_flags & SC_OP_TXAGGR))
return -EOPNOTSUPP;
ath9k_ps_wakeup(sc);
ret = ath_tx_aggr_start(sc, sta, tid, ssn);
if (!ret)
ieee80211_start_tx_ba_cb_irqsafe(vif, sta->addr, tid);
ath9k_ps_restore(sc);
break;
case IEEE80211_AMPDU_TX_STOP:
ath9k_ps_wakeup(sc);
ath_tx_aggr_stop(sc, sta, tid);
ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
ath9k_ps_restore(sc);
break;
case IEEE80211_AMPDU_TX_OPERATIONAL:
ath9k_ps_wakeup(sc);
ath_tx_aggr_resume(sc, sta, tid);
ath9k_ps_restore(sc);
break;
default:
ath_err(ath9k_hw_common(sc->sc_ah), "Unknown AMPDU action\n");
}
local_bh_enable();
return ret;
}
static int ath9k_get_survey(struct ieee80211_hw *hw, int idx,
struct survey_info *survey)
{
struct ath_softc *sc = hw->priv;
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
struct ieee80211_supported_band *sband;
struct ieee80211_channel *chan;
unsigned long flags;
int pos;
spin_lock_irqsave(&common->cc_lock, flags);
if (idx == 0)
ath_update_survey_stats(sc);
sband = hw->wiphy->bands[IEEE80211_BAND_2GHZ];
if (sband && idx >= sband->n_channels) {
idx -= sband->n_channels;
sband = NULL;
}
if (!sband)
sband = hw->wiphy->bands[IEEE80211_BAND_5GHZ];
if (!sband || idx >= sband->n_channels) {
spin_unlock_irqrestore(&common->cc_lock, flags);
return -ENOENT;
}
chan = &sband->channels[idx];
pos = chan->hw_value;
memcpy(survey, &sc->survey[pos], sizeof(*survey));
survey->channel = chan;
spin_unlock_irqrestore(&common->cc_lock, flags);
return 0;
}
static void ath9k_set_coverage_class(struct ieee80211_hw *hw, u8 coverage_class)
{
struct ath_softc *sc = hw->priv;
struct ath_hw *ah = sc->sc_ah;
mutex_lock(&sc->mutex);
ah->coverage_class = coverage_class;
ath9k_hw_init_global_settings(ah);
mutex_unlock(&sc->mutex);
}
static void ath9k_flush(struct ieee80211_hw *hw, bool drop)
{
struct ath_softc *sc = hw->priv;
int timeout = 200; /* ms */
int i, j;
ath9k_ps_wakeup(sc);
mutex_lock(&sc->mutex);
cancel_delayed_work_sync(&sc->tx_complete_work);
if (drop)
timeout = 1;
for (j = 0; j < timeout; j++) {
int npend = 0;
if (j)
usleep_range(1000, 2000);
for (i = 0; i < ATH9K_NUM_TX_QUEUES; i++) {
if (!ATH_TXQ_SETUP(sc, i))
continue;
npend += ath9k_has_pending_frames(sc, &sc->tx.txq[i]);
}
if (!npend)
goto out;
}
if (!ath_drain_all_txq(sc, false))
ath_reset(sc, false);
ieee80211_wake_queues(hw);
out:
ieee80211_queue_delayed_work(hw, &sc->tx_complete_work, 0);
mutex_unlock(&sc->mutex);
ath9k_ps_restore(sc);
}
struct ieee80211_ops ath9k_ops = {
.tx = ath9k_tx,
.start = ath9k_start,
.stop = ath9k_stop,
.add_interface = ath9k_add_interface,
.change_interface = ath9k_change_interface,
.remove_interface = ath9k_remove_interface,
.config = ath9k_config,
.configure_filter = ath9k_configure_filter,
.sta_add = ath9k_sta_add,
.sta_remove = ath9k_sta_remove,
.conf_tx = ath9k_conf_tx,
.bss_info_changed = ath9k_bss_info_changed,
.set_key = ath9k_set_key,
.get_tsf = ath9k_get_tsf,
.set_tsf = ath9k_set_tsf,
.reset_tsf = ath9k_reset_tsf,
.ampdu_action = ath9k_ampdu_action,
.get_survey = ath9k_get_survey,
.rfkill_poll = ath9k_rfkill_poll_state,
.set_coverage_class = ath9k_set_coverage_class,
.flush = ath9k_flush,
};