ath9k: Fix race in reset-work usage

Using work_pending() to defer certain operations when
a HW-reset work has been queued is racy since the check
would return false when the work item is actually in
execution. Use SC_OP_HW_RESET instead to fix this race.
Also, unify the reset debug statistics maintenance.

Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This commit is contained in:
Rajkumar Manoharan
2012-07-17 17:16:42 +05:30
committed by John W. Linville
szülő 6dcc344469
commit 124b979bae
7 fájl változott, egészen pontosan 41 új sor hozzáadva és 34 régi sor törölve

Fájl megtekintése

@@ -202,7 +202,7 @@ static void ath_mci_cal_msg(struct ath_softc *sc, u8 opcode, u8 *rx_payload)
case MCI_GPM_BT_CAL_REQ:
if (mci_hw->bt_state == MCI_BT_AWAKE) {
ar9003_mci_state(ah, MCI_STATE_SET_BT_CAL_START);
ieee80211_queue_work(sc->hw, &sc->hw_reset_work);
ath9k_queue_reset(sc, RESET_TYPE_MCI);
}
ath_dbg(common, MCI, "MCI State : %d\n", mci_hw->bt_state);
break;