On Mon, 2009-12-28 at 16:57 +0800, Zhu Yi wrote:
> Make addba_resp_timer aware the HT_AGG_STATE_REQ_STOP_BA_MSK mask
> so that when ___ieee80211_stop_tx_ba_session() is issued the timer
> will quit. Otherwise when suspend happens before the timer expired,
> the timer handler will be called immediately after resume and
> messes up driver status.
Want to reference the bug report still?
In any case
Acked-by: Johannes Berg <johannes@xxxxxxxxxxxxxxxx>
> Cc: Johannes Berg <johannes@xxxxxxxxxxxxxxxx>
> Signed-off-by: Zhu Yi <yi.zhu@xxxxxxxxx>
> ---
> net/mac80211/agg-tx.c | 3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
> index ceda366..a57bd73 100644
> --- a/net/mac80211/agg-tx.c
> +++ b/net/mac80211/agg-tx.c
> @@ -179,7 +179,8 @@ static void sta_addba_resp_timer_expired(unsigned
> long data)
>
> /* check if the TID waits for addBA response */
> spin_lock_bh(&sta->lock);
> - if ((*state & (HT_ADDBA_REQUESTED_MSK | HT_ADDBA_RECEIVED_MSK)) !=
> + if ((*state & (HT_ADDBA_REQUESTED_MSK | HT_ADDBA_RECEIVED_MSK |
> + HT_AGG_STATE_REQ_STOP_BA_MSK)) !=
> HT_ADDBA_REQUESTED_MSK) {
> spin_unlock_bh(&sta->lock);
> *state = HT_AGG_STATE_IDLE;
Attachment:
signature.asc
Description: This is a digitally signed message part
