Re: [PATCH v7 2/4] ath9k: always call ath_reset from workqueue context
From: Felix Fietkau <hidden>
Date: 2011-08-30 06:44:05
On 2011-08-30 8:20 AM, Mohammed Shafi wrote:
On Tue, Aug 30, 2011 at 11:39 AM, Felix Fietkau[off-list ref] wrote:quoted
On 2011-08-30 7:57 AM, Mohammed Shafi wrote:quoted
On Tue, Aug 30, 2011 at 12:10 AM, Felix Fietkau[off-list ref] wrote:quoted
This makes it much easier to add further rework to avoid race conditions between reset and other work items. Move other functions to make ath_reset static. Signed-off-by: Felix Fietkau[off-list ref] --- drivers/net/wireless/ath/ath9k/ath9k.h | 3 +- drivers/net/wireless/ath/ath9k/beacon.c | 4 +- drivers/net/wireless/ath/ath9k/init.c | 1 + drivers/net/wireless/ath/ath9k/main.c | 152 ++++++++++++++++--------------- drivers/net/wireless/ath/ath9k/xmit.c | 6 +- 5 files changed, 86 insertions(+), 80 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/xmit.c b/drivers/net/wireless/ath/ath9k/xmit.c index 49b93c2..f60706b 100644 --- a/drivers/net/wireless/ath/ath9k/xmit.c +++ b/drivers/net/wireless/ath/ath9k/xmit.c @@ -582,7 +582,7 @@ static void ath_tx_complete_aggr(struct ath_softc *sc, struct ath_txq *txq, rcu_read_unlock(); if (needreset) - ath_reset(sc, false); + ieee80211_queue_work(sc->hw,&sc->hw_reset_work);Hi Felix, retry_tx is set to 'false' in ath_reset here, but ath_reset_work always has it true. had i missed some thing?Since the reset work is deferred now, it can be allowed to retry tx now instead of dropping all packets in the queue.oh ok, in this path we need reset when the chip becomes deaf/mute(as mentioned in the comment), so i am not sure retrying tx will help rather than doing immediate reset of the chip. thanks.
Well, tx will be retried only after the full reset, so it should be fine. I'll make sure the next version will bail out of the tx completion tasklet when a reset is necessary, so that the reset can run right away. - Felix