2011-02-18 13:00:47

by Stanislaw Gruszka

Subject: [RFC] ath9k correct interrupts ?

I'm not sure if that patch is correct or not. It possibly fix
change from commit 4df3071ebd92ef7115b409da64d0eb405d24a631
"ath9k_hw: optimize interrupt mask changes" that looks
suspicious for me. But from other hand this patch
could be totally wrong. Felix?

diff --git a/drivers/net/wireless/ath/ath9k/mac.c b/drivers/net/wireless/ath/ath9k/mac.c
index 180170d..2915b11 100644
--- a/drivers/net/wireless/ath/ath9k/mac.c
+++ b/drivers/net/wireless/ath/ath9k/mac.c
@@ -885,7 +885,7 @@ void ath9k_hw_set_interrupts(struct ath_hw *ah, enum ath9k_int ints)
struct ath_common *common = ath9k_hw_common(ah);

if (!(ints & ATH9K_INT_GLOBAL))
- ath9k_hw_enable_interrupts(ah);
+ ath9k_hw_disable_interrupts(ah);

ath_dbg(common, ATH_DBG_INTERRUPT, "0x%x => 0x%x\n", omask, ints);

@@ -963,7 +963,8 @@ void ath9k_hw_set_interrupts(struct ath_hw *ah, enum ath9k_int ints)
REG_CLR_BIT(ah, AR_IMR_S5, AR_IMR_S5_TIM_TIMER);
}

- ath9k_hw_enable_interrupts(ah);
+ if (ints & ATH9K_INT_GLOBAL)
+ ath9k_hw_enable_interrupts(ah);

return;
}

2011-02-18 18:04:47

by Felix Fietkau

Subject: Re: [RFC] ath9k correct interrupts ?

On 2011-02-18 2:00 PM, Stanislaw Gruszka wrote:
> I'm not sure if that patch is correct or not. It possibly fix
> change from commit 4df3071ebd92ef7115b409da64d0eb405d24a631
> "ath9k_hw: optimize interrupt mask changes" that looks
> suspicious for me. But from other hand this patch
> could be totally wrong. Felix?
Looks good to me

- Felix