Re: 3.4-rc ath9k regression (Re: [ath9k-devel] 3.3.1 ath9k regression)

From: Felipe Contreras
Date: Sat Apr 14 2012 - 17:15:19 EST


On Thu, Apr 12, 2012 at 10:28 PM, Sergio Correia <lists@xxxxxxxx> wrote:
> Hello Felix,
>
> On Thu, Apr 12, 2012 at 2:11 PM, Felix Fietkau <nbd@xxxxxxxxxxx> wrote:
>> On 2012-04-12 5:35 PM, Michael Leun wrote:
>>> On Thu, 12 Apr 2012 16:58:34 +0300
>>> Felipe Contreras <felipe.contreras@xxxxxxxxx> wrote:
>>>
>>>> On Thu, Apr 12, 2012 at 4:46 PM, Felix Fietkau <nbd@xxxxxxxxxxx>
>>>> wrote:
>>>> > On 2012-04-09 5:29 PM, Michael Leun wrote:
>>>> >> On Mon, 9 Apr 2012 19:52:45 +0530
>>>>
>>>> >> Seems obvious to me, that this line
>>>> >>
>>>> >> if (sc->ps_idle && (sc->ps_flags & PS_WAIT_FOR_TX_ACK))
>>>> >>
>>>> >> makes the real difference.
>>>> > Please try this patch with plain 3.3.1 or 3.4-rc2 to see if it fixes
>>>> > this issue:
>>>>
>>>> Nope, not here (3.3.1).
>>>
>>> Also negative for 3.4-rc2, but that SLEEP<>AWAKE noise at the end
>>> stopped. Still need "iwconfig wlan0 essid <myssid>" to make it ping
>>> again (that not included in log below).
>> OK, so my fix is definitely worth submitting, since it gets rid of some
>> unnecessary sleep mode transitions. About the iwconfig command: are you
>> not using wpa_supplicant?
>>
>>> Due to Felipes result and 3.4-rc2 didn't repeat my test with 3.3.1, if
>>> you think it would add useful information I could.
>> Maybe it would be useful to get some more detailed logs from that.
>>
>
> your patch seems to work for me, with 3.3.1:
>
> this is the log after I disable/re-enable the interface:

That is curious, are you saying it doesn't work initially?

Maybe we are talking about two different issues, but my issue is
easily reproducible 100%; I boot, use iwconfig to ad-hoc mode, set the
essid, ip link up, and run wpa_supplicant, at which point it hangs.
It's a strange hang; the rest of the system works fine, but
wpa_supplicant doesn't respond any more and I hear some noise, as if
the CPU was at 100%, but top doesn't show that. With my distro's 3.3.1
kernel I do see the CPU at 100%, although I was trying to connect to
a different (non ad-hoc) network.

It it's helpful I would gladly test different combinations of patches
and networks, just let me know what would you like me to test.

Cheers.

--
Felipe Contreras
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/