Host mode instability & device lockups

Oleksij Rempel linux at rempel-privat.de
Wed May 1 13:05:53 EDT 2013


Am 01.05.2013 18:49, schrieb Andy Ross:
> On 05/01/2013 09:35 AM, Adrian Chadd wrote:
>> Well, does bringing the interface down/up fix it? I guess not.
>>
>> I wonder if the firmware has crashed here. Unfortunately you'll need
>> to add a couple UART pins to your NIC to enable/see debugging output.
>
> Yeah.  I saw the posts about the SMT pins, but let's just say I'm
> going to need a lot more than five test units to make that work. ;)
>
> I guess I was hoping for an answer like "here's the watchdog code, it
> runs out of a timer interrupt and checks this word for an update, such
> that if it doesn't see XXX from the host every NN ms it will reset and
> emit YYY from endpoint 2" or whatever.  Then I could write a script to
> check for that in the usbmon output and maybe see if I could get the
> driver to recover.

That would lead to workaround, so no body will care about broken 
firmware. Or some one will as why restart firwmare every hour. If it is 
an firmware issue, we need to fix it.

> As it is, I'm at least hopeful I can get the driver fixed such that it
> doesn't hang the machine.

Did just removed module or did rmmmod --force?


-- 
Regards,
Oleksij



More information about the ath9k_htc_fw mailing list