Re: [PATCH 3/5] tm6000: bugfix interrupt reset

[Mauro Carvalho Chehab <mchehab@redhat.com>]

On 06-12-2011 06:12, Thierry Reding wrote:
> * Thierry Reding wrote:
>> * Mauro Carvalho Chehab wrote:
>>> That means that all we need is to get rid of TM6000_QUIRK_NO_USB_DELAY.
>>
>> I've just reviewed my patches again and it seems that no-USB-delay quirk
>> patch was only partially applied. The actual location where it was introduced
>> was in tm6000_read_write_usb() to allow the msleep(5) to be skipped, which on
>> some devices isn't required and significantly speeds up firmware upload. So I
>> don't think we should get rid of it.
>>
>> If it helps I can rebase the code against your branch (which one would that
>> be exactly?) and resend the rest of the series.
>
> Looking more closely, I think my original patch was applied wrongly. If you
> look at the original patch:
>
> 	http://article.gmane.org/gmane.linux.drivers.video-input-infrastructure/37552
>
> and look at the applied version in this commit:
>
> 	42845708363fc92a190f5c47e6fe750e3919f867
>
> Then you see that the hunk from the tm6000_read_write_usb() function was
> applied to the tm6000_reset() function instead.

Hmmm... probably merge conflicts. It is rare, but such things sometimes happen, if
two functions are very similar, and there were enough changes on a driver for the
patch logic to do the wrong thing.

In any case, please send me the patch again, against my tree.

Thanks,
Mauro

>
> Thierry