Hi Felipe,
On Monday 10 April 2017 05:53 PM, Felipe Balbi wrote:
>
> Hi,
>
> Roger Quadros <rogerq@xxxxxx> writes:
>>> From: Janusz Dziedzic <januszx.dziedzic@xxxxxxxxx>
>>>
>>> In the case of bounced ep0 requests, we must delay DMA operation until
>>> after ->complete() otherwise we might overwrite contents of req->buf.
>>>
>>> This caused problems with RNDIS gadget.
>>>
>>> Signed-off-by: Janusz Dziedzic <januszx.dziedzic@xxxxxxxxx>
>>> Signed-off-by: Felipe Balbi <felipe.balbi@xxxxxxxxxxxxxxx>
>>> ---
>>
>> Sekhar found out that this fixes RNDIS for us on v4.9.
>> Could you please tag this for stable? Thanks.
>
> Was this a regression on v4.9 or was it already broken? Was it broken
> forever or which commit regressed it? If it falls into
> "has-never-worked-before" then I'm not sure we should tag stable. If we
> can blame a commit, then I can send the patch to stable, no issues.
Mainline v4.9 works. v4.9.20 is broken because commit d62145929992
("usb: dwc3: gadget: always unmap EP0 requests") was backported it.
I think the $SUBJECT patch needs to be propagated to all the stable
kernels to which commit d62145929992 was back-ported to.
Thanks,
Sekhar
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
