RE: [PATCH V2 2/2] usb: musb: Stop bulk endpoint while queue is rotated

[Andrew Goodbody <andrew.goodbody@cambrionix.com>]

> From: Sergei Shtylyov [mailto:sergei.shtylyov@cogentembedded.com]
> On 05/23/2016 04:40 PM, Sergei Shtylyov wrote:
> 
> >>>> Ensure that the endpoint is stopped by clearing REQPKT before
> >>>> clearing DATAERR_NAKTIMEOUT before rotating the queue on the
> >>>> dedicated bulk endpoint.
> >>>> This addresses an issue where a race could result in the endpoint
> >>>> receiving data before it was reprogrammed resulting in a warning
> >>>> about such data from musb_rx_reinit before it was thrown away.
> >>>> The data thrown away was a valid packet that had been correctly
> >>>> ACKed which meant the host and device got out of sync.
> >>>>
> >>>> Signed-off-by: Andrew Goodbody
> <andrew.goodbody@cambrionix.com>
> >>>> Cc: stable@vger.kernel.org
> >>>> ---
> >>>> V2 added comment about clearing REQPKT before
> DATAERR_NAKTIMEOUT
> >>>>
> >>>>  drivers/usb/musb/musb_host.c | 6 ++++++
> >>>>  1 file changed, 6 insertions(+)
> >>>>
> >>>> diff --git a/drivers/usb/musb/musb_host.c
> >>>> b/drivers/usb/musb/musb_host.c index 2966596..676cb98 100644
> >>>> --- a/drivers/usb/musb/musb_host.c
> >>>> +++ b/drivers/usb/musb/musb_host.c
> >>>> @@ -997,6 +997,12 @@ static void musb_bulk_nak_timeout(struct
> musb
> >>> *musb, struct musb_hw_ep *ep,
> >>>>          /* clear nak timeout bit */
> >>>>          rx_csr = musb_readw(epio, MUSB_RXCSR);
> >>>>          rx_csr |= MUSB_RXCSR_H_WZC_BITS;
> >>>> +        /*
> >>>> +         * Need to stop the transaction by clearing REQPKT before
> >>>
> >>>     Transcation? Maybe transfer?
> >>
> >> The quote from the TRM is "If the DATAERR_NAKTIMEOUT bit is set, the
> >> controller can be directed either to continue trying this transaction
> >> (until it times out again) by clearing the DATAERR_NAKTIMEOUT bit or
> >> to abort the transaction by clearing REQPKT bit before clearing the
> DATAERR_NAKTIMEOUT bit."
> >> So 'transaction' is correct.
> >
> >    OK.
> >
> >>>> +         * DATAERR_NAKTIMEOUT ref TRM section 16.3.8.2.2.1.2
> >>>
> >>>     Which TRM? Do you understand that the MUSB core is used by
> >>> multiple SoCs?
> >>> I'd recommend referring just to the "abstract" manual or the MUSB
> >>> programmer's guide (section 9.2.2 if you want an exact ref.).
> >
> >> That would be the AM335x Sitara Processors TRM. I don't have the MUSB
> >> programmer's guide, is it available online somewhere?
> >
> >    AFAIK, no.
> >
> >> I do not understand what you mean by the "abstract" manual.
> >
> >    Just saying "the manual" or "the manuals".
> >
> >> Would you be OK with "ref MUSB Programmer's Guide section 9.2.2"
> >
> >    Yes.
> 
>     The full document name is "MUSBMHDRC USB 2.0 HIGH-SPEED DUAL-ROLE
> CONTROLLER Programmer's Guide".

I moved the comment up to replace the one at the top so the patch now looks like this

	if (is_in) {
 		dma = is_dma_capable() ? ep->rx_channel : NULL;
 
-		/* clear nak timeout bit */
+		/*
+		 * Need to stop the transaction by clearing REQPKT first
+		 * then the NAK Timeout bit ref MUSBMHDRC USB 2.0 HIGH-SPEED
+		 * DUAL-ROLE CONTROLLER Programmer's Guide, section 9.2.2
+		 */
 		rx_csr = musb_readw(epio, MUSB_RXCSR);
 		rx_csr |= MUSB_RXCSR_H_WZC_BITS;
+		rx_csr &= ~MUSB_RXCSR_H_REQPKT;
+		musb_writew(epio, MUSB_RXCSR, rx_csr);
 		rx_csr &= ~MUSB_RXCSR_DATAERROR;
 		musb_writew(epio, MUSB_RXCSR, rx_csr);


Are you OK with that now?
Andrew