On Mon, Mar 28, 2011 at 05:18, Felipe Balbi wrote:
> On Mon, Mar 28, 2011 at 05:06:51AM -0400, Mike Frysinger wrote:
>> the musb ip on blackfin parts has an anomaly with DMA mode 1 transmits
>> where if the transfer is not a multiple of the maximum packet size,
>> the last short packet might be corrupted. Âto work around this issue,
>> we transfer the full sized packets using dma mode 1 and then use dma
>> mode 0 for the last short packet.
>>
>> so in common code:
>> --- a/drivers/usb/musb/musb_host.c
>> +++ b/drivers/usb/musb/musb_host.c
>> @@ -672,7 +672,10 @@ static bool musb_tx_dma_program(struct dma_controller *dma,
>> Â Â wmb();
>>
>> Â Â if (!dma->channel_program(channel, pkt_size, mode,
>> - Â Â Â Â Â urb->transfer_dma + offset, length)) {
>> + Â Â Â Â Â urb->transfer_dma + offset,
>> + Â Â Â Â Â (channel->desired_mode == 0) ? length :
>> + Â Â Â Â Â Â Â length - (length % qh->maxpacket)))
>> + Â {
>> Â Â Â Â dma->channel_release(channel);
>> Â Â Â Â hw_ep->tx_channel = NULL;
>>
>> @@ -1750,7 +1753,10 @@ void musb_host_rx(struct musb *musb, u8 epnum)
>> Â Â Â Â Â Â Â*/
>> Â Â Â Â Â Â ret = c->channel_program(
>> Â Â Â Â Â Â Â Â dma, qh->maxpacket,
>> - Â Â Â Â Â Â Â dma->desired_mode, buf, length);
>> + Â Â Â Â Â Â Â dma->desired_mode, buf,
>> + Â Â Â Â Â Â Â (dma->desired_mode == 0)
>> + Â Â Â Â Â Â Â ? length
>> + Â Â Â Â Â Â Â : length - (length % qh->maxpacket));
>>
>> Â Â Â Â Â Â if (!ret) {
>> Â Â Â Â Â Â Â Â c->channel_release(dma);
>>
>> any ideas on how to get this workaround merged ? Âanother idea i had
>> was that we could add a hook to the top of dma_channel_program to let
>> targets muck with the arguments ... the Blackfin code would just need
>> to do something like this at the start of the func:
>> + Â if (musb_channel->transmit && channel->desired_mode == 1)
>> + Â Â Â len = len - (len % packet_sz);
>
> is that IP-related or Blackfin-only ? I mean, does this bug affects all
> MUSB of a particular RTL version or only the Blackfin ones ?
our understanding of the issue is that it is Blackfin specific. seems
to be related to our internal bus/dma connections to the musb ip.
-mike
--
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
