On Tue, Jun 21, 2016 at 6:10 PM, Bryan O'Donoghue <pure.logic@xxxxxxxxxxxxxxxxx> wrote: > On Mon, 2016-06-20 at 12:17 +0300, Andy Shevchenko wrote: >> Bryan, since this series doesn't include any DMA management quirks I >> didn't >> address it. It is a matter of the following series which I would send >> later. > > OK I'm not 100% happy with that but I trust you to do it. I investigated the issue with DMA you pointed me to. Seems Quark UART / DMA requires not only handshake interface IDs, but also Channel # can't be arbitrary. So, I attach the one patch here to try. I will fix the series to follow the observing. -- With Best Regards, Andy Shevchenko
Attachment:
0001-TBD-serial-8250_dma-acquire-both-DMA-channels.patch
Description: Binary data
