Hi, Robert Bielik <Robert.Bielik@xxxxxxxxx> writes: >> >> I guess UAC1 doesn't need feedback endpoints, right? Seems like that >> >> should be something specific to UAC2. At least for now. >> > >> > It seems like it is needed there aswell, see >> > http://www.usb.org/developers/docs/devclass_docs/audio10.pdf (3.7.2.2) >> > (Feedback endpoint == Synch Endpoint) >> >> fair enough, then it could be done generically in u_audio.c > > Ok, but I guess for the generic case (when it is possible to calculate > the feedback value) that lower levels need to provide f.i. the high > resolution timestamps of SOF IRQs to upper levels ? In the ALSA > subsystem case, the calculation is straightforward and static. > > I don't know if you saw my proposal for a new UAC2 gadget upper level > implementation, where the gadget should be exposed as a JACK > client. In this case, there would be need for those SOF timestamps to > be able to calculate the feedback value. Enabling SOF interrupts will be a big pain :-) Well, enabling the interrupt itself is a no-brainer, but it'll cause terrible CPU overload. I haven't gone through the Audio spec to know what it is that the host side is expecting from the feedback endpoint, but I'll try to do that later today. -- balbi
Attachment:
signature.asc
Description: PGP signature
