I spent some time this weekend dealing with various users of the “DVICO Dual Digital 4” who were seeing that zl10353 regression I have fixed on four other boards so far. I can say some choice words about the person who submitted the patch that caused the breakage, or the person who blindly accepted it upstream without considering the implications for all the other products, but there isn’t much value in pointing fingers at this point. I just don’t have the time/energy to go through and fix *all* the effected products, so I have to deal with them as people report the problem.
The big project I’ve been working on for the last few days has been the HVR-1600 ALSA audio support. This will result in an ALSA device being present which can be used in conjunction with capturing raw video. Andy Walls was kind enough to point me to some skeleton code he wrote a few months ago and never had the time to get working. I cleaned it up a bit and now need to get the DMA code working properly. Because I don’t want to break compatibility with the existing PCM device, I need to be careful in how the stream gets multiplexed to both the ALSA subsystem and the original device node. This creates some interesting implications on things like mmap and the buffer management for the firmware queues.
If all goes well, I should basically have it “working” this week and just need to test and nail down any edge cases.