>Ok. Sorry to be so pedantic, it must be some small small error, but... > >1. what does snd_pcm_avail_update() do and what does its return value mean?
it checks on the h/w pointer to discover how much data/space is actually available. it returns the space (in units of frames). if there has been an xrun, it returns -EPIPE. >1b. why does the value of 'avail' returned by snd_pcm_mmap_begin differ from > avail_update's value? IIRC, it reflects issues to do with the layout of the h/w buffer. the results from mmap_begin() reflect the amount you can touch in one single contiguous operation (e.g. memcpy). keep in mind that the h/w buffer is a ring buffer, and the space available may actually be split across the buffer boundary. if you look at the audioengine code again, you'll see how i use this. we invoke the engine's clients to process each contiguous block. >1c. what good does calling avail_update do? its the only reliable (and cheap) way to find out the status when you wake up from poll/select. consider what would happen if for whatever reason, the kernel failed to schedule your thread when the last h/w interrupt made your thread ready to run again. you still wake up, but its after two interrupts (hopefully), and so the amount of data is larger than you may be expecting. avail_update() tells you of this situation. >The following output prompted these questions. The loop is still not blocking >btw. lets see the code. _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
