No subject

i.e, wake_up
is not taken as indication of completion of the purpose.

If you have time, could u please confirm/correct my understanding?
Thanks.

>> >> An example of such situation is snd_pcm_lib_read1/write1 consuming so=
me space
>> >> of the period and going to sleep from wait_for_avail_min upon syncing=
 with
>> >> the DMA pointer. Clearly just the remainder of period size is needed,=
 but
>> >> wake_up is called only after _two_ periodic interrupts from that poin=
t.
>> >
>> > In that case, the original behavior is correct.
>> going by current implementation, that is correct, but is that desirable?
>
> Correct =3D working as designed.
>
>
> Takashi
>