21 Nov
2019
21 Nov
'19
10:45 p.m.
OK, that makes sense. Thanks for the suggestion. Regarding your previous comment about adding flush_work() to the
sync_stop()
op, would that still be required?
Yes, that's needed no matter which way is used; the pending work must be synced at some point.
Thanks, Takashi. I will follow up with the suggested SOF changes after your patches have been applied.
Thanks, Ranjani