On 09/09/14 13:43, Rob Clark wrote: > Although, I wonder if some waitqueue scheme would be a bit more sane.. > ie. end of apply_worker, if there is nothing more queued up, signal > the event that _flush() is waiting on.
Maybe, but we would still need either the separate apply_lock, or unlock the crtc->mutex to allow the workqueue to run. Tomi -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20140909/5b7ff192/attachment.sig>
