I like this! For the series: Reviewed-by: Kenneth Graunke <kenn...@whitecape.org>
However, there's one thing you missed: since you made it emit_end_query for (potentially) both OQ -and- timestamp on batchbuffer end, you should reserve enough space for the extra PIPE_CONTROLs emitted by write_timestamp().
This means adding an extra (4+5)*4 = 36 bytes to BATCH_RESERVED in intel_batchbuffer.h, and updating the comment accordingly.
Also, a tiny nit: the comment on brw_emit_query_end() is now incomplete/out of date (it does more than getting an ending PS_DEPTH_COUNT now).
Thanks! _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
