On 15/06/17 08:51, Kenneth Graunke wrote:
On Wednesday, June 14, 2017 5:38:31 PM PDT Lionel Landwerlin wrote:
Due to an underlying hardware race condition, we have no guarantee
that all the reports coming from the OA buffer related to the workload
we're trying to measure have landed to memory
On Wednesday, June 14, 2017 5:38:31 PM PDT Lionel Landwerlin wrote:
> Due to an underlying hardware race condition, we have no guarantee
> that all the reports coming from the OA buffer related to the workload
> we're trying to measure have landed to memory by the time all the work
> submitted has
Due to an underlying hardware race condition, we have no guarantee
that all the reports coming from the OA buffer related to the workload
we're trying to measure have landed to memory by the time all the work
submitted has completed. That means we need to keep on reading the OA
stream until we
