Quoting Chris Wilson (2020-02-27 17:04:42) > Quoting Lionel Landwerlin (2020-02-27 12:43:56) > > We still need to wait for the initial OA configuration to happen > > before we enable OA report writes to the OA buffer. > > I can confirm this fixes the hang Lionel reported on Skylake [still odd > that we can only get this to be an issue on skl]. However, Lionel > mentioned that we should be more careful and ensure the emit_oa_config() > request is scheduled last. We're currently looking at different ways we > can do that.
Fwiw, this patch works and fixes the problem. I prefer not adding a single use temporary to i915_perf_stream, but I leave that choice to Lionel. Reviewed-by: Chris Wilson <[email protected]> -Chris _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
