Quoting Ville Syrjala (2020-03-02 14:39:42) > From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > Pull the code to do the CS timestamp ns<->ticks conversion into > helpers and use them all over. Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > The check in i915_perf_noa_delay_set() seems a bit dubious, > so we switch it to do what I assume it wanted to do all along > (ie. make sure the resulting delay in CS timestamp ticks > doesn't exceed 32bits)? Yes. The MI_MATH operates on just 32b of the timestamp, and so wants a delay in that range. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
