[igt-dev] ✓ Fi.CI.IGT: success for tests/perf_pmu: Avoid RT thread for accuracy test

[Chris Wilson]

Quoting Tvrtko Ursulin (2018-03-28 17:56:53)
> 
> Solid results from this simplified version so far from two runs. But 
> it's impossible to know whether it improves (or worsens) the long term 
> results on KBL, which had occasional fails, or GLK, which had more 
> frequent fails but has disappeared from the farm?

But I have been noticing that we've been drifting from our target ;)

I'm not opposed, I just grew a little attached to that calibration loop.
I really would like to see if we can just drop RT and keep everything
else :)
-Chris