I am going to kick off a bisect on one of the failing benchmarks since they all seem to be failing with the same stack trace.  It seems like a generic trace, so I am not sure the issue, but they are all failing similarly so hopefully a bisect will shed some light on the CL.  


Link to failing step log:

Here is the link to a failing builder:  https://uberchromegw.corp.google.com/i/chromium.perf/builders/Win%2010%20High-DPI%20Perf%20%284%29/builds/50

I am going to choose the smoothness.tough_image_decode_cases to run the bisect on.


If the test is disabled, please downgrade to Pri-2.
 I didn't disabled any tests yet since there are at least 14 different failing benchmarks.  I am hoping the bisect will shed more light on the outstanding issue.