We've noticed it in the drilldown for Linux64/Sunspider Here's a link to it in Perfherder: https://treeherder.mozilla.org/perf.html#/graphs?series=mozilla-central,1730499,1,10&series=mozilla-central,1730527,1,10&selected=mozilla-central,1730527,371811,543822127,10 03:06:04 INFO - raptor-output PERFHERDER_DATA: {"framework": {"name": "raptor"}, "suites": [{"extraOptions": [], "name": "raptor-sunspider-firefox", "lowerIsBetter": false, "alertThreshold": 2.0, "value": 219.78000000000003, "subtests": [{"name": "benchmark_score", "lowerIsBetter": false, "alertThreshold": 2.0, "replicates": [], "value": 0, "unit": "score"}, This is a minor issue but wanted to track it.

interesting bug, this is always 0.0 (raptor sunspider, not jsshell-bench version). We create a placeholder for this here: https://searchfox.org/mozilla-central/source/testing/raptor/raptor/output.py#322 I suspect we could remove this small block of code and the benchmark would run and report just fine.

please review my code

I pushed the change to Try: https://treeherder.mozilla.org/#/jobs?repo=try&author=igoldan@mozilla.com&fromchange=ee30f38d0d0013f4bedac606332998333525b1cd

Pushed by jmaher@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/15fdc0e2feb8 benchmark_score for Sunspider Chrome is unnecessary r=jmaher jmaher