All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=782742

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=526f6b7697dee4b5414f37714a74e6f122efb5e69e1a2067be268e3741330ab4


Bot(s) for this bug's original alert(s):

linux-release

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14837476f80000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14837476f80000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1128c656f80000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/1128c656f80000

Simon: Looks like pinpoint is getting '0' for both runs on this test - is there an issue with the testing / pulling values?

Looks like there might be a bug reading the histogram output, specifically the histogram with name='decode-gif' has a stories diagnostic, with 'decode-gif.html', but there's no story specified in the job.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14deea77040000

hans: I'm really sorry the bisect has had so much trouble with this perf regression. But in the bisect in #8, it really looks like the clang roll from November, r513697, regressed this benchmark. (I posted a screenshot showing how I came to the conclusion that it's the clang roll, sorry the new bisector is a bit confusing for this bug.)

Is there anything that can be done at this point?

Deleted: pinpoint-explanation.png 195 KB

	pinpoint-explanation.png 195 KB View Download

I'm not sure when we'll get to it, but let's leave this open as the benchmark doesn't seem to have recovered.

Thanks! Taking the Performance-Sheriff label off this bug, since it doesn't sound like it makes sense for us to ping for updates.

Looking at the performance graph now, this change doesn't stand out at all.