All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=748451

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=b4c8e38bbd83b60e829660590e223a398d2085618c2405545266b41a42dbe344


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

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8973104705436901760

=== Auto-CCing suspected CL author pcc@chromium.org ===

Hi pcc@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Peter Collingbourne
  Commit : 063509f9e490079536cc8cb019d50ab40a075f13
  Date   : Wed Jul 12 17:23:33 2017
  Subject: Roll clang 305735:307486.

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : image_decoding.image_decoding_measurement
  Metric       : ImageDecoding_avg/ImageDecoding_avg
  Change       : 0.60% | 112.932803571 -> 113.679428571

Revision             Result                   N
chromium@485946      112.933 +- 2.56958       14      good
chromium@486001      113.043 +- 2.78033       21      good
chromium@486003      113.059 +- 0.843576      6       good
chromium@486004      116.784 +- 1.01955       6       bad       <--
chromium@486005      116.62 +- 0.931657       6       bad
chromium@486008      116.04 +- 0.650624       6       bad
chromium@486015      114.781 +- 1.42835       6       bad
chromium@486029      116.208 +- 1.16521       6       bad
chromium@486056      113.679 +- 2.04489       14      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests image_decoding.image_decoding_measurement

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8973104705436901760


For feedback, file a bug with component Speed>Bisection

Doesn't seem actionable, may be related to  issue 741831 .

So size went up _and_ perf went down? Maybe that new pass isn't very good for arm?