All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=690241

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgqLel9ggM


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

Android Builder

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988180337111639264

=== BISECT JOB RESULTS ===
NO Perf regression found, tests failed to produce values

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : (MonochromePublic.apk)
  Metric       : MonochromePublic.apk_Specifics/normalized apk size


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 --story-filter=normalized.apk.size resource_sizes (MonochromePublic.apk)

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8988180337111639264

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6416675454320640


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

Bisect failed.

There is only one CL in the roll, the v8 roll at r448620. Do we care about 80kib regression?

We certainly care, but there's not much we can do until bisects are fixed. (well, we could manually build all revisions, but that's not a good long-term solution).

In this case, I can tell from the log that it's:
https://codereview.chromium.org/2679063003

Since it was reverted in this roll, and a similar 80kb improvement was seen in the v8 roll where it landed (rev 448310). Also - the commit description states that it saves 80kb :P.

In this case, I think the appropriate action is to note the change in:
https://docs.google.com/spreadsheets/d/1GrRkszV7Oy5pVsaMb5Eb6s8izW9t4dElBxIH3iGq93o/edit#gid=1894856744

It's valuable to collect a the list of things that can reduce our size.

In addition, comment on the review to see if a reland is being attempted (which I've now done).

-RVG

I wasn't planning on relanding this CL since it regresses RegExp.prototype.exec microbenchmarks by around 5%. Unfortunate, the space savings would've been very nice..

There's a tracking bug here for large builtins in V8: https://bugs.chromium.org/p/v8/issues/detail?id=5737 

 Issue 692410  has been merged into this issue.