All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=770546

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=44f4fe0a942016ed66087816d5f6c2119dadb511398072bb602c839c031ae3b7


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966935590799816112

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

Hi lalitm@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 : Lalit Maganti
  Commit : fdc0a261697b6e767f45acb3d61fb5dc005d205a
  Date   : Thu Sep 21 12:30:10 2017
  Subject: memory-infra: Enable LevelDB global dumps in background mode

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : tracing.tracing_with_background_memory_infra
  Metric       : MojoLevelDB_memory_dump_cpu_overhead_avg/http___www.yahoo.com_
  Change       : 414.99% | 0.0101666666667 -> 0.0523571428571

Revision             Result                       N
chromium@501704      0.0101667 +- 0.0032914       6       good
chromium@503369      0.0123125 +- 0.00717417      8       good
chromium@503421      0.0107143 +- 0.00639754      7       good
chromium@503422      0.05 +- 0.0193391            6       bad       <--
chromium@503423      0.0465 +- 0.0150333          5       bad
chromium@503425      0.0485833 +- 0.0100602       6       bad
chromium@503428      0.0512857 +- 0.011808        7       bad
chromium@503434      0.0561667 +- 0.0361778       6       bad
chromium@503447      0.0476 +- 0.0131605          5       bad
chromium@503473      0.056 +- 0.0168523           5       bad
chromium@503578      0.0495 +- 0.0108397          7       bad
chromium@503786      0.13605 +- 0.585129          10      bad
chromium@504202      0.0522 +- 0.0132212          5       bad
chromium@505034      0.0523571 +- 0.020527        7       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 --story-filter=http...www.yahoo.com. tracing.tracing_with_background_memory_infra

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966935590799816112


For feedback, file a bug with component Speed>Bisection

I believe that such a regression will be expected because we are intentionally adding more data to memory dumps so it makes sense that the CPU overhead would also rise.

Yes it is expected. I thought tracing.tracing_with_background_memory_infra was not part of the sheriff queue. It should only send alert emails to me (there is a memory-infra sheriff email id).
shend@ is it not true anymore?

Hi ssid, apologies for the confusion, I'm not actually sheriffing. I landed a patch recently and I ran a few bisects to see if it caused any regressions. I must've ran this bisect by mistake as it's not relevant to my patch. I'm closing this now. Sorry for the confusion!

 Issue 770895  has been merged into this issue.