All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=898916

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=be59e312538092bf52bce79178eb94c415f477d7c7ba34e24da914bc5479e69e


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

Android Nexus5X WebView Perf
Win 7 Nvidia GPU Perf

memory.desktop - Benchmark documentation link:
  None

system_health.memory_desktop - Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

memory.top_10_mobile - Benchmark documentation link:
  None

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/117d5d11e40000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/117d5d11e40000

alexilin, looks like your change (https://chromium.googlesource.com/chromium/src/+/81c4128e3f75163f0a99c021c3bfc8ce5c0d7a0f) deal with memory, could you take a look?

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/125b9e3de40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/125b9e3de40000

[DOMStorage] Enable onion soup codepath by dmurph@chromium.org
https://chromium.googlesource.com/chromium/src/+/80ce30280a48850aad465138b84fe209474733e8
memory:chrome:all_processes:reported_by_chrome:malloc:effective_size: 1.817e+07 → 1.828e+07 (+1.122e+05)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

This onion soup change modifies where memory is stored in a couple places. We have both regressions and improvements and they seem to equal out. I'm assuming this is part of those movements.