All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=922815

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=7870535b747f437eae8bfbdb09506815f13641d5d27413f7b6f6b338db82b7ec


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

mac-10_12_laptop_low_end-perf

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

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/15939e42540000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/15939e42540000

Switch Mac OS HarfBuzz hb_face creation to table copying by drott@chromium.org
https://chromium.googlesource.com/chromium/src/+/023f5763fc17c435b386782af5b7217ebcb09b15
memory:chrome:all_processes:reported_by_os:system_memory:private_footprint_size: 2.119e+08 → 1.983e+08 (-1.357e+07)

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

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

Thanks, yes, this is basically recovering from  issue 908552 .