All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=901750

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=265f7cfbd2d4977dfbfbe251e2b5808bc02fc18122560b1bb53803e543332780


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

linux-perf
mac-10_12_laptop_low_end-perf
mac-10_13_laptop_high_end-perf

v8.browsing_desktop - Benchmark documentation link:
  None

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

memory.desktop - Benchmark documentation link:
  None

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1784c305e40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1784c305e40000

[snapshot] Remove the builtins snapshot by jgruber@chromium.org
https://chromium.googlesource.com/v8/v8/+/4ef0e79cba592ee16d279587dc8393bfb229f995
v8-gc-incremental-finalize: 0.1367 → 0.1685 (+0.03172)

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

Benchmark documentation link:
  None

Interesting.. I'd be surprised if this CL also caused the memory regression in old space:

https://chromeperf.appspot.com/report?sid=6c01d89781701ef17eb32cbf075896374b33b0d86659f3f5ad2915ceb5d4a517&start_rev=603012&end_rev=605286

Kicking off a bisect there.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1588a96de40000

The gc-time regression is explained by addtl dispatch table iteration:

https://chromium-review.googlesource.com/c/v8/v8/+/1304539/8/src/heap/heap.cc#3809

+delphick fyi, pinpoint says dispatch table iteration costs 23% in gc-incremental-finalize (wow).

📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1588a96de40000

[Presentation API] Remove from insecure contexts. by mfoltz@chromium.org
https://chromium.googlesource.com/chromium/src/+/592f6faa792259bb3515fea429699c3e126a143e
memory:chrome:all_processes:reported_by_chrome:v8:effective_size: 2.78e+06 → 2.771e+06 (-8652)

inspector: move injected script source to native by kozyatinskiy@chromium.org
https://chromium.googlesource.com/v8/v8/+/7e079c660b686d916ed5cd88bfa4ecf597300193
memory:chrome:all_processes:reported_by_chrome:v8:effective_size: 2.774e+06 → 3.018e+06 (+2.436e+05)

[snapshot] Remove the builtins snapshot by jgruber@chromium.org
https://chromium.googlesource.com/v8/v8/+/4ef0e79cba592ee16d279587dc8393bfb229f995
memory:chrome:all_processes:reported_by_chrome:v8:effective_size: 3.018e+06 → 3.018e+06 (+10.93)

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

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

I'm not familiar with this benchmark , but  it appears to have improved after my commit. Versus the other two. 

memory:chrome:all_processes:reported_by_chrome:v8:effective_size increases with

inspector: move injected script source to native
By kozyatinskiy@chromium.org

Extracted gc-incremental-finalize things (see #6) to  https://crbug.com/902230 .

kozy@ is no longer working on it, over to jgruber@.

Looking the pinpoint results again, the regression actually comes from 

[snapshot] Remove the builtins snapshot
By jgruber@chromium.org

Chromiumdash: https://chromiumdash.appspot.com/commit/4ef0e79cba592ee16d279587dc8393bfb229f995

Overall the signals are mostly positive, and I only see a single real site that seems to have regressed by a bit. I don't think it's worth spending time on this.