Issue 576552 in chromium: Performance Regression: 5-10% page load time across most boards

[chro...@googlecode.com]

Status: Untriaged
Owner: ----
CC: rohi...@chromium.org
Labels: Type-Bug-Regression Pri-1 OS-Chrome Performance Cros-Perf-Monitor
M-49

New issue 576552 by mshe...@chromium.org: Performance Regression: 5-10%
page load time across most boards
https://code.google.com/p/chromium/issues/detail?id=576552

Overview of Regression:
Across ~all boards there has been a 5-10% regression in page load time
(page_cycler.typical_25/cold_times.page_load_time) in 49-dev starting w/
7678.0.0

Size of Regression:
The regression ranges in size from roughly 5-10%

Affected Device(s)/Platform(s):
This regression appears across all devices

Affected Build(s):
Chrome OS: builds >= 7678.0.0
Chrome version: ~49.0.2574.0
Chrome base commit: 361527

Crosbolt Link:
https://cros-goldeneye.corp.google.com/console/listCrosbolt?baselineMilestoneChannel=47%20%3A%20STABLE&comparisonMilestoneChannel=49%20%3A%20DEV&comparisonLimitResults=3&analysisType=BUILD_TO_BUILD&selectedDevices&graphSKU=snow_arm_v7l_rev4_2Gb&graphTest=page_cycler.typical_25%2Fcold_times.page_load_time

Good Graphs to View (click cell to view):
- Page Load Time / snow
- Page Load Time / jaq
- Page Load Time / heli
And any or all other red (or yellow) "Scroll Smoothness Percentage" cells

For a regression across all Chrome OS platforms, is there a similar
regression on Chromeperf for OS X and/or Win systems:
There is a simlar regression on Chromeperf for OS X @ commit position 361485
https://chromeperf.appspot.com/report?sid=5414556054bd9d6da0fe448143b8eceb10836f40fcf98a82a7ac831c5d5f529d&start_rev=360463&end_rev=362131

Notes:
While the Crosbolt graphs are fairly noisy in spots, there is a definite
step at 7678.0

--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings

[chro...@googlecode.com]

Updates:
Blockedon: chromium:564223

Comment #1 on issue 576552 by mshe...@chromium.org: Performance Regression:

5-10% page load time across most boards
https://code.google.com/p/chromium/issues/detail?id=576552

Based on Chromeperf graph, this is likely blocked by the following Issue
564223.

[chro...@googlecode.com]

Updates:
Cc: rsch...@chromium.org

Comment #2 on issue 576552 by mshe...@chromium.org: Performance Regression:

5-10% page load time across most boards
https://code.google.com/p/chromium/issues/detail?id=576552

Hey Ryan,

Just wanted to confirm with you that when we find regressions that
duplicate Chromeperf regressions that have bugs filed against them, that we
should file our issue as blocked on that issue?

Thanks!

Best,
Matt

[chro...@googlecode.com]

Comment #5 on issue 576552 by mshe...@google.com: Performance Regression:

5-10% page load time across most boards
https://code.google.com/p/chromium/issues/detail?id=576552

Issue chrome-os-partner:51437 has been merged into this issue.