New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 645553 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 645569
Owner:
Last visit > 30 days ago
Closed: Sep 2016
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

3%-48.5% regression in memory.top_10_mobile at 417088:417296

Project Member Reported by rsch...@chromium.org, Sep 9 2016

Issue description

See the link to graphs below.
 
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Sep 10 2016


===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@417087  1763573  2501.64  5  good
chromium@417140  1737359  849.23   5  good
chromium@417153  1737359  1997.92  5  good
chromium@417166  2598174  1543.51  5  bad
chromium@417192  2600304  1436.47  5  bad
chromium@417296  2605383  3084.4   5  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 645553

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests memory.top_10_mobile
Test Metric: foreground-memory:chrome:all_processes:reported_by_os:system_memory:ashmem:proportional_resident_size_avg/http_m_intl_taobao_com_group_purchase_html
Relative Change: 47.73%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1611
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001972179483550768


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6732840547385344

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!
Narrowed to 417153-417166. Trying some bisects there.
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Sep 13 2016

Mergedinto: 645569
Status: Duplicate (was: Assigned)

===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : cc: Change preraster/predecode split to be half of max skewport extent.
Author  : vmpstr
Commit description:
  
This patch iterates on the raster/decode split to include skewport into
consideration. Specifically, in addition to limiting preraster to 1000
pixels, we also limit it by half of the skewport extent, where skewport
extent is the largest padding in any direction away from the viewport.

The tough scrolling cases seems to indicate that checkerboards
remain roughly the same with this patch.

R=enne
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2317913002
Cr-Commit-Position: refs/heads/master@{#417164}
Commit  : 1a4f3af8e0b96d1517939be11395132f9faa5e8c
Date    : Thu Sep 08 02:16:50 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@417153  8115978  36701.7  5  good
chromium@417160  8134820  30773.7  5  good
chromium@417163  8139079  18997.4  5  good
chromium@417164  8960409  28659.9  5  bad    <--
chromium@417165  8943534  16038.2  5  bad
chromium@417166  8947138  18844.9  5  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 645553

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests memory.top_10_mobile
Test Metric: foreground-memory:chrome:all_processes:reported_by_os:system_memory:ashmem:proportional_resident_size_avg/https_mobile_twitter_com_justinbieber_skip_interstitial_true
Relative Change: 10.24%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/4117
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001682273294115936


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5782628492378112

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Sep 13 2016


===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@417087  1880227  621.334  5  good
chromium@417140  1856061  224.566  5  good
chromium@417153  1855979  342.672  5  good
chromium@417160  1864007  16113.9  5  good
chromium@417163  1859748  466.99   5  good
chromium@417165  1933476  224.019  5  bad
chromium@417166  1933558  366.212  5  bad
chromium@417192  1933721  409.5    5  bad
chromium@417296  1937653  224.019  5  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 645553

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests memory.top_10_mobile
Test Metric: foreground-memory:chrome:all_processes:reported_by_os:system_memory:ashmem:proportional_resident_size_avg/http_yandex_ru_touchsearch_text_science
Relative Change: 3.05%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1614
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001682251324288080


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5859217422942208

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

Sign in to add a comment