New issue
Advanced search Search tips

Issue 616936 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner: ----
Closed: Aug 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

4.1% regression in startup.warm.blank_page at 397045:397076

Project Member Reported by briander...@chromium.org, Jun 2 2016

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=616936

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg3MDHpQoM


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

chromium-rel-win7-single
Project Member

Comment 2 by sheriffbot@chromium.org, Jul 7 2016

Labels: -M-53 M-54 MovedFrom-53
Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Last bisect has narrowed down the range to 397026:397035 but mothing in that range jumps out to me! Kicking another bisect with that range.

http://test-results.appspot.com/revision_range?start=397026&end=397035

2016-07-03 18:08:34 
===== BISECT JOB RESULTS =====
Status: started


===== TESTED REVISIONS =====
Revision         Mean   Std Dev  N  Good?
chromium@397026  552.6  6.10737  5  good
chromium@397035  808.4  223.742  5  bad
chromium@397043  643.6  4.219    5  bad
chromium@397064  718.8  18.6869  5  bad

Bisect job ran on: winx64_zen_perf_bisect
Bug ID: 616936

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests startup.warm.blank_page
Test Metric: foreground_tab_request_start/foreground_tab_request_start
Relative Change: 30.08%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64_zen_perf_bisect/builds/241
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9008138049139558800

That bisect doesn't look super clean. Either way, trying another.
Cc: hirosh...@chromium.org
Owner: hirosh...@chromium.org

=== Auto-CCing suspected CL author hiroshige@chromium.org ===

Hi hiroshige@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


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


===== SUSPECTED CL(s) =====
Subject : Do not initiate revalidation of Resource with redirects from MemoryCache
Author  : hiroshige
Commit description:
  
BUG= 613971 

Review-Url: https://codereview.chromium.org/2011283002
Cr-Commit-Position: refs/heads/master@{#397064}
Commit  : 64ad58e1d8451857f6a763651727c9d39375ea3b
Date    : Wed Jun 01 05:16:57 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@397026  659.583  60.8612  12  good
chromium@397043  666.583  42.9703  12  good
chromium@397054  506.2    5.54076  5   good
chromium@397059  508.8    7.04982  5   good
chromium@397062  494.2    7.39594  5   good
chromium@397063  546.8    12.7945  5   good
chromium@397064  731.75   35.3018  8   bad    <--

Bisect job ran on: winx64_zen_perf_bisect
Bug ID: 616936

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests startup.warm.blank_page
Test Metric: foreground_tab_load_complete/foreground_tab_load_complete
Relative Change: 2.46%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64_zen_perf_bisect/builds/362
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005127659133418576


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5900869707497472

| 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!
Perf sheriff ping: reminder to follow up on possible performance issues
Friendly sheriff ping: Have you had a chance to look into this regression? Thanks!
Hmm. My change only affects revalidation of resources that contains redirects, and thus is unlikely to affect the test. When tested on Linux, the code I changed doesn't seem to be executed (inserted CHECK(0) there and wasn't executed).

Started bisects again...
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Sep 10 2016


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


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@397049  136.884  0.826668  5  good
chromium@397062  137.674  0.564901  5  good
chromium@397069  137.8    0.385686  5  good
chromium@397071  138.642  0.478629  5  bad
chromium@397074  139.547  0.87089   5  bad

Bisect job ran on: winx64nvidia_perf_bisect
Bug ID: 616936

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests startup.warm.blank_page
Test Metric: first_main_frame_load_time/first_main_frame_load_time
Relative Change: 1.95%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64nvidia_perf_bisect/builds/1863
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001942002198859504


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6138928912924672

| 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 15 by 42576172...@developer.gserviceaccount.com, Sep 10 2016


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


===== TESTED REVISIONS =====
Revision         Mean   Std Dev  N  Good?
chromium@397026  537.6  21.3612  5  good
chromium@397027  536.4  6.22896  5  good
chromium@397035  641.2  27.335   5  bad
chromium@397043  655.0  8.30662  5  bad
chromium@397064  759.0  31.9687  5  bad

Bisect job ran on: winx64_zen_perf_bisect
Bug ID: 616936

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests startup.warm.blank_page
Test Metric: foreground_tab_load_complete/foreground_tab_load_complete
Relative Change: 41.18%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64_zen_perf_bisect/builds/473
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001943254950245936


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4574400840466432

| 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 16 by 42576172...@developer.gserviceaccount.com, Sep 10 2016


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


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@397026  503.407  21.5411  27  good
chromium@397027  498.583  15.9343  12  good
chromium@397035  520.37   24.5422  27  bad
chromium@397043  512.0    28.0651  18  bad
chromium@397064  526.75   10.3475  8   bad

Bisect job ran on: winx64_zen_perf_bisect
Bug ID: 616936

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests startup.warm.blank_page
Test Metric: foreground_tab_request_start/foreground_tab_request_start
Relative Change: 3.03%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64_zen_perf_bisect/builds/474
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9001943244892946416


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5860786293964800

| 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!
Perf sheriff ping
Only one bot, but there is a ref build result and it appears that the ref build performance didn't regress; started another bisect job but if we can't reproduce it with bisect bots then I'm not sure how we'll find the cause.
Project Member

Comment 20 by 42576172...@developer.gserviceaccount.com, Oct 17 2016

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64_zen_perf_bisect/builds/553
Failure reason: the build has failed.
Additional errors:
The metric was not found in the test output.
Either of the initial "good" or "bad" revisions failed to be tested or built.

Owner: ----
Status: Available (was: Assigned)
Unassign myself as my change is not executed in the tests (Comment #13).
Project Member

Comment 24 by 42576172...@developer.gserviceaccount.com, May 18 2017

Owner: hirosh...@chromium.org

=== Auto-CCing suspected CL author hiroshige@chromium.org ===

Hi hiroshige@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : hiroshige
  Commit : 64ad58e1d8451857f6a763651727c9d39375ea3b
  Date   : Wed Jun 01 05:16:57 2016
  Subject: Do not initiate revalidation of Resource with redirects from MemoryCache

Bisect Details
  Configuration: winx64_zen_perf_bisect
  Benchmark    : startup.warm.blank_page
  Metric       : foreground_tab_request_start/foreground_tab_request_start
  Change       : 13.88% | 513.833333333 -> 585.166666667

Revision             Result                  N
chromium@397026      513.833 +- 38.7406      6      good
chromium@397043      512.333 +- 43.8102      6      good
chromium@397054      451.833 +- 77.1416      6      good
chromium@397059      484.0 +- 59.716         6      good
chromium@397062      451.5 +- 41.7073        6      good
chromium@397063      505.222 +- 76.7565      9      good
chromium@397064      585.167 +- 71.7136      6      bad       <--

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=2 --also-run-disabled-tests startup.warm.blank_page

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8979227173964723680

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5810328853544960


| 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 Speed>Bisection.  Thank you!
Owner: ----
Unassign myself again (According to my Comment #13 my changed code path is not executed in the tests).

Project Member

Comment 26 by 42576172...@developer.gserviceaccount.com, May 19 2017


=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : startup.warm.blank_page
  Metric       : first_main_frame_load_time/first_main_frame_load_time
  Change       : 1.43% | 132.5 -> 133.571428571

Suspected Commit Range
  2 commits in range
  https://chromium.googlesource.com/chromium/src/+log/d64a9dbc5611dcab7a850f8bbadfcc12d6e7b465..04b26cac6e20af2aff50b8a064691c605d79ae13


Revision             Result                  N
chromium@397042      132.5 +- 4.41588        14       good
chromium@397043      ---                     ---      build failure
chromium@397044      135.0 +- 13.2665        21       bad
chromium@397050      134.071 +- 4.11443      14       bad
chromium@397062      133.762 +- 7.3355       21       bad
chromium@397071      133.571 +- 3.6645       14       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=2 --also-run-disabled-tests startup.warm.blank_page

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8979227152916851216

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5869298486083584


| 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 Speed>Bisection.  Thank you!
Status: WontFix (was: Available)
Doesn't look like we can find the cause of this. Metric has gone back down.

Sign in to add a comment