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

Issue 650590 link

Starred by 1 user
Status: WontFix
Owner:
█ hpayer@chromium.org
Last visit > 30 days ago
Closed: Oct 2016
Cc:
jochen@chromium.org
█ hpayer@chromium.org
sullivan@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

22.6%-58.5% regression in memory.long_running_idle_gmail_tbmv2 at 406346:420666

Project Member Reported by rmcilroy@chromium.org, Sep 27 2016 
See the link to graphs below.
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Sep 27 2016 


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


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev  N   Good?
chromium@420258  58638255  3426697  18  good
chromium@420286  59418878  2835549  18  bad

Bisect job ran on: win_perf_bisect
Bug ID: 650590

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_tbmv2
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:allocated_objects_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:allocated_objects_size_avg
Relative Change: 3.47%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/6943
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9000373627025253744


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5782205563928576

| 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 7 by 42576172...@developer.gserviceaccount.com, Oct 1 2016

Cc: hpayer@chromium.org
Owner: hpayer@chromium.org

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

Hi hpayer@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 : [heap] Make typed slot set state and operations atomic.
Author  : hpayer
Commit description:
  
BUG= chromium:648568 

Review-Url: https://codereview.chromium.org/2360513002
Cr-Commit-Position: refs/heads/master@{#39596}
Commit  : 6f06c36c3e1ecb41f1d301ad2159366f876d53c8
Date    : Wed Sep 21 13:20:48 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev  N   Good?
chromium@420258                26.6095  2.42349  18  good
chromium@420272                27.5632  1.63299  18  good
chromium@420276                27.0686  2.02282  8   good
chromium@420278                27.986   1.14252  12  good
chromium@420278,v8@b097c6c4f1  26.7632  2.76428  12  good
chromium@420278,v8@d2c975d217  27.0502  2.25366  12  good
chromium@420278,v8@6f06c36c3e  30.4078  3.56475  12  bad    <--
chromium@420278,v8@5784773feb  31.0301  2.38306  12  bad
chromium@420279                30.6321  1.01804  8   bad
chromium@420286                29.5341  3.09554  18  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 650590

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_tbmv2
Test Metric: v8-gc-total_max/v8-gc-total_max
Relative Change: 5.26%
Score: 98.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/934
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9000082541739721504


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5569308833349632

| 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, Oct 1 2016

Cc: jochen@chromium.org
Owner: jochen@chromium.org

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

Hi jochen@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 : Reland of "Release streamed script resources after it was compiled"
Author  : jochen
Commit description:
  
Original issue's description:
> Release streamed script resources after it was compiled
>
> Otherwise, we'd hold on to the resources until the embedder frees them
> which might take a long time
>
> R=marja@chromium.org,verwaest@chromium.org
> BUG=
>
> Committed: https://crrev.com/877dac34465c018bb534b7781fbe242ae4e33c32
> Cr-Commit-Position: refs/heads/master@{#38999}

TBR=marja@chromium.org,verwaest@chromium.org
BUG= chromium:642347 

Review-Url: https://codereview.chromium.org/2296733002
Cr-Commit-Position: refs/heads/master@{#39018}
Commit  : 60c75fbe4124215a7e0a5beabe24f16bd977c1cb
Date    : Tue Aug 30 14:11:20 2016


===== TESTED REVISIONS =====
Revision                       Mean      Std Dev  N   Good?
chromium@406345                79314725  2902577  27  good
chromium@413506                79566041  2746446  27  good
chromium@415296                80033682  2340599  12  good
chromium@415352                80008024  1269519  12  good
chromium@415380                80182607  1743667  8   good
chromium@415384                80871534  1976817  12  good
chromium@415385                79518979  2193043  12  good
chromium@415385,v8@e2361954e5  80753212  3966967  12  good
chromium@415385,v8@60c75fbe41  75474989  2209105  12  bad    <--
chromium@415385,v8@bdf5566281  76209115  2181244  8   bad
chromium@415386                76669252  2660103  8   bad
chromium@415387                75225693  2300245  5   bad
chromium@415394                76199342  2463188  5   bad
chromium@415408                75899019  3353533  12  bad
chromium@415520                75594159  3609101  12  bad
chromium@415743                76540869  2135715  8   bad
chromium@416190                76618843  3305933  12  bad
chromium@417086                74205068  3886990  12  bad
chromium@420666                76570264  3490701  18  bad

Bisect job ran on: win_perf_bisect
Bug ID: 650590

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_tbmv2
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:effective_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:effective_size_avg
Relative Change: 1.23%
Score: 99.5

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/6956
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9000084713294660832


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4547869812457472

| 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!

Comment 9 by rmcilroy@chromium.org, Oct 3 2016

Owner: hpayer@chromium.org
Hmm, seems more likely to be Hannes's CL given it is a regression in GC metrics.

Comment 10 by hpayer@chromium.org, Oct 5 2016

Owner: jochen@chromium.org
Two regressions here. The time metrics regressed by the slot set CL. The space metrics by Jochen's CL. I am moving the slot CL regressions to a different bug and adding Jochen as an owner for the space regressions.

Comment 11 by jochen@chromium.org, Oct 6 2016

Owner: hpayer@chromium.org
my cl reduces memory by 500k, why is that "bad"

Comment 12 by rmcilroy@chromium.org, Oct 6 2016

Cc: sullivan@chromium.org
Status: WontFix (was: Assigned)
Hmm strange, the graphs show a memory increase however the bisects show a memory reduction on your CL like you say.

I just noticed that these alerts are all on the chromium-win-clang bot, which is an FYI bot that doesn't have any bisectors (see https://github.com/catapult-project/catapult/issues/2883). The Dashboard UI suggests using the android_webview_aosp_perf_bisect bot (which is what these bisecs above were testing) which is completely wrong, so the bisects aren't useful. Closing as WontFix since this only occurred on FYI bot.
►

Sign in to add a comment