New issue
Advanced search Search tips

Issue 665319 link

Starred by 0 users
Status: WontFix
Owner:
jgruber@chromium.org
Closed: Dec 2016
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

4.6% regression in system_health.memory_desktop at 429439:430822

Project Member Reported by jgruber@chromium.org, Nov 15 2016 
See the link to graphs below.

Comment 1 by jgruber@chromium.org, Nov 15 2016 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=665319

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgi7PvtwkM


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

chromium-rel-mac10
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Nov 15 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N    Good?
chromium@429438  29871663  473903947  265  good
chromium@430822  30766160  498335411  265  bad

Bisect job ran on: mac_10_10_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2479
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995944619864408784


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5819891112738816

| 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 5 by 42576172...@developer.gserviceaccount.com, Nov 15 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N    Good?
chromium@430477  30684944  419450777  212  good
chromium@430592  31860570  564763605  265  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/1028
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995944576586507344


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5829676709183488

| 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 7 by rbyers@chromium.org, Nov 18 2016

Labels: Performance
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Nov 20 2016 

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2485
Failure reason: the build has failed due to infrastructure failure.

Comment 9 by jgruber@chromium.org, Nov 21 2016

Owner: rmcilroy@chromium.org
Speculatively assigning to rmcilroy for 

cf5b0c5 [Interpreter] Switch back to optimizing after 2 ticks for Ignition code. by rmcilroy ยท 13 days ago

For some reason bisects fail even though the regression is clear on browse_news and load_news benchmarks.

Comment 10 by rmcilroy@chromium.org, Nov 21 2016

Owner: jgruber@chromium.org
I'm pretty sure this is nothing to do with the "Switch back to optimizing after 2 ticks for Ignition code" - this regression is in the object space (not the code space which might be affected by the change above). Also the set of code going through Ignition is pretty small (only Turbofanned code), so it's highly unlikely.

The regression range is massive (over 1300 Chromium CLs and tens of V8 rolls) so it could be anything (unless there was a specific reason you thought it was the CL above jgruber@ ?). I'll kick off another bisect and assign back to jgruber@.
Project Member

Comment 12 by 42576172...@developer.gserviceaccount.com, Nov 21 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N    Good?
chromium@429438  29814885  416850590  212  good
chromium@430822  29783765  287432392  106  bad

Bisect job ran on: mac_10_10_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2487
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995384485772792880


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6397025178877952

| 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 14 by 42576172...@developer.gserviceaccount.com, Dec 6 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N    Good?
chromium@429438  29825409  415681790  212  good
chromium@430822  30850727  395183456  159  bad

Bisect job ran on: mac_10_10_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg
Relative Change: 3.44%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2500
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994042356093401344


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5883100758802432

| 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 17 by 42576172...@developer.gserviceaccount.com, Dec 9 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N   Good?
chromium@429438  36020733  252470408  32  good
chromium@430822  40248178  268120578  24  bad

Bisect job ran on: mac_10_10_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/load_news
Relative Change: 11.74%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2509
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993752457027070752


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6211898926694400

| 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 18 by 42576172...@developer.gserviceaccount.com, Dec 9 2016 


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


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean      Std Dev    N   Good?
chromium@429438  30198870  132218151  35  good
chromium@430822  34612817  115038097  24  bad

Bisect job ran on: mac_10_10_perf_bisect
Bug ID: 665319

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/browse_news
Relative Change: 14.62%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_bisect/builds/2508
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8993752501413083840


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5262129622417408

| 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 19 by jgruber@chromium.org, Dec 13 2016

Status: WontFix (was: Assigned)
Closing, bisects fail to reproduce.
Project Member

Comment 21 by 42576172...@developer.gserviceaccount.com, Apr 10 2017 

Bisect failed: None
Failure reason: incorrect bisect configuation.
Project Member

Comment 23 by 42576172...@developer.gserviceaccount.com, Apr 11 2017 


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: mac_10_10_perf_bisect
  Benchmark    : system_health.memory_desktop
  Metric       : memory:chrome:renderer_processes:reported_by_chrome:v8:heap:allocated_objects_size_avg/browse_news

Revision             Result                     N
chromium@429438      30198870 +- 132218151      35      good
chromium@430822      34612817 +- 115038097      24      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests system_health.memory_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982632334095844128

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5262129622417408


| 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!
►

Sign in to add a comment