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

Issue 664191 link

Starred by 1 user

Issue metadata

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

Blocked on:
issue 664510
issue 664743
issue 665439



Sign in to add a comment

5.1%-33.3% regression in memory.top_10_mobile at 431017:431050

Project Member Reported by rmcilroy@chromium.org, Nov 10 2016

Issue description

See the link to graphs below.
 
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Nov 11 2016


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


=== 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.

===== TESTED REVISIONS =====
Revision         Mean  Std Dev  N  Good?
chromium@431022  N/A   N/A      0  good
chromium@431075  N/A   N/A      0  bad

Bisect job ran on: android_nexus7_perf_bisect
Bug ID: 664191

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.search.yahoo.com.search..ylt.p.google memory.top_10_mobile
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_search_yahoo_com_search__ylt_p_google
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/3467
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8996282337085463152


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5775603960119296

| 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!
Blockedon: 664510
Project Member

Comment 8 by 42576172...@developer.gserviceaccount.com, Nov 12 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@431022  4614590  479492   27  good
chromium@431075  4573911  411975   27  bad

Bisect job ran on: android_nexus7_perf_bisect
Bug ID: 664191

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.search.yahoo.com.search..ylt.p.google memory.top_10_mobile
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_search_yahoo_com_search__ylt_p_google
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/3469
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8996200184971116032


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6165001532866560

| 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!
Blockedon: 664743
Cc: heimbuef@google.com verwa...@chromium.org
I've kicked off a bisect again. Looking at the V8 rolls in the range, the only CL which seems relevent is "[Parser] Added SMI cache" 

Toon: Do is it possible that the SMI cache CL could be regressing V8 effective size memory?
It won't regress it by a lot. There's an 8KB cache in the parser, that improves memory usage a lot if you reuse lots of smis like in asm.js. We can easily make the cache smaller though; I presume the first 10 would already be good enough...
This is a regression of about 1MB or so, so unless there is a bug in the SMI cache, it seems like it's probably not that.
Project Member

Comment 15 by 42576172...@developer.gserviceaccount.com, Nov 14 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@431022  78491633  2553627  27  good
chromium@431075  78607573  3494677  27  bad

Bisect job ran on: android_nexus7_perf_bisect
Bug ID: 664191

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.www.baidu.com.s.word.google memory.top_10_mobile
Test Metric: memory:chrome:all_processes:reported_by_os:system_memory:proportional_resident_size_avg/background/after_http_www_baidu_com_s_word_google
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/3470
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8996006812565594400


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6348045975814144

| 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, Nov 15 2016


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


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev  N   Good?
chromium@431022                4787623  222748   18  good
chromium@431049                4812721  682320   27  good
chromium@431062                4798846  442090   18  good
chromium@431066                4829685  433505   18  good
chromium@431066,v8@0700bc1bf8  4819817  376403   12  good
chromium@431066,v8@472cb045c5  4766608  665838   27  unknown
chromium@431066,v8@4cfe2e82c6  4691244  92658.5  8   bad
chromium@431067                4731737  403889   18  bad
chromium@431068                4777676  600437   27  unknown
chromium@431069                4745638  694009   27  bad
chromium@431075                4722731  235095   18  bad

Bisect job ran on: android_nexus7_perf_bisect
Bug ID: 664191

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=https.www.google.co.uk.hl.en.q.science memory.top_10_mobile
Test Metric: memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_https_www_google_co_uk_hl_en_q_science
Relative Change: 1.36%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/3471
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8996006769390045504


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5897327225602048

| 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!
Blockedon: 665439
Project Member

Comment 19 by 42576172...@developer.gserviceaccount.com, Nov 16 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@430986  20579482  1547575  27  good
chromium@431126  20446400  1251971  27  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 664191

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.m.intl.taobao.com.group.purchase.html memory.top_10_mobile
Test Metric: memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/foreground/http_m_intl_taobao_com_group_purchase_html
Relative Change: None

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/4350
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8995841925576972288


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5270225946148864

| 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!
Status: WontFix (was: Untriaged)
These regressions happened before M56 branch. M56 is now in stable. These regressions made it to the stable channel. Marking wontfix.
Labels: Performance-Memory
Project Member

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


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

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/foreground/http_m_intl_taobao_com_group_purchase_html

Revision             Result                   N
chromium@430986      20579482 +- 1547575      27      good
chromium@431126      20446400 +- 1251971      27      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=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.m.intl.taobao.com.group.purchase.html memory.top_10_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982632207638604176

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5270225946148864


| 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!
Project Member

Comment 25 by 42576172...@developer.gserviceaccount.com, Apr 11 2017


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

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_search_yahoo_com_search__ylt_p_google


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=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.search.yahoo.com.search..ylt.p.google memory.top_10_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982596457404448000

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5775603960119296


| 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