Issue metadata
Sign in to add a comment
|
1.4% regression in storage.indexeddb_endure at 450438:450531 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Feb 17 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987419787395724128
,
Feb 17 2017
=== Auto-CCing suspected CL author yzshen@chromium.org === Hi yzshen@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 : yzshen Commit : 2859a2ac06ab5d9df6706cc45525dc4a2085051c Date : Tue Feb 14 22:24:25 2017 Subject: Mojo C++ bindings: Simplify associated interface API. Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : storage.indexeddb_endure Metric : vm_private_dirty_final_browser/vm_private_dirty_final_browser Change : 1.94% | 25284.9090909 -> 25774.9090909 Revision Result N chromium@450437 25284.9 +- 134.353 6 good chromium@450484 25319.9 +- 305.058 6 good chromium@450490 25282.8 +- 322.32 6 good chromium@450493 25363.0 +- 174.977 6 good chromium@450494 25851.1 +- 52.6601 6 bad <-- chromium@450495 25841.6 +- 113.714 6 bad chromium@450496 25778.2 +- 285.175 6 bad chromium@450508 25865.8 +- 219.329 6 bad chromium@450531 25774.9 +- 362.969 6 bad 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 storage.indexeddb_endure Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8987419787395724128 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=4961172140326912 | 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!
,
Feb 17 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36 commit 9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36 Author: yzshen <yzshen@chromium.org> Date: Fri Feb 17 23:07:50 2017 Mojo C++ bindings: change some std::unique_ptr<base::Lock> to base::Optional<base::Lock>. This helps to reduce heap allocation. BUG= 651103 , 693476 Review-Url: https://codereview.chromium.org/2707483002 Cr-Commit-Position: refs/heads/master@{#451394} [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/connector.h [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/lib/connector.cc [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/lib/may_auto_lock.h [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/lib/multiplex_router.cc [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/lib/multiplex_router.h [modify] https://crrev.com/9ac6e6466cc0df7e1a3ad4488c5c8bdc2db4da36/mojo/public/cpp/bindings/lib/scoped_interface_endpoint_handle.cc
,
Feb 22 2017
I think the memory usage has dropped to its original level. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by alexclarke@chromium.org
, Feb 17 2017