Issue metadata
Sign in to add a comment
|
2.9% regression in system_health.memory_mobile at 472365:472398 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
May 18 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8979252897059224928
,
May 19 2017
=== Auto-CCing suspected CL author rockot@chromium.org === Hi rockot@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 : Ken Rockot Commit : f378d95eebc60fd6568c4524abaec2aaa377c7d0 Date : Wed May 17 07:59:27 2017 Subject: Make Service Manager name resolution synchronous Bisect Details Configuration: android_webview_nexus6_aosp_perf_bisect Benchmark : system_health.memory_mobile Metric : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_search/load_search_google Change : 1.18% | 22124093.3333 -> 22384464.0 Revision Result N chromium@472364 22124093 +- 166883 6 good chromium@472373 22150939 +- 147634 6 good chromium@472375 22093186 +- 296372 9 good chromium@472376 22284065 +- 410170 14 bad <-- chromium@472377 22369890 +- 306775 9 bad chromium@472381 22341077 +- 295883 6 bad chromium@472398 22384464 +- 213884 6 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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.search.google system_health.memory_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8979252897059224928 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5951162576011264 | 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!
,
May 19 2017
I don't think this is accurate, so I'm guessing it's maybe just noise? There's nothing about the change that should increase memory usage - in fact it should almost certainly have improved any interesting performance metric as it deletes code, reduces task scheduling overhead, and generally reduces complexity of the system. https://chromeperf.appspot.com/group_report?rev=472376
,
May 19 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8979147265389544080
,
May 19 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8979147260583501152
,
May 19 2017
If the bisect is wrong, please unassign yourself. A lot of the time there are real regressions that need bisects on different platforms to reproduce. I kicked off a few more.
,
May 20 2017
=== Auto-CCing suspected CL author kbr@chromium.org === Hi kbr@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 : kbr Commit : ebbd3a5c5eee3ec38736e60c53f2ea1bc3faed6c Date : Wed May 17 10:55:21 2017 Subject: Add command_line to SystemInfo data structure. Bisect Details Configuration: android_webview_arm64_aosp_perf_bisect Benchmark : system_health.memory_mobile Metric : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/blank_about/blank_about_blank Change : 1.69% | 16570306.6667 -> 16850370.6667 Revision Result N chromium@472397 16570307 +- 67497.3 6 good chromium@472411 16559896 +- 109630 6 good chromium@472413 16586520 +- 72509.0 6 good chromium@472414 16571160 +- 70052.4 6 good chromium@472415 16863000 +- 131798 6 bad <-- chromium@472418 16869485 +- 113942 6 bad chromium@472424 16850371 +- 57829.6 6 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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=blank.about.blank system_health.memory_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8979147260583501152 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=6409362169397248 | 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!
,
May 20 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : kbr Commit : ebbd3a5c5eee3ec38736e60c53f2ea1bc3faed6c Date : Wed May 17 10:55:21 2017 Subject: Add command_line to SystemInfo data structure. Bisect Details Configuration: android_webview_arm64_aosp_perf_bisect Benchmark : system_health.memory_mobile Metric : memory:webview:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/background_social/background_social_facebook Change : 2.15% | 20076312.0 -> 20508781.3333 Revision Result N chromium@472397 20076312 +- 122889 6 good chromium@472411 20164717 +- 120377 6 good chromium@472413 20245443 +- 56243.9 6 good chromium@472414 20291693 +- 104304 6 good chromium@472415 20557251 +- 64839.0 6 bad <-- chromium@472418 20524995 +- 79535.4 6 bad chromium@472424 20508781 +- 143730 6 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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=background.social.facebook system_health.memory_mobile Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8979147265389544080 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5891625605136384 | 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!
,
May 23 2017
I agree looking at the graphs that it seems there was an increase in memory around this time, but my change adds a simple string to a data structure returned via DevTools to Telemetry's harness at the beginning of time, and I can't imagine how that could have led to a persistent increase in the browser's memory usage on these benchmarks, even if Telemetry's calling this API at the beginning of each page. Closing as WontFix.
,
Jun 16 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/958a49a3e9f2f2fe39e5059f73f660e89de8fba3 commit 958a49a3e9f2f2fe39e5059f73f660e89de8fba3 Author: Kenneth Russell <kbr@chromium.org> Date: Fri Jun 16 23:11:38 2017 Avoid redundant eglMakeCurrent calls. The previous fix for initialization of a new virtual context caused the real context to be released. Avoid this by adding a new entry point to release only the virtual context so it can be quickly made current again. BUG= 718809 , 724071 Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel Change-Id: Ieb8c4bfd1db5dfa90168f3a859182efa9c39d0d3 Reviewed-on: https://chromium-review.googlesource.com/534779 Reviewed-by: John Bauman <jbauman@chromium.org> Commit-Queue: Kenneth Russell <kbr@chromium.org> Cr-Commit-Position: refs/heads/master@{#480211} [modify] https://crrev.com/958a49a3e9f2f2fe39e5059f73f660e89de8fba3/gpu/command_buffer/service/gl_context_virtual.cc [modify] https://crrev.com/958a49a3e9f2f2fe39e5059f73f660e89de8fba3/gpu/command_buffer/service/gl_context_virtual.h [modify] https://crrev.com/958a49a3e9f2f2fe39e5059f73f660e89de8fba3/gpu/ipc/service/gpu_command_buffer_stub.cc [modify] https://crrev.com/958a49a3e9f2f2fe39e5059f73f660e89de8fba3/ui/gl/gl_context.cc [modify] https://crrev.com/958a49a3e9f2f2fe39e5059f73f660e89de8fba3/ui/gl/gl_context.h
,
Jun 17 2017
I realize now I probably shouldn't have referenced this bug in the CL above. I thought that this one had the same culprit CL as Issue 718809 . |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by tdres...@chromium.org
, May 18 2017