Consistent crash on smoothness.image_decoding_cases |
|||
Issue descriptionhttps://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus7v2%20Perf%20%282%29/builds/1592/steps/smoothness.image_decoding_cases/logs/stdio Started failing consistently on N7v2 after #384560 Launching a return_code bisect to attempt finding a culprit cl
,
Apr 19 2016
===== BISECT JOB RESULTS ===== Status: completed ===== SUSPECTED CL(s) ===== Subject : Revert of Propagate is_official_build to version script. (patchset #1 id:1 of https://codereview.chromium.org/1838173005/ ) Author : skyostil Commit description: Reason for revert: Made all the Android perf bots red: https://bugs.chromium.org/p/chromium/issues/detail?id=599830 Original issue's description: > Propagate is_official_build to version script. > > OFFICIAL_BUILD is a special variable in the version script which > takes its value from the environment variable. This means that local > builds with is_official_build will not actually result in > ChromeVersionConstants.isOfficialBuild() returning true. > > This CL propagates the is_official_build truthiness to the script so > official builds no longer depend solely on the environment variable. > > BUG= 599268 > > Committed: https://crrev.com/eab85df0275b825e428954c4030e358ce968ecd9 > Cr-Commit-Position: refs/heads/master@{#384340} TBR=agrieve@chromium.org,thestig@chromium.org,wnwen@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG= 599268 Review URL: https://codereview.chromium.org/1851843002 Cr-Commit-Position: refs/heads/master@{#384610} Commit : 756e9b934bf1d706dd59ab39bed2af842b67063b Date : Fri Apr 01 16:57:35 2016 ===== TESTED REVISIONS ===== Revision Exit Code Std. Dev. Num Values Good? chromium@384560 1 N/A 20 good chromium@384604 1 N/A 20 good chromium@384607 1 N/A 20 good chromium@384609 1 N/A 20 good chromium@384610 0 N/A 20 bad <- chromium@384615 0 N/A 20 bad chromium@384626 0 N/A 20 bad chromium@384648 0 N/A 20 bad Bisect job ran on: android_nexus7_perf_bisect Bug ID: 604515 Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --also-run-disabled-tests smoothness.image_decoding_cases Test Metric: frame_times/frame_times Relative Change: 100.00% Score: 80.0 Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/2918 Job details: https://chromeperf.appspot.com/buildbucket_job_status/9015006735431727184 Not what you expected? We'll investigate and get back to you! https://chromeperf.appspot.com/bad_bisect?try_job_id=604515 | 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!
,
May 5 2016
According to that bisect result, the test was *fixed* by skyostil's change. Although it appears that in the latest builds, that test is failing. Still not sure why. https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus7v2%20Perf%20%282%29
,
May 5 2016
New return_code bisect job: https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/2958
,
May 5 2016
===== BISECT JOB RESULTS ===== Status: completed ===== SUSPECTED CL(s) ===== Subject : Revert of Propagate is_official_build to version script. (patchset #1 id:1 of https://codereview.chromium.org/1838173005/ ) Author : skyostil Commit description: Reason for revert: Made all the Android perf bots red: https://bugs.chromium.org/p/chromium/issues/detail?id=599830 Original issue's description: > Propagate is_official_build to version script. > > OFFICIAL_BUILD is a special variable in the version script which > takes its value from the environment variable. This means that local > builds with is_official_build will not actually result in > ChromeVersionConstants.isOfficialBuild() returning true. > > This CL propagates the is_official_build truthiness to the script so > official builds no longer depend solely on the environment variable. > > BUG= 599268 > > Committed: https://crrev.com/eab85df0275b825e428954c4030e358ce968ecd9 > Cr-Commit-Position: refs/heads/master@{#384340} TBR=agrieve@chromium.org,thestig@chromium.org,wnwen@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG= 599268 Review URL: https://codereview.chromium.org/1851843002 Cr-Commit-Position: refs/heads/master@{#384610} Commit : 756e9b934bf1d706dd59ab39bed2af842b67063b Date : Fri Apr 01 16:57:35 2016 ===== TESTED REVISIONS ===== Revision Exit Code Std Dev N Good? chromium@384500 1 N/A 20 good chromium@384569 1 N/A 20 good chromium@384604 1 N/A 20 good chromium@384609 1 N/A 20 good chromium@384610 0 N/A 20 bad <-- chromium@384611 0 N/A 20 bad chromium@384613 0 N/A 20 bad chromium@384621 0 N/A 20 bad chromium@384638 0 N/A 20 bad chromium@384775 0 N/A 20 bad chromium@385050 0 N/A 20 bad chromium@385600 0 N/A 20 bad Bisect job ran on: android_nexus7_perf_bisect Bug ID: 604515 Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests smoothness.image_decoding_cases Test Metric: frame_times/frame_times Relative Change: 100.00% Score: 80.0 Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus7_perf_bisect/builds/2958 Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013481417327685360 Not what you expected? We'll investigate and get back to you! https://chromeperf.appspot.com/bad_bisect?try_job_id=5820501671280640 | 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!
,
May 6 2016
It unfortunately looks like there's two failures overlapping here. Prior to build 1577 smoothness.image_decoding_cases was failing on both ToT and the reference build. After the fix there, the reference build got fixed. I'm guessing there was additional breakage somewhere between #1577 and the first build where both the ref and ToT tests were broken. Of course, that doesn't explain why the bisect returns success after #384610... sigh. Original breakage in: https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus7v2%20Perf%20%282%29/builds/1491 or https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus7v2%20Perf%20%282%29/builds/1492
,
May 6 2016
(Sorry, original breakage in between the linked builds, and #1577, I mean).
,
Jun 1 2016
The bot is green and the benchmark is also enabled on Nexus7. Closing this bug. |
|||
►
Sign in to add a comment |
|||
Comment 1 by robert...@chromium.org
, Apr 18 2016