New issue
Advanced search Search tips

Issue 895274 link

Starred by 1 user
Status: WontFix
Owner:
r...@igalia.com
Closed: Dec 27
Cc:
npm@chromium.org
r...@igalia.com
e...@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 895738


Sign in to add a comment

49.1% regression in rendering.mobile/queueing_durations at 598969:599042

Project Member Reported by npm@chromium.org, Oct 15 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Oct 15 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=895274

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=94a4010eb463fe07a159c2f2e18e0c717565c75e376c7694cb07b623537cda34


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

android-nexus5x-perf

rendering.mobile - Benchmark documentation link:
  https://bit.ly/rendering-benchmarks
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Oct 15

Cc: r...@igalia.com
Owner: r...@igalia.com
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/10d1ddbce40000

Fix perf regression in LayoutBoxModelObject::RelativePositionOffset() by rego@igalia.com
https://chromium.googlesource.com/chromium/src/+/4226ddf99103e493d7afb23a4c7902ee496108b6
queueing_durations: 0.2369 → 0.3612 (+0.1243)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Benchmark documentation link:
  https://bit.ly/rendering-benchmarks

Comment 4 by r...@igalia.com, Oct 16 

I cannot yet run try jobs on pinpoint: https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/1YGzRWESzug

Could someone run it for me with the following CL https://chromium-review.googlesource.com/c/chromium/src/+/1282124 ?

Comment 6 by npm@chromium.org, Oct 16 

Done
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Oct 16 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/15f3134ce40000

Comment 8 by npm@chromium.org, Oct 16 

Hopefully you're able to see the job results, ping me if not.

Comment 9 by r...@igalia.com, Oct 16 

Yeah I can see them, thanks. I didn't have time to analyze them though.

Comment 10 by r...@igalia.com, Oct 31

Blockedon: 895738
Project Member

Comment 12 by 42576172...@developer.gserviceaccount.com, Nov 8 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/1186a8b5e40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Nov 8 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/14a5b027e40000
Project Member

Comment 16 by 42576172...@developer.gserviceaccount.com, Nov 9 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/128a8640140000
Project Member

Comment 18 by 42576172...@developer.gserviceaccount.com, Nov 10 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/166eacade40000
Project Member

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

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/14ea9700140000
Project Member

Comment 22 by 42576172...@developer.gserviceaccount.com, Nov 21 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/14ebe5ebe40000
Project Member

Comment 24 by 42576172...@developer.gserviceaccount.com, Nov 22 

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1508abf4140000

Fix perf regressions in LayoutBoxModelObject::RelativePositionOffset() by rego@igalia.com
https://chromium-review.googlesource.com/c/chromium/src/+/1347356/1
queueing_durations: 0.3955 → No values

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Benchmark documentation link:
  https://bit.ly/rendering-benchmarks
Project Member

Comment 26 by 42576172...@developer.gserviceaccount.com, Nov 22 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/116a0f8fe40000
Project Member

Comment 29 by 42576172...@developer.gserviceaccount.com, Dec 3 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/154edf42140000

unsupported operand type(s) for +: 'float' and 'NoneType'
Project Member

Comment 30 by 42576172...@developer.gserviceaccount.com, Dec 4 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/15102290140000

unsupported operand type(s) for +: 'float' and 'NoneType'

Comment 31 by r...@igalia.com, Dec 10

Cc: e...@chromium.org
I've been trying different things but I didn't find a solution.

I guess we might need to assume that this regression is due to the new code to fix  issue #835607 .
Could this be fine? Do you have any idea to explore?

Comment 32 by r...@igalia.com, Dec 27

Status: WontFix (was: Assigned)
As I didn't manage to fix the problem and got not feedback on the previous comment I'm closing this.

Please reopen if you think there's a better solution to fix the original problem without causing performance regressions here.
►

Sign in to add a comment