eae@: would you mind triaging this? Sorry if you are not the right person.

Hmm, I've been unable to reproduce this on either windows or mac. Don't have a linux machine handy at the moment.

Are any layouts happening once the page has loaded?

How can I tell if a layout is happening?

This is what it looks like:
https://drive.google.com/file/d/0B2LVVIKSxUVYNHVPR0FZWm42cDQ/view?usp=sharing
Note I am not touching any input devices while this is happening.

Capture a timeline and look for layout calls.

Yes, there's a ton of layouts. Here's a link to the timeline data:
https://drive.google.com/file/d/0B2LVVIKSxUVYUkFXOE9ERmZFQTQ/view?usp=sharing

Thanks, interestingly that looks very different from the profile I captured (which mostly has a js driven animation).

Cannot reproduce this on Linux 54.0.2816.0.

Also can't reproduce on Windows Version 54.0.2824.0 canary (64-bit)

Bisect might be helpful

Considering mstensho@ cannot reproduce this in the same build, it could be something about my setup. Also, I did a remote desktop into my machine yesterday, and this was not happening over remote desktop. However it is happening every time during normal use. I thought perhaps an extension is causing this, but this is happening in an incognito window as well.

Would it be possible for you to run bisect-builds to find a regression range? https://www.chromium.org/developers/bisect-builds-py

How do I determine the relevant revisions? https://src.chromium.org/viewvc/chrome/branches/ most recent branch is ~2 years old.

We don't use SVN anymore. You can use http://omahaproxy.appspot.com/ to find the branch_base_position for stable/beta/dev, and it also has a way to map a version number to a branch position (use that with the version that you know isn't working).

Yeah, I tried that, but it looked like branch base positions are out of the revision range that bisect-builds accepts.

54.0.2816.0 yields Branch Base Position: 409090. 52.0.2743.116 yields 394939. I ran 
bisect-builds.py -a linux -g 394939 -b 409090 --use-local-cache --verify-range

And got:

Downloading list of known revisions...
Saved revisions 15734-382086 to /usr/local/google/home/mfomitchev/bisect/.bisect-builds-cache.json
Traceback (most recent call last):
  File "./bisect-builds.py", line 1172, in <module>
    sys.exit(main())
  File "./bisect-builds.py", line 1133, in main
    evaluator, opts.verify_range)
  File "./bisect-builds.py", line 740, in Bisect
    raise RuntimeError(msg)
RuntimeError: We don't have enough builds to bisect. revlist: []

You need linux64 instead of linux

Ah, thanks.

When I give it 409090 - it downloads 409086, and it works fine in that revision. The next revision it has is 409092, and it works fine in that revision as well.

So either it was an issue which was introduced and fixed between 409086 and 409092, which seems unlikely, or it's something in my profile. I've tried using --user-data-dir to point bisect-builds to my local profile, but when I do it - the script just always launches my local version of chrome rather than the one it downloads for some reason.

Hm... I think it does that if you already have a browser window open with that profile. Try closing all Chrome windows before running the bisect script?

Able to reproduce the issue on windows 7, Linux Ubuntu 14.04 and Mac 10.11.6 using chrome version 52.0.2743.116 and canary 54.0.2829.0 with the below steps

1. Go to URL https://in.answers.yahoo.com/question/index?qid=20111119013305AAgHjDr
2.Open console and observed the output of
 var body = document.body
html = document.documentElement
body.scrollHeight
3.Refresh the page and again see the output 
4.Output is different for each time page loaded

Please find the attached screen cast for the same.This issue is observed from M35 old builds.Marking it as Untriaged.
This is working fine on Firefox getting same output(574) for each time.

Thanks,

Deleted: 635945.mp4 4.0 MB

	635945.mp4 4.0 MB View Download

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available. If you change it back, also remove the "Hotlist-Recharge-Cold" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

No longer reproduces.