User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:60.0) Gecko/20100101 Firefox/60.0 Build ID: 20180605171542 Steps to reproduce: 1) Create a new file called "frame.hmtl", insert a the normal head things (lang, title...) 2) Create a new file called "site.html", insert a the normal head things, add a test text then a iframe element to create a iframe from "frame.html" and after the iframe I add some text. Actual results: After a few seconds the content of the iframe goes over the normal content (see picture bug.png) Expected results: nothing

I managed to reproduce this issue as well. Could not reproduce on chrome.

What more information do you need?

Bogdan, did you have a question for me on this or a request for action? :) Thanks. I saved the files locally and and it doesn't happy to me on desktop Firefox Nightly 63. jonkjolk, which version of Firefox on Android did you use?

Device: - Nokia 6 (Android 7.1.1); @Karl It was a request for action since it seems to be a web compatibility issue, I've managed to reproduce this on all versions of FF for Android(Nightly, Beta, and Release). Please let me know if there is any more information that I could provide. Note: This issue is not reproducible on Desktop versions of FF.

@Karl I use the latest version available in the Google Playstore(61.0)

ok let's see I will put it online to be able to test a bit more. These are the files on Android. http://www.la-grange.net/tmp/test/site.html ok I was able to reproduce and its weird because it takes time before it happens. For me nothing is happening, if I do not interact with the content. 1. Let the page 2. Scroll the iframe content up and down again 3. wait. No interaction. After ~15s the iframe content is overflowing its iframe. This is a serious rendering issue. :) I took a profile which exhibit the scroll at the beginning and the spontaneous overflowing. https://perfht.ml/2tSGyei

Sounds like an APZ bug. Is this a regression? Botond, have you seen anything like this? Might be container-ful scrolling related.

This is the same as bug 1339925 - according to my bisection at https://bugzilla.mozilla.org/show_bug.cgi?id=1339925#c2 it's likely a regression from Markus' clipping patches.

Marking bug 1339925 as a dependency for now, so we don't forget to re-test this page after that bug is fixed.

Like bug 1339925, this is also fixed by the patch in bug 1486695.

Update: the patch in bug 1486695 was modified, and it no longer fixes bug 1339925.