+++ This bug was initially created as a clone of Bug #1137890 +++ +++ This bug was initially created as a clone of Bug #1123921 +++ Try run in bug 1137890 shows that enabling containerless-root on B2G passes all tests. Local testing also showed no strange/broken behaviour, so I think we're good to turn it on.

Comment on attachment 8591280 [details] [diff] [review] Patch Ship it!

Comment on attachment 8591280 [details] [diff] [review] Patch My bad, I tested this more and realized it does actually break behaviour in the browser for some pages (e.g. people.mozilla.com/~kgupta/grid.html) when zoomed.

The clipping change from bug 1147038 does fix half of the problem I was seeing, but the scrollbars are still mispositioned. Hopefully fixing bug 1133905 will take care of this.

Comment on attachment 8591280 [details] [diff] [review] Patch This should be good now.

This is causing a smoke test blocker. Please see bug 1166326. We may need this backed out.

Backed out in https://hg.mozilla.org/integration/mozilla-inbound/rev/e7e537fd67ec The sheriff assured me this would make the next nightly.

This seems to cause the regression to youtube video app's video playback. When I scroll the youtube video app's list during video playback, video's rendering position is not correct.

(In reply to Sotaro Ikeda [:sotaro] from comment #12) > This seems to cause the regression to youtube video app's video playback. > When I scroll the youtube video app's list during video playback, video's > rendering position is not correct. I am going to check it by local back-out.

(In reply to Sotaro Ikeda [:sotaro] from comment #12) > This seems to cause the regression to youtube video app's video playback. > When I scroll the youtube video app's list during video playback, video's > rendering position is not correct. mstange said that bug 1148582 seems to fix it.

This is the "clip not moving with async scroll" issue. I haven't checked which patches in bug 1148582 will actually fix it, I just know that we're expecting that bug to take care of all our clipping problems.

(In reply to Sotaro Ikeda [:sotaro] from comment #13) > (In reply to Sotaro Ikeda [:sotaro] from comment #12) > > This seems to cause the regression to youtube video app's video playback. > > When I scroll the youtube video app's list during video playback, video's > > rendering position is not correct. > > I am going to check it by local back-out. The local back-out fixed the problem.

(In reply to Sotaro Ikeda [:sotaro] from comment #12) > This seems to cause the regression to youtube video app's video playback. > When I scroll the youtube video app's list during video playback, video's > rendering position is not correct. kats, is it expected regression until bug 1148582 fix?

Redirecting to tn, since he's the one actually working on this.

The landing for this is causing a smoke test blocker. Please see bug 1170611. We may need this backed out.

Backed out in https://hg.mozilla.org/integration/mozilla-inbound/rev/e0515cb0cb5a

(In reply to Markus Stange [:mstange] from comment #15) > This is the "clip not moving with async scroll" issue. I haven't checked > which patches in bug 1148582 will actually fix it, I just know that we're > expecting that bug to take care of all our clipping problems. It doesn't appear to be related to bug 1148582. There is only one actively scrolling scrollframe involved, and only one set of metrics.

(In reply to Timothy Nikkel (:tn) from comment #21) > There is only one actively > scrolling scrollframe involved, and only one set of metrics. But two clips: The video ImageLayer has its own overflow:hidden clip, and then there is the clip of the actively scrolling scrollframe - during scrolling, we want to move the overflow:hidden clip but not the clip that clips the video to the scrollframe.

Oh right, forgot about non-scrolling clips.