Scroll the test case: - Absolute layer is jittery during scrolling - Events pass through it to the scroller Both are signs of a missing positioning node.
Created attachment 365725 [details] test
<rdar://problem/49207055>
Actually we're incorrectly making a "Moves" node for the nested position:absolute.
Created attachment 367291 [details] Patch
*** Bug 196820 has been marked as a duplicate of this bug. ***
Created attachment 367292 [details] Patch
Regressions: Unexpected text-only failures (1) fast/scrolling/ios/overflow-scroll-overlap-6.html [ Failure ] You'll need to rebase for this progression.
Comment on attachment 367292 [details] Patch Attachment 367292 [details] did not pass ios-sim-ews (ios-simulator-wk2): Output: https://webkit-queues.webkit.org/results/11852526 New failing tests: fast/scrolling/ios/overflow-scroll-overlap-6.html
Created attachment 367312 [details] Archive of layout-test-results from ews123 for ios-simulator-wk2 The attached test failures were seen while running run-webkit-tests on the ios-sim-ews. Bot: ews123 Port: ios-simulator-wk2 Platform: Mac OS X 10.13.6
https://trac.webkit.org/changeset/244215/webkit
Actually, this is wrong: --- a/trunk/LayoutTests/fast/scrolling/ios/overflow-scroll-overlap-6-expected.txt +++ b/trunk/LayoutTests/fast/scrolling/ios/overflow-scroll-overlap-6-expected.txt @@ -2,5 +2,5 @@ case 1: -case 2: Scrollable 2 -case 3: Scrollable 3 +case 2: +case 3: Case 3 should remain scrollable.
(In reply to Antti Koivisto from comment #11) > Actually, this is wrong: > > --- > a/trunk/LayoutTests/fast/scrolling/ios/overflow-scroll-overlap-6-expected.txt > +++ > b/trunk/LayoutTests/fast/scrolling/ios/overflow-scroll-overlap-6-expected.txt > @@ -2,5 +2,5 @@ > > case 1: > -case 2: Scrollable 2 > -case 3: Scrollable 3 > +case 2: > +case 3: > > Case 3 should remain scrollable. Maybe the hit-testing code needs adjusting for nested absolutes.
> Maybe the hit-testing code needs adjusting for nested absolutes. Hit-testing code respects whatever positioning nodes are generated. You may have regressed something.
I'll investigate via bug 196868.