190500 2018-10-11 17:11:21 -0700 Bad behavior of scrolling, position:fixed inside flattened iframe 2019-05-23 15:21:08 -0700 1 1 1 Unclassified WebKit Layout and Rendering WebKit Nightly Build Unspecified Unspecified RESOLVED CONFIGURATION CHANGED InRadar P2 Normal --- 196290 1 simon.fraser webkit-unassigned ajuma bfulgham fred.wang koivisto simon.fraser webkit-bug-importer zalan oldest_to_newest 1468263 0 352107 simon.fraser 2018-10-11 17:11:21 -0700 Created attachment 352107 Testcase Attached test case, when loaded on iOS, shows bad problems where the position of the logging box jumps around, and the content doesn't get fully tilted, and it can't be scrolled. 1468264 1 webkit-bug-importer 2018-10-11 17:11:47 -0700 <rdar://problem/45211964> 1517363 2 simon.fraser 2019-03-15 21:48:44 -0700 Even with frame flattening off, I see two issues with this test content on iPad: 1. On first load, the position:fixed in the iframe is missing 2. After scrolling the iframe for a bit, we lose all the tiles in the logging element. My guess is that GraphicsLayer geometry isn't getting updated somehow, triggering incorrect visible rects. Maybe this is related to boundsOrigin vs. position differences. 1519602 3 fred.wang 2019-03-21 05:00:40 -0700 @smfr: Not sure how it relates to bug 194433, but AFAIK my remaining tests for position: fixed/sticky when async frame scrolling is enabled still don't pass. 1521728 4 simon.fraser 2019-03-27 10:26:55 -0700 https://trac.webkit.org/changeset/243539/webkit fixes position:fixed inside the iframe. I still see missing tiles in the scroller sometimes. 1538808 5 koivisto 2019-05-23 15:21:08 -0700 Appears to be fixed. 352107 2018-10-11 17:11:21 -0700 2018-10-11 17:11:21 -0700 Testcase iframe-testcase.zip application/zip 2115 simon.fraser UEsDBBQACAAIAC56S00AAAAAAAAAAAAAAAATABAAaG9zdGluZy1pZnJhbWUuaHRtbFVYDABa5r9b 98u/W/UBUABlUE0LwjAMvfdXxIFHbRUFmXUX9awHLx47W7dC90E61CH+d9tVh8N3SMLLyyMJH+0O 29P5uIe8KUxC+DcpIRNCwIHbpjUq6WoPfUVRKHj2hEchMNNlDDPGWP2AuQvrgSBXOsubGJbsr3XX ssljWLHxkE8rlAq9p3O0ldESMhTtUFQLKXWZOdXi1/cVVqf97oTTcBNPK9l2jOt/brF42URWF7VR kyq1Cm8Kp/4TUcJp0LhBGiadUfejN1BLBwgC//RmuQAAADsBAABQSwMECgAAAAAARYlLTQAAAAAA AAAAAAAAAAkAEABfX01BQ09TWC9VWAwAYea/W2Hmv1v1ARQAUEsDBBQACAAIAC56S00AAAAAAAAA AAAAAAAeABAAX19NQUNPU1gvLl9ob3N0aW5nLWlmcmFtZS5odG1sVVgMAFrmv1v3y79b9QFQAGNg FWNnYGJg8E1MVvAPVohQgAKQGAMnEBsB8XUgBvIZQXwigGNISBCEBdbxBYhF0ZQwQ8UFGBikkvNz 9RILCnJS9XISi0tKi1NTUhJLUpUDgqFmsAAJIJYDqctNTC7KzwVKF+sVp+akJpdk5ucFJealp0LV coAIBgYZNLVlmcWZSTmpnnkpqRUMDLeO7Y8Gqf4iGmQKos2tjI0MAFBLBwjrEdbbnAAAAAkBAABQ SwMEFAAIAAgAq3tLTQAAAAAAAAAAAAAAABQAEABzaW1wbGUtb2JzZXJ2ZXIuaHRtbFVYDABa5r9b ws6/W/UBUACVVltvmzAUfu+v8JimEDUhpPdR0odkndSpWqe2L9O2B4MNWHUwMiaXVf3vOy6EBnCy za2C853j71x9gv/u093s8fu3a5SoOb86OPDLp59QTK4OECx/ThVGKZ7TibVgdJkJqSwUilTRVE2s JSMqmZy5rlWp52rNabnXKxBkjZ7rr3ollMWJ8tCpCytbXdbCl3pXb94rLGOqWgyZyJliIvUQDnLB C0UvG/JXpzw0Pm/Qb9s+cjsiJTLAL7qCAIdPsRRFSoah4EJ6KOBFy2IgVsM8wUQsPeTC31G2Ahvw kTO+oHJ/jFzEMUvjnUFGbEVJ014E+R/m7DeFKM2RdGFOI3PkVbou3A/mZBlyIiCoiOto81AKzpvS 4ZIGT0wNN1rDUgli9MC5Ikz+lmAZB9h2B6j6d8b9droloXIoMWFFbop1Dm3DIHWnbUFlIJZ43S6K P9rqXR9cZpl6a+SoSENdDoQJeRS3Irbzfi1sVm6BJdrUdIKICIs53BUH+viaU72drm+I3atUeq3Y KthRdKVm5TVDhxOUo0Nk/Uwts3KZ4EeRgcETfa/cvS1XB8OAX+Y0VJTMMOe6EjYYlIzujK6Ov9f2 /E0SVlzQWipBPXC9InU4TWOADgGsoDZLJCSydQq1fI1EtFHstxxpmlRsDpdhY2rt6O+OEp/13bGP +y0jzbNSCDXV/Qe9tFma6cvD3VcnB+NpzKK1XRK/Ke8nDbQOnJxxBgfvIcfebtKu8n7yum5QxJJ6 j8dt5f+gxvDpoa20dqTAtYcMh6rAHE1n9/+Q2XLW63sy7aTD7re9ftlq8Hqr+0YEOZWNoVu3OxEP VBWZvau3N0fhFqV0iW62gr2rRLbhygzQM1KJpHkiOPHQj05C3EEXcsYm8MgEHpvAExN4agLPTOC5 CbwwgR+74LiB/EIvrdp0xmH1M75nGpYa7WGwKYhTbexSrW+abkuWwu+vA813vQDaW5bD7ISCwaDF pDfYFH+AIsxzWlHA0K8GvT8qX3p8/coCj0xSxMjEqkasdeWPAAIBYYtXQemKpU8CpB/VyVH5EvUH UEsHCOMzruoWAwAAXQkAAFBLAwQUAAgACACre0tNAAAAAAAAAAAAAAAAHwAQAF9fTUFDT1NYLy5f c2ltcGxlLW9ic2VydmVyLmh0bWxVWAwAWua/W8LOv1v1AVAAY2AVY2dgYmDwTUxW8A9WiFCAApAY AycQGwHxDSAG8hm5GIgCjiEhQRAWWMcXIBZDU8IMFRdgYJBKzs/VSywoyEnVy0ksLiktTk1JSSxJ VQ4IhprBAiRYGRjkQOpyE5OL8nOB0sV6xak5qcklmfl5QYl56alQtSAXMzIwyKCpLcsszkzKSfXM S0mtYGBQOLo/GqQ6+LOUNYg2NLUytDQAAFBLBwhMgqyAoAAAAAoBAABQSwECFQMUAAgACAAuektN Av/0ZrkAAAA7AQAAEwAMAAAAAAAAAABApIEAAAAAaG9zdGluZy1pZnJhbWUuaHRtbFVYCABa5r9b 98u/W1BLAQIVAwoAAAAAAEWJS00AAAAAAAAAAAAAAAAJAAwAAAAAAAAAAED9QQoBAABfX01BQ09T WC9VWAgAYea/W2Hmv1tQSwECFQMUAAgACAAuektN6xHW25wAAAAJAQAAHgAMAAAAAAAAAABApIFB AQAAX19NQUNPU1gvLl9ob3N0aW5nLWlmcmFtZS5odG1sVVgIAFrmv1v3y79bUEsBAhUDFAAIAAgA q3tLTeMzruoWAwAAXQkAABQADAAAAAAAAAAAQKSBOQIAAHNpbXBsZS1vYnNlcnZlci5odG1sVVgI AFrmv1vCzr9bUEsBAhUDFAAIAAgAq3tLTUyCrICgAAAACgEAAB8ADAAAAAAAAAAAQKSBoQUAAF9f TUFDT1NYLy5fc2ltcGxlLW9ic2VydmVyLmh0bWxVWAgAWua/W8LOv1tQSwUGAAAAAAUABQCPAQAA ngYAAAAA