Summary: | REGRESSION (r232356): After zooming a page in and out, it's possible to temporarily have missing tiles (google.com, apple.com) | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Antti Koivisto <koivisto> | ||||||
Component: | Layout and Rendering | Assignee: | Nobody <webkit-unassigned> | ||||||
Status: | RESOLVED FIXED | ||||||||
Severity: | Normal | CC: | bfulgham, commit-queue, simon.fraser, zalan | ||||||
Priority: | P2 | Keywords: | InRadar | ||||||
Version: | WebKit Nightly Build | ||||||||
Hardware: | Unspecified | ||||||||
OS: | Unspecified | ||||||||
Attachments: |
|
Description
Antti Koivisto
2018-07-11 07:02:17 PDT
I mean <rdar://problem/41863803> Created attachment 344839 [details]
patch
Comment on attachment 344839 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=344839&action=review > Source/WebKit/WebProcess/WebPage/WebPage.h:1669 > + float m_lastTransactionPageScaleFactor { 0 }; We store page scale as a double in RemoteLayerTreeTransaction and we've certainly had bugs with float/double precision mismatch before, so should this be a double, and/or do matching with epsilon? > Source/WebKit/WebProcess/WebPage/WebPage.h:1670 > + uint64_t m_lastTransactionIDWithScaleChange { 0 }; I wish we had a typedef for transactionIDs :( Created attachment 344865 [details]
patch
Comment on attachment 344865 [details] patch Clearing flags on attachment: 344865 Committed r233780: <https://trac.webkit.org/changeset/233780> All reviewed patches have been landed. Closing bug. |