Pages frequently fail to enter the back/forward cache due to loading ad frames (e.g. on Yahoo.com). Page cache error is usually "Quick redirect coming".
<rdar://problem/56132022>
Created attachment 380690 [details] Patch
Created attachment 380691 [details] Patch
Created attachment 380712 [details] Patch
Comment on attachment 380712 [details] Patch Clearing flags on attachment: 380712 Committed r251019: <https://trac.webkit.org/changeset/251019>
All reviewed patches have been landed. Closing bug.
Comment on attachment 380712 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=380712&action=review > Source/WebCore/ChangeLog:24 > + (WebCore::FrameLoader::stopForPageCache): > + Stop policy checks & cancel scheduled navigations after stopping loads. Stopping loads may > + run script which may in theory schedule new navigations. This is hardening. Since this is hardening, I think It needs a comment; since un-hardening it won’t lead to a test failing, presumably only the comment will prevent us from un-hardening it.
(In reply to Darin Adler from comment #7) > Comment on attachment 380712 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=380712&action=review > > > Source/WebCore/ChangeLog:24 > > + (WebCore::FrameLoader::stopForPageCache): > > + Stop policy checks & cancel scheduled navigations after stopping loads. Stopping loads may > > + run script which may in theory schedule new navigations. This is hardening. > > Since this is hardening, I think It needs a comment; since un-hardening it > won’t lead to a test failing, presumably only the comment will prevent us > from un-hardening it. Good idea, I will add a comment.
(In reply to Chris Dumez from comment #8) > (In reply to Darin Adler from comment #7) > > Comment on attachment 380712 [details] > > Patch > > > > View in context: > > https://bugs.webkit.org/attachment.cgi?id=380712&action=review > > > > > Source/WebCore/ChangeLog:24 > > > + (WebCore::FrameLoader::stopForPageCache): > > > + Stop policy checks & cancel scheduled navigations after stopping loads. Stopping loads may > > > + run script which may in theory schedule new navigations. This is hardening. > > > > Since this is hardening, I think It needs a comment; since un-hardening it > > won’t lead to a test failing, presumably only the comment will prevent us > > from un-hardening it. > > Good idea, I will add a comment. Landed in <https://trac.webkit.org/changeset/251029>.