The commit r268886 introduced a page load time regression on iOS. In particular, loading nytimes and facebook seems to be slower.
<rdar://problem/71543823>
How was the regression tested? Bisecting? Reverting? What is the metric for page load time?
I have a speculation on how to fix it, working on a patch.
Created attachment 416140 [details] Patch
(In reply to Noam Rosenthal from comment #3) > I have a speculation on how to fix it, working on a patch. Great, thanks! I can test this patch.
(In reply to Per Arne Vollan from comment #5) > (In reply to Noam Rosenthal from comment #3) > > I have a speculation on how to fix it, working on a patch. > > Great, thanks! I can test this patch. Thanks! Let me know if there's a way I can test it myself :)
(In reply to Noam Rosenthal from comment #6) > (In reply to Per Arne Vollan from comment #5) > > (In reply to Noam Rosenthal from comment #3) > > > I have a speculation on how to fix it, working on a patch. > > > > Great, thanks! I can test this patch. > Thanks! > Let me know if there's a way I can test it myself :) It looks like this patch recovers all, or most of the initial regression. Great work!
Committed r270845: <https://trac.webkit.org/changeset/270845> All reviewed patches have been landed. Closing bug and clearing flags on attachment 416140 [details].