Loading Safari causes the above assertion to trigger. This is because the disk image caching callback is happening on the main (UI) thread instead of the web thread when using WebKit1.
<rdar://problem/18342344>
Created attachment 238390 [details] Patch
Comment on attachment 238390 [details] Patch r=me
Comment on attachment 238390 [details] Patch Clearing flags on attachment: 238390 Committed r173782: <http://trac.webkit.org/changeset/173782>
All reviewed patches have been landed. Closing bug.
Comment on attachment 238390 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=238390&action=review > Source/WebCore/loader/cocoa/DiskCacheMonitorCocoa.mm:111 > + CFCachedURLResponseCallBackBlock blockToRun = ^ (const CFCachedURLResponseRef response) This const is incorrect and unneeded. Also would be nice to have consistent syntax with the definition of "block" above.
Comment on attachment 238390 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=238390&action=review >> Source/WebCore/loader/cocoa/DiskCacheMonitorCocoa.mm:111 >> + CFCachedURLResponseCallBackBlock blockToRun = ^ (const CFCachedURLResponseRef response) > > This const is incorrect and unneeded. Also would be nice to have consistent syntax with the definition of "block" above. You're right. I am not sure how I ended up with a const there. Will fix.
(In reply to comment #7) > (From update of attachment 238390 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=238390&action=review > > >> Source/WebCore/loader/cocoa/DiskCacheMonitorCocoa.mm:111 > >> + CFCachedURLResponseCallBackBlock blockToRun = ^ (const CFCachedURLResponseRef response) > > > > This const is incorrect and unneeded. Also would be nice to have consistent syntax with the definition of "block" above. > > You're right. I am not sure how I ended up with a const there. Will fix. Fixed in r173858 - <http://trac.webkit.org/changeset/173858>.