It is ridiculous that fixing a "regression caused one fail" caused 6 crashes. But now it happened, see this URL for details: http://build.webkit.sed.hu/builders/x86-64%20Linux%20Qt%20Release%20WebKit2%20%28Amazon%20EC2%29/builds/9585
That's what I was afraid of. Looks like an early return is missing somewhere, I'm going to take a look.
just a note: I prefer early testing ;-) (before commiting buggy patches)
Created attachment 167988 [details] Patch Hopefully that is the last iteration, this is getting pretty embarassing.
Comment on attachment 167988 [details] Patch Clearing flags on attachment: 167988 Committed r130894: <http://trac.webkit.org/changeset/130894>
All reviewed patches have been landed. Closing bug.