On Tue, 24 Jan 2017 19:45:35 -0800, tonyc wrote: > On Fri, 14 Oct 2016 18:21:08 -0700, brian.carpenter@gmail.com wrote: > > I have a smaller test case for this issue: > > > > perl -e '$q0=q/00|0()0|0()\1/;0=~$q0' > > Your original test case produced a stack overflow (deep recursion) for > me before or after Hugo's 129377 patch in > 2dfc11ec3af312f4fa3eb244077c79dbb5fc2d85. > > This test case is fixed by Hugo's patch. > > On Mon, 17 Oct 2016 04:02:43 -0700, hv wrote: > > On Fri Oct 14 18:21:08 2016, brian.carpenter@gmail.com wrote: > > > I have a smaller test case for this issue: > > > > > > perl -e '$q0=q/00|0()0|0()\1/;0=~$q0' > > > > This is rather another case of #129377, and is fixed by my proposed > > patch there. > > So merging this into 129377. I'm not sure merging was the right thing to do - the "another case of #129377" applies only to Brian's followup claim to have a smaller test case, not to the original report. That's fine though, I can create a clearer new ticket based on the first half of my original analysis - I think we already have ample references to the second (stackbusting) issue there. Hugo --- via perlbug: queue: perl5 status: pending release https://rt.perl.org/Ticket/Display.html?id=129377