On Tue, Feb 05, 2008 at 07:33:46PM -0800, Zev Benjamin wrote: > # New Ticket Created by Zev Benjamin > # Please include the string: [perl #50554] > # in the subject line of all future correspondence about this issue. > # <URL: http://rt.perl.org/rt3/Ticket/Display.html?id=50554 > > > > I unfortunately can't reduce this to a more self-contained test-case > right now. I add the following function to one of the files in > languages/perl6/src/builtins: > > .sub blah > $I0 = 1 > .return ($I0) > .end > > I then recompiled perl6. The following works fine: > > $ ./perl6 > > my $a = blah(); say $a; > 1 > > However, omitting the final semicolon causes a segfault: > $ ./perl6 > > my $a = blah(); say $a > Segmentation fault (core dumped) > > Additionally, omitting the "()" after "blah" stops the segfault from > occurring. Strangely, specifying --target=pir seems to indicate that > all three variations produce identical pir.
What happens if you run using "parrot perl6.pbc" instead of the perl6 executable? If you still get the segfault, how about trying parrot with the -G option? How about if the code to be executed is read from a file instead of being run via interactive mode? My initial thought is that this is our perpetual GC bug manifesting itself, especially since the PIR is identical for each of them. Pm
