On Thu, Sep 22, 2005 at 04:33:19AM -0700, Hugo van der Sanden via RT wrote: > If the same code succeeded with 5.8.6, it is likely that independent > changes have affected stack usage, and my guess is that lengthening > the test string slightly would be enough to cause 5.8.6 also to fail.
You're right; by doubling the length of the string I get: Complex regular subexpression recursion limit (32766) exceeded at chomplogs-test2.pl line 12. under 5.8.6. It still doesn't dump core, but is clearly heading in that direction. Both machines have a ulimit stack size of 65536K, but the perl-5.8.6 machine is FreeBSD-5.4 whereas the perl-5.8.7 machine is FreeBSD-4.11, so there may be other differences which explain the larger stack usage. On the FreeBSD-4.11/perl-5.8.7 machine, I get Bus error (core dumped) and gdb backtrace shows S_regmatch calls 2516 levels deep. I would have thought that 64MB of stack would have been plenty for that. $ perl -e 'print `sh -c "ulimit -a"`' cpu time (seconds, -t) unlimited file size (512-blocks, -f) unlimited data seg size (kbytes, -d) 524288 stack size (kbytes, -s) 65536 core file size (512-blocks, -c) unlimited max memory size (kbytes, -m) unlimited locked memory (kbytes, -l) unlimited max user processes (-u) 5547 open files (-n) 11095 virtual mem size (kbytes, -v) unlimited sbsize (bytes, -b) unlimited Regards, Brian.
