New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Large number of string evals makes interpreter emit syntax error #11112
Comments
From andreas@faafeng.comCreated by aff@affdesktop.domain.localThe perl interpreter will eventually emit an error as if the running perl v5.8.8, x86_64-linux-thread-multi, Linux 2.6.18 The problem may be related to PL_evalseq overflow but this has not been Note that it requires significant CPU time to reproduce this bug Typical error messages are : Variable "$self" will not stay shared at (eval 999973501) line 15. Example code: $ cat eval.pl print 'start: ' . scalar localtime(); for ( print __END__ $ perl eval.pl This is perl, v5.8.8 built for x86_64-linux-thread-multi ------------------------------------------------------------------------------- $ perl eval.pl This is perl, v5.8.8 built for x86_64-linux-thread-multi ------------------------------------------------------------------------------- $ ./perl -Ilib -l eval.pl $ ./perl -v This is perl 5, version 13, subversion 9 (v5.13.9-206-g3dcd47c) built Perl Info
|
From @CorionHello all, as another data point, I can add Strawberry Perl 5.12 (MinGW) on Windows In an attempt to more quickly reproduce the error (as the current There is a bit more discussion and different programs that pump up the http://perlmonks.org/?node_id=884800 but nothing that is quick enough for conveniently adding this to the -max [1] Code that does NOT trigger the error: #!/opt/perl/bin/perl use strict; my $count = 0; while (1) { exit if ! $continue--; |
The RT System itself - Status changed from 'new' to 'open' |
From vadim.konovalov@alcatel-lucent.com
wow... I confirm (windows, Activestate perl 5.12.1), and this indeed takes 7 hours to run and 1_073_xxx_xxx iterations. D:\TESTS\perl>perl -w starnge-eval.pl Regards, |
From @iabynOn Sat, Feb 05, 2011 at 05:39:24PM +0100, Konovalov, Vadim (Vadim)** CTR ** wrote:
I've found the problem and am currently working up a fix. -- |
From @iabynOn Sun, Feb 06, 2011 at 02:35:30PM +0000, Dave Mitchell wrote:
Now fixed with commits d1bfb64 The issue involved the wrapping of the PL_cop_seqmax variable (a U32), The rest of the commits solve the general problem of wrapping by cleaning -- |
@iabyn - Status changed from 'open' to 'resolved' |
From @csjewellI thought that number looked familiar - that's just under 1Gi (2^30 = I wonder if that helps any... On Sat, 05 Feb 2011 17:39 +0100, "Konovalov, Vadim (Vadim)** CTR **"
"Your random numbers are not that random" -- perl-5.10.1.tar.gz/util.c Strawberry Perl for Windows betas: http://strawberryperl.com/beta/ |
Migrated from rt.perl.org#83364 (status was 'resolved')
Searchable as RT83364$
The text was updated successfully, but these errors were encountered: