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
Assertion failure in S_ReANY #16939
Comments
From @dur-randirCreated by @dur-randirWhile fuzzing perl v5.29.9-63-g2496d8f3f7 built with afl and run for(0..6){0if split$p;($$v0l0e)&=($f,$v0l0e)=0;($e)=0} to cause an assertion failure perl: ./inline.h:182: struct regexp *S_ReANY(const REGEXP *const): GDB stack trace is following #0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50 While it looks like one from stack-not-refcounted pack, I don't see commit 6ef7fe5 PATCH: [perl #131551] Too deep regex compilation recursion This patch, started by Yves Orton, and refined in consultation with Tony Perl Info
|
From @khwilliamsonOn Sun, 07 Apr 2019 09:51:11 -0700, randir wrote:
I am unable to reproduce this on Linux, -O0, g++ 7.3 |
The RT System itself - Status changed from 'new' to 'open' |
From @dur-randirOn Wed, 10 Apr 2019 15:54:24 -0700, khw wrote:
Here's a full -V from a -O0 perl that triggers this for me, built with gcc 8.3.0: Summary of my perl5 (revision 5 version 29 subversion 10) configuration: Characteristics of this binary (from libperl): |
From @dur-randirIt also crashes for me with gcc 7.4.0 |
From @hvdsOn Wed, 10 Apr 2019 15:54:24 -0700, khw wrote:
I can reproduce it here with gcc 4.8.4, both at the specified commit and current blead (a719c52). I can simplify it a bit to: For this variant, looping (0..2) instead gives just a bad free: I suspect these are all symptoms of corruption via stack refcounting on $x. Interestingly, building with clang and address-sanitizer I get a completely different result:
|
From @dur-randirOn Thu, 11 Apr 2019 05:34:20 -0700, hv wrote:
Good, so it can be shoved under some table) |
From @hvdsOn Thu, 11 Apr 2019 06:24:51 -0700, randir wrote:
Bad, because attempting to diagnose it properly would eat up time and give little back. I don't know for sure that's the issue, but suspecting it is already makes me reluctant to spend more time on it. Hugo |
Migrated from rt.perl.org#134003 (status was 'open')
Searchable as RT134003$
The text was updated successfully, but these errors were encountered: