in reply to Re^7: Parallel Modules ?
in thread Parallel Modules ?

I found the root cause of the problem. use encoding "utf8" caused the segmentation fault. I removed it. Problem is gone. However, I got "corrupted double-linked list" at run time. It does not always happen. Sometimes the program works fine. Sometimes, I got "corrupted double-linked list". Program died when it happened. Below are the messages. I have no clue what causes it. Any idea?

*** glibc detected *** /usr/bin/perl: corrupted double-linked list: 0x
+000000000bbec430 ***
======= Backtrace: =========
/lib64/libc.so.6[0x366cc70583]
/lib64/libc.so.6[0x366cc7222c]
/lib64/libc.so.6(cfree+0x4b)[0x366cc7273b]
/usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so(perl_
+destruct+0xffa)[0x366dc3966a]
/usr/bin/perl(main+0xb3)[0x401773]
/lib64/libc.so.6(__libc_start_main+0xf4)[0x366cc1d994]
/usr/bin/perl[0x401609]
======= Memory map: ========
00400000-00403000 r-xp 00000000 08:03 75519                           
+   /usr/bin/perl
00602000-00604000 rw-p 00002000 08:03 75519                           
+   /usr/bin/perl
0a327000-0c5c3000 rw-p 0a327000 00:00 0                               
+   [heap]
422bd000-422be000 ---p 422bd000 00:00 0 
422be000-42cbe000 rw-p 422be000 00:00 0 
42cbe000-42cbf000 ---p 42cbe000 00:00 0 
42cbf000-436bf000 rw-p 42cbf000 00:00 0 
436bf000-436c0000 ---p 436bf000 00:00 0 
436c0000-440c0000 rw-p 436c0000 00:00 0 
3513c00000-3513d38000 r-xp 00000000 08:03 2523842                     
+   /usr/lib/libxml2.so.2.7.8
3513d38000-3513f37000 ---p 00138000 08:03 2523842                     
+   /usr/lib/libxml2.so.2.7.8
3513f37000-3513f41000 rw-p 00137000 08:03 2523842                     
+   /usr/lib/libxml2.so.2.7.8
3513f41000-3513f42000 rw-p 3513f41000 00:00 0 
366c800000-366c81c000 r-xp 00000000 08:03 1638708                     
+   /lib64/ld-2.5.so
366ca1b000-366ca1c000 r--p 0001b000 08:03 1638708                     
+   /lib64/ld-2.5.so
366ca1c000-366ca1d000 rw-p 0001c000 08:03 1638708                     
+   /lib64/ld-2.5.so
366cc00000-366cd4d000 r-xp 00000000 08:03 1638709                     
+   /lib64/libc-2.5.so
366cd4d000-366cf4d000 ---p 0014d000 08:03 1638709                     
+   /lib64/libc-2.5.so
366cf4d000-366cf51000 r--p 0014d000 08:03 1638709                     
+   /lib64/libc-2.5.so
366cf51000-366cf52000 rw-p 00151000 08:03 1638709                     
+   /lib64/libc-2.5.so
366cf52000-366cf57000 rw-p 366cf52000 00:00 0 
366d000000-366d082000 r-xp 00000000 08:03 1638712                     
+   /lib64/libm-2.5.so
366d082000-366d281000 ---p 00082000 08:03 1638712                     
+   /lib64/libm-2.5.so
366d281000-366d282000 r--p 00081000 08:03 1638712                     
+   /lib64/libm-2.5.so
366d282000-366d283000 rw-p 00082000 08:03 1638712                     
+   /lib64/libm-2.5.so
366d400000-366d402000 r-xp 00000000 08:03 1638608                     
+   /lib64/libdl-2.5.so
366d402000-366d602000 ---p 00002000 08:03 1638608                     
+   /lib64/libdl-2.5.so
366d602000-366d603000 r--p 00002000 08:03 1638608                     
+   /lib64/libdl-2.5.so
366d603000-366d604000 rw-p 00003000 08:03 1638608                     
+   /lib64/libdl-2.5.so
366d800000-366d816000 r-xp 00000000 08:03 1638711                     
+   /lib64/libpthread-2.5.so
366d816000-366da15000 ---p 00016000 08:03 1638711                     
+   /lib64/libpthread-2.5.so
366da15000-366da16000 r--p 00015000 08:03 1638711                     
+   /lib64/libpthread-2.5.so
366da16000-366da17000 rw-p 00016000 08:03 1638711                     
+   /lib64/libpthread-2.5.so
366da17000-366da1b000 rw-p 366da17000 00:00 0 
366dc00000-366dd2b000 r-xp 00000000 08:03 202553                      
+   /usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so
366dd2b000-366df2b000 ---p 0012b000 08:03 202553                      
+   /usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so
366df2b000-366df34000 rw-p 0012b000 08:03 202553                      
+   /usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi/CORE/libperl.so
366df34000-366df36000 rw-p 366df34000 00:00 0 
3670000000-367000d000 r-xp 00000000 08:03 1638714                     
+   /lib64/libgcc_s-4.1.2-20080825.so.1
367000d000-367020d000 ---p 0000d000 08:03 1638714                     
+   /lib64/libgcc_s-4.1.2-20080825.so.1
367020d000-367020e000 rw-p 0000d000 08:03 1638714                     
+   /lib64/libgcc_s-4.1.2-20080825.so.1
3673800000-3673815000 r-xp 00000000 08:03 1638453                     
+   /lib64/libnsl-2.5.so
3673815000-3673a14000 ---p 00015000 08:03 1638453                     
+   /lib64/libnsl-2.5.so
3673a14000-3673a15000 r--p 0Abort

[download]


Here is my system info:
uname -a
Linux 2.6.18-164.el5 #1 SMP Tue Aug 18 15:51:48 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux
perl -v
This is perl, v5.8.8 built for x86_64-linux-thread-multi
Copyright 1987-2006, Larry Wall
Modules I used

use strict;
use warnings;
use threads;
use Scalar::Util 'reftype';
use Data::Dumper;
use CGI qw(:standard Vars);
use LWP::UserAgent;
use XML::Simple;

[download]

Replies are listed 'Best First'.
Re^9: Parallel Modules ?
by BrowserUk (Patriarch) on Dec 01, 2011 at 01:07 UTC
    I have no clue what causes it. Any idea?

    Not a clue I'm afraid. I don't use gcc/linux.

    The first thing I would do were I you, is install a recent copy of Perl and relevant libraries, and then try your code there.

    From memory, 5.8.8 wasn't a bad release of perl as far a threading was concerned, but an aweful lot of bugs have been detected and fixed in the intervening nearly 6 years. If you are lucky, the problem will have just 'gone away'. If not, you will at least stand some chance of getting any bug reports you raise taken seriously.

    Then you need to start stripping away large chunks of your program, whilst maintaining the same error, until you have something small and independent of your own setup and data that demonstrates the problem, so that people can reproduce your problem on their systems.

    But trying it on the newest releases -- of perl and all involved modules -- should be your first priority.

    You could also try describing your program requirements here and see if someone can't come up with a working solution you could use.

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.

    The start of some sanity?
[reply]

In Section Seekers of Perl Wisdom