bug-gawk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug-gawk] : in debug mode, every eval causes double free of memory


From: Andrew J. Schorr
Subject: Re: [bug-gawk] : in debug mode, every eval causes double free of memory
Date: Wed, 7 May 2014 09:28:25 -0400
User-agent: Mutt/1.5.21 (2010-09-15)

Hi,

On Wed, May 07, 2014 at 08:16:59AM +0200, Jan Chaloupka wrote:
> it is up-to date version from repo, i.e. gawk-4.1.1.tar.xz . Also
> tested on the latest source codes from git from 2014-05-05 16:50:23,
> hash bce72232ae0d79d8af6b5aa1986b78c04d6c7c28 . Crash occurs again.

I'm afraid I don't see it either.  Using the current git master branch on
a Fedora 19 x86_64 system:

bash-4.2$ valgrind gawk -f /dev/null --debug
==17042== Memcheck, a memory error detector
==17042== Copyright (C) 2002-2012, and GNU GPL'd, by Julian Seward et al.
==17042== Using Valgrind-3.8.1 and LibVEX; rerun with -h for copyright info
==17042== Command: gawk -f /dev/null --debug
==17042== 
gawk> eval ""
==17042== Syscall param sendmsg(msg.msg_name) points to uninitialised byte(s)
==17042==    at 0x5B44E20: __sendmsg_nocancel (in /usr/lib64/libc-2.17.so)
==17042==    by 0x4E48306: readline (in /usr/lib64/libreadline.so.6.2)
==17042==    by 0x429DB9: zzparse (command.y:1041)
==17042==    by 0x435BA1: debug_prog (debug.c:2834)
==17042==    by 0x40D3BD: main (main.c:741)
==17042==  Address 0x7feffef62 is on thread 1's stack
==17042== 
gawk> ==17042== 
==17042== HEAP SUMMARY:
==17042==     in use at exit: 116,333 bytes in 293 blocks
==17042==   total heap usage: 414 allocs, 121 frees, 139,739 bytes allocated
==17042== 
==17042== LEAK SUMMARY:
==17042==    definitely lost: 280 bytes in 1 blocks
==17042==    indirectly lost: 0 bytes in 0 blocks
==17042==      possibly lost: 0 bytes in 0 blocks
==17042==    still reachable: 116,053 bytes in 292 blocks
==17042==         suppressed: 0 bytes in 0 blocks
==17042== Rerun with --leak-check=full to see details of leaked memory
==17042== 
==17042== For counts of detected and suppressed errors, rerun with: -v
==17042== Use --track-origins=yes to see where uninitialised values come from
==17042== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 2 from 2)

What do you see?

Regards,
Andy



reply via email to

[Prev in Thread] Current Thread [Next in Thread]