Re: Core in a multithread application in AIX using RogueWave
- From: Paul Pluzhnikov <ppluzhnikov-nsp@xxxxxxxxxxx>
- Date: Mon, 05 Mar 2007 12:16:10 -0800
"jmartin" <jorge.jose.martin@xxxxxxx> writes:
I get a core un our application in a point that belongs to internal
libraries.
No, you don't. You get a crash inside malloc/free implementation.
With 99.9999% probability, this is a result of heap corruption.
The compilation is theorically thread safe using the -
qthreaded compiler and linker option. I have seem that the RogueWave
libraries were not built using that option. Could be that?
Yes, theoretically it could be that.
But an application bug is much more likely.
Use "heap debugging" tools. Start with this one:
http://www.redbooks.ibm.com/redbooks/SG245674/wwhelp/wwhimpl/java/html/wwhelp.htm
and go to 4.2.4 The debug malloc allocator.
Other tools that may help: Purify, Valgrind (linux-only), Insure++,
ZeroFault.
Cheers,
--
In order to understand recursion you must first understand recursion.
Remove /-nsp/ for email.
.
- Follow-Ups:
- References:
- Core in a multithread application in AIX using RogueWave
- From: jmartin
- Core in a multithread application in AIX using RogueWave
- Prev by Date: Re: How to Disable IPv4 on AIX
- Next by Date: Re: How to Disable IPv4 on AIX
- Previous by thread: Core in a multithread application in AIX using RogueWave
- Next by thread: Re: Core in a multithread application in AIX using RogueWave
- Index(es):
Relevant Pages
|
