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

Re: latest glibc required.



Hi Jakub
I found that if the object is build on RHEL machine and run on RHEL machine it
works fine but
If the object is build on linux7.2 and run on RHEl3 machine it hangs giving the
message
"#0  0xbaf7bf39 in free () from /lib/tls/libc.so.6"
Does it indicate the application is correrct and there is something missing on RHEL
?
If I use the workaround for "glibc" on RHEL machine the same object works fine it
indicates that the problem was in "glibc".
Please help me I am unable to know whether  the bug is in application or the
environment.


Jakub Jelinek wrote:

> On Wed, Jul 07, 2004 at 02:44:43PM +0530, Anant Kantak wrote:
> >
>
> > Hi
> > I got the problem while running the linux7.2 objects on the rhel platform.
> > The problem is, execution hang with following message
> >  "#0  0xbaf7bf39 in free () from /lib/tls/libc.so.6"
>
> This most likely is an application bug.
> Hang in free usually means double free somehwere or overflowing a malloc
> buffer.
> Try ElectricFence on the application, or MALLOC_CHECK_=3, or valgrind.
>
> > Refering to link "http://bugs.freehep.org/browse/AIDAJNI-5?page=history";
> > I found that the problem is due to the "libc" is having bug, which will be
> > fixed in the next release of the "glibc". The workaround for the above problem
> > is "setenv LD_ASSUME_KERNEL 2.4.1  which disable NPTL in glibc.
>
> The above link mentions #90301, which was fixed by
> 2003-04-19  Ulrich Drepper  <drepper redhat com>
>
>         * stdlib/cxa_finalize.c (__cxa_finalize): Don't call
>         UNREGISTER_ATFORK if d == NULL.
>
> RHEL3 uses glibc CVS snapshot from 2003-09-26, so obviously this fix is in.
>
>         Jakub



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