[Pdns-dev] Re: bindbackend_leak patch also missing

[Norbert Sendetzky]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Monday 22 December 2003 12:42, Norbert Sendetzky wrote:
> On Monday 22 December 2003 12:31, you wrote:
> > I'm going through your patches now. This one won't be applied,
> > bindbackends generally don't get destroyed anyhow and I have
> > reason to believe you can't copy pthread_mutex objects bitwise.
>
> Depends on the presencs of pointers inside pthread_mutex_t. At
> least the Linux implementation doesn't use pointers, but I don't
> know about other implementation.

It's not pthread_mutex_t but pthread_rwlock_t, but that doesn't make a 
difference.

I think you can apply the patch, because pthread_rwlock_destroy() is 
not called in the destructor. Therefore you get an exact copy of the 
pthread_rwlock_t object and any pointers inside will be still valid 
even if the original object is destoyed.


Norbert

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj/yzrYACgkQxMLs5v5/7eA6VwCeKUM++Wsh5SDYlHIcXWhneQI3
eYIAoLdjw5oOCXmW2IHxuLyzzQR0yN8Y
=0H4v
-----END PGP SIGNATURE-----