On Wed, 2009-06-10 at 20:54 +0200, Ruediger Pluem wrote:
> On 06/10/2009 05:56 AM, Bojan Smojver wrote:
> > On Wed, 2009-06-10 at 13:51 +1000, Bojan Smojver wrote:
> >> Actually, that LDAP thing is not required.
> >
> > So, here are the two patches. One for trunk (for now, until this gets
> > fixed properly) and one for 1.3.x/1.4.x.
> >
>
>
> I currently don't get the dbm/apr_dbm.c thing. Shouldn't we do a
>
> return APR_SUCCESS
>
> after the while loop below like in the other cases?
>
> - if (!drivers)
> - {
> + if (apr_atomic_inc32(&initialised)) {
> + apr_atomic_set32(&initialised, 1); /* prevent wrap-around */
> +
> + while (apr_atomic_read32(&in_init)) /* wait until we get fully
> inited */
> + ;
> + }
> + else {
> apr_pool_t *parent;
I think not, because after the if/else, we still have work to do. We
need to find the symbol for our DB engine (either from the hash or by
dso_load), which depends on type, which can be different from call to
call. No?
--
Bojan
