I think the big picture on this one is that the caller of
ap_wait_or_timeout() needs to pass in storage for the ap_proc_t so we
can avoid the storage leak.
ap_wait_or_timeout() can still return NULL or valid ap_proc_t, but the
valid ap_proc_t storage would be passed in as a parm.
> gstein 00/05/23 16:32:37
>
> Modified: src/main mpm_common.c
> Log:
> use the correct size for 'ret'
>
> Revision Changes Path
> 1.7 +1 -1 apache-2.0/src/main/mpm_common.c
>
> Index: mpm_common.c
> ===================================================================
> RCS file: /home/cvs/apache-2.0/src/main/mpm_common.c,v
> retrieving revision 1.6
> retrieving revision 1.7
> diff -u -r1.6 -r1.7
> --- mpm_common.c 2000/05/23 23:26:51 1.6
> +++ mpm_common.c 2000/05/23 23:32:35 1.7
> @@ -180,7 +180,7 @@
> {
> struct timeval tv;
> ap_status_t rv;
> - ap_proc_t *ret = ap_pcalloc(p, sizeof(ret));
> + ap_proc_t *ret = ap_pcalloc(p, sizeof(*ret));
>
> ++wait_or_timeout_counter;
> if (wait_or_timeout_counter == INTERVAL_OF_WRITABLE_PROBES) {
>
>
>
>
--
Jeff Trawick | trawick@ibm.net | PGP public key at web site:
http://www.geocities.com/SiliconValley/Park/9289/
Born in Roswell... married an alien...
|