* ulogd - polling handler - code clarification
@ 2012-07-30 22:18 Gomathivinayagam Muthuvinayagam
2012-07-31 11:27 ` Pablo Neira Ayuso
0 siblings, 1 reply; 2+ messages in thread
From: Gomathivinayagam Muthuvinayagam @ 2012-07-30 22:18 UTC (permalink / raw)
To: netfilter-devel
Hi,
I was going through the ulogd_inpflow_NFCT.c : polling_handler code. I
need few clarifications.
If someone can answer for that, that will be great.
725 case NFCT_T_UPDATE:
726 id = hashtable_hash(cpi->ct_active, ct);
727 ts = (struct ct_timestamp *)
728 hashtable_find(cpi->ct_active, ct, id);
729 if (ts)
730 nfct_copy(ts->ct, ct, NFCT_CP_META);
731 else {
732 ts = calloc(sizeof(struct ct_timestamp), 1);
733 if (ts == NULL)
734 return NFCT_CB_CONTINUE;
735
736 ts->ct = ct;
737 gettimeofday(&ts->time[START], NULL);
738
739 ret = hashtable_add(cpi->ct_active,
&ts->hashnode, id);
740 if (ret < 0) {
741 free(ts);
742 return NFCT_CB_CONTINUE;
743 }
744 return NFCT_CB_STOLEN; // Here, why not
NFCT_CB_CONTINUE is used?
745 }
746 break;
Here in Line no 744, why NFCT_CB_STOLEN is used instead of
NFCT_CB_CONTINUE ? I am not able to understand the reason.
Thanks & Regards,
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: ulogd - polling handler - code clarification
2012-07-30 22:18 ulogd - polling handler - code clarification Gomathivinayagam Muthuvinayagam
@ 2012-07-31 11:27 ` Pablo Neira Ayuso
0 siblings, 0 replies; 2+ messages in thread
From: Pablo Neira Ayuso @ 2012-07-31 11:27 UTC (permalink / raw)
To: Gomathivinayagam Muthuvinayagam; +Cc: netfilter-devel
On Mon, Jul 30, 2012 at 03:18:32PM -0700, Gomathivinayagam Muthuvinayagam wrote:
> Hi,
>
> I was going through the ulogd_inpflow_NFCT.c : polling_handler code. I
> need few clarifications.
> If someone can answer for that, that will be great.
>
>
> 725 case NFCT_T_UPDATE:
> 726 id = hashtable_hash(cpi->ct_active, ct);
> 727 ts = (struct ct_timestamp *)
> 728 hashtable_find(cpi->ct_active, ct, id);
> 729 if (ts)
> 730 nfct_copy(ts->ct, ct, NFCT_CP_META);
> 731 else {
> 732 ts = calloc(sizeof(struct ct_timestamp), 1);
> 733 if (ts == NULL)
> 734 return NFCT_CB_CONTINUE;
> 735
> 736 ts->ct = ct;
> 737 gettimeofday(&ts->time[START], NULL);
> 738
> 739 ret = hashtable_add(cpi->ct_active,
> &ts->hashnode, id);
> 740 if (ret < 0) {
> 741 free(ts);
> 742 return NFCT_CB_CONTINUE;
> 743 }
> 744 return NFCT_CB_STOLEN; // Here, why not
> NFCT_CB_CONTINUE is used?
> 745 }
> 746 break;
>
> Here in Line no 744, why NFCT_CB_STOLEN is used instead of
> NFCT_CB_CONTINUE ? I am not able to understand the reason.
NFCT_CB_STOLEN means that the object is not released once the callback
code is done.
That object is released once we notice that the entry is not there in
the kernel anymore.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2012-07-31 11:27 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-07-30 22:18 ulogd - polling handler - code clarification Gomathivinayagam Muthuvinayagam
2012-07-31 11:27 ` Pablo Neira Ayuso
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).