d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix a comment and some commented-out code in ip_mloopback to

Browse Source 
reflect how multicast loopback really works.
This commit is contained in:
Bill Fenner 1997-02-28 19:40:48 +00:00
parent 5ee71a0d7b
commit e1596dff1e
1 changed files with 10 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
sys/netinet/ip_output.c
View File

@ -31,7 +31,7 @@
* SUCH DAMAGE.
*
* @(#)ip_output.c 8.3 (Berkeley) 1/21/94
* $Id$
* $Id: ip_output.c,v 1.51 1997/02/22 09:41:36 peter Exp $
*/
#define _IP_VHL
@ -1310,18 +1310,17 @@ ip_mloopback(ifp, m, dst)
}
/*
* NB:
* We can't simply call ip_input() directly because
* the ip_mforward() depends on the `input interface'
* being set to something unreasonable so that we don't
* attempt to forward the looped-back copy.
* It's also not clear whether there are any lingering
* reentrancy problems in other areas which might be
* exposed by this code. For the moment, we'll err
* on the side of safety by continuing to abuse
* loinput().
* It's not clear whether there are any lingering
* reentrancy problems in other areas which might
* be exposed by using ip_input directly (in
* particular, everything which modifies the packet
* in-place). Yet another option is using the
* protosw directly to deliver the looped back
* packet. For the moment, we'll err on the side
* of safety by continuing to abuse looutput().
*/
#ifdef notdef
copym->m_pkthdr.rcvif = &loif[0];
copym->m_pkthdr.rcvif = ifp;
ip_input(copym)
#else
(void) looutput(ifp, copym, (struct sockaddr *)dst, NULL);