Hi,
...
> (Meanwhile, Michal, can I get a Signed-off-by: line from you for these
> patches? Thanks a lot.)
no problem :-)
There is a leak of a socket's multicast source filter list structure
on closing a socket with a multicast source filter set on an interface
that does not exist any more.
This patch fixes it:
Signed-off-by: Michal Ruzicka <[EMAIL PROTECTED]>
--- linux-2.6.17.8/net/ipv4/igmp.c.orig 2006-08-11 11:45:56.000000000 +0200
+++ linux-2.6.17.8/net/ipv4/igmp.c 2006-08-11 11:51:56.000000000 +0200
@@ -2202,13 +2202,13 @@
struct in_device *in_dev;
inet->mc_list = iml->next;
- if ((in_dev = inetdev_by_index(iml->multi.imr_ifindex)) !=
NULL) {
- (void) ip_mc_leave_src(sk, iml, in_dev);
+ in_dev = inetdev_by_index(iml->multi.imr_ifindex);
+ (void) ip_mc_leave_src(sk, iml, in_dev);
+ if (in_dev != NULL) {
ip_mc_dec_group(in_dev,
iml->multi.imr_multiaddr.s_addr);
in_dev_put(in_dev);
}
sock_kfree_s(sk, iml, sizeof(*iml));
-
}
rtnl_unlock();
}
Due to a bug in IP_DROP_MEMBERSHIP setsockopt handling code it is not
possible to leave a multicast group joined on an interface that
does not exist any more.
This patch makes leaving a multicast group possible even in that case:
Signed-off-by: Michal Ruzicka <[EMAIL PROTECTED]>
--- linux-2.6.17.8/net/ipv4/igmp.c.leak 2006-08-11 11:50:46.000000000 +0200
+++ linux-2.6.17.8/net/ipv4/igmp.c 2006-08-11 11:52:33.000000000 +0200
@@ -1799,19 +1799,15 @@
rtnl_lock();
in_dev = ip_mc_find_dev(imr);
- if (!in_dev) {
- rtnl_unlock();
- return -ENODEV;
- }
ifindex = imr->imr_ifindex;
for (imlp = &inet->mc_list; (iml = *imlp) != NULL; imlp = &iml->next) {
if (iml->multi.imr_multiaddr.s_addr == group &&
iml->multi.imr_ifindex == ifindex) {
- (void) ip_mc_leave_src(sk, iml, in_dev);
-
*imlp = iml->next;
- ip_mc_dec_group(in_dev, group);
+ (void) ip_mc_leave_src(sk, iml, in_dev);
+ if (in_dev != NULL)
+ ip_mc_dec_group(in_dev, group);
rtnl_unlock();
sock_kfree_s(sk, iml, sizeof(*iml));
return 0;
Michal
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
