Re: linux kernel's IPV6_MULTICAST_HOPS default is 64; should be 1?

[Brian Haley <brian.haley@hp.com>]

David Miller wrote:
> From: David Stevens <dlstevens@us.ibm.com>
> Date: Tue, 4 May 2010 00:48:46 -0700
> 
>> It's set to -1 by default, but the common code for unicast and
>> multicast in getsockopt is falling through to use the dst_entry.
>>
>> I believe (though I haven't actually tried it recently) it actually
>> uses "1" for the default value for multicast;

No, on-the-wire it's actually 64.

> It doesn't, all of the uses in the ipv6 stack say something like:
> 
> 	if (multicast)
> 		hlimit = np->mcast_hops;
> 	else
> 		hlimit = np->hop_limit;
> 	if (hlimit < 0)
> 		hlimit = ip6_dst_hoplimit(dst);
> 
> Therefore, the change suggested by Elliot and which I committed is the
> way to get the correct behavior and fix this.

Not exactly.  It fixes the case where it's wrong by default, but
the corner case of setting it to -1 via setsockopt() says:

    x == -1:       use kernel default

But that will revert back to the kernel using 64 on the next transmit.
I can work on an update to this that makes a new mcast_hops per-interface
setting and makes ip6_dst_hoplimit() aware of it.  Or even easier, just
have setsockopt() trap the -1 and set np->mcast_hops to 1.  Built but
untested patch below.

-Brian

--


Specifying -1 for setsockopt(IPV6_MULTICAST_HOPS) should set the socket
value back to the system default value of IPV6_DEFAULT_MCASTHOPS (1).

Signed-off-by: Brian Haley <brian.haley@hp.com>

diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index bd43f01..fa6875b 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -486,7 +486,10 @@ done:
 			goto e_inval;
 		if (val > 255 || val < -1)
 			goto e_inval;
-		np->mcast_hops = val;
+		if (val == -1)
+			np->mcast_hops = IPV6_DEFAULT_MCASTHOPS;
+		else
+			np->mcast_hops = val;
 		retv = 0;
 		break;