Call for testers: NetBSD IGMPv2 merge

5 msg	nmdm(4) does not call .l_close
2 msg	tcp messages I've never seen now in Current and...
2 msg	net/twinkle stuck in _umtx_op syscall
2 msg	Minor miracle
1 msg	emu10kx and related Creative drivers
2 msg	Can't build -current on a 6.2 box
2 msg	setting dumpdev while boot on 7-CURRENT
12 msg	Can't create TCP connections to certain IP addr...
2 msg	snd_emu10k1 mixer
4 msg	ZFS on a notebook/512M settings
5 msg	Are SATA port multipliers supported in -CURRENT?
6 msg	installworld breaks my FreeBSD/powerpc system!!!
1 msg	page fault in g_event (probably zfs related?)
8 msg	FreeBSD on intel Mac
5 msg	ZFS on root: slave mountpoints mount failed

Call for testers: NetBSD IGMPv2 merge
\ Bruce M. Simpson (25 Jul 2007)
. \ Andre Oppermann (25 Jul 2007)

3 msg	Interrupt storm on atapci1+
4 msg	Error while upgrading FreeBSD-Release6.1 to CUR...
5 msg	Removing NET_NEEDS_GIANT: first patch
5 msg	zsh oddities with recent -current

Subject:	Call for testers: NetBSD IGMPv2 merge
Group:	Freebsd-current
From:	Bruce M. Simpson
Date:	25 Jul 2007

Hi,

I have mostly merged NetBSD's IGMPv2 implementation with a few refinements.
This work is being funded by a secret santa. :-)

This code is easier to read (and thus maintain), and fixes a minor
memory leak in the IPv4 stack. It is also a jumping off point for the
IGMPv3 work. KAME patch sets up to date have been against NetBSD, though
I don't intend to use their patches directly.

Patch against -CURRENT, URL:
http://people.freebsd.org/~bms/dump/igmp-netbsd-sync.patch

I have not merged the change which suppresses host membership reports
for the link-local multicast scope space 224.0.0.0/8, my concern is that
this breaks IGMP snooping switches on immediately attached layer 2
networks, even though this range is not globally routable. (*)

Testing and review much appreciated. If enough people are willing to
soak test this code before 7.0 point release is branched then it might
justify going in in time for 7.0.

Regards,
BMS
(*) That is not a typo - 224.0.0.0/8 *not* the whole Class D range
224.0.0.0/4.
_______________________________________________
freebsd-current mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe"