additional heuristic check for safer operation.
…his bug made showing a lot of BCEs corrupt on screen.
in MLDv1-compat-mode (the 2006-09-06's commit is not enough)
MLDv1-compat-mode (a bug due to a commit 2006-09-04)
…eceived in mldv1-compat-mode
…change an MLDv2 report robustness-variables times.
loopbacked ethernet frame still goes to the physical I/F's BPF. This closes the todo issue of the following item in CHANGELOG.2000. > Tue May 30 01:29:04 JST 2000 firstname.lastname@example.org ... > XXX if_simloop() call from ip6_output() causes junk to be injected > to bpf buffer. we may need some fix separately from this commit. (suggested by Max Laier)
…ILTER. (RFC3768 5.1.2 and 5.2.2)
- returns an error code when the MSF state change fails due to a lack of memory - fixed a bug that an MLDv2 State-Change Report is not always advertised robustness-variables times. - MLDv1-compat-timer properly checks the current compat-timer status
…n MLDv2 report robustness-variables times.
(Proposed by YuJung Lee)
…plication in userland applications. (suggested by Hitoshi Asaeda)
…command is issued on had.
bsd make command.
mip6_tunnel_input(): - eliminate 'proto' parameter and calculate proto value since darwin takes only 2 parameters - use protocol_input() to inject decapsulated ipv6 packet instead of IF_ENQUEUE() as other BSD does. With these fixes, MIP6 can work on darwin.