mcast: fix leaked igmp packets on multicast cleanup #1391
Conversation
This reverts commit fa03d37. This commit caused us to not send IGMP leave messages if the inpcb went away. In other words: we freed pending packets whenever the socket closed rather than when the interface (or address) goes away. Reviewed by: glebius Sponsored by: Rubicon Communications, LLC ("Netgate") Differential Revision: https://reviews.freebsd.org/D43032 (cherry picked from commit c196e43)
When we release a multicast address (e.g. on interface shutdown) we may
still have packets queued in inm_scq. We have to free those, or we'll
leak memory.
Reviewed by: glebius
Sponsored by: Rubicon Communications, LLC ("Netgate")
Differential Revision: https://reviews.freebsd.org/D43033
(cherry picked from commit c2e3404)
fichtner
force-pushed
the
igmp_leave_leak
branch
from
d171786
to
e2e0761
Compare
|
Normally we don't take MFC PRs. However, this one looks like it's easy enough to followup on so I'll allow it for now. |
|
Thanks. To avoid chasing up committers for MFC on operational issues would it make sense to make this semi-recurring? One change of note is ad874544d9f that would at least benefit 14.2 in the long run. |
|
Yea, the original committer is responsible for MFC requests. So that's the first line of attack on that. |
|
Fair enough, thanks for responding. I'll chase Vincenzo then. |
|
These two commits had already been merged. I really don't understand why github doesn't reflect this. |
|
Even if they did match, it would only close it. |
The commits cherry-picked here have not been added to stable/14.
They have been known to fix IGMP leave messages which are currently
broken on both 14.0 and 14.1. The necessary MFC has been absent from
them for unknown reasons.
Differential Revision: https://reviews.freebsd.org/D46190