>are you referring to:
>
>ChangeSet@1.1310.1.32, 2003-06-18 18:01:12-07:00, dlstevens@us.ibm.com
[IPV4/IPV6]: Fix IGMP device refcount leaks, with help from
yoshfuji@linux-ipv6.org.
Yes.
>i just pulled the latest 2.5 sources and i dont see any changes
>that would seem to correct the bug i am seeing. the timers still
>use __in_dev_put() to drop the references. this wont call
>in_dev_finish_destroy() if it drops the last reference (which is
>fairly likely given the duration of the timers).
But you didn't try reproducing the problem, apparently...
The reference count can't go to zero where __in_dev_put()'s are used, and
the timer isn't running when the interface is downed-- it's deleted as part
of ip_mc_down().
Try it out. I predict it'll work, but I'm certainly interested in hearing
if you have a test case that doesn't. :-)
+-DLS
-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
