Dan Smith wrote: > OL> Did you also address this ? > > Sorry, I meant to put something in the intro mail about this. > > OL> [The reason sk_free() expects no sk_socket is because the socket > OL> at this point must have been released already via > OL> proto_ops->release(). The callback is assumed to orphan the > OL> socket]. > > Right, which is why we're not seeing the warning. I've got a > follow-on patch to the current set started but wanted to avoid packing > more onto these two patches. If you'd like me to resubmit the set and > include that patch (when finished) I'll do that. >
I prefer not to pull in code that we know has a leak, but rather wait for the fixed version. I'll take the 2nd patch (post-file deferqueue) now. Oren. _______________________________________________ Containers mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/containers _______________________________________________ Devel mailing list [email protected] https://openvz.org/mailman/listinfo/devel
