On Thu, Jun 07, 2018 at 02:35:39PM +0200, Phil Sutter wrote: > Yes, I agree with Michal. IIRC, flushing a specific primary along with > all it's secondaries from an interface is not even supported by > iproute2, so no need to optimize for that I guess. OTOH, if your > solution allowed to get rid of that nasty loop in ipaddr_flush(), I owe > you one extra beer at the next occasion. :)
I'm afraid it will have to stay as fallback for older kernels not supporting flush requests. But there would be no need to actually use it. If we know RTM_DELADDR request for zero address is guaranteed to fail with current and older kernels, we could do - use RTM_DELADDR with IFA_F_FLUSH and zero address - if it fails, get the list and run the loop If not, it could still be - use RTM_DELADDR with IFA_F_FLUSH and zero address - get the list of addresses (empty if first step worked) - run the loop Michal Kubecek
