Re: Regression: net/ipv6/mld running system out of memory (not a leak)

On Mon, Mar 09, 2020 at 01:31:16PM +0100, Rafał Miłecki wrote:

quoted

2. Should we move ipv6_dev_mc_inc() from ipv6_add_dev() to ipv6_mc_up()?
I don't know yet, this dependents on whether we could add multicast address
on non-Ethernen dev.

I'm not the one to answer them surely with my limited net subsystem
understanding :( Any idea how to proceed with this? I assume your patch
is still the right step, do you think you can send it officially now?

quoted

quoted

we call ipv6_mc_leave_localaddr() without ipv6_mc_join_localaddr()
called first which seems unintuitive.

This doesn't matter much yet. As we will check if we have the address
in __ipv6_dev_mc_dec(), if not, we just return. But yes, form logic, this
looks asymmetric.

Right, just slightly unintuitive asymmetric code.

Hi Rafał,

I investigated this issue and am going to enable ipv6_mc_up for non-Ethernet
interface. The patch I sent you before will be post as a RFC for net-next.

Thanks
Hangbin