On Thu, Oct 10, 2019 at 12:07 PM Guillaume Nault <gna...@redhat.com> wrote:
>
> In rtnl_net_notifyid(), we certainly can't pass a null GFP flag to
> rtnl_notify(). A GFP_KERNEL flag would be fine in most circumstances,
> but there are a few paths calling rtnl_net_notifyid() from atomic
> context or from RCU critical section. The later also precludes the use
> of gfp_any() as it wouldn't detect the RCU case. Also, the nlmsg_new()
> call is wrong too, as it uses GFP_KERNEL unconditionally.
>
> Therefore, we need to pass the GFP flags as parameter. The problem then
> propagates recursively to the callers until the proper flags can be
> determined. The problematic call chains are:
>
> * ovs_vport_cmd_fill_info -> peernet2id_alloc -> rtnl_net_notifyid
>
> * rtnl_fill_ifinfo -> rtnl_fill_link_netnsid -> peernet2id_alloc
> -> rtnl_net_notifyid
>
> For openvswitch, ovs_vport_cmd_get() and ovs_vport_cmd_dump() prevent
> ovs_vport_cmd_fill_info() from using GFP_KERNEL. It'd be nice to move
> the call out of the RCU critical sections, but struct vport doesn't
> have a reference counter, so that'd probably require taking the ovs
> lock. Also, I don't get why ovs_vport_cmd_build_info() used GFP_ATOMIC
> in nlmsg_new(). I've changed it to GFP_KERNEL for consistency, as this
> functions seems to be allowed to sleep (as stated in the comment, it's
> called from a workqueue, under the protection of a mutex).
>
It is safe to change GFP flags to GFP_KERNEL in ovs_vport_cmd_build_info().
The patch looks good to me.
> For core networking, rtmsg_ifinfo_build_skb() prevents rtnl_fill_ifinfo()
> from using GFP_KERNEL. So we have to reuse rtmsg_ifinfo_build_skb()'s GFP
> flags there. Actually, the only caller preventing the use of GFP_KERNEL,
> is __dev_notify_flags() which calls rtmsg_ifinfo() with GFP_ATOMIC (see
> commit 7f29405403d7 ("net: fix rtnl notification in atomic context")).
> I'd have expected this function to be allowed to sleep, but I haven't
> dug deeper yet.
>
> Before I spend more time on this, do we have a chance to make
> ovs_vport_cmd_fill_info() and __dev_notify_flags() sleepable?
> I'd like to avoid passing GFP flags along these call chains, if at all
> possible.
>
> Found by code inspection.
>
> Fixes: 9a9634545c70 ("netns: notify netns id events")
> Signed-off-by: Guillaume Nault <gna...@redhat.com>
