netdev - Re: [PATCH v1 net-next] dev: Pass netdevice_tracker to dev_get_by_flags

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAAVpQUAocrVbJTk08_VxThuNpgXAC7Zvcn4OHM6FdNKXbs6k2A@mail.gmail.com>
Date: Wed, 9 Jul 2025 18:38:30 -0700
From: Kuniyuki Iwashima <kuniyu@...gle.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: "David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Paolo Abeni <pabeni@...hat.com>, Andrew Lunn <andrew+netdev@...n.ch>, 
	David Ahern <dsahern@...nel.org>, Simon Horman <horms@...nel.org>, 
	Kuniyuki Iwashima <kuni1840@...il.com>, netdev@...r.kernel.org
Subject: Re: [PATCH v1 net-next] dev: Pass netdevice_tracker to dev_get_by_flags_rcu().

On Wed, Jul 9, 2025 at 5:16 PM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Wed,  9 Jul 2025 19:01:32 +0000 Kuniyuki Iwashima wrote:
> > diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
> > index 53cf68e0242bf..fa7f0c22167b4 100644
> > --- a/net/ipv6/anycast.c
> > +++ b/net/ipv6/anycast.c
> > @@ -69,6 +69,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
> >       struct ipv6_pinfo *np = inet6_sk(sk);
> >       struct ipv6_ac_socklist *pac = NULL;
> >       struct net *net = sock_net(sk);
> > +     netdevice_tracker dev_tracker;
> >       struct net_device *dev = NULL;
> >       struct inet6_dev *idev;
> >       int err = 0, ishost;
> > @@ -112,8 +113,8 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
> >                       goto error;
> >               } else {
> >                       /* router, no matching interface: just pick one */
> > -                     dev = dev_get_by_flags_rcu(net, IFF_UP,
> > -                                                IFF_UP | IFF_LOOPBACK);
> > +                     dev = netdev_get_by_flags_rcu(net, &dev_tracker, IFF_UP,
> > +                                                   IFF_UP | IFF_LOOPBACK);
> >               }
> >               rcu_read_unlock();
> >       }
> > @@ -159,7 +160,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
> >  error_idev:
> >       in6_dev_put(idev);
> >  error:
> > -     dev_put(dev);
> > +     netdev_put(dev, &dev_tracker);
>
> Hmmm.. not sure this is legal.. We could have gotten the reference from
> dev_get_by_index() or a bare dev_hold() -- I mean there are two other
> ways of acquiring dev in this function. Either all or none of them
> have to be tracker aware, we can't mix?

Oh sorry, I totally forgot to update other parts..
Will squash this in v2.

diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index fa7f0c22167b4..f8a8e46286b8e 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -80,7 +80,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex,
const struct in6_addr *addr)
  return -EINVAL;

  if (ifindex)
- dev = dev_get_by_index(net, ifindex);
+ dev = netdev_get_by_index(net, ifindex, &dev_tracker, GFP_KERNEL);

  if (ipv6_chk_addr_and_flags(net, addr, dev, true, 0, IFA_F_TENTATIVE)) {
  err = -EINVAL;
@@ -105,7 +105,7 @@ int ipv6_sock_ac_join(struct sock *sk, int
ifindex, const struct in6_addr *addr)
  rt = rt6_lookup(net, addr, NULL, 0, NULL, 0);
  if (rt) {
  dev = dst_dev(&rt->dst);
- dev_hold(dev);
+ netdev_hold(dev, &dev_tracker, GFP_ATOMIC);
  ip6_rt_put(rt);
  } else if (ishost) {
  rcu_read_unlock();