Hi all,
Today's linux-next merge of the net-next tree got a conflict in:
net/ax25/af_ax25.c
between commit:
d7c4c9e075f8c ("ax25: fix incorrect dev_tracker usage")
from the net tree and commit:
d62607c3fe459 ("net: rename reference+tracking helpers")
from the net-next tree.
I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging. You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.
diff --cc net/ax25/af_ax25.c
index 5b5363c99ed50,bbac3cb4dc99d..0000000000000
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c
@@@ -102,7 -102,8 +102,8 @@@ again
ax25_disconnect(s, ENETUNREACH);
s->ax25_dev = NULL;
if (sk->sk_socket) {
- dev_put_track(ax25_dev->dev, &ax25_dev->dev_tracker);
+ netdev_put(ax25_dev->dev,
- &ax25_dev->dev_tracker);
++ &ax25->dev_tracker);
ax25_dev_put(ax25_dev);
}
ax25_cb_del(s);
@@@ -1065,7 -1066,7 +1066,7 @@@ static int ax25_release(struct socket *
del_timer_sync(&ax25->t3timer);
del_timer_sync(&ax25->idletimer);
}
- dev_put_track(ax25_dev->dev, &ax25->dev_tracker);
- netdev_put(ax25_dev->dev, &ax25_dev->dev_tracker);
++ netdev_put(ax25_dev->dev, &ax25->dev_tracker);
ax25_dev_put(ax25_dev);
}
@@@ -1146,7 -1147,7 +1147,7 @@@ static int ax25_bind(struct socket *soc
if (ax25_dev) {
ax25_fillin_cb(ax25, ax25_dev);
- dev_hold_track(ax25_dev->dev, &ax25->dev_tracker, GFP_ATOMIC);
- netdev_hold(ax25_dev->dev, &ax25_dev->dev_tracker, GFP_ATOMIC);
++ netdev_hold(ax25_dev->dev, &ax25->dev_tracker, GFP_ATOMIC);
}
done:
On Tue, Aug 2, 2022 at 8:19 AM <[email protected]> wrote:
>
> Hi all,
>
> Today's linux-next merge of the net-next tree got a conflict in:
>
> net/ax25/af_ax25.c
>
> between commit:
>
> d7c4c9e075f8c ("ax25: fix incorrect dev_tracker usage")
>
> from the net tree and commit:
>
> d62607c3fe459 ("net: rename reference+tracking helpers")
>
> from the net-next tree.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging. You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
>
> diff --cc net/ax25/af_ax25.c
> index 5b5363c99ed50,bbac3cb4dc99d..0000000000000
> --- a/net/ax25/af_ax25.c
> +++ b/net/ax25/af_ax25.c
> @@@ -102,7 -102,8 +102,8 @@@ again
> ax25_disconnect(s, ENETUNREACH);
> s->ax25_dev = NULL;
> if (sk->sk_socket) {
> - dev_put_track(ax25_dev->dev, &ax25_dev->dev_tracker);
> + netdev_put(ax25_dev->dev,
> - &ax25_dev->dev_tracker);
This part seems wrong.
Commit d7c4c9e075f8c only changed the two spots, one in ax25_release()
and one in ax25_bind()
> ++ &ax25->dev_tracker);
> ax25_dev_put(ax25_dev);
> }
> ax25_cb_del(s);
> @@@ -1065,7 -1066,7 +1066,7 @@@ static int ax25_release(struct socket *
> del_timer_sync(&ax25->t3timer);
> del_timer_sync(&ax25->idletimer);
> }
> - dev_put_track(ax25_dev->dev, &ax25->dev_tracker);
> - netdev_put(ax25_dev->dev, &ax25_dev->dev_tracker);
> ++ netdev_put(ax25_dev->dev, &ax25->dev_tracker);
> ax25_dev_put(ax25_dev);
> }
>
> @@@ -1146,7 -1147,7 +1147,7 @@@ static int ax25_bind(struct socket *soc
>
> if (ax25_dev) {
> ax25_fillin_cb(ax25, ax25_dev);
> - dev_hold_track(ax25_dev->dev, &ax25->dev_tracker, GFP_ATOMIC);
> - netdev_hold(ax25_dev->dev, &ax25_dev->dev_tracker, GFP_ATOMIC);
> ++ netdev_hold(ax25_dev->dev, &ax25->dev_tracker, GFP_ATOMIC);
> }
>
> done:
On Wed, Aug 03, 2022 at 12:42:30AM -0700, Eric Dumazet wrote:
> On Tue, Aug 2, 2022 at 8:19 AM <[email protected]> wrote:
> > s->ax25_dev = NULL;
> > if (sk->sk_socket) {
> > - dev_put_track(ax25_dev->dev, &ax25_dev->dev_tracker);
> > + netdev_put(ax25_dev->dev,
> > - &ax25_dev->dev_tracker);
> This part seems wrong.
> Commit d7c4c9e075f8c only changed the two spots, one in ax25_release()
> and one in ax25_bind()
It is, I fixed it up later.