[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20211217130325.305961eb@hermes.local>
Date: Fri, 17 Dec 2021 13:03:25 -0800
From: Stephen Hemminger <stephen@...workplumber.org>
To: Yevhen Orlov <yevhen.orlov@...ision.eu>
Cc: netdev@...r.kernel.org,
Volodymyr Mytnyk <volodymyr.mytnyk@...ision.eu>,
Taras Chornyi <taras.chornyi@...ision.eu>,
Mickey Rachamim <mickeyr@...vell.com>,
Serhiy Pshyk <serhiy.pshyk@...ision.eu>,
Taras Chornyi <tchornyi@...vell.com>,
Oleksandr Mazur <oleksandr.mazur@...ision.eu>,
"David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next 6/6] net: marvell: prestera: Implement initial
inetaddr notifiers
On Fri, 17 Dec 2021 21:54:38 +0200
Yevhen Orlov <yevhen.orlov@...ision.eu> wrote:
> switch (event) {
> case NETDEV_UP:
> + if (re) {
> + NL_SET_ERR_MSG_MOD(extack, "rif_entry already exist");
> + return -EEXIST;
> + }
> + re = prestera_rif_entry_create(port->sw, &re_key,
> + prestera_fix_tb_id(kern_tb_id),
> + port_dev->dev_addr);
> + if (!re) {
> + NL_SET_ERR_MSG_MOD(extack, "Can't create rif_entry");
> + return -EINVAL;
> + }
> + dev_hold(port_dev);
How are these device references cleaned up on module removal?
Powered by blists - more mailing lists