[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <b4d25621-f89a-4b48-9a00-33274f130e8c@lunn.ch>
Date: Fri, 9 Jun 2023 16:25:14 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Christian Marangi <ansuelsmth@...il.com>
Cc: Pavel Machek <pavel@....cz>, Lee Jones <lee@...nel.org>,
"David S. Miller" <davem@...emloft.net>,
Yang Li <yang.lee@...ux.alibaba.com>,
linux-leds@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] leds: trigger: netdev: add additional specific link
speed mode
> + if (trigger_data->net_dev != NULL) {
> + struct ethtool_link_ksettings cmd;
> +
cmd is a stack variable, so contains random junk:
> trigger_data->carrier_link_up = netif_carrier_ok(trigger_data->net_dev);
>
> + if (trigger_data->carrier_link_up) {
> + rtnl_lock();
> + __ethtool_get_link_ksettings(trigger_data->net_dev, &cmd);
/* Internal kernel helper to query a device ethtool_link_settings. */
int __ethtool_get_link_ksettings(struct net_device *dev,
struct ethtool_link_ksettings *link_ksettings)
{
ASSERT_RTNL();
if (!dev->ethtool_ops->get_link_ksettings)
return -EOPNOTSUPP;
If the op is not implemented, it just returns.
> + rtnl_unlock();
> +
> + trigger_data->link_speed = cmd.base.speed;
and now you are accessing the random junk.
> + __ethtool_get_link_ksettings(trigger_data->net_dev, &cmd);
> +
> + trigger_data->link_speed = cmd.base.speed;
You have this code three times. I suggest you pull it out into a
little helper, and within the helper, deal with the return code, and
set speed to 0 if it is unknown.
Andrew
Powered by blists - more mailing lists