netdev - Re: [RFC] managing PHY carrier from user space

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]

Date:   Tue, 11 Sep 2018 09:56:21 -0700
From:   Florian Fainelli <f.fainelli@...il.com>
To:     Joakim Tjernlund <Joakim.Tjernlund@...inera.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "andrew@...n.ch" <andrew@...n.ch>
Subject: Re: [RFC] managing PHY carrier from user space

On 09/11/2018 09:41 AM, Joakim Tjernlund wrote:
> I am looking for a way to induce carrier state from user space, primarily
> for Fixed PHYs as these are always up. ifplugd/dhcp etc. does not behave properly
> if the link is up when it really isn't.

Was my suggestion in my email to you somehow not working? This is
obviously not acceptable for upstream, there is no reason, even for a
fixed PHY, to attempt to mangle with the carrier state for any
reasonable production purposes.

> 
> I came up with a new 'phy_carrier' attribute in /sys/class/net/eth0/phydev
> where I can induce carrier state:
> 
> diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
> index a1e7ea4d4b16..f82beeabdd75 100644
> --- a/drivers/net/phy/phy_device.c
> +++ b/drivers/net/phy/phy_device.c
> @@ -612,10 +612,39 @@ phy_has_fixups_show(struct device *dev, struct device_attribute *attr,
>  }
>  static DEVICE_ATTR_RO(phy_has_fixups);
>  
> +static ssize_t
> +phy_carrier_show(struct device *dev, struct device_attribute *attr,
> +                char *buf)
> +{
> +       struct phy_device *phydev = to_phy_device(dev);
> +       struct net_device *netdev = phydev->attached_dev;
> +
> +       return sprintf(buf, "%d\n", netif_carrier_ok(netdev));
> +}
> +
> +static ssize_t phy_carrier_store(struct device *dev, struct device_attribute *attr,
> +                                const char *buf, size_t len)
> +{
> +       struct phy_device *phydev = to_phy_device(dev);
> +       struct net_device *netdev = phydev->attached_dev;
> +       bool enable;
> +
> +       if (strtobool(buf, &enable))
> +               return -EINVAL;
> +
> +       if (enable)
> +               netif_carrier_on(netdev);
> +       else
> +               netif_carrier_off(netdev);
> +       return len;
> +}
> +static DEVICE_ATTR_RW(phy_carrier);
> +
>  static struct attribute *phy_dev_attrs[] = {
>         &dev_attr_phy_id.attr,
>         &dev_attr_phy_interface.attr,
>         &dev_attr_phy_has_fixups.attr,
> +       &dev_attr_phy_carrier.attr,
>         NULL,
>  };
>  ATTRIBUTE_GROUPS(phy_dev);
> 
> I would like to know if this acceptable for linux proper or if there is
> a better way to do this?



> 
>  Jocke
> 


-- 
Florian