linux-kernel - Re: [PATCH 1/3] net: phy: Check phy

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:   Thu, 7 Jun 2018 12:34:54 -0500
From:   Brandon Maier <brandon.maier@...kwellcollins.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     netdev@...r.kernel.org, f.fainelli@...il.com, davem@...emloft.net,
        michal.simek@...inx.com,
        Clayton Shotwell <clayton.shotwell@...kwellcollins.com>,
        Kristopher Cory <kristopher.cory@...kwellcollins.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] net: phy: Check phy_driver ready before accessing

On Thu, Jun 7, 2018 at 11:52 AM, Andrew Lunn <andrew@...n.ch> wrote:
> FYI: net-next is closed at the moment. Please resubmit these in two
> weeks time.

Ah, I didn't see networking/netdev-FAQ.txt. I'll resubmit these then.

> I'm sure there are more issues like this in the code.  e.g. there is
> no attempt made to hold a reference to the child phy. So it could be
> unbound. priv->phy_drv->read_status(phydev) is then going to do bad
> things.
>

Agreed. Another thing that looks suspicious to me is the driver
overrides the private data of the device it's attaching too, in the
`priv->phy_dev->priv = priv` bit. Seems like that could cause all
sorts of driver corruption problems.

But fixing that is going to require more drastic changes to how this
driver works. So it'd be worth applying this patch in the mean time.