netdev - Re: [PATCH net v1] net: dsa: mv88e6xxx: Make unsupported C45 reads return 0xffff

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]

Message-ID: <0d9e0412-6ca3-407a-b2a1-b18ab4c20714@lunn.ch>
Date: Mon, 22 Jan 2024 14:39:10 +0100
From: Andrew Lunn <andrew@...n.ch>
To: Vladimir Oltean <vladimir.oltean@....com>
Cc: netdev-maintainers <edumazet@...gle.com>, kuba@...nel.org,
	pabeni@...hat.com, davem@...emloft.net,
	netdev <netdev@...r.kernel.org>, stable@...r.kernel.org,
	Tim Menninger <tmenninger@...estorage.com>
Subject: Re: [PATCH net v1] net: dsa: mv88e6xxx: Make unsupported C45 reads
 return 0xffff

On Mon, Jan 22, 2024 at 02:24:57PM +0200, Vladimir Oltean wrote:
> Hi Andrew,
> 
> On Sat, Jan 20, 2024 at 08:21:25PM +0100, Andrew Lunn wrote:
> > When there is no device on the bus for a given address, the pull up
> > resistor on the data line results in the read returning 0xffff. The
> > phylib core code understands this when scanning for devices on the
> > bus, and a number of MDIO bus masters make use of this as a way to
> > indicate they cannot perform the read.
> > 
> > Make us of this as a minimal fix for stable where the mv88e6xxx
> 
> s/us/use/
> 
> Also, what is the "proper" fix if this is the minimal one for stable?

Hi Vladimir

I have a patchset for net-next, once it opens. I looked at how C22 and
C45 differ in handling error codes. C22 allows the MDIO bus driver to
return -ENODEV to indicate its impossible for a device to be at a
given address. The scan code then skips that address and continues to
the next address. Current C45 code would turn that -ENODEV into an
-EIO and consider it fatal. So i change the C45 code to allow for
-ENODEV in the same way, and change the mv88e6xxx driver to return
-ENODEV if there are is no C45 read op.

Since making the handling of the error codes uniform is more than a
simple fix, i decided on a minimal fix for net.

Thanks for the comments on the commit message, i will address them
soon.

	Andrew