netdev - Re: [PATCH net-next v2 7/7] net: dsa: mv88e6xxx: Add per port devlink regions

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: <20200927011131.56ol5vlldq3yttmr@skbuf>
Date:   Sun, 27 Sep 2020 04:11:31 +0300
From:   Vladimir Oltean <olteanv@...il.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     David Miller <davem@...emloft.net>,
        netdev <netdev@...r.kernel.org>,
        Florian Fainelli <f.fainelli@...il.com>,
        Vladimir Oltean <vladimir.oltean@....com>,
        Jiri Pirko <jiri@...dia.com>, Jakub Kicinski <kuba@...nel.org>
Subject: Re: [PATCH net-next v2 7/7] net: dsa: mv88e6xxx: Add per port
 devlink regions

On Sun, Sep 27, 2020 at 03:03:00AM +0200, Andrew Lunn wrote:
> > I was meaning to ask this since the global regions patchset, but I
> > forgot.
> >
> > Do we not expect to see, under the same circumstances, the same region
> > snapshot on a big endian and on a little endian system?
>
> We have never had any issues with endinness with MDIO. PHY/DSA drivers
> work with host endian. The MDIO bus controller does what it needs to
> do when shifting the bits out, as required by class 22 or 45.
>
> netlink in general assume host endian, as far as i know. So a big
> endian and a little endian snapshot are going to be different.

If the binary form of the snapshot is supposed to be consumed and done
with immediately after the netlink communication is over, sure. But it
is presented to the user, and in fact this is even the way it is
presented by default, with devlink, there's no pretty-printing.

> Arnd did an interesting presentation for LPC. He basically shows that
> big endian is going away, with the exception of IBM big iron. I don't
> expect an IBM Z to have a DSA switch!

Tell that to my T1040 chip with an 8-port Seville switch driven by the
Ocelot driver.

Also, armv8 can also boot in big-endian mode.