[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201120024311.5021d6b7@nic.cz>
Date: Fri, 20 Nov 2020 02:43:11 +0100
From: Marek Behun <marek.behun@....cz>
To: Andrew Lunn <andrew@...n.ch>
Cc: Pavana Sharma <pavana.sharma@...i.com>, lkp@...el.com,
ashkan.boldaji@...i.com, clang-built-linux@...glegroups.com,
davem@...emloft.net, f.fainelli@...il.com,
gregkh@...uxfoundation.org, kbuild-all@...ts.01.org,
kuba@...nel.org, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org, robh+dt@...nel.org,
devicetree@...r.kernel.org, vivien.didelot@...il.com
Subject: Re: [PATCH v10 4/4] net: dsa: mv88e6xxx: Add support for mv88e6393x
family of Marvell
Hi Andrew,
On Fri, 20 Nov 2020 02:29:06 +0100
Andrew Lunn <andrew@...n.ch> wrote:
> > + if (speed >= 2500 && port > 0 && port < 9)
> > + return -EOPNOTSUPP;
>
> Maybe i'm missing something, but it looks like at this point you can
> call
>
> return mv88e6xxx_port_set_speed_duplex(chip, port, speed, true, true, duplex);
He can't. That function does not support speed 5000. You can't simply
add it, because it clashes with register value for speed 2500 on
previous switches (Peridot, Topaz).
Amethyst reg val Peridot + Topaz reg val
2500 SPD_1000 | ALT_BIT SPD_10000 | ALT_BIT
5000 SPD_10000 | ALT_BIT not supported
10000 SPD_UNFORCED SPD_UNFORCED
When I sent my proposal for Amethyst I somehow did it, and you
commented [1]:
> This is getting more and more complex. Maybe it is time to refactor it?
And I agree :)
Marek
[1] https://www.spinics.net/lists/netdev/msg678090.html
Powered by blists - more mailing lists