[Top] [All Lists]

Re: [1/3] MIPS: octeon: Add interface mode detection for Octeon II

To: David Daney <>
Subject: Re: [1/3] MIPS: octeon: Add interface mode detection for Octeon II
From: Aaro Koskinen <>
Date: Wed, 4 Jun 2014 21:53:03 +0300
Cc: Alex Smith <>, Aaro Koskinen <>, David Daney <>,
In-reply-to: <>
List-archive: <>
List-help: <>
List-id: linux-mips <>
List-owner: <>
List-post: <>
List-software: Ecartis version 1.0.0
List-subscribe: <>
List-unsubscribe: <>
Original-recipient: rfc822;
References: <> <> <> <>
User-agent: Mutt/1.5.23 (2014-03-12)

On Wed, Jun 04, 2014 at 10:12:39AM -0700, David Daney wrote:
> On 06/04/2014 08:58 AM, Alex Smith wrote:
> >On 04/06/14 15:47, Aaro Koskinen wrote:
> >>On Thu, May 29, 2014 at 11:10:01AM +0100, Alex Smith wrote:
> >>>Add interface mode detection for Octeon II. This is necessary to detect
> >>>the interface modes correctly on the UBNT E200 board. Code is taken
> >>>from the UBNT GPL source release, with some alterations: SRIO, ILK and
> >>>RXAUI interface modes are removed and instead return disabled as these
> >>>modes are not currently supported.
> >>>
> >>>Tested-by: David Daney <>
> >>>Signed-off-by: Alex Smith <>
> >>
> >>I tried booting ebb6800 board with these patches.
> There seem to be problems (I think in the interrupt controller code) for
> cn68xx based systems in the kernel.  So I couldn't get it to boot
> on my ebb6800 even to the point it tries to initialize the networking
> hardware.

I tested two boards with 3.15-rc8. They boot fine to shell, it's just
modprobing octeon-ethernet that is causing issues.

> >>It hangs somewhere in __cvmx_helper_xaui_enable() with XAUI port. Looking
> >>at the UBNT GPL package, xaui init is quite different with 68XX specific
> >>code paths.  Maybe those bits should be added too, or then disable XAUI
> >>support as well?
> >
> >Probably the best thing to do for now would be to disable it. Does it
> >boot successfully for you if you switch CVMX_HELPER_INTERFACE_MODE_XAUI
> >to disabled?
> ... I don't think it matters.  The patch Alex et al. came up with is an
> improvement over what is already there.  The fact that there are still some
> configurations that don't work can be addressed with follow-on patches.

Fair enough; previously octeon-ethernet panic()ed on ebb6800 and now it
silently hangs - so it's not really any serious regression.

If these patches already make it work on ER Pro then good, at least
it's a some kind of improvement towards a proper Octeon II support.


<Prev in Thread] Current Thread [Next in Thread>