[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <72BC0C8BD7BB6F45988A99382E5FBAE544490C96@hhmail02.hh.imgtec.org>
Date: Tue, 7 Apr 2015 17:16:55 +0000
From: James Hartley <James.Hartley@...tec.com>
To: Andrew Bresticker <abrestic@...omium.org>,
Arnd Bergmann <arnd@...db.de>
CC: "David S. Miller" <davem@...emloft.net>,
Giuseppe Cavallaro <peppe.cavallaro@...com>,
"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Govindraj Raja <Govindraj.Raja@...tec.com>,
Damien Horsley <Damien.Horsley@...tec.com>
Subject: RE: [PATCH V2 2/2] stmmac: Add IMG Pistachio platform glue layer
> -----Original Message-----
> From: abrestic@...gle.com [mailto:abrestic@...gle.com] On Behalf Of
> Andrew Bresticker
> Sent: 06 April 2015 23:28
> To: Arnd Bergmann; James Hartley
> Cc: David S. Miller; Giuseppe Cavallaro; devicetree@...r.kernel.org;
> netdev@...r.kernel.org; linux-kernel@...r.kernel.org; Govindraj Raja
> Subject: Re: [PATCH V2 2/2] stmmac: Add IMG Pistachio platform glue layer
>
> On Mon, Apr 6, 2015 at 3:10 PM, Arnd Bergmann <arnd@...db.de> wrote:
> > On Monday 06 April 2015 14:42:38 Andrew Bresticker wrote:
> >> At the moment, the only additional setup required for the DWMAC on
> >> the IMG Pistachio SoC is to request and enable a separate gate clock
> >> for the register interface.
> >>
> >> Signed-off-by: Andrew Bresticker <abrestic@...omium.org>
> >> Signed-off-by: Govindraj Raja <govindraj.raja@...tec.com>
> >> Cc: James Hartley <james.hartley@...tec.com>
> >
> > Why do you need a special glue driver for that?
> >
> > Since you don't do anything special, I'd say it should be possible to
> > handle this with just the default driver.
>
> Right, at the moment we only need to request and enable a second clock,
> which could be added to the core driver. If we don't expect to have to
> extend this driver to deal with other configurations (which IIRC there were
> some concerns about when I wrote this), then I'm fine with folding this change
> into the core driver - James?
Weren't the concerns about USB not Ethernet? I don't remember anything that would stop us extending the driver for eth.
>
> Thanks,
> Andrew
James.
Powered by blists - more mailing lists