[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110317100010.GF29758@n2100.arm.linux.org.uk>
Date: Thu, 17 Mar 2011 10:00:10 +0000
From: Russell King - ARM Linux <linux@....linux.org.uk>
To: Andrew Victor <avictor.za@...il.com>
Cc: nicolas.ferre@...el.com, Jamie Iles <jamie@...ieiles.com>,
plagnioj@...osoft.com, linux-arm-kernel@...ts.infradead.org,
netdev@...r.kernel.org
Subject: Re: [PATCHv2 1/9] at91: provide macb clks with "pclk" and "hclk"
name
On Thu, Mar 17, 2011 at 11:22:37AM +0200, Andrew Victor wrote:
> hi Russell,
>
> >> There is no reference to a "pclk" or "hclk" in the AT91 architecture.
> >> So to avoid possible confusion, maybe create two "fake" clocks both
> >> parented to "macb_clk", and add a comment they're only for
> >> compatibility with the AVR32.
> >
> > It doesn't matter what's in the documentation.
> >
> > What matters more than conforming to documentation is keeping the drivers
> > in a clean and maintainable state without throwing lots of ifdefs into
> > them.
>
> I'm not saying the drivers need ifdefs, they should request both
> "pclk" and "hclk" as suggested.
>
> What I was suggesting is the platform clock setup on AT91 as:
> macb_clk
> |
> +-- hclk
> +-- pclk
>
> rather than:
> pclk
> |
> +-- hclk
And what I've been saying all along is to make pclk a _dummy_ clock on
the platform it doesn't exist for, rather than making it related in some
way to another clock given to the peripheral.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists