On Mon, Oct 08, 2012 at 11:27:06AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 11:03 Mon 08 Oct , Johannes Stezenbach wrote:
> > On Sun, Oct 07, 2012 at 03:03:03PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > allow the driver to request it's clock and enable it
> > ...
> > > + clk = clk_get(dev, NULL);
> > > + if (IS_ERR(clk)) {
> > > + ret = PTR_ERR(clk);
> > > + dev_err(dev, "no clk ret = %d\n", ret);
> > > + return ret;
> > > + }
> >
> > Just a minor nit: s/allow/require/. Users of the driver may
> > now have to define a dummy clock. Since the actual clock freqeuency
> > is not used by the driver you could also make it optional?
> >
> > And while at it, it's "its", not "it's". :-)
> srry but there is no user at hte mainline and I'm preparing the ST support
> where I nned to enabld the clock and this IP require 2 clock one for the gmac
> and one for phy
>
> so no please provide a clock on generic drivers I do want to use the clk
> framework
That is not the issue I tried to point out. The problem is that
your commit description is inaccurate.
> btw which SoC do you use because I see no code for it
SoC that doesn't exist yet (in development).
Johannes
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
