Re: [PATCH 11/18] omap2: convert clock.c to use new PRCM functions and symbolic constants

[Tony Lindgren <tony@atomide.com>]

* Igor Stoppa <igor.stoppa@nokia.com> [070525 02:04]:
> Sorry, I accidentally pressed send before completing 
> 
> 
> On Fri, 2007-05-25 at 11:59 +0300, Igor Stoppa wrote:
> > Hi,
> > 
> > On Fri, 2007-05-25 at 02:23 -0600, ext Paul Walmsley wrote:
> > > 
> > >  	long long dpll_clk;
> > >  	int dpll_mult, dpll_div, amult;
> > > +	u32 dpll;
> > > +
> > > +	dpll = cm_read_mod_reg(PLL_MOD, CM_CLKSEL1);
> > > +	dpll_mult = dpll & OMAP24XX_DPLL_MULT_MASK;
> > > +	dpll_mult >>= OMAP24XX_DPLL_MULT_SHIFT;		/* 10 bits */
> > > +	dpll_div = dpll & OMAP24XX_DPLL_DIV_MASK;
> > > +	dpll_div >>= OMAP24XX_DPLL_DIV_SHIFT;		/* 4 bits */
> > 
> > 
> > 
> 
> Why not this?
> 
> #define OMAP24XX_DPLL_MULT_LEN	10
> #define OMAP24XX_DPLL_MULT_OFFS	12
> 
> #define get_bitfield(value, name)\
> (((value) >> name##_OFFS) & ((1 << name##_LEN) - 1))
> 
> dpll = cm_read_mod_reg(PLL_MOD, CM_CLKSEL1);
> dpll_mult = get_bitfield(dpll, OMAP24XX_DPLL_MULT);
> dpll_div = get_bitfield(dpll, OMAP24XX_DPLL_DIV);

That could be cleaned up further, but I'll push Paul's patches
today. We can always do additional clean-up, but let's first
wait and make sure things still work for all omap2 boards.

Regards,

Tony