Re: [PATCH] clk: tegra: Fix PLLE programming

[Peter De Schrijver]

On Mon, Mar 31, 2014 at 09:46:22PM +0200, Thierry Reding wrote:
> * PGP Signed by an unknown key
> 
> On Mon, Mar 31, 2014 at 10:51:38AM -0600, Stephen Warren wrote:
> > On 03/31/2014 08:45 AM, Thierry Reding wrote:
> > > From: Thierry Reding <treding@xxxxxxxxxx>
> > > 
> > > PLLE has M, N and P divider shift and width parameters that differ from
> > > the defaults. Furthermore, when clearing the M, N and P divider fields
> > > the corresponding masks were never shifted, thereby clearing only the
> > > lowest bits of the register. This lead to a situation where the PLLE
> > > programming would only work if the register hadn't been touched before.
> > 
> > > diff --git a/drivers/clk/tegra/clk-pll.c b/drivers/clk/tegra/clk-pll.c
> > 
> > > @@ -730,8 +730,10 @@ static int clk_plle_enable(struct clk_hw *hw)
> > >  	if (pll->params->flags & TEGRA_PLLE_CONFIGURE) {
> > >  		/* configure dividers */
> > >  		val = pll_readl_base(pll);
> > > -		val &= ~(divm_mask(pll) | divn_mask(pll) | divp_mask(pll));
> > > -		val &= ~(PLLE_BASE_DIVCML_WIDTH << PLLE_BASE_DIVCML_SHIFT);
> > > +		val &= ~(divp_mask(pll) << PLLE_BASE_DIVP_SHIFT |
> > > +			 divn_mask(pll) << PLLE_BASE_DIVN_SHIFT |
> > > +			 divm_mask(pll) << PLLE_BASE_DIVM_SHIFT);
> > 
> > Shouldn't those shift values also be a macro/inline like
> > divm_shift(pll), since ...
> > 
> > > +static struct div_nmp pll_e_nmp = {
> > > +	.divn_shift = PLLE_BASE_DIVN_SHIFT,
> > > +	.divn_width = PLLE_BASE_DIVN_WIDTH,
> > > +	.divm_shift = PLLE_BASE_DIVM_SHIFT,
> > > +	.divm_width = PLLE_BASE_DIVM_WIDTH,
> > > +	.divp_shift = PLLE_BASE_DIVP_SHIFT,
> > > +	.divp_width = PLLE_BASE_DIVP_WIDTH,
> > > +};
> > 
> > ... that table contains parameters for both width and shift values, not
> > just width values?
> 
> Yes, that could be done. At some point I was also thinking about simply
> converting the masks to be shifted masks, but then realized that the
> patche would become rather intrusive.
> 
> Perhaps, though, to make the code slightly more terse, in addition to
> div{n,m,p}_shift() macros I could add div{n,m,p}_mask_shifted() macros
> as well to combine both.
> 

Problem is that there are 2 shift values in case the PLL has an override bit.
The corresponding PMC register is not always layed out in the same way than
the CAR register.

Cheers,

Peter.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/