Re: [PATCH 2/2] drm/msm/dsi: Get PHY ref clock from the DT

[Matthias Kaehlcke]

On Tue, Nov 06, 2018 at 03:11:48PM -0800, Stephen Boyd wrote:
> Quoting Matthias Kaehlcke (2018-11-02 14:45:34)
> > @@ -630,7 +632,8 @@ static int pll_10nm_register(struct dsi_pll_10nm *pll_10nm)
> >         char clk_name[32], parent[32], vco_name[32];
> >         char parent2[32], parent3[32], parent4[32];
> >         struct clk_init_data vco_init = {
> > -               .parent_names = (const char *[]){ "xo" },
> > +               .parent_names = (const char *[]){
> > +                       __clk_get_name(pll_10nm->vco_ref_clk) },
> 
> I find this syntax odd, in addition to needing to check for NULL here as
> Sean pointed out. Preferably just have it be the address of the
> character pointer instead of making an anonymous array and then casting
> that inline, i.e
> 
> 		.parent_names = &ref_clk_name,

Ok

I'm not convinced the check for NULL is needed though, see my reply to Sean.

> >                 .num_parents = 1,
> >                 .name = vco_name,
> >                 .flags = CLK_IGNORE_UNUSED,
> > @@ -786,6 +789,12 @@ struct msm_dsi_pll *msm_dsi_pll_10nm_init(struct platform_device *pdev, int id)
> >         pll_10nm->id = id;
> >         pll_10nm_list[id] = pll_10nm;
> >  
> > +       pll_10nm->vco_ref_clk = devm_clk_get(&pdev->dev, "ref");
> > +       if (IS_ERR(pll_10nm->vco_ref_clk)) {
> > +               dev_err(&pdev->dev, "couldn't get 'ref' clock\n");
> 
> This might be because of probe defer, which may be annoying to see this
> failure many times.

Ok, will skip the logging for -EPROBE_DEFER

Cheers

Matthias