RE: [PATCH v4 2/3] pinctrl: renesas: Renesas RZ/N1 pinctrl driver

From: Phil Edworthy
Date: Mon Sep 24 2018 - 08:33:06 EST


Hi Geert,

On 24 September 2018 12:59 Geert Uytterhoeven wrote:
> On Wed, Sep 19, 2018 at 4:24 PM Phil Edworthy wrote:
> > This provides a pinctrl driver for the Renesas RZ/N1 device family.
> >
> > Based on a patch originally written by Michel Pollet at Renesas.
> >
> > Signed-off-by: Phil Edworthy <phil.edworthy@xxxxxxxxxxx>
>
> Thanks for your patch!
>
> > --- /dev/null
> > +++ b/drivers/pinctrl/pinctrl-rzn1.c
>
> > +/*
> > + * Structure detailing the HW registers on the RZ/N1 devices.
> > + * Both the Level 1 mux registers and Level 2 mux registers have the
> > +same
> > + * structure. The only difference is that Level 2 has additional MDIO
> > +registers
> > + * at the end.
> > + */
> > +struct rzn1_pinctrl_regs {
> > + union {
> > + u32 conf[170];
> > + u8 pad0[0x400];
>
> This looks a bit confusing, and isn't really padding, as you use a union.
> What about getting rid of the union, and making it real padding?
>
> u32 conf[170];
> u32 pad0[86];
>
> > + };
> > + u32 status_protect; /* 0x400 */
> > + /* MDIO mux registers, level2 only */
> > + u32 l2_mdio[2];
> > +};
>
> BTW, while using a struct instead of register offset definitions has its merits,
> it also has its drawbacks, like the need for the "0x400" comment.
> You don't have to change it, though.
>
> > +static const struct rzn1_pin_group *rzn1_pinctrl_find_group_by_name(
> > + const struct rzn1_pinctrl *ipctl, const char *name) {
> > + const struct rzn1_pin_group *grp = NULL;
> > + int i;
>
> unsigned int i;
> (rzn1_pinctrl.ngroups is unsigned int)
>
> > +
> > + for (i = 0; i < ipctl->ngroups; i++) {
> > + if (!strcmp(ipctl->groups[i].name, name)) {
> > + grp = &ipctl->groups[i];
> > + break;
> > + }
> > + }
> > +
> > + return grp;
> > +}
>
> > +static int rzn1_pinconf_set(struct pinctrl_dev *pctldev, unsigned int pin,
> > + unsigned long *configs, unsigned int
> > +num_configs) {
> > + struct rzn1_pinctrl *ipctl = pinctrl_dev_get_drvdata(pctldev);
> > + enum pin_config_param param;
> > + int i;
>
> unsigned int i;
>
> > + u32 arg;
> > + u32 l1, l1_cache;
> > + u32 drv;
> > +
> > + if (pin >= ARRAY_SIZE(ipctl->lev1->conf))
> > + return -EINVAL;
> > +
> > + l1 = readl(&ipctl->lev1->conf[pin]);
> > + l1_cache = l1;
> > +
> > + for (i = 0; i < num_configs; i++) {
>
> > +static int rzn1_pinconf_group_get(struct pinctrl_dev *pctldev,
> > + unsigned int selector,
> > + unsigned long *config) {
> > + struct rzn1_pinctrl *ipctl = pinctrl_dev_get_drvdata(pctldev);
> > + struct rzn1_pin_group *grp = &ipctl->groups[selector];
> > + unsigned long old = 0;
> > + int i;
>
> unsigned int i;
>
> > +
> > + dev_dbg(ipctl->dev, "group get %s selector:%d\n", grp->name,
> > + selector);
>
> %u to format unsigned int.
>
> > +
> > + for (i = 0; i < grp->npins; i++) {
>
> > +static int rzn1_pinconf_group_set(struct pinctrl_dev *pctldev,
> > + unsigned int selector,
> > + unsigned long *configs,
> > + unsigned int num_configs) {
> > + struct rzn1_pinctrl *ipctl = pinctrl_dev_get_drvdata(pctldev);
> > + struct rzn1_pin_group *grp = &ipctl->groups[selector];
> > + int ret, i;
>
> unsigned int i;
>
> > +
> > + dev_dbg(ipctl->dev, "group set %s selector:%d
> > + configs:%p/%d\n",
>
> %u
>
> > + grp->name, selector, configs, num_configs);
> > +
> > + for (i = 0; i < grp->npins; i++) {
> > + unsigned int pin = grp->pins[i];
> > +
> > + ret = rzn1_pinconf_set(pctldev, pin, configs, num_configs);
> > + if (ret)
> > + return ret;
> > + }
> > +
> > + return 0;
> > +}
>
>
> > +static int rzn1_pinctrl_parse_functions(struct device_node *np,
> > + struct rzn1_pinctrl *ipctl,
> > +u32 index)
>
> Why u32 instead of plain unsigned int?
>
> > +{
> > + struct device_node *child;
> > + struct rzn1_pmx_func *func;
> > + struct rzn1_pin_group *grp;
> > + u32 i = 0;
>
> Why not plain unsigned int?
>
> > + int ret;
> > +
> > + func = &ipctl->functions[index];
> > +
> > + /* Initialise function */
> > + func->name = np->name;
> > + func->num_groups = rzn1_pinctrl_count_function_groups(np);
> > + if (func->num_groups == 0) {
> > + dev_err(ipctl->dev, "no groups defined in %pOF\n", np);
> > + return -EINVAL;
> > + }
> > + dev_dbg(ipctl->dev, "function %s has %d groups\n",
> > + np->name, func->num_groups);
> > +
> > + func->groups = devm_kmalloc_array(ipctl->dev,
> > + func->num_groups, sizeof(char *),
> > + GFP_KERNEL);
> > + if (!func->groups)
> > + return -ENOMEM;
> > +
> > + if (of_property_count_u32_elems(np, RZN1_PINS_PROP) > 0) {
> > + func->groups[i] = np->name;
> > + grp = &ipctl->groups[ipctl->ngroups];
> > + grp->func = func->name;
> > + ret = rzn1_pinctrl_parse_groups(np, grp, ipctl);
> > + if (ret < 0)
> > + return ret;
> > + i++;
> > + ipctl->ngroups++;
> > + }
> > +
> > + for_each_child_of_node(np, child) {
> > + func->groups[i] = child->name;
> > + grp = &ipctl->groups[ipctl->ngroups];
> > + grp->func = func->name;
> > + ret = rzn1_pinctrl_parse_groups(child, grp, ipctl);
> > + if (ret < 0)
> > + return ret;
> > + i++;
> > + ipctl->ngroups++;
> > + }
> > +
> > + dev_dbg(ipctl->dev, "function %s parsed %d/%d groups\n",
>
> %u/%u
>
> > + np->name, i, func->num_groups);
> > +
> > + return 0;
> > +}
> > +
> > +static int rzn1_pinctrl_probe_dt(struct platform_device *pdev,
> > + struct rzn1_pinctrl *ipctl) {
> > + struct device_node *np = pdev->dev.of_node;
> > + struct device_node *child;
> > + unsigned int maxgroups = 0;
> > + u32 nfuncs = 0;
> > + u32 i = 0;
>
> Why not plain unsigned int, for both?

Thanks for your review and comments, you are right in all of them.
I'll fix them and re-post.

Thanks
Phil

> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-
> m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds