On Tue, Jun 07, 2016 at 04:12:29PM +0200, Heiko Stübner wrote:
> Hi Guenter,
>
> Am Dienstag, 7. Juni 2016, 06:19:45 schrieb Guenter Roeck:
> > On Tue, Jun 7, 2016 at 2:54 AM, Heiko Stübner <heiko@xxxxxxxxx> wrote:
> > > Hi Frank,
> > >
> > > Am Montag, 6. Juni 2016, 17:20:04 schrieb Frank Wang:
> > >> The newer SoCs (rk3366, rk3399) take a different usb-phy IP block
> > >> than rk3288 and before, and most of phy-related registers are also
> > >> different from the past, so a new phy driver is required necessarily.
> > >>
> > >> Signed-off-by: Frank Wang <frank.wang@xxxxxxxxxxxxxx>
> > >> ---
> > >
> > > [...]
> > >
> > >> +static int rockchip_usb2phy_probe(struct platform_device *pdev)
> > >> +{
> > >> + struct device *dev = &pdev->dev;
> > >> + struct device_node *np = dev->of_node;
> > >> + struct device_node *child_np;
> > >> + struct phy_provider *provider;
> > >> + struct rockchip_usb2phy *rphy;
> > >> + const struct of_device_id *match;
> > >> + int index, ret;
> > >> +
> > >> + rphy = devm_kzalloc(dev, sizeof(*rphy), GFP_KERNEL);
> > >> + if (!rphy)
> > >> + return -ENOMEM;
> > >> +
> > >> + match = of_match_device(dev->driver->of_match_table, dev);
> > >> + if (!match || !match->data) {
> > >> + dev_err(dev, "phy configs are not assigned!\n");
> > >> + return -EINVAL;
> > >> + }
> > >> +
> > >> + if (!dev->parent || !dev->parent->of_node)
> > >> + return -EINVAL;
> > >> +
> > >> + rphy->grf = syscon_node_to_regmap(dev->parent->of_node);
> > >> + if (IS_ERR(rphy->grf))
> > >> + return PTR_ERR(rphy->grf);
> > >> +
> > >> + rphy->dev = dev;
> > >> + rphy->phy_cfg = match->data;
> > >> + platform_set_drvdata(pdev, rphy);
> > >> +
> > >> + ret = rockchip_usb2phy_clk480m_register(rphy);
> > >> + if (ret) {
> > >> + dev_err(dev, "failed to register 480m output clock\n");
> > >> + return ret;
> > >> + }
> > >> +
> > >> + rphy->vbus_host = devm_regulator_get_optional(dev, "vbus_host");
> > >> + if (IS_ERR(rphy->vbus_host)) {
> > >> + ret = PTR_ERR(rphy->vbus_host);
> > >> + if (ret == -EPROBE_DEFER)
> > >> + return ret;
> > >> +
> > >> + dev_info(dev, "vbus_host regulator is not assigned!\n");
> > >> + rphy->vbus_host = NULL;
> > >> + } else {
> > >> + ret = regulator_enable(rphy->vbus_host);
> > >> + if (ret)
> > >> + return ret;
> > >> + }
> > >> +
> > >> + index = 0;
> > >> + for_each_available_child_of_node(np, child_np) {
> > >> + struct rockchip_usb2phy_port *rport = &rphy->ports[index];
> > >> + struct phy *phy;
> > >> +
> > >> + phy = devm_phy_create(dev, child_np,
> > >> &rockchip_usb2phy_ops);
> > >> + if (IS_ERR(phy)) {
> > >> + dev_err(dev, "failed to create phy\n");
> > >> + ret = PTR_ERR(phy);
> > >> + goto put_child;
> > >> + }
> > >> +
> > >> + rport->phy = phy;
> > >> +
> > >> + /* initialize otg/host port separately */
> > >> + if (!of_node_cmp(child_np->name, "host-port")) {
> > >> + ret = rockchip_usb2phy_host_port_init(rphy, rport,
> > >> + child_np);
> > >> + if (ret)
> > >> + goto put_child;
> > >> + }
> > >> +
> > >> + phy_set_drvdata(rport->phy, rport);
> > >> +
> > >> + /* to prevent out of boundary */
> > >> + if (++index >= rphy->phy_cfg->num_ports)
> > >> + break;
> > >> + }
> > >> +
> > >> + provider = devm_of_phy_provider_register(dev, of_phy_simple_xlate);
> > >> + return PTR_ERR_OR_ZERO(provider);
> > >> +
> > >> +put_child:
> > >> + of_node_put(child_np);
> > >> + return ret;
> > >> +}
> > >> +
> > >> +static const struct rockchip_usb2phy_cfg rk3366_phy_cfgs = {
> > >> + .num_ports = 2,
> > >> + .clkout_ctl = { 0x0724, 15, 15, 1, 0 },
> > >> + .port_cfgs = {
> > >> + [USB2PHY_PORT_HOST] = {
> > >> + .phy_sus = { 0x0728, 15, 0, 0, 0x1d1 },
> > >> + .ls_det_en = { 0x0680, 4, 4, 0, 1 },
> > >> + .ls_det_st = { 0x0690, 4, 4, 0, 1 },
> > >> + .ls_det_clr = { 0x06a0, 4, 4, 0, 1 },
> > >> + .utmi_ls = { 0x049c, 14, 13, 0, 1 },
> > >> + .utmi_hstdet = { 0x049c, 12, 12, 0, 1 }
> > >> + }
> > >> + },
> > >> +};
> > >
> > > I also realized that the rk3399 has two of those phy-blocks, so I think
> > > we'll need to adapt your data storage mechanism a tiny bit. Maybe
> > > something like the following:
> > >
> > > In the dts:
> > >
> > > grf: syscon@ff770000 {
> > >
> > > compatible = "rockchip,rk3366-grf", "syscon", "simple-mfd";
> > >
> > > ...
> > >
> > > u2phy: usb2-phy@700 {
> > >
> > > compatible = "rockchip,rk3366-usb2phy";
> > >
> > > reg = <0x700 0x2c>;
> > > ...
> > >
> > > };
> > >
> > > };
> > >
> > > In the driver
> > >
> > > static const struct rockchip_usb2phy_cfg rk3366_phy_cfgs[] = {
> > >
> > > {
> > >
> > > .reg = 0x700,
> > > .num_ports = 2,
> > > .clkout_ctl = { 0x0724, 15, 15, 1, 0 },
> > > .port_cfgs = {
> > >
> > > [USB2PHY_PORT_HOST] = {
> > >
> > > .phy_sus = { 0x0728, 15, 0, 0,
> > > 0x1d1 },
> > > ...
> > >
> > > }
> > >
> > > },
> > >
> > > },
> > >
> > > };
> > >
> > > and in _probe then simply use the correct array entry matching the dts reg
> > > property.
> > >
> > > On the rk3399 this would then probably be
> > >
> > > u2phy0: phy@e450 {
> > >
> > > compatible = "rockchip,rk3366-usb2phy";
> > >
> > > reg = <0xe450 0x10>;
> > > ...
> > >
> > > };
> > >
> > > u2phy1: phy@e460 {
> > >
> > > compatible = "rockchip,rk3366-usb2phy";
> > >
> > > reg = <0xe460 0x10>;
> > > ...
> > >
> > > };
> > >
> > > This is slightly similar to what we do with the emmc-phy. But also as the
> > > usbphys differ in a lot of way, this is really only used to match to the
> > > data block.
> > >
> > > This only affects the match-data selection and the rest of the driver does
> > > not need to be changed for this.
> >
> > Would this replace the host-port property ?
>
> nope ... each phy-block has two ports (host + otg). Franks rk3366 has one phy-
> block while the rk3399 has two phy blocks (with two ports on each).
>
Ah, now I understand. Yes, I think you are right, it would make sense
to add this now.
Guenter
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
