Hi Chris,
On Sun, Aug 21, 2016 at 9:52 PM, Chris <zywang33 at 126.com> wrote:
> Hi Guenter
>
[ ... ]
>>> +static int rockchip_usb3_phy_power_on(struct phy *phy)
>>> +{
>>> + struct rockchip_typec_phy *tcphy = phy_get_drvdata(phy);
>>> + struct rockchip_usb3phy_port_cfg *cfg = &tcphy->port_cfgs;
>>> + const struct usb3phy_reg *reg = &cfg->pipe_status;
>>> + int timeout, new_mode, ret = 0;
>>> + u32 val;
>>> +
>>> + mutex_lock(&tcphy->lock);
>>> +
>>> + new_mode = tcphy_get_mode(tcphy);
>>> + if (new_mode < 0) {
>>> + ret = new_mode;
>>> + goto unlock_ret;
>>> + }
>>> +
>>> + /* DP-only mode; fall back to USB2 */
>>> + if (!(new_mode & (MODE_DFP_USB | MODE_UFP_USB)))
>>> + goto unlock_ret;
>>> +
>>
>> The rest of the code (calling tcphy_phy_init(), waiting for pipe
>> ready) is not needed in this case ?
>
>
> If there is only DP, the tcphy_phy_init will be called by DP controller, and
> the pipe is for USB3, so do not need do anything, just return a "fake"
> success here.
> And the usb phy power off do nothing. Just make USB controller happier.
>
Yes, you are right. It is a bit confusing (and I had overlooked) that
MODE_DFP_USB is set for both USB2 and USB3 if DP isn't active, but if
DP is active it is only set for USB3.
Guenter
