From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chen Yu Subject: Re: [PATCH v2 07/10] usb: dwc3: Registering a role switch in the DRD code. Date: Tue, 19 Feb 2019 11:37:33 +0800 Message-ID: <39e822a2-54fb-f4ac-4471-ed8312acf53e@huawei.com> References: <20190218112310.17860-1-chenyu56@huawei.com> <20190218112310.17860-8-chenyu56@huawei.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Jun Li Cc: liuyu712@hisilicon.com, linux-usb@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, john.stultz@linaro.org, suzhuangluan@hisilicon.com, kongfei@hisilicon.com, wanghu17@hisilicon.com, butao@hisilicon.com, chenyao11@huawei.com, fangshengzhou@hisilicon.com, lipengcheng8@huawei.com, songxiaowei@hisilicon.com, xuyiping@hisilicon.com, xuyoujun4@huawei.com, yudongbin@hisilicon.com, zangleigang@hisilicon.com, Felipe Balbi , Greg Kroah-Hartman , Heikki Krogerus , Li Jun List-Id: devicetree@vger.kernel.org Hi, On 2019/2/19 10:57, Jun Li wrote: >> >> +static int dwc3_usb_role_switch_set(struct device *dev, enum usb_role role) >> +{ >> + u32 mode; >> + >> + switch (role) { >> + case USB_ROLE_HOST: >> + mode = DWC3_GCTL_PRTCAP_HOST; >> + break; >> + case USB_ROLE_DEVICE: >> + mode = DWC3_GCTL_PRTCAP_DEVICE; >> + break; >> + default: >> + mode = DWC3_GCTL_PRTCAP_HOST; > > I'd propose let dwc3 stay at device mode at default case. > Setting mode to DWC3_GCTL_PRTCAP_HOST at default case due to the requirement of Hikey960 board. When there is no cable connected to type-C port, dwc3 should be set to HOST mode and the hub on the Hikey960 will be power on. Here is the details of the Hikey960: https://www.96boards.org/documentation/consumer/hikey/hikey960/hardware-docs/hardware-user-manual.md.html But I will check the code to see if it is possible to make dwc3 set to device mode at default case. >> + break; >> + }; >> + >> + dwc3_set_mode(dev_get_drvdata(dev), mode); >> + return 0; >> +} >> + >> +static enum usb_role dwc3_usb_role_switch_get(struct device *dev) >> +{ >> + struct dwc3 *dwc = dev_get_drvdata(dev); >> + unsigned long flags; >> + enum usb_role role; >> + >> + spin_lock_irqsave(&dwc->lock, flags); >> + role = dwc->current_otg_role; >> + spin_unlock_irqrestore(&dwc->lock, flags); >> + >> + return role; >> +} >> + >> int dwc3_drd_init(struct dwc3 *dwc) >> { >> + struct usb_role_switch_desc dwc3_role_switch = {0}; >> int ret, irq; >> >> dwc->edev = dwc3_get_extcon(dwc); >> @@ -522,6 +556,15 @@ int dwc3_drd_init(struct dwc3 *dwc) >> dwc3_set_mode(dwc, DWC3_GCTL_PRTCAP_OTG); >> } >> >> + dwc3_role_switch.fwnode = dev_fwnode(dwc->dev); >> + dwc3_role_switch.set = dwc3_usb_role_switch_set; >> + dwc3_role_switch.get = dwc3_usb_role_switch_get; >> + dwc->role_sw = usb_role_switch_register(dwc->dev, &dwc3_role_switch); >> + if (ret) { >> + dwc3_drd_exit(dwc); >> + return PTR_ERR(dwc->role_sw); >> + } >> + > > My understanding is, if you use usb role switch, you don't need either edev or > OTG block event, so this should be like: > > if (device_property_read_bool(dwc->dev, "usb-role-switch")) { > usb role switch register; > } else if (edev) { > ... > } else { > ... > } Yes. Thanks! > > Jun >> return 0; >> } >> >> @@ -557,4 +600,6 @@ void dwc3_drd_exit(struct dwc3 *dwc) >> >> if (!dwc->edev) >> free_irq(dwc->otg_irq, dwc); >> + >> + usb_role_switch_unregister(dwc->role_sw); >> } >> -- >> 2.15.0-rc2 >> > > . > Thanks Yu Chen