linux-kernel - Re: [PATCH V4 3/6] soc: qcom: eud: Add driver support for Embedded USB Debugger(EUD)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <bf41cb03-a836-11ed-c203-d78d420eca99@quicinc.com>
Date:   Mon, 31 Jan 2022 11:19:31 +0530
From:   Souradeep Chowdhury <quic_schowdhu@...cinc.com>
To:     Bjorn Andersson <bjorn.andersson@...aro.org>
CC:     <linux-arm-msm@...r.kernel.org>, <linux-usb@...r.kernel.org>,
        <devicetree@...r.kernel.org>, <pure.logic@...us-software.ie>,
        <greg@...ah.com>, <robh@...nel.org>,
        <linux-kernel@...r.kernel.org>, <quic_tsoni@...cinc.com>,
        <quic_psodagud@...cinc.com>, <quic_satyap@...cinc.com>,
        <quic_pheragu@...cinc.com>, <quic_rjendra@...cinc.com>,
        <quic_sibis@...cinc.com>, <quic_saipraka@...cinc.com>
Subject: Re: [PATCH V4 3/6] soc: qcom: eud: Add driver support for Embedded
 USB Debugger(EUD)


On 1/28/2022 9:41 PM, Bjorn Andersson wrote:
> On Thu 27 Jan 04:01 PST 2022, Souradeep Chowdhury wrote:
>
>> On 1/26/2022 10:17 AM, Bjorn Andersson wrote:
>>> On Fri 21 Jan 07:53 CST 2022, Souradeep Chowdhury wrote:
> [..]
>>>> +		return PTR_ERR(chip->base);
>>>> +
>>>> +	chip->mode_mgr = devm_platform_ioremap_resource(pdev, 1);
>>>> +	if (IS_ERR(chip->mode_mgr))
>>>> +		return PTR_ERR(chip->mode_mgr);
>>>> +
>>>> +	chip->irq = platform_get_irq(pdev, 0);
>>>> +	ret = devm_request_threaded_irq(&pdev->dev, chip->irq, handle_eud_irq,
>>>> +			handle_eud_irq_thread, IRQF_ONESHOT, NULL, chip);
>>>> +	if (ret)
>>>> +		return dev_err_probe(chip->dev, ret, "failed to allocate irq\n");
>>>> +
>>>> +	enable_irq_wake(chip->irq);
>>>> +
>>>> +	platform_set_drvdata(pdev, chip);
>>>> +
>>>> +	return 0;
>>> Per the updated binding, the EUD would now be a usb-role-switch as well
>>> and when not enabled should simply propagate the incoming requests. So I
>>> was expecting this to register as a usb_role_switch as well...
>> Can you please elaborate on this?
>>
>> Do I need to define a separate 'usb_role_switch_desc' here and register
>> using 'usb_role_switch_register'?
>>
>> Also what should be the set method in this case for usb_role_switch_desc?
>>
> My expectation is that in normal operation pmic_glink will provide role
> switching requests and then as you enable the EUD it will force the role
> to gadget.
>
> So my suggestion was that you make eud a role-switch and as long as EUD
> is disabled you just pass through the role-switch vote from pmic_glink
> onto the dwc3.
>
> Perhaps I'm misunderstanding how this is really working.

As per the code, the role-switching on dwc3 is enabled from 
drivers/usb/dwc3/drd.c on setting the usb mode to "otg " ,

that is being done on the 5th patch of this series. There is no other 
entity that is switching role for dwc3 other than EUD.

Also the role-switch registration happens from drivers/usb/dwc3/drd.c as 
follows

static int dwc3_setup_role_switch(struct dwc3 *dwc)
{
         struct usb_role_switch_desc dwc3_role_switch = {NULL};
         u32 mode;

         dwc->role_switch_default_mode = 
usb_get_role_switch_default_mode(dwc->dev);
         if (dwc->role_switch_default_mode == USB_DR_MODE_HOST) {
                 mode = DWC3_GCTL_PRTCAP_HOST;
         } else {
                 dwc->role_switch_default_mode = USB_DR_MODE_PERIPHERAL;
                 mode = DWC3_GCTL_PRTCAP_DEVICE;
         }

         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;
         dwc3_role_switch.driver_data = dwc;
         dwc->role_sw = usb_role_switch_register(dwc->dev, 
&dwc3_role_switch);
         if (IS_ERR(dwc->role_sw))
                 return PTR_ERR(dwc->role_sw);

         dwc3_set_mode(dwc, mode);
         return 0;
}

Can you please point out the pmic_glink driver upstream and how it is 
related in this case?

>
> Regards,
> Bjorn