From: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
To: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
Cc: Thinh Nguyen <Thinh.Nguyen@synopsys.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Bjorn Andersson <bjorn.andersson@oss.qualcomm.com>,
"linux-arm-msm@vger.kernel.org" <linux-arm-msm@vger.kernel.org>,
"linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 3/4] usb: dwc3: qcom: Facilitate autosuspend during host mode
Date: Mon, 23 Jun 2025 23:59:11 +0000 [thread overview]
Message-ID: <20250623235856.b2jwgf5j6yup2sww@synopsys.com> (raw)
In-Reply-To: <20250610091357.2983085-4-krishna.kurapati@oss.qualcomm.com>
On Tue, Jun 10, 2025, Krishna Kurapati wrote:
> When in host mode, it is intended that the controller goes to suspend
> state to save power and wait for interrupts from connected peripheral
> to wake it up. This is particularly used in cases where a HID or Audio
> device is connected. In such scenarios, the usb controller can enter
> auto suspend and resume action after getting interrupts from the
> connected device.
>
> Allow autosuspend for and xhci device and allow userspace to decide
> whether to enable this functionality.
>
> a) Register to usb-core notifications in set_role vendor callback to
> identify when root hubs are being created. Configure them to
> use_autosuspend.
>
> b) Identify usb core notifications where the HCD is being added and
> enable autosuspend for that particular xhci device.
>
> Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
> ---
> drivers/usb/dwc3/dwc3-qcom.c | 62 ++++++++++++++++++++++++++++++++----
> 1 file changed, 56 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c
> index d40b52e2ba01..17bbd5a06c08 100644
> --- a/drivers/usb/dwc3/dwc3-qcom.c
> +++ b/drivers/usb/dwc3/dwc3-qcom.c
> @@ -95,6 +95,8 @@ struct dwc3_qcom {
> * internally by mutex lock.
> */
> enum usb_role current_role;
> +
> + struct notifier_block xhci_nb;
> };
>
> #define to_dwc3_qcom(d) container_of((d), struct dwc3_qcom, dwc)
> @@ -647,6 +649,39 @@ static int dwc3_qcom_setup_irq(struct dwc3_qcom *qcom, struct platform_device *p
> return 0;
> }
>
> +static int dwc3_xhci_event_notifier(struct notifier_block *nb,
> + unsigned long event, void *ptr)
> +{
> + struct dwc3_qcom *qcom = container_of(nb, struct dwc3_qcom, xhci_nb);
> + struct dwc3 *dwc = &qcom->dwc;
> + struct usb_bus *ubus = ptr;
> + struct usb_hcd *hcd;
> +
> + if (!dwc->xhci)
> + goto done;
> +
> + hcd = platform_get_drvdata(dwc->xhci);
> + if (!hcd)
> + goto done;
> +
> + if (event != USB_BUS_ADD)
> + goto done;
> +
> + if (strcmp(dev_name(ubus->sysdev), dev_name(dwc->sysdev)) != 0)
Can this be false? If possible, I'd like to avoid these pointers and
strcmp here.
> + goto done;
> +
> + if (event == USB_BUS_ADD) {
This condition is redundant when you have the check a few lines above.
> + /*
> + * Identify instant of creation of primary hcd and
> + * mark xhci as autosuspend capable at this point.
> + */
> + pm_runtime_use_autosuspend(&dwc->xhci->dev);
> + }
> +
> +done:
> + return NOTIFY_DONE;
> +}
> +
> static void dwc3_qcom_set_role_notifier(struct dwc3 *dwc, enum usb_role next_role)
> {
> struct dwc3_qcom *qcom = to_dwc3_qcom(dwc);
> @@ -659,12 +694,22 @@ static void dwc3_qcom_set_role_notifier(struct dwc3 *dwc, enum usb_role next_rol
> return;
> }
>
> - if (qcom->current_role == USB_ROLE_DEVICE &&
> - next_role != USB_ROLE_DEVICE)
> + if (qcom->current_role == USB_ROLE_NONE) {
> + if (next_role == USB_ROLE_DEVICE) {
> + dwc3_qcom_vbus_override_enable(qcom, true);
> + } else if (next_role == USB_ROLE_HOST) {
> + qcom->xhci_nb.notifier_call = dwc3_xhci_event_notifier;
> + usb_register_notify(&qcom->xhci_nb);
> + }
> + } else if (qcom->current_role == USB_ROLE_DEVICE &&
> + next_role != USB_ROLE_DEVICE) {
> dwc3_qcom_vbus_override_enable(qcom, false);
> - else if ((qcom->current_role != USB_ROLE_DEVICE) &&
> - (next_role == USB_ROLE_DEVICE))
> - dwc3_qcom_vbus_override_enable(qcom, true);
> + } else if (qcom->current_role == USB_ROLE_HOST) {
> + if (next_role == USB_ROLE_NONE)
> + usb_unregister_notify(&qcom->xhci_nb);
> + else if (next_role == USB_ROLE_DEVICE)
> + dwc3_qcom_vbus_override_enable(qcom, true);
We don't unregister the notifier when switching from host to device?
> + }
>
> pm_runtime_mark_last_busy(qcom->dev);
> pm_runtime_put_sync(qcom->dev);
> @@ -774,6 +819,8 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
>
> if (qcom->mode == USB_DR_MODE_HOST) {
> qcom->current_role = USB_ROLE_HOST;
> + qcom->xhci_nb.notifier_call = dwc3_xhci_event_notifier;
> + usb_register_notify(&qcom->xhci_nb);
> } else if (qcom->mode == USB_DR_MODE_PERIPHERAL) {
> qcom->current_role = USB_ROLE_DEVICE;
> dwc3_qcom_vbus_override_enable(qcom, true);
> @@ -794,7 +841,7 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
> ret = dwc3_core_probe(&probe_data);
> if (ret) {
> ret = dev_err_probe(dev, ret, "failed to register DWC3 Core\n");
> - goto clk_disable;
> + goto unregister_notify;
> }
>
> ret = dwc3_qcom_interconnect_init(qcom);
> @@ -817,6 +864,9 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
> dwc3_qcom_interconnect_exit(qcom);
> remove_core:
> dwc3_core_remove(&qcom->dwc);
> +unregister_notify:
> + if (qcom->mode == USB_DR_MODE_HOST)
> + usb_unregister_notify(&qcom->xhci_nb);
> clk_disable:
> clk_bulk_disable_unprepare(qcom->num_clocks, qcom->clks);
>
> --
> 2.34.1
>
BR,
Thinh
next prev parent reply other threads:[~2025-06-23 23:59 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-10 9:13 [PATCH v2 0/4] usb: dwc3: Modify role-switching QC drd usb controllers Krishna Kurapati
2025-06-10 9:13 ` [PATCH v2 1/4] usb: dwc3: core: Introduce glue callbacks for flattened implementations Krishna Kurapati
2025-06-17 1:29 ` Thinh Nguyen
2025-06-23 23:24 ` Thinh Nguyen
2025-06-24 13:03 ` Krishna Kurapati
2025-06-10 9:13 ` [PATCH v2 2/4] usb: dwc3: qcom: Implement glue callbacks to facilitate runtime suspend Krishna Kurapati
2025-06-10 10:58 ` Dmitry Baryshkov
2025-06-10 16:36 ` Krishna Kurapati
2025-06-10 17:23 ` Dmitry Baryshkov
2025-06-11 14:44 ` Krishna Kurapati
2025-06-23 23:32 ` Thinh Nguyen
2025-06-24 13:04 ` Krishna Kurapati
2025-07-11 23:29 ` Thinh Nguyen
2025-07-13 9:29 ` Krishna Kurapati
2025-07-30 1:23 ` Thinh Nguyen
2025-07-30 2:16 ` Krishna Kurapati
2025-08-01 1:01 ` Thinh Nguyen
2025-08-05 11:18 ` Krishna Kurapati
2025-08-05 23:18 ` Thinh Nguyen
2025-08-06 0:18 ` Krishna Kurapati
2025-08-06 0:50 ` Thinh Nguyen
2025-06-10 9:13 ` [PATCH v2 3/4] usb: dwc3: qcom: Facilitate autosuspend during host mode Krishna Kurapati
2025-06-10 11:00 ` Dmitry Baryshkov
2025-06-10 16:40 ` Krishna Kurapati
2025-06-10 17:24 ` Dmitry Baryshkov
2025-06-23 23:59 ` Thinh Nguyen [this message]
2025-06-24 13:08 ` Krishna Kurapati
2025-06-10 9:13 ` [PATCH v2 4/4] usb: dwc3: qcom: Remove extcon functionality from glue Krishna Kurapati
2025-06-10 11:04 ` Dmitry Baryshkov
2025-06-10 10:01 ` [PATCH v2 0/4] usb: dwc3: Modify role-switching QC drd usb controllers Krishna Kurapati
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250623235856.b2jwgf5j6yup2sww@synopsys.com \
--to=thinh.nguyen@synopsys.com \
--cc=bjorn.andersson@oss.qualcomm.com \
--cc=gregkh@linuxfoundation.org \
--cc=krishna.kurapati@oss.qualcomm.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®