Re: [PATCH] USB: Fix the issue of task recovery failure caused by USB status when S4 wakes up

[Alan Stern]

[Please make sure that the lines in your email message don't extend 
beyond 76 columns or so.]

Lots of things here seem to be wrong.

On Fri, Sep 06, 2024 at 11:05:48AM +0800, Duan Chenghao wrote:
> When a device is inserted into the USB port and an S4 wakeup is initiated,

There is no such thing as an S4 wakeup.  Do you mean wakeup from an S4 
suspend state?

> after the USB-hub initialization is completed, it will automatically enter suspend mode.

What will enter suspend mode?  The hub that the device was plugged into?
That should not happen.  The hub initialization code should detect that 
a new device was plugged in and prevent the hub from suspending.

> Upon detecting a device on the USB port, it will proceed with resume and set the hcd to the HCD_FLAG_WAKEUP_PENDING state.

HCD_FLAG_WAKEUP_PENDING is not a state.  It is a flag.

> During the S4 wakeup process, peripherals are put into suspend mode, followed by task recovery.

What do you mean by "task recovery"?  We don't need to recover any 
tasks.

What do you mean by "peripherals are put into suspend mode"?  That's not 
what happens.  Peripherals are set back to full power.

> However, upon detecting that the hcd is in the HCD_FLAG_WAKEUP_PENDING state,
> it will return an EBUSY status, causing the S4 suspend to fail and subsequent task recovery to not proceed.

What will return an EBUSY status?

Why do you say that S4 suspend will fail?  Aren't you talking about S4 
wakeup?

Can you provide a kernel log that explains these points and shows what 
problem you are trying to solve?

> This patch makes two modifications in total:
> 1. The set_bit and clean_bit operations for the HCD_FLAG_WAKEUP_PENDING flag of Hcd,
> which were previously split between the top half and bottom half of the interrupt,
> are now unified and executed solely in the bottom half of the interrupt.
> This prevents the bottom half tasks from being frozen during the S4 process,
> ensuring that the clean_bit process can proceed without interruption.

The name is "clear_bit" (with an 'r'), not "clean_bit".

> 2. Add a condition to the set_bit operation for the hcd status HCD_FLAG_WAKEUP_PENDING.
> When the hcd status is HC_STATE_SUSPENDED, perform the setting of the aforementioned status bit.
> This prevents a subsequent set_bit from occurring after the clean_bit if the hcd is in the resuming process.

hcd_bus_resume() clears that HCD_FLAG_WAKEUP_PENDING bit after calling 
hcd->driver->bus_resume().  After that point, usb_hcd_resume_root_hub() 
won't be called, so how can HCD_FLAG_WAKEUP_PENDING get set again?

Alan Stern

> Signed-off-by: Duan Chenghao <duanchenghao@xxxxxxxxxx>
> ---
>  drivers/usb/core/hcd.c | 1 -
>  drivers/usb/core/hub.c | 3 +++
>  2 files changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
> index 1ff7d901fede..a6bd0fbd82f4 100644
> --- a/drivers/usb/core/hcd.c
> +++ b/drivers/usb/core/hcd.c
> @@ -2389,7 +2389,6 @@ void usb_hcd_resume_root_hub (struct usb_hcd *hcd)
>  	spin_lock_irqsave (&hcd_root_hub_lock, flags);
>  	if (hcd->rh_registered) {
>  		pm_wakeup_event(&hcd->self.root_hub->dev, 0);
> -		set_bit(HCD_FLAG_WAKEUP_PENDING, &hcd->flags);
>  		queue_work(pm_wq, &hcd->wakeup_work);
>  	}
>  	spin_unlock_irqrestore (&hcd_root_hub_lock, flags);
> diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> index 4b93c0bd1d4b..7f847c4afc0d 100644
> --- a/drivers/usb/core/hub.c
> +++ b/drivers/usb/core/hub.c
> @@ -3835,11 +3835,14 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
>  
>  int usb_remote_wakeup(struct usb_device *udev)
>  {
> +	struct usb_hcd  *hcd = bus_to_hcd(udev->bus);
>  	int	status = 0;
>  
>  	usb_lock_device(udev);
>  	if (udev->state == USB_STATE_SUSPENDED) {
>  		dev_dbg(&udev->dev, "usb %sresume\n", "wakeup-");
> +		if (hcd->state == HC_STATE_SUSPENDED)
> +			set_bit(HCD_FLAG_WAKEUP_PENDING, &hcd->flags);
>  		status = usb_autoresume_device(udev);
>  		if (status == 0) {
>  			/* Let the drivers do their thing, then... */
> -- 
> 2.34.1
> 
>