Re: [PATCH v3] usb: dwc3: Stop active transfers before halting the controller

[Wesley Cheng]

On 9/24/2020 11:06 PM, Felipe Balbi wrote:
> 
> Hi,
> 
> Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> writes:
>>>> Hence, the reason if there was already a pending IRQ triggered, the
>>>> dwc3_gadget_disable_irq() won't ensure the IRQ is handled.  We can do
>>>> something like:
>>>> if (!is_on)
>>>> 	dwc3_gadget_disable_irq()
>>>> synchronize_irq()
>>>> spin_lock_irqsave()
>>>> if(!is_on) {
>>>> ...
>>>>
>>>> But the logic to only apply this on the pullup removal case is a little
>>>> messy.  Also, from my understanding, the spin_lock_irqsave() will only
>>>> disable the local CPU IRQs, but not the interrupt line on the GIC, which
>>>> means other CPUs can handle it, unless we explicitly set the IRQ
>>>> affinity to CPUX.
>>>
>>> Yeah, the way I understand this can't really happen. But I'm open to
>>> being educated. Maybe Alan can explain if this is really possibility?
>>

Hi Felipe/Alan,

Thanks for the detailed explanations and inputs.  Useful information to
have!

>> It depends on the details of the hardware, but yes, it is possible in
>> general for an interrupt handler to run after you have turned off the
>> device's interrupt-request line.  For example:
>>
>> 	CPU A				CPU B
>> 	---------------------------	----------------------
>> 	Gets an IRQ from the device
>> 	Calls handler routine		spin_lock_irq
>> 	  spin_lock_irq			Turns off the IRQ line
>> 	  ...spins...			spin_unlock_irq
>> 	  Rest of handler runs
>> 	  spin_unlock_irq
>>
>> That's why we have synchronize_irq().  The usual pattern is something
>> like this:
>>
>> 	spin_lock_irq(&priv->lock);
>> 	priv->disconnected = true;
>> 	my_disable_irq(priv);
>> 	spin_unlock_irq(&priv->lock);
>> 	synchronize_irq(priv->irq);
>>
>> And of course this has to be done in a context that can sleep.
>>
>> Does this answer your question?
> 
> It does, thank you Alan. It seems like we don't need a call to
> disable_irq(), only synchronize_irq() is enough, however it should be
> called with spinlocks released, not held.
> 

I mean...I'm not against using the synchronize_irq() +
dwc3_gadget_disable_irq() route, since that will address the concern as
well.  It was just with the disable/enable IRQ route, I didn't need to
explicitly check the is_on flag again, since I didn't need to worry
about overwriting the DEVTEN reg (for the pullup enable case).  Will
include this on the next version.

Thanks
Wesley Cheng

> Thanks
> 

-- 
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project