Re: [PATCH] usb: dwc3: gadget: Avoid race between dwc3 interrupt handler and irq thread handler

From: Felipe Balbi
Date: Tue Dec 27 2016 - 06:09:18 EST



Hi,

Janusz Dziedzic <janusz.dziedzic@xxxxxxxxx> writes:
> 2016-12-26 9:01 GMT+01:00 Baolin Wang <baolin.wang@xxxxxxxxxx>:
>> On some platfroms(like x86 platform), when one core is running the USB gadget
>> irq thread handler by dwc3_thread_interrupt(), meanwhile another core also can
>> respond other interrupts from dwc3 controller and modify the event buffer by
>> dwc3_interrupt() function, that will cause getting the wrong event count in
>> irq thread handler to make the USB function abnormal.
>>
>> We should add spin_lock/unlock() in dwc3_check_event_buf() to avoid this race.
>>
> Interesting, I always think we mask interrupt in dwc3_interrupt() by setting
> DWC3_GEVNTSIZ_INTMASK
> And unmask interrupt when we end dwc3_thread_interrupt().
>
> So, we shouldn't get any IRQ from HW during dwc3_thread_interrupt(),
> or I miss something?
> Do you have some traces that indicate this masking will not work correctly?

that's the very question I have. We are already masking interrupts from
this controller. The only thing this could race with is usb_ep_queue(),
but that gets nowhere close to anything we're doing in the top half
handler, so there's really no danger of anything bad happening.

I'd like to see traces as well.

--
balbi

Attachment: signature.asc
Description: PGP signature