Re: [PATCH] usb: dwc3: gadget: Disable gadget IRQ during pullup disable

From: Felipe Balbi <balbi@kernel.org>
To: Wesley Cheng <wcheng@codeaurora.org>, gregkh@linuxfoundation.org
Cc: linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
	jackp@codeaurora.org, Wesley Cheng <wcheng@codeaurora.org>
Subject: Re: [PATCH] usb: dwc3: gadget: Disable gadget IRQ during pullup disable
Date: Thu, 10 Jun 2021 14:09:19 +0300	[thread overview]
Message-ID: <87h7i60ye8.fsf@kernel.org> (raw)
In-Reply-To: <1621571037-1424-1-git-send-email-wcheng@codeaurora.org>

[-- Attachment #1: Type: text/plain, Size: 1791 bytes --]

Wesley Cheng <wcheng@codeaurora.org> writes:

> Current sequence utilizes dwc3_gadget_disable_irq() alongside
> synchronize_irq() to ensure that no further DWC3 events are generated.
> However, the dwc3_gadget_disable_irq() API only disables device
> specific events.  Endpoint events can still be generated.  Briefly
> disable the interrupt line, so that the cleanup code can run to
> prevent device and endpoint events. (i.e. __dwc3_gadget_stop() and
> dwc3_stop_active_transfers() respectively)
>
> Without doing so, it can lead to both the interrupt handler and the
> pullup disable routine both writing to the GEVNTCOUNT register, which
> will cause an incorrect count being read from future interrupts.
>
> Fixes: ae7e86108b12 ("usb: dwc3: Stop active transfers before halting the controller")
> Signed-off-by: Wesley Cheng <wcheng@codeaurora.org>
> ---
>  drivers/usb/dwc3/gadget.c | 11 +++++------
>  1 file changed, 5 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
> index 49ca5da..89aa9ac 100644
> --- a/drivers/usb/dwc3/gadget.c
> +++ b/drivers/usb/dwc3/gadget.c
> @@ -2260,13 +2260,10 @@ static int dwc3_gadget_pullup(struct usb_gadget *g, int is_on)
>  	}
>  
>  	/*
> -	 * Synchronize any pending event handling before executing the controller
> -	 * halt routine.
> +	 * Synchronize and disable any further event handling while controller
> +	 * is being enabled/disabled.
>  	 */
> -	if (!is_on) {
> -		dwc3_gadget_disable_irq(dwc);
> -		synchronize_irq(dwc->irq_gadget);
> -	}
> +	disable_irq(dwc->irq_gadget);
>  
>  	spin_lock_irqsave(&dwc->lock, flags);

spin_lock_irqsave() is already disabling interrupt, right? Why do we
need another call to disable_irq()?

-- 
balbi

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 511 bytes --]