Date:   Fri, 3 Feb 2023 11:45:31 +0100
From:   Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To:     Anastasia Belova <abelova@astralinux.ru>
Cc:     Jakob Koschel <jakobkoschel@gmail.com>, linux-usb@vger.kernel.org,
        linux-kernel@vger.kernel.org, lvc-project@linuxtesting.org
Subject: Re: [PATCH] goku_udc: Add check for NULL in goku_irq
Message-ID: <Y9zly1vrj9z4c1qT@kroah.com>
References: <20230203101828.14799-1-abelova@astralinux.ru>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230203101828.14799-1-abelova@astralinux.ru>
Precedence: bulk

On Fri, Feb 03, 2023 at 01:18:28PM +0300, Anastasia Belova wrote:
> Before dereferencing dev->driver check it for NULL.
> 
> If an interrupt handler is called after assigning
> NULL to dev->driver, but before resetting dev->int_enable,
> NULL-pointer will be dereferenced.
> 
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> 
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Signed-off-by: Anastasia Belova <abelova@astralinux.ru>
> ---
>  drivers/usb/gadget/udc/goku_udc.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/gadget/udc/goku_udc.c b/drivers/usb/gadget/udc/goku_udc.c
> index bdc56b24b5c9..896bba8b47f1 100644
> --- a/drivers/usb/gadget/udc/goku_udc.c
> +++ b/drivers/usb/gadget/udc/goku_udc.c
> @@ -1616,8 +1616,9 @@ static irqreturn_t goku_irq(int irq, void *_dev)
>  pm_next:
>  		if (stat & INT_USBRESET) {		/* hub reset done */
>  			ACK(INT_USBRESET);
> -			INFO(dev, "USB reset done, gadget %s\n",
> -				dev->driver->driver.name);
> +			if (dev->driver)
> +				INFO(dev, "USB reset done, gadget %s\n",
> +					dev->driver->driver.name);

How can this ever happen?  Can you trigger this somehow?  If not, I
don't think this is going to be possible (also what's up with printk
from an irq handler???)

Odds are, no one actually has this hardware anymore, right?

thanks,

greg k-h