linux-kernel - Re: [PATCH] usb: gadget: udc: add null check before pointer dereference

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 02 May 2017 10:29:01 -0500
From:   "Gustavo A. R. Silva" <garsilva@...eddedor.com>
To:     Alan Stern <stern@...land.harvard.edu>
Cc:     Felipe Balbi <balbi@...nel.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jussi Kivilinna <jussi.kivilinna@...tian.com>,
        Tim Harvey <tharvey@...eworks.com>,
        Colin Ian King <colin.king@...onical.com>,
        Raz Manor <Raz.Manor@...ens.com>, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Peter Senna Tschudin <peter.senna@...il.com>
Subject: Re: [PATCH] usb: gadget: udc: add null check before pointer
 dereference

Hi Alan,

Quoting Alan Stern <stern@...land.harvard.edu>:

> On Mon, 1 May 2017, Gustavo A. R. Silva wrote:
>
>> Add null check before dereferencing dev->regs pointer inside
>> net2280_led_shutdown() function.
>>
>> Addresses-Coverity-ID: 101783
>> Signed-off-by: Gustavo A. R. Silva <garsilva@...eddedor.com>
>> ---
>>  drivers/usb/gadget/udc/net2280.c | 8 +++++---
>>  1 file changed, 5 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/usb/gadget/udc/net2280.c  
>> b/drivers/usb/gadget/udc/net2280.c
>> index 3828c2e..1898a4b 100644
>> --- a/drivers/usb/gadget/udc/net2280.c
>> +++ b/drivers/usb/gadget/udc/net2280.c
>> @@ -3573,7 +3573,11 @@ static void net2280_remove(struct pci_dev *pdev)
>>  	BUG_ON(dev->driver);
>>
>>  	/* then clean up the resources we allocated during probe() */
>> -	net2280_led_shutdown(dev);
>> +
>> +	if (dev->regs) {
>> +		net2280_led_shutdown(dev);
>> +		iounmap(dev->regs);
>> +	}
>>  	if (dev->requests) {
>>  		int		i;
>>  		for (i = 1; i < 5; i++) {
>> @@ -3588,8 +3592,6 @@ static void net2280_remove(struct pci_dev *pdev)
>>  		free_irq(pdev->irq, dev);
>>  	if (dev->quirks & PLX_PCIE)
>>  		pci_disable_msi(pdev);
>> -	if (dev->regs)
>> -		iounmap(dev->regs);
>>  	if (dev->region)
>>  		release_mem_region(pci_resource_start(pdev, 0),
>>  				pci_resource_len(pdev, 0));
>
> No, you must not move the iounmap() call, because an interrupt could
> theoretically occur at any time.
>

Yeah, I was suspicious about it.

> Either just live with an extra test of dev->regs, or else move the
> net2280_led_shutdown() call later.
>

In this case I think it is safe to move the net2280_led_shutdown()  
call, as the function is only turning off the LEDs.

I'll send a patch shortly.
Thank you
--
Gustavo A. R. Silva