linux-kernel - Re: [PATCH 1/2] usb: gadget: udc: avoid use of freed pointer

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]

Message-ID: <20170214155351.Horde.RbjzVc4YaibN9ftOcW_ubLW@webmail.embeddedor.com>
Date:   Tue, 14 Feb 2017 15:53:51 -0600
From:   "Gustavo A. R. Silva" <garsilva@...eddedor.com>
To:     Michal Nazarewicz <mina86@...a86.com>
Cc:     balbi@...nel.org, gregkh@...uxfoundation.org, bhelgaas@...gle.com,
        heikki.krogerus@...ux.intel.com, linux-geode@...ts.infradead.org,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] usb: gadget: udc: avoid use of freed pointer

Hi Michal,

Quoting Michal Nazarewicz <mina86@...a86.com>:

> On Mon, Feb 13 2017, Gustavo A. R. Silva wrote:
>> Rewrite udc_free_dma_chain() function to avoid use of pointer after free.
>>
>> Addresses-Coverity-ID: 1091172
>> Reviewed-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
>> Signed-off-by: Gustavo A. R. Silva <garsilva@...eddedor.com>
>
> Acked-by: Michal Nazarewicz <mina86@...a86.com>
>
>> ---
>>  drivers/usb/gadget/udc/amd5536udc.c | 20 +++++++++++---------
>>  1 file changed, 11 insertions(+), 9 deletions(-)
>>
>> diff --git a/drivers/usb/gadget/udc/amd5536udc.c  
>> b/drivers/usb/gadget/udc/amd5536udc.c
>> index ea03ca7..ded97a3 100644
>> --- a/drivers/usb/gadget/udc/amd5536udc.c
>> +++ b/drivers/usb/gadget/udc/amd5536udc.c
>> @@ -611,21 +611,23 @@ udc_alloc_request(struct usb_ep *usbep, gfp_t gfp)
>>  static int udc_free_dma_chain(struct udc *dev, struct udc_request *req)
>>  {
>>  	int ret_val = 0;
>> -	struct udc_data_dma	*td;
>> -	struct udc_data_dma	*td_last = NULL;
>> +	struct udc_data_dma *td = req->td_data;
>>  	unsigned int i;
>>
>> +	dma_addr_t addr_aux = 0x00;
>
> Perhaps call it ‘addr_next’ or ‘next’?
>
>> +	dma_addr_t addr = (dma_addr_t)td->next;
>> +	td->next = 0x00;
>> +
>>  	DBG(dev, "free chain req = %p\n", req);
>>
>>  	/* do not free first desc., will be done by free for request */
>> -	td_last = req->td_data;
>> -	td = phys_to_virt(td_last->next);
>> -
>>  	for (i = 1; i < req->chain_len; i++) {
>> -		pci_pool_free(dev->data_requests, td,
>> -			      (dma_addr_t)td_last->next);
>> -		td_last = td;
>> -		td = phys_to_virt(td_last->next);
>> +		td = phys_to_virt(addr);
>> +		addr_aux = (dma_addr_t)td->next;
>> +		td->next = 0x00;
>
> This is unnecessary.
>
>> +		pci_pool_free(dev->data_requests, td, addr);
>> +		td = NULL;
>
> Ditto.
>
>> +		addr = addr_aux;
>>  	}
>>
>>  	return ret_val;
>> --
>> 2.5.0
>>
>

Thanks for your comments, I will send version 2 shortly.
--
Gustavo A. R. Silva