linux-kernel - Re: [PATCH] usb: dwc2: gadget: fix dwc2_check_param_tx_fifo

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] [day] [month] [year] [list]

Date:   Thu, 30 Nov 2017 13:12:55 +0000
From:   Amelie DELAUNAY <amelie.delaunay@...com>
To:     Minas Harutyunyan <Minas.Harutyunyan@...opsys.com>,
        John Youn <John.Youn@...opsys.com>,
        Felipe Balbi <balbi@...nel.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
CC:     "linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] usb: dwc2: gadget: fix dwc2_check_param_tx_fifo_sizes

Hi Minas,

On 11/30/2017 10:43 AM, Minas Harutyunyan wrote:
> Hi Amelie,
> 
> On 11/27/2017 6:42 PM, Amelie Delaunay wrote:
>> In case of OTG mode, with an OTG adapter plugged, the
>> dwc2_check_param_tx_fifo_sizes function reads DPTXFSIZN registers while
>> the controller is in Host mode, and, because DPTXFSIZN registers are
>> Device specific registers, the values read are 0.
>> Then, g_tx_fifo_size[fifo] values are considered invalid because in the
>> comparison test, they need to be greater than min and lower than dptxfszn
>> which is 0.
>> To fix this issue, force the controller in device mode if needed before
>> reading the DPTXFSIZN registers, and restore the previous mode if needed
>> after reading.
>>
>> Fixes: 3c6aea7344c3 ("usb: dwc2: gadget: Add checking for g-tx-fifo-size
>> parameter")
>> Signed-off-by: Amelie Delaunay <amelie.delaunay@...com>
>> ---
>>    drivers/usb/dwc2/params.c | 10 ++++++++++
>>    1 file changed, 10 insertions(+)
>>
>> diff --git a/drivers/usb/dwc2/params.c b/drivers/usb/dwc2/params.c
>> index ef73af6..302fefb 100644
>> --- a/drivers/usb/dwc2/params.c
>> +++ b/drivers/usb/dwc2/params.c
>> @@ -469,6 +469,7 @@ static void dwc2_check_param_tx_fifo_sizes(struct dwc2_hsotg *hsotg)
>>    	int fifo;
>>    	int min;
>>    	u32 total = 0;
>> +	bool forced;
>>    	u32 dptxfszn;
>>    
>>    	fifo_count = dwc2_hsotg_tx_fifo_count(hsotg);
>> @@ -483,6 +484,12 @@ static void dwc2_check_param_tx_fifo_sizes(struct dwc2_hsotg *hsotg)
>>    		dwc2_set_param_tx_fifo_sizes(hsotg);
>>    	}
>>    
>> +	/*
>> +	 * Reading DPTXFSIZN registers requires the controller to be in device
>> +	 * mode. The mode will be forced, if necessary, to read these values.
>> +	 */
>> +	forced = dwc2_force_mode_if_needed(hsotg, false);
>> +
>>    	for (fifo = 1; fifo <= fifo_count; fifo++) {
>>    		dptxfszn = (dwc2_readl(hsotg->regs + DPTXFSIZN(fifo)) &
>>    			FIFOSIZE_DEPTH_MASK) >> FIFOSIZE_DEPTH_SHIFT;
>> @@ -495,6 +502,9 @@ static void dwc2_check_param_tx_fifo_sizes(struct dwc2_hsotg *hsotg)
>>    			hsotg->params.g_tx_fifo_size[fifo] = dptxfszn;
>>    		}
>>    	}
>> +
>> +	if (forced)
>> +		dwc2_clear_force_mode(hsotg);
>>    }
>>    
>>    #define CHECK_RANGE(_param, _min, _max, _def) do {			\
>>
> 
> I'll recommend to use "[PATCH] usb: dwc2: Fix TxFIFOn sizes and total
> TxFIFO size issues" instead.

I've tested your patch and it fixes the issue too. Thank you for warning 
me of the existence of this patch. Mine can be dropped.

Regards,
Amelie
> 
> Thanks,
> Minas
>