linux-kernel - Re: [PATCH] usb: core: introduce per-port over-current counters

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:   Mon, 19 Feb 2018 17:05:34 +0100
From:   Richard Leitner <richard.leitner@...data.com>
To:     Greg KH <gregkh@...uxfoundation.org>,
        Richard Leitner <dev@...l1n.net>
CC:     <linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <stern@...land.harvard.edu>, <linux@...ck-us.net>,
        <mathias.nyman@...ux.intel.com>, <johan@...nel.org>,
        <felipe.balbi@...ux.intel.com>, <ekorenevsky@...il.com>,
        <peter.chen@....com>, <drake@...lessm.com>, <joe@...ches.com>
Subject: Re: [PATCH] usb: core: introduce per-port over-current counters


On 02/19/2018 04:59 PM, Greg KH wrote:
> On Mon, Feb 19, 2018 at 01:01:07PM +0100, Richard Leitner wrote:
>> From: Richard Leitner <richard.leitner@...data.com>
>>
>> For some userspace applications information on the number of
>> over-current conditions at specific USB hub ports is relevant. Therefore
>> introduce a oc_counter in the usb port struct which is exported via
>> sysfs.
>>
>> Signed-off-by: Richard Leitner <richard.leitner@...data.com>
>> ---
>> Tested on an i.MX6DL based board.
>> ---
>>  drivers/usb/core/hub.c  |  4 +++-
>>  drivers/usb/core/hub.h  |  1 +
>>  drivers/usb/core/port.c | 10 ++++++++++
>>  3 files changed, 14 insertions(+), 1 deletion(-)
> 
> When you add/remove/modify a sysfs attribute, you always have to
> document in Documentation/ABI/

Ok. Thank you. Seems I missed that, Sorry!

>>
>> diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
>> index c5c1f6cf3228..448fba1e1827 100644
>> --- a/drivers/usb/core/hub.c
>> +++ b/drivers/usb/core/hub.c
>> @@ -5104,8 +5104,10 @@ static void port_event(struct usb_hub *hub, int port1)
>>  
>>  	if (portchange & USB_PORT_STAT_C_OVERCURRENT) {
>>  		u16 status = 0, unused;
>> +		port_dev->oc_count++;
>>  
>> -		dev_dbg(&port_dev->dev, "over-current change\n");
>> +		dev_dbg(&port_dev->dev, "over-current change #%u\n",
>> +			port_dev->oc_count);
>>  		usb_clear_port_feature(hdev, port1,
>>  				USB_PORT_FEAT_C_OVER_CURRENT);
>>  		msleep(100);	/* Cool down */
>> diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h
>> index 2a700ccc868c..b5cf567bf9e2 100644
>> --- a/drivers/usb/core/hub.h
>> +++ b/drivers/usb/core/hub.h
>> @@ -100,6 +100,7 @@ struct usb_port {
>>  	unsigned int is_superspeed:1;
>>  	unsigned int usb3_lpm_u1_permit:1;
>>  	unsigned int usb3_lpm_u2_permit:1;
>> +	unsigned int oc_count;
>>  };
>>  
>>  #define to_usb_port(_dev) \
>> diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c
>> index 1a01e9ad3804..0bfe410eb8a7 100644
>> --- a/drivers/usb/core/port.c
>> +++ b/drivers/usb/core/port.c
>> @@ -41,6 +41,15 @@ static ssize_t connect_type_show(struct device *dev,
>>  }
>>  static DEVICE_ATTR_RO(connect_type);
>>  
>> +static ssize_t oc_count_show(struct device *dev, struct device_attribute *attr,
>> +			     char *buf)
>> +{
>> +	struct usb_port *port_dev = to_usb_port(dev);
>> +
>> +	return sprintf(buf, "%u\n", port_dev->oc_count);
>> +}
>> +static DEVICE_ATTR_RO(oc_count);
> 
> I don't see what userspace can do with this number, as there's not much
> it can do with it.

I've answered this question to Felipe already:

On 02/19/2018 03:12 PM, Richard Leitner wrote:
> In our case we have a series of USB hardware (using the cp210x driver) which
> communicates using a proprietary protocol. These devices sometimes trigger an
> over-current situation on some hubs. In case of such an over-current situation
> the USB devices offer an interface for reducing the max used power.
> As these conditions are quite rare and imply performance reductions of the
> device we don't want to use reduce the max power always.
> 
> With this patch I want to give the user-space application the possibility to
> react adequately to such over-current situations.
> 
> I hope this explains my motivation for this patch in a comprehensible way.

I'm of course always open for a better/cleaner/safer way of doing this ;-)

Thank you!

regards;Richard.L