linux-kernel - Re: [PATCH net-next] qmi_wwan: Add rawip module param

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: <42522712-b739-3db4-9111-a4b378a65809@axis.com>
Date:   Tue, 12 Sep 2023 09:44:26 +0200
From:   Stefan x Nilsson <stefan.x.nilsson@...s.com>
To:     Bjørn Mork <bjorn@...k.no>
CC:     "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>, <netdev@...r.kernel.org>,
        <linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <kernel@...s.com>
Subject: Re: [PATCH net-next] qmi_wwan: Add rawip module param

On 9/12/23 09:22, Bjørn Mork wrote:
> Stefan x Nilsson <stefan.x.nilsson@...s.com> writes:
> 
>> Certain QMI modems will start communicating in rawip mode after
>> bootup, and will not work properly if communication starts off in
>> ethernet mode. So add a module parameter, rawip_as_default, that
>> can be used to load the qmi driver in rawip mode.
>>
>> The advantage compared to changing rawip at a later point using
>> sysfs is that the os will not detect the device and start talking
>> to it while the driver is still in incorrect mode.
>>
>> Signed-off-by: Stefan x Nilsson <stefan.x.nilsson@...s.com>
>> ---
>>   drivers/net/usb/qmi_wwan.c | 11 +++++++++++
>>   1 file changed, 11 insertions(+)
>>
>> diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
>> index 344af3c5c836..968c60ececf8 100644
>> --- a/drivers/net/usb/qmi_wwan.c
>> +++ b/drivers/net/usb/qmi_wwan.c
>> @@ -46,6 +46,10 @@
>>    * commands on a serial interface
>>    */
>>   
>> +/* Module parameters */
>> +static bool rawip_as_default;
>> +module_param(rawip_as_default, bool, 0644);
>> +
>>   /* driver specific data */
>>   struct qmi_wwan_state {
>>   	struct usb_driver *subdriver;
>> @@ -843,6 +847,13 @@ static int qmi_wwan_bind(struct usbnet *dev, struct usb_interface *intf)
>>   	}
>>   	dev->net->netdev_ops = &qmi_wwan_netdev_ops;
>>   	dev->net->sysfs_groups[0] = &qmi_wwan_sysfs_attr_group;
>> +
>> +	/* Set the driver into rawip mode if requested by module param */
>> +	if (rawip_as_default) {
>> +		info->flags |= QMI_WWAN_FLAG_RAWIP;
>> +		qmi_wwan_netdev_setup(dev->net);
>> +	}
>> +
>>   err:
>>   	return status;
>>   }
>>
> 
> NAK
> 
> There is no reason to start communicating with the device before
> changing the framing, using the existing sysfs interface.
> 
> This seems to be a workaround for some userspace bug.  I don't think we
> yet another userspace knob for that.  And certainly not one that applies
> to every device.
> 
> 
> 
> Bjørn

Right, got the message and I see your point. Thanks for the review.

Would a quirk on the affected modem be an better/acceptable solution?

Best Regards
Stefan x Nilsson