linux-kernel - Re: [PATCH] option: Improve Quectel EP06 detection

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 for Android: free password hash cracker in your pocket

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

Date:   Thu, 13 Sep 2018 01:25:43 +0700
From:   Lars Melin <larsm17@...il.com>
To:     Kristian Evensen <kristian.evensen@...il.com>
Cc:     Johan Hovold <johan@...nel.org>, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] option: Improve Quectel EP06 detection

On 9/12/2018 23:57, Kristian Evensen wrote:
> On Wed, Sep 12, 2018 at 6:32 PM Lars Melin <larsm17@...il.com> wrote:
>> You have chosen a platform which has limited usb resources and want to
>> solve that problem by adjusting the device driver?
> 
> No, you asked for a good reason for why disabling and not just
> ignoring an interface makes sense, and I think that supporting
> multiple EP06 on platforms with limited endpoints qualifies as a
> reason. My motivation behind this patch and modifying the driver, is
> to make the driver work with the different options/combinations
> supported by the modem. The platforms I am working on merely triggered
> the error and inspired the change.
> 
>> Why don't you just unbind those interfaces which you are not using and
>> which are eating up your usb resources?
> 
> As far as I know, unbinding interfaces from the driver does not free
> up the memory allocated to the interface by/on the USB controller. I
> also tried, just in case, and the output from lsusb is the same
> regardless of bind/unbind.
> 
> Btw, the patch for the QMI driver has been accepted, since you
> mentioned that driver earlier. So the assumption about interface four
> is removed from there.
> 
> BR,
> Kristian
> 

That the patch has qmi-wwan patch has been accepted does not change the 
fact that you are solving your problems in the wrong end.
You are using the OEM re-branding AT cmd to change the interface 
composition without changing the vid and pid at the same time, this is a 
big donut and the whole reason for why you have submitted patches which 
shouldn't be needed.


rgds
/Lars