Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <710ed88d-462b-17ed-ae4d-906ddd0dee8a@gmail.com>
Date:   Sun, 1 Oct 2023 11:11:13 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.15.1
Subject: Re: [PATCH] Input: powermate - fix use-after-free in
 powermate_config_complete
To:     Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc:     linux-input@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20230916-topic-powermate_use_after_free-v1-1-2ffa46652869@gmail.com>
 <CAPnbTwKqNghcoPj-FGQQxo0xr-AYTm8pYBYCUgyKT6VxZpZCOA@mail.gmail.com>
 <ZQjKwQDKmU8L9C9e@google.com>
Content-Language: en-US
From:   Javier Carrasco <javier.carrasco.cruz@gmail.com>
In-Reply-To: <ZQjKwQDKmU8L9C9e@google.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi Dmitry,

On 19.09.23 00:10, Dmitry Torokhov wrote:
> On Mon, Sep 18, 2023 at 06:51:49AM +0200, Javier Carrasco Cruz wrote:
>> Hi,
>>
>> There's an obvious error in the patch I introduced when cleaningup
>> (urb->status should be used instead of just status). I will send a v2.
> 
> I think what we need is call to usb_kill_urb(pm->config) in
> powermate_disconnect(), right after call to input_unregister_device().
> 
> Thanks.
> That is definitely a more meaningful and elegant solution, so I will
check it out and eventually send a v2 with it if everything seems ok. On
the other hand usb_kill_urb() is already used on pm->irq before calling
input_unregister_device(), so I would move the existing usb_kill_urb to
have both calls right after the unregister_device call for code
consistency, if that is alright.

Thanks and best regards.