Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Sender: Larry Finger <larry.finger@gmail.com>
Message-ID: <fb51e801-d1b9-a436-8fd2-e05a69f0c24b@lwfinger.net>
Date:   Fri, 18 Feb 2022 11:45:31 -0600
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.6.0
Subject: Re: [PATCH RFC] staging: r8188eu: comment about the chip's packet
 format
Content-Language: en-US
To:     Dan Carpenter <dan.carpenter@oracle.com>,
        Martin Kaiser <martin@kaiser.cx>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Phillip Potter <phil@philpotter.co.uk>,
        Michael Straube <straube.linux@gmail.com>,
        linux-staging@lists.linux.dev, linux-kernel@vger.kernel.org
References: <20220218092252.853807-1-martin@kaiser.cx>
 <20220218102227.GE2407@kadam>
From:   Larry Finger <Larry.Finger@lwfinger.net>
In-Reply-To: <20220218102227.GE2407@kadam>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2/18/22 04:22, Dan Carpenter wrote:
> On Fri, Feb 18, 2022 at 10:22:53AM +0100, Martin Kaiser wrote:
>> The structs phy_rx_agc_info and phy_status_rpt define parts of the
>> header data that the r8188eu chipset sends to this driver via usb.
>>
>> Add a comment to clarify that we cannot modify the content of these
>> structures and remove seemingly unused fields.
>>
>> Signed-off-by: Martin Kaiser <martin@kaiser.cx>
>> ---
>>
>> Dear all,
>>
>> I experimented with "cleaning up" these structures and related code before
>> discovering that their content comes from usb packets we receive from the
>> r8188eu chipset.
>>
>> Would it make sense to add a word of warning to prevent others from
>> repeating this exercise?
> 
> Just the fact that these structs are endian means they're from the
> firmware or the network.
> 
> If a struct has a pointer in it, then it's rarely part of the UAPI but
> if it has endian data then it probably is.

Additionally, do not change any struct with the __packed attribute. Most will 
have endian variables as well, but not all will.

Testing after modifying any struct is a necessity.

Larry