linux-kernel - Re: [PATCH] USB: serial: Correct return value on read

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] [day] [month] [year] [list]

Message-ID: <CAGD+ZM-zPrug0D_no9qgm1mA8gw_MgT1hy8xzMa4wazeme+RxA@mail.gmail.com>
Date:   Mon, 27 Nov 2017 14:21:25 +0000
From:   saks cuso <gimcuan@...il.com>
To:     Johan Hovold <johan@...nel.org>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] USB: serial: Correct return value on read

On 27 November 2017 at 10:21, Johan Hovold <johan@...nel.org> wrote:
> On Sun, Nov 26, 2017 at 04:18:51PM +0000, Gimcuan Hui wrote:
>> It's meaningless to return buf[0] on read. Because the caller of this
>> interface checks the return value negative or not. Instead, we should
>> return the result variable.
>
> It's not really meaningless, it's just that the return value is no
> longer being used the way it was originally intended. I think
> "redundant" and/or "confusing" would be a more appropriate description.
>

yes,  i will use "confusing" instead.

>> Signed-off-by: Gimcuan Hui <gimcuan@...il.com>
>> ---
>>  drivers/usb/serial/ark3116.c | 4 +---
>>  1 file changed, 1 insertion(+), 3 deletions(-)
>>
>> diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c
>> index 3c544782f60b..bfdbc7164e7b 100644
>> --- a/drivers/usb/serial/ark3116.c
>> +++ b/drivers/usb/serial/ark3116.c
>> @@ -101,11 +101,9 @@ static int ark3116_read_reg(struct usb_serial *serial,
>>                               reg, result);
>>               if (result >= 0)
>>                       result = -EIO;
>> -
>> -             return result;
>>       }
>>
>> -     return buf[0];
>> +     return result;
>
> Also we do not want to return the value of result on success as that
> would always be 1 (i.e. the buffer size).
>
> You could change this function, and also the write_reg one, to return 0
> on success since this is a more common pattern.
>
> Please also rephrase your commit summary (Subject) since you're not
> really "correcting" (as in fixing) anything here. This is more a of nice
> clean up, even if it could potentially also prevent future bugs.
>
> Thanks,
> Johan

agree, v2 will be post soon.

thank you for correcting.