Message-ID: <1513175961.26281.20.camel@suse.com>
Subject: Re: [PATCH] USB: serial: ark3116.c: Remove unused TIOCSSERIAL case
 from ioctl
From: Oliver Neukum <oneukum@suse.com>
To: Mikhail Zaytsev <flashed@mail.ru>
Cc: linux-kernel@vger.kernel.org, linux-usb@vger.kernel.org
Date: Wed, 13 Dec 2017 15:39:21 +0100
In-Reply-To: <20171213153027.7aaab62b@zaytsev.tver.pg>
References: <20171213123004.4619000a@zaytsev.tver.pg>
         <1513160248.26281.5.camel@suse.com>
         <20171213143142.1c312b57@zaytsev.tver.pg>
         <1513165248.26281.11.camel@suse.com>
         <20171213153027.7aaab62b@zaytsev.tver.pg>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1121
Lines: 46

Am Mittwoch, den 13.12.2017, 15:30 +0300 schrieb Mikhail Zaytsev:
> On Wed, 13 Dec 2017 12:40:48 +0100 Oliver Neukum <oneukum@suse.com> wrote:
> 
> > 
> > They give you nothing. If you are looking at a vendor ID nothing but the
> > bare number makes sense. You are just making peoples' life harder when
> > they have to look up that definition. A symbolic name is fine if it gives
> > meaning. Even if the information you give is that the value is magic
> > and therefore not understood. But a vendor ID is an arbitrary yet
> > meaningful number. There is no point in hiding it.
> 
> Thanks. I hear you, Oliver. What about:
> 
> -		serstruct.baud_base = 460800;
> 
> Is it a magic number? I think yes.
> 

Hi,

yes sure. That is a candidate for a symbolic name. Though if you use
it once, I see no benefit, but it does not hurt either. The member
is named and that is the important thing.

A line like

if (rate > 38400) return -EINVAL;

is not so good

if (rate > MAX_BAUD) return -EINVAL;

better

But:

device->maxbaudrate = 38400

is better than

device->maxbaudrate = MAX_BAUD

You see the point?

	Regards
		Oliver