linux-kernel - Re: checkpatch guide for newbies

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

Message-ID: <alpine.DEB.2.02.1309260749580.1984@localhost6.localdomain6>
Date:	Thu, 26 Sep 2013 07:53:40 +0200 (CEST)
From:	Julia Lawall <julia.lawall@...6.fr>
To:	Alexander Holler <holler@...oftware.de>
cc:	Al Viro <viro@...IV.linux.org.uk>,
	Bjorn Helgaas <bhelgaas@...gle.com>,
	Peter Senna Tschudin <peter.senna@...il.com>,
	Dan Carpenter <dan.carpenter@...cle.com>,
	kernel-janitors@...r.kernel.org,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: checkpatch guide for newbies

On Thu, 26 Sep 2013, Alexander Holler wrote:

> Am 26.09.2013 05:04, schrieb Al Viro:
> > On Thu, Sep 26, 2013 at 04:57:32AM +0200, Alexander Holler wrote:
> > > Am 26.09.2013 04:52, schrieb Alexander Holler:
> > > 
> > > > I'm aware of people which do nest 8 levels deep just to avoid a return,
> > > > break or goto.
> > > > 
> > > > But trying to limit that by limiting the line length is like ...
> > > > (choose your own own misguided comparison, it's too late for me I
> > > > currently only meorize some of those which don't make sense in english)
> > > 
> > > But I'm still able to offer a solution: ;)
> > > 
> > > limit the number of tabs, not the line length (at least not to 80).
> > 
> > With that limited (and it's visually harder to keep track of), what's
> > the problem with 80-column limit on line length?  Just how long do
> > you want those "descriptive names" to be?
> 
> Oh, personally I don't have any limit there. ;) I like descriptive function
> and variable names whenever they make sense. And often they make comments
> uneccessary and therefor prevent errors because those descriptive names are
> visible whenever the function or variable is used, and comments usually appear
> only once and get forgotten when scrolled out of the screen.
> 
> But just take a function like
> 
> void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
>                                 struct timespec *wtom, struct timespec
> *sleep);
> 
> I like such function names ;) (ok I wouldn't have use those and), but it's
> hard to press this into 80 characters, especially when the arguments should
> have some meaning too (e.g. what does wtom stand for?)
> 
> If you use that somewhere you get
> 
>         get_xtime_and_monotonic_and_sleep_offset(a, b, c)
> 
> using silly names and that already is a 58 characters long. So only 22 are
> left to distribute over 3 variable names. And now think what happens if that
> wouldn't be a void function.

Personally, I prefer to use my screen real estate for multiple 80-column 
windows, so I can see different parts of the code at once.  Anything that 
goes over 80 columns is very hard to read.

Perhaps it is a bad example, but I don't even find this very long name 
very understandable.  Monotonic is an adjective and xtime and sleep are 
nouns, so I don't understand how it all fits together.  Maybe cramming a 
lot of information into a variable name is not always so successful... 
Actually, I really appreciate comments on functions, that explain the 
purpose of the function, and the constraints on its usage.

julia
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/