linux-kernel - Re: Coding Style: Reverse XMAS tree declarations ? (was Re: [PATCH net-next v6 02/10] dpaa

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: <20161104194824.GA5533@gracie>
Date:   Fri, 4 Nov 2016 13:48:24 -0600
From:   David VomLehn <vomlehn@...as.net>
To:     Randy Dunlap <rdunlap@...radead.org>
Cc:     Joe Perches <joe@...ches.com>, David Miller <davem@...emloft.net>,
        madalin.bucur@....com, Andrew Morton <akpm@...ux-foundation.org>,
        Jonathan Corbet <corbet@....net>, netdev@...r.kernel.org,
        linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
        oss@...error.net, ppc@...dchasers.com, pebolle@...cali.nl,
        joakim.tjernlund@...nsmode.se
Subject: Re: Coding Style: Reverse XMAS tree declarations ? (was Re: [PATCH
 net-next v6 02/10] dpaa_eth: add support for DPAA Ethernet)

On Fri, Nov 04, 2016 at 10:05:15AM -0700, Randy Dunlap wrote:
> On 11/03/16 23:53, Joe Perches wrote:
> > On Thu, 2016-11-03 at 15:58 -0400, David Miller wrote:
> >> From: Madalin Bucur <madalin.bucur@....com>
> >> Date: Wed, 2 Nov 2016 22:17:26 +0200
> >>
> >>> This introduces the Freescale Data Path Acceleration Architecture
> >>> +static inline size_t bpool_buffer_raw_size(u8 index, u8 cnt)
> >>> +{
> >>> +     u8 i;
> >>> +     size_t res = DPAA_BP_RAW_SIZE / 2;
> >>
> >> Always order local variable declarations from longest to shortest line,
> >> also know as Reverse Christmas Tree Format.
> > 
> > I think this declaration sorting order is misguided but
> > here's a possible change to checkpatch adding a test for it
> > that does this test just for net/ and drivers/net/
> 
> I agree with the misguided part.
> That's not actually in CodingStyle AFAICT. Where did this come from?
> 
> 
> thanks.
> -- 
> ~Randy

This puzzles me. The CodingStyle gives some pretty reasonable rationales
for coding style over above the "it's easier to read if it all looks the
same". I can see rationales for other approaches (and I am not proposing
any of these):

alphabetic order        Easier to search for declarations
complex to simple       As in, structs and unions, pointers to simple
                        data (int, char), simple data. It seems like I
                        can deduce the simple types from usage, but more
                        complex I need to know things like the
                        particular structure.
group by usage          Mirror the ontological locality in the code

Do we have a basis for thinking this is easier or more consistent than
any other approach?
-- 
David VL