netdev - Re: [PATCH v2 5/5] net/core: Allow the compiler to verify declaration and definition consistency

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: <20190327000307.GB16554@bistromath.localdomain>
Date:   Wed, 27 Mar 2019 01:03:07 +0100
From:   Sabrina Dubroca <sd@...asysnail.net>
To:     Al Viro <viro@...iv.linux.org.uk>
Cc:     Bart Van Assche <bvanassche@....org>,
        David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
        Willem de Bruijn <willemb@...gle.com>
Subject: Re: [PATCH v2 5/5] net/core: Allow the compiler to verify
 declaration and definition consistency

2019-03-26, 18:17:58 +0000, Al Viro wrote:
> Dumping everything into widely-included files is a Bloody Bad Idea(tm);
> it makes reasoning about the code much harder.
> 
> If anything, we should trim the hell out of those; details that matter
> only to a well-defined subset of the kernel should be local to it.
> Consider, for example, include/net/af_unix.h.  The stuff defined in
> there:
> 
[snip example]
> 
> Dumping internal details into include/* makes life much harder
> when working with the code, trying to understand it, etc.
> The usual reasons to separate interfaces and internals do
> apply in the kernel.
> 
> Note, BTW, that stale junk (extern for a function removed at some
> point, etc.) tends to stay around, confusing the hell out of readers.
> And include/* tends to be considerably more sticky in that respect.
> 
> The bottom line: keep public headers tidy; internal details belong
> with the code working with those.

Uh, yeah, makes sense. Thanks for the detailed example, I didn't think
the include/* situation was that bad.

-- 
Sabrina