[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <a98a68c8-8b08-b3e9-da7b-ff57a2614f96@users.sourceforge.net>
Date: Sun, 1 Jul 2018 11:22:36 +0200
From: SF Markus Elfring <elfring@...rs.sourceforge.net>
To: Julia Lawall <julia.lawall@...6.fr>,
Kees Cook <keescook@...omium.org>, linux-mm@...ck.org,
kernel-hardening@...ts.openwall.com
Cc: kernel-janitors@...r.kernel.org,
LKML <linux-kernel@...r.kernel.org>,
Matthew Wilcox <willy@...radead.org>,
Rasmus Villemoes <linux@...musvillemoes.dk>,
Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [PATCH v3 12/16] treewide: Use array_size() for kmalloc()-family
>> * The repetition of such a constraint in subsequent SmPL rules could be avoided
>> if inheritance will be used for this metavariable.
>
> This is quite incorrect.
I suggest to consider additional software design options.
> Inheritance is only possible when a match of the previous rule has succeeded.
I agree with this information.
> If a rule never applies in a given file, the rules that inherit from it
> won't apply either.
I would like to point the possibility out to specify a source code search
which will find interesting function calls at least by an inital SmPL rule.
> Furthermore, what is inherited is the value, not the constraint.
This technical detail can be fine.
> If the original binding of alloc only ever matches kmalloc,
> then the inherited references will only match kmalloc too.
Can the desired search pattern be extended in significant ways?
Regards,
Markus
Powered by blists - more mailing lists