linux-kernel - Re: [PATCH] staging: lustre: ldlm: use designated initializers

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: <CAGXu5jJ5-SJyznhn2U03OxnkucJBKFsVuaWoCevJHK2V3vANjw@mail.gmail.com>
Date:   Tue, 20 Dec 2016 11:46:35 -0800
From:   Kees Cook <keescook@...omium.org>
To:     Dan Carpenter <dan.carpenter@...cle.com>
Cc:     "Hammond, John" <john.hammond@...el.com>,
        Bruce Korb <bruce.korb@...il.com>,
        "devel@...verdev.osuosl.org" <devel@...verdev.osuosl.org>,
        "Dilger, Andreas" <andreas.dilger@...el.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "Drokin, Oleg" <oleg.drokin@...el.com>,
        Vitaly Fertman <vitaly_fertman@...atex.com>,
        "Liu, Emoly" <emoly.liu@...el.com>,
        "Faccini, Bruno" <bruno.faccini@...el.com>,
        Lustre Development List <lustre-devel@...ts.lustre.org>
Subject: Re: [PATCH] staging: lustre: ldlm: use designated initializers

On Tue, Dec 20, 2016 at 11:07 AM, Dan Carpenter
<dan.carpenter@...cle.com> wrote:
> On Tue, Dec 20, 2016 at 02:57:17PM +0000, Hammond, John wrote:
>> "{ NULL }" is valid ISO C, but unfortunately "{}" is not.
>
> In the kernel we don't care.  We use lots of GCC extensions.

We depend on the compiler to do "incomplete zero-initialization" of
structures that are not mentioned in an initializer. The reason { NULL
} works is because the first field in the structure can take a NULL
value, and then the rest are zero-initialized by the compiler. { } is
the same thing, but doesn't use ordered initialization. If this style
is truly unacceptable to you, then { .somefield = NULL } can work, or
as you point out, if it's being initialized later, the static
initializer can be dropped entirely.

-Kees

-- 
Kees Cook
Nexus Security