linux-kernel - Re: [PATCH] drm/sti: mark symbols static where possible

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]

Date:   Thu, 08 Sep 2016 11:56:12 +0200
From:   Arnd Bergmann <arnd@...db.de>
To:     Emil Velikov <emil.l.velikov@...il.com>
Cc:     Baoyou Xie <baoyou.xie@...aro.org>,
        Benjamin Gaignard <benjamin.gaignard@...aro.org>,
        Vincent Abriou <vincent.abriou@...com>,
        David Airlie <airlied@...ux.ie>, xie.baoyou@....com.cn,
        "Linux-Kernel@...r. Kernel. Org" <linux-kernel@...r.kernel.org>,
        ML dri-devel <dri-devel@...ts.freedesktop.org>
Subject: Re: [PATCH] drm/sti: mark symbols static where possible

On Thursday, September 8, 2016 10:35:17 AM CEST Emil Velikov wrote:
> On 7 September 2016 at 12:05, Baoyou Xie <baoyou.xie@...aro.org> wrote:
> > We get 2 warnings when building kernel with W=1:
> As you're going through DRM I was wondering if you have a rough number
> of warnings we get at the various W levels 1,2,...

I've looked at the W=1 warnings overall, and the count I got a
month ago was 648 warnings for drivers/gpu/::

    471 -Werror=missing-prototypes
     12 -Werror=type-limits
    124 -Werror=unused-but-set-variable
     41 -Werror=unused-const-variable=

vs for the whole kernel

   2033 -Werror=missing-prototypes
     58 -Werror=suggest-attribute=format
    167 -Werror=type-limits
   1398 -Werror=unused-but-set-variable
   1526 -Werror=unused-const-variable=

but that was after I had already fixed some of the other warnings
locally. It shouldn't be hard to fix all of them for any given
subsystem, often a single line change gets rid of a number
of individual warnings.

My basic idea however is not to do it by subsystem but instead
do it one warning at a time for the entire kernel and then enable
that warning by default without W=1.

> Hope you'll have the time/interest to sort some of the W>1 ones as well 

I suggested to Baoyou that he starts looking at missing-prototype
warnings across the kernel, as these are likely to find the most
actual bugs out of the W=1 warnings we get.

	Arnd