Re: [PATCH] format-security: move static strings to const

From: Jani Nikula <jani.nikula@linux.intel.com>
To: Kees Cook <keescook@chromium.org>,
	Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mips@linux-mips.org,
	linux-decnet-user@lists.sourceforge.net, kernel@stlinux.com,
	Qianqian Xie <xieqianqian@huawei.com>,
	Mugunthan V N <mugunthanvnm@ti.com>,
	Tony Lindgren <tony@atomide.com>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
	Eric Dumazet <edumazet@google.com>,
	"Maciej W. Rozycki" <macro@linux-mips.org>,
	Christian Gromm <christian.gromm@microchip.com>,
	Daniel Vetter <daniel.vetter@intel.com>,
	Ingo Molnar <mingo@kernel.org>,
	Andrey Shvetsov <andrey.shvetsov@k2l.de>,
	devel@driverdev.osuosl.org, Kejian Yan <yankejian@huawei.com>,
	Felipe Balbi <felipe.balbi@linux.intel.com>,
	Russell King <linux@armlinux.org.uk>,
	Jarod Wilson <jarod@redhat.com>,
	linux-hippi@sunsite.dk, Daode Huang <huangdaode@hisilicon.com>,
	linux-serial@vger.kernel.org, Jiri Slaby <jslaby@s>
Subject: Re: [PATCH] format-security: move static strings to const
Date: Thu, 06 Apr 2017 11:48:48 +0300	[thread overview]
Message-ID: <87mvbtzztb.fsf@intel.com> (raw)
In-Reply-To: <20170405214711.GA5711@beast>

On Thu, 06 Apr 2017, Kees Cook <keescook@chromium.org> wrote:
> While examining output from trial builds with -Wformat-security enabled,
> many strings were found that should be defined as "const", or as a char
> array instead of char pointer. This makes some static analysis easier,
> by producing fewer false positives.
>
> As these are all trivial changes, it seemed best to put them all in
> a single patch rather than chopping them up per maintainer.

> diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
> index f6d4d9700734..1ff9d5912b83 100644
> --- a/drivers/gpu/drm/drm_fb_helper.c
> +++ b/drivers/gpu/drm/drm_fb_helper.c
> @@ -2331,7 +2331,7 @@ EXPORT_SYMBOL(drm_fb_helper_hotplug_event);
>  int __init drm_fb_helper_modinit(void)
>  {
>  #if defined(CONFIG_FRAMEBUFFER_CONSOLE_MODULE) && !defined(CONFIG_EXPERT)
> -	const char *name = "fbcon";
> +	const char name[] = "fbcon";

I'd always write the former out of habit. Why should I start using the
latter? What makes it better?

What keeps the kernel from accumulating tons more of the former?

Here's an interesting comparison of the generated code. I'm a bit
surprised by what gcc does, I would have expected no difference, like
clang. https://godbolt.org/g/OdqUvN

The other changes adding const in this patch are, of course, good.

BR,
Jani.

-- 
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel