Re: [PATCH] Kbuild: Move -Wmaybe-uninitialized to W=1

From: Paul Bolle <pebolle@tiscali.nl>
To: Borislav Petkov <bp@alien8.de>
Cc: linux-kbuild@vger.kernel.org, Sam Ravnborg <sam@ravnborg.org>,
	lkml <linux-kernel@vger.kernel.org>, Michael Matz <matz@suse.de>,
	x86-ml <x86@kernel.org>
Subject: Re: [PATCH] Kbuild: Move -Wmaybe-uninitialized to W=1
Date: Thu, 10 Jul 2014 13:03:14 +0200	[thread overview]
Message-ID: <1404990194.19776.5.camel@x220> (raw)
In-Reply-To: <20140710104223.GD2970@pd.tnic>

On Thu, 2014-07-10 at 12:42 +0200, Borislav Petkov wrote:
> I haven't even wished for it but just to prove everybody my point: just
> built rc4+ from Linus' repo with gcc 4.9.0, see below.
> 
> Now all of a sudden there's more noise, maybe because this is a
> different .config. However, I doubt those are real bugs.

Would you mind sharing that .config?

> A cursory look through those shows that they're not really bugs -
> gcc simply can't know with all the ifdeffery, partial usage based on

Nit: I'd expect ifdeffery to make the code only difficult to understand
for the human reader. By the time gcc will be doing flow analysis the
preprocessor has long handled all ifdeffery.

> conditionals, etc, etc.

Thanks,

Paul Bolle