From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751787Ab3IRNpq (ORCPT ); Wed, 18 Sep 2013 09:45:46 -0400 Received: from eddie.linux-mips.org ([78.24.191.182]:43391 "EHLO cvs.linux-mips.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751655Ab3IRNpm (ORCPT ); Wed, 18 Sep 2013 09:45:42 -0400 Date: Wed, 18 Sep 2013 15:45:33 +0200 From: Ralf Baechle To: linux-kernel@vger.kernel.org, Arnd Bergmann , David Howells , Markos Chandras Cc: linux-mips@linux-mips.org Subject: Re: [PATCH v2] MIPS: ath79: Avoid using unitialized 'reg' variable Message-ID: <20130918134533.GN22468@linux-mips.org> References: <1377082042-4219-1-git-send-email-markos.chandras@imgtec.com> <20130903133839.GA10563@linux-mips.org> <5225EC3B.1070701@imgtec.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <5225EC3B.1070701@imgtec.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Sep 03, 2013 at 03:03:39PM +0100, Markos Chandras wrote: > >Was this triggered by CONFIG_BUG=n? > > > > Ralf > > > > Hi Ralf, > > Yes it was triggered by CONFIG_BUG=n So here's a small test case to demonstrate the issue: /* * Definition of BUG taken from asm-generic/bug.h for the CONFIG_BUG=n case */ #define BUG() do {} while(0) int foo(int arg) { int res; if (arg == 1) res = 23; else if (arg -= 2) res = 42; else BUG(); return res; } [ralf@h7 linux-mips]$ gcc -O2 -Wall -c bug.c bug.c: In function ‘foo’: bug.c:17:2: warning: ‘res’ may be used uninitialized in this function [-Wmaybe-uninitialized] return res; ^ It's fairly obvious to see what's happening here - GCC doesn't know that the else case can not be reached, thus razorsharply concludes that res may be used uninitialized. I think the definition of BUG should be changed to something like #define BUG() unreachable() This has the disadvantage of of expanding into a while (1) loop for older compilers - but that's only for older compilers, relativly minor in bloat and last I checked BUG() wasn't performance critical ;-) Ralf Signed-off-by: Ralf Baechle include/asm-generic/bug.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index 7d10f96..6f78771 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -108,7 +108,7 @@ extern void warn_slowpath_null(const char *file, const int line); #else /* !CONFIG_BUG */ #ifndef HAVE_ARCH_BUG -#define BUG() do {} while(0) +#define BUG() unreachable() #endif #ifndef HAVE_ARCH_BUG_ON