From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <50E07220.8000602@xenomai.org>
Date: Sun, 30 Dec 2012 17:56:00 +0100
From: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
MIME-Version: 1.0
References: <50E062E8.6030903@openwide.fr>
In-Reply-To: <50E062E8.6030903@openwide.fr>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Subject: Re: [Xenomai] Xenomai 2.6.2 user-space compilation error on ARM <
 armv6
List-Id: Discussions about the Xenomai project <xenomai.xenomai.org>
List-Unsubscribe: <http://www.xenomai.org/mailman/options/xenomai>,
	<mailto:xenomai-request@xenomai.org?subject=unsubscribe>
List-Archive: <http://www.xenomai.org/pipermail/xenomai>
List-Post: <mailto:xenomai@xenomai.org>
List-Help: <mailto:xenomai-request@xenomai.org?subject=help>
List-Subscribe: <http://www.xenomai.org/mailman/listinfo/xenomai>,
	<mailto:xenomai-request@xenomai.org?subject=subscribe>
To: Romain Naour <romain.naour@openwide.fr>
Cc: xenomai@xenomai.org

On 12/30/2012 04:51 PM, Romain Naour wrote:

> Hi Gilles,
> 
> First of all, thank you for the new release of Xenomai :)
> 
> I'm compiling Xenomai 2.6.2 and Linux kernel 3.5.3 for mini2440 with buildroot 2012.11.
> 
> During compilation I get the following error:
> error: #error "SMP not supported below armv6, compile with -march=armv6 or above"
> 
> This error is caused by the SMP mode which is now enabled by default on ARM.
> So I added the option --disable-smp on the buildroot's command line for ./configure
> (buildroot assumes that the SMP mode is not enabled when it compile Xenomai for arm) 
> 
> This option became mandatory for ARM < armv6 to pass this test: (xenomai-2.6/include/asm-arm/feature.h)
> 
> #if __LINUX_ARM_ARCH__ < 6 && defined(CONFIG_SMP)
> #error "SMP not supported below armv6, compile with -march=armv6 or above"
> #endif
> 
> The error message may be advisable to add the option --disable-smp ?
> However, since your objective is to allow to run the same binary on UP and SMP system, it is safe to remove this test for ARM < armv6 now ?


Hi Romain,

thanks for the report, this test only makes sense if compiling with the
"ad-hoc" atomic operations, with the default setting which is to use gcc
atomic operations, we can remove the test.

Will fix for 2.6.2.1, thanks for the report.

Regards.

-- 
                                                                Gilles.