From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <4B27C8B0.3030902@domain.hid> Date: Tue, 15 Dec 2009 18:34:40 +0100 From: Jan Kiszka MIME-Version: 1.0 References: <1260281665.792922.9257.nullmailer@domain.hid> <4B1E6B1B.5080306@domain.hid> <1260370202.738468.27358.nullmailer@domain.hid> <4B1FEC63.9070100@domain.hid> <1260439679.561436.26932.nullmailer@domain.hid> <4B20E2D7.8060103@domain.hid> <4B241D1B.1030002@domain.hid> In-Reply-To: <4B241D1B.1030002@domain.hid> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai-help] xenomai 2.4.10 - invalid opcode: 0000 List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Petr Cervenka Cc: xenomai-help Jan Kiszka wrote: > Jan Kiszka wrote: >> ... >> Reproducible with your .config under kvm, will investigate later. >> > > It looks like we have a problem with CONFIG_PREEMPT on x86-SMP at the > moment (2.6.30&.31, maybe also older kernels). You may try with > CONFIG_PREEMPT_VOLUNTARY/NONE, but I can't promise that this heals the > issue, maybe it just makes it far less probable. > > The corruption is easy to reproduce - but always at different spots, > thus it's very hard to catch. Still investigating. > I think I found all gremlins, the last one [1] was probably the major contributor. Could you try if the patches in [2] help you? Works for me so far. Jan [1]http://permalink.gmane.org/gmane.linux.kernel.adeos.general/1491 [2]http://git.kiszka.org/?p=ipipe-2.6.git;a=shortlog;h=refs/heads/queues/2.6.31-x86 -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux