From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <51767C09.3070502@siemens.com> Date: Tue, 23 Apr 2013 14:18:17 +0200 From: Jan Kiszka MIME-Version: 1.0 References: <51757858.6040604@bollue.de> In-Reply-To: <51757858.6040604@bollue.de> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai] Sporadic freeze using RTnet List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kai Bollue Cc: xenomai@xenomai.org On 2013-04-22 19:50, Kai Bollue wrote: > Hello, > > after having used Xenomai with RTnet for quite some time, we switched to > new hardware and now experience sporadic freezes when communicating over > RTnet. (To trigger the problem, it is sufficient to call rtping in a > loop.) After several minutes (up to about an hour), the system stops > responding. Output over a serial console while this happens is > Does sysrq still work then? sysrq-l would be interesting e.g. > - Either nothing at all, or > > - > [ 481.189349] INFO: rcu_bh detected stalls on CPUs/tasks: { 2} > (detected by 0, t=60042 jiffies) > [ 481.291572] INFO: Stall ended before state dump start > [ 541.224383] INFO: rcu_preempt detected stalls on CPUs/tasks: { 2} > (detected by 0, t=60002 jiffies) > [ 541.331818] INFO: Stall ended before state dump start > (repeated), or > > - > [ 2333.757994] Xenomai: Unbalanced lock/unlock Ah, here's where the problem start. CONFIG_IPIPE_TRACE_PANIC may provide further information. Enable it via the I-pipe tracer. Then, before triggering the lockup, increase the backtrace points to, say, 2000: echo 2000 > /proc/ipipe/trace/back_trace_points Jan -- Siemens AG, Corporate Technology, CT RTC ITP SDP-DE Corporate Competence Center Embedded Linux