From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <4F4FBC3C.3070600@domain.hid> Date: Thu, 01 Mar 2012 19:13:16 +0100 From: Gilles Chanteperdrix MIME-Version: 1.0 References: <4F3146AA.5070902@domain.hid> <4F41144E.1020004@domain.hid> <4F4FB9F3.8030806@domain.hid> In-Reply-To: <4F4FB9F3.8030806@domain.hid> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai-core] high duration of __ipipe_handle_irq List-Id: Xenomai life and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Roberto Bielli Cc: xenomai@xenomai.org On 03/01/2012 07:03 PM, Roberto Bielli wrote: > Hi, > > i have this problem with xenomai ( 2.5.6 - arm freescale imx25 ): > > I have a xenomai application that uses more rt_tasks, and i have a > strange behaviour. > When i run my application the period of *__ipipe_handle_irq* is very > long ( ~3ms in some cases ) in the moment > that come the *__ipipe_mach_timerint *interrupt and so the other > interrupts are very late. > > Do you know something about this ? I am sorry, but I do not understand your mail. __ipiep_handle_irq is not periodic and this sentence: When i run my application the period of *__ipipe_handle_irq* is very long ( ~3ms in some cases ) in the moment that come the *__ipipe_mach_timerint *interrupt and so the other is not understandable. Please try to describe your problem more clearly, or better, use the I-pipe tracer and record a trace of the problem, and post the relevant part of the trace here. -- Gilles.