From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <4E568CFC.2050203@domain.hid> Date: Thu, 25 Aug 2011 19:57:16 +0200 From: Gilles Chanteperdrix MIME-Version: 1.0 References: <4E54EAE4.2060705@domain.hid> <4E54F5B4.5070100@domain.hid> <4E552695.7090401@domain.hid> In-Reply-To: Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai-help] Hanging List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: suresh reddy Cc: Xenomai help On 08/25/2011 11:25 AM, suresh reddy wrote: > Dear Giles, It is Gilles. > > 1. taking timing results using rt_timer-tsc on pandaboard for a test program > using xenomai native services. > > 2. taking timing results using gettimeofday on pandaboard for same test > program using linuxthreads. > > > Can i compare these two timing results ? your system is likely using nptl, not linuxthreads. And clock_gettime instead of gettimeofday will give you precision down to the nanosecond, like rt_timer_tsc2ns. But yes, you can compare the results, if you bear in mind that real-time is about worst case latencies, so results obtained by running tests for a few seconds or minutes on an idle machine are meaningless. Also note that since Xenomai has a posix api, you can have the exact same code running with and without xenomai. -- Gilles.