From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <476A4F33.2050107@domain.hid> Date: Thu, 20 Dec 2007 12:17:07 +0100 From: Philippe Gerum MIME-Version: 1.0 References: <4B2EEE69E41E524EB7FCDC62A15D68D8018CFEBF@ARVMAIL1.mra.roland-man.biz> In-Reply-To: <4B2EEE69E41E524EB7FCDC62A15D68D8018CFEBF@ARVMAIL1.mra.roland-man.biz> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: Philippe Gerum Subject: Re: [Xenomai-help] Uniqueness of registry objects Reply-To: rpm@xenomai.org List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Thomas.Debes@domain.hid Cc: xenomai@xenomai.org Thomas.Debes@domain.hid wrote: > Hi, > > I am using the native skin together with mutexes and message queues. If > I use the same name for both objects (mutex="test" , queuename="test") I > can not register the second one because it fails with error -17 (file > exists) in rt_queue_create. I assume that the registry service doesn't > make a difference between the various RT object types. Can you confirm > this behaviour? Yes. If it is by design would it be better to differ between > the various RT object types (mutex, queue, pipe etc.)? > This uniqueness was deliberately required so that we could implement remote object registration in the future, for supporting distributed systems. In such a case, it's better that peers do not have to know about each other's exact nature. In this case, one would just want to establish a link between peers based on a common protocol, regardless of what actually implements each communication endpoint. -- Philippe.