From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Schnell Subject: Re: AMP on an SMP system Date: Fri, 02 Aug 2013 14:13:21 +0200 Message-ID: <51FBA261.10301@lumino.de> References: <51FB6EE1.3090708@lumino.de> <20130802114225.GR3880@pengutronix.de> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <20130802114225.GR3880@pengutronix.de> Sender: linux-embedded-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="iso-8859-1"; format="flowed" To: Robert Schwebel Cc: linux-embedded@vger.kernel.org On 08/02/2013 01:42 PM, Robert Schwebel wrote: > Before hacking around (which might also lead to interesting solutions= ), > I would start using a kernel with preempt-rt support and play with th= e > cpu affinity: > > http://lxr.linux.no/#linux+v3.10.4/Documentation/kernel-parameters.tx= t#L1257 > Robert ! Nice to see you here (I do own your "Embedded Linux Handbuch f=FCr=20 Entwickler" :-) ) Thanks for the pointer ! I do already know "preempt-rt", but I was not aware of cpu affinity. So this might help. In fact I need a way to do very guaranteed low latency. regarding the=20 high clock rate (about 1 GHz) modern ARM chips can provide, maybe=20 preempt-rt with the cpu affinity might be a decent way to go. The raining questions include - how to calculate the maximum latency that can be guaranteed ? (i.e.= =20 does the Kernel impose any spinlocks and interrupt disables on the woul= d=20 be AMP subsystem ?) - how to assign an interrupt (e.g. a dedicated timer) to the subsyste= m ? - Do the interrupts immediately call the ISR of the cpu "under=20 affinity" or is some additional latency imposed by the Kernel (and how=20 many cpu cycles at max are needed to enter the ISR) ? Thanks, -Michael