From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tim Sander Subject: Re: 3.12.6-rt9 on AM335x (armv7) Date: Tue, 07 Jan 2014 23:58:57 +0100 Message-ID: <10785809.PZUZGWPMgY@hydra> References: <52C44448.3030405@ma-fu.de> <15611103.51LOkxSQt9@dabox> <52CC6F94.6070906@ma-fu.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-rt-users To: Matthias Fuchs Return-path: Received: from lvps176-28-13-145.dedicated.hosteurope.de ([176.28.13.145]:41266 "EHLO lvps176-28-13-145.dedicated.hosteurope.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753007AbaAGW5q convert rfc822-to-8bit (ORCPT ); Tue, 7 Jan 2014 17:57:46 -0500 In-Reply-To: <52CC6F94.6070906@ma-fu.de> Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hi Matthias > without hackbench and frieds (idle system) I get a max. latency of ab= out > 4ms. Running hackbench (default arguments, -l 1000), it still increa= ses up > to 5ms: >=20 > ... > # Total: 000000193 > # Min Latencies: 00024 > # Avg Latencies: 00205 > # Max Latencies: 05058 > ... >=20 > The obvious PM stuff is disabled. No dynamic frequency scaling, etc. > Really weired stuff. I already mentioned the 3.2 system with the same= CPU in > the OSADL lab. It comes with max. latency of 118us. So I really expec= t > something similiar. Yeah 118 =B5s should be the right ballpark, while still a little high f= or my=20 taste. All cortex a8 seem to have rather high latency. The A9's seem mu= ch=20 better in this respect. Have you seen the following: https://www.osadl.org/Realtime-Preempt-Kernel.kernel-rt.0.html#latencyf= ighting Cyclictest has tracing support which might help finding the cause of th= e long=20 latency.=20 Best regards Tim -- To unsubscribe from this list: send the line "unsubscribe linux-rt-user= s" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html