From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gilles Chanteperdrix Subject: Re: i.MX28 milliseconds latencies, interrupts disabled in arch_cpu_idle? Date: Sat, 19 Apr 2014 12:36:07 +0200 Message-ID: <53525197.5040904@xenomai.org> References: <53503BD8.8000506@meduna.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "linux-rt-users@vger.kernel.org" To: Stanislav Meduna Return-path: Received: from sd-51317.dedibox.fr ([62.210.215.82]:60493 "EHLO sd-51317.dedibox.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751043AbaDSKvs (ORCPT ); Sat, 19 Apr 2014 06:51:48 -0400 In-Reply-To: <53503BD8.8000506@meduna.org> Sender: linux-rt-users-owner@vger.kernel.org List-ID: Le 17/04/2014 22:38, Stanislav Meduna a =C3=A9crit : > Hi, >=20 > on a i.MX28: >=20 > Linux nxt 3.12.15-rt25+ #83 PREEMPT RT Thu Apr 17 20:54:00 CEST 2014 = armv5tejl GNU/Linux > > (...) > > Maybe it is just some tracing infrastructure interference - the effec= t > of enabling the tracing is huge on this platform for some reason. > Just enabling the irqsoff tracing makes the machine unsuitable > for any serious load. =46or the record, latencies on imx28 are high because: 1- it has an arm926 core, which uses a VIVT cache, and Linux chooses to flush cache at each mm switch to avoid aliasing issues; 2- the arm926 idle routine, in order to work around a hardware bug flushes the instruction cache before calling the "wfi" instruction. #2 can be avoided by using nohlt #1 can be avoided by using the ARM FCSE extension. A patch exists for #1 which is currently integrated into the Adeos patc= h but could exist independently. I could post the patch here for Linux vanilla here if there is enough interest, and let you do the merge with the preempt_rt patch. Regards. --=20 Gilles. -- 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