From mboxrd@z Thu Jan 1 00:00:00 1970 From: Venkat Subbiah Subject: Re: task switch from net-rx to idle when there is napi processing to be done Date: Thu, 26 Jan 2012 12:35:40 -0800 Message-ID: <4F21B91C.7080206@caviumnetworks.com> References: <4F1F6B66.5010005@caviumnetworks.com> <20120125085523.GG6305@pengutronix.de> <4F1FE7EB.3030601@caviumnetworks.com> <1327591757.22710.101.camel@gandalf.stny.rr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "Subbiah, Venkat" , =?ISO-8859-15?Q?Uwe_Kleine-K=F6nig?= , RT To: Steven Rostedt Return-path: Received: from mail3.caviumnetworks.com ([12.108.191.235]:8796 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752186Ab2AZUfl (ORCPT ); Thu, 26 Jan 2012 15:35:41 -0500 In-Reply-To: <1327591757.22710.101.camel@gandalf.stny.rr.com> Sender: linux-rt-users-owner@vger.kernel.org List-ID: On 01/26/2012 07:29 AM, Steven Rostedt wrote: > On Wed, 2012-01-25 at 03:30 -0800, Venkat Subbiah wrote: >> On 01/25/2012 12:55 AM, Uwe Kleine-K=F6nig wrote: >>> Hello, >>> >>> On Tue, Jan 24, 2012 at 06:39:34PM -0800, Venkat Subbiah wrote: >>>> In the process of debugging a napi ethernet driver performance >>>> issue, what I am noticing is >>>> >>>> 1. While the driver is in the middle of a napi packet processing >>>> loop, there is a task switch from >>>> sirq-net-rx to idle even though there is pending napi processing t= o be done. >>> I didn't check your logs below, but maybe this is related to the de= fault >>> settings in /proc/sys/kernel/sched_rt_period_us and >>> /proc/sys/kernel/sched_rt_runtime_us? That is 0.05s per second is >>> reserved for non-RT tasks tasks such that a run-away realtime proce= ss >>> will not lock up the machine. >>> >>> To verify that, try >>> >>> echo -1> /proc/sys/kernel/sched_rt_runtime_us >>> >>> . >> Thanks for you response. That was it. Setting this to -1 does the ex= pected. > Note, you should also have seen a warning in the logs when an RT task= is > throttled. Did you see such a thing? I didn't get to the console,but it is in the kernel logs. > > -- Steve > > > -- 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