From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: Using jiffies for tcp_time_stamp? Date: Fri, 25 May 2012 21:00:32 +0200 Message-ID: <1337972432.3347.70.camel@edumazet-glaptop> References: <4FBFAC30.8050508@genband.com> <1337964880.3347.52.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Dave Taht , Chris Friesen , netdev@vger.kernel.org To: =?UTF-8?Q?Sre=C4=87ko_Juri=C4=87-Kavelj?= Return-path: Received: from mail-ey0-f174.google.com ([209.85.215.174]:49183 "EHLO mail-ey0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755467Ab2EYTAh (ORCPT ); Fri, 25 May 2012 15:00:37 -0400 Received: by eaak11 with SMTP id k11so319760eaa.19 for ; Fri, 25 May 2012 12:00:36 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 2012-05-25 at 20:35 +0200, Sre=C4=87ko Juri=C4=87-Kavelj wrote: > From what I've seen in the code, NO_HZ doesn't make jiffies go away, > it simply doesn't use regular CONFIG_HZ interrupt to update, but > updates them when has an opportunity? HZ=3D1000 makes jiffies 10 times more precise, and with NO_HZ, generate= s no extra timer interrupts. This also makes timers workload smoothed, instead of spikes.