From mboxrd@z Thu Jan 1 00:00:00 1970 From: Raz Subject: hpet0 versus hpet2 Date: Sun, 29 Jun 2014 15:49:31 +0300 Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 To: linux-rt-users Return-path: Received: from mail-ob0-f178.google.com ([209.85.214.178]:49084 "EHLO mail-ob0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752600AbaF2Mtw (ORCPT ); Sun, 29 Jun 2014 08:49:52 -0400 Received: by mail-ob0-f178.google.com with SMTP id wn1so7574983obc.37 for ; Sun, 29 Jun 2014 05:49:51 -0700 (PDT) Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hey Rt folks I performed a benchmark that compares the jitter of hpet2 to hpet0. hpet0 is reserved to the kernel use ( hpet_reserve_timer(&hd, 0). and hd.hd_irq[0] = HPET_LEGACY_8254; in arch/x86/kernel/hpet.c). I removed these two lines. The bellow is a histogram compares the default hpet2 to hpet0 in atom D525. The test performed is a simple udp sender, it sends a single udp packet each 1ms interval. Time stamping were made by Beckhoff ET2000 ,so that the timestamps taken by a reliable and accurate tool. Question is whether it is possible to push a new kernel boot command line option that gives the user hpet0 ? hpet0 -50 0 -40 0 -20 2 -10 14 -5 143 -1 3992 1 95217 5 544 10 74 20 14 30 0 40 0 50 0 More 0 hpet2 Bin Frequency -50 1 -40 0 -20 0 -10 0 -5 0 -1 1 1 18 5 71 10 854 20 93098 30 237 40 1069 50 0 More 0