From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753607AbdF3W0r (ORCPT ); Fri, 30 Jun 2017 18:26:47 -0400 Received: from mail.kernel.org ([198.145.29.99]:60354 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753543AbdF3W0n (ORCPT ); Fri, 30 Jun 2017 18:26:43 -0400 DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 0B53C22BDB Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.org Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=acme@kernel.org From: Arnaldo Carvalho de Melo To: Ingo Molnar Cc: linux-kernel@vger.kernel.org, Adrian Hunter , Andi Kleen , Arnaldo Carvalho de Melo Subject: [PATCH 29/30] perf intel-pt: Do not use TSC packets for calculating CPU cycles to TSC Date: Fri, 30 Jun 2017 19:25:13 -0300 Message-Id: <20170630222514.11832-30-acme@kernel.org> X-Mailer: git-send-email 2.9.4 In-Reply-To: <20170630222514.11832-1-acme@kernel.org> References: <20170630222514.11832-1-acme@kernel.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Adrian Hunter CBR (core-to-bus ratio) packets provide an indication of CPU frequency. A more accurate measure can be made by counting the cycles (given by CYC packets) in between other timing packets (either MTC or TSC). Using TSC packets has at least 2 issues: 1) timing might have stopped (e.g. mwait) or 2) TSC packets within PSB+ might slip past CYC packets. For now, simply do not use TSC packets for calculating CPU cycles to TSC. That leaves the case where 2 MTC packets are used, otherwise falling back to the CBR value. Signed-off-by: Adrian Hunter Cc: Andi Kleen Link: http://lkml.kernel.org/r/1495786658-18063-37-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/intel-pt-decoder/intel-pt-decoder.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c index 5dea06289db5..aa1593ce551d 100644 --- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c @@ -711,6 +711,12 @@ static int intel_pt_calc_cyc_cb(struct intel_pt_pkt_info *pkt_info) break; case INTEL_PT_TSC: + /* + * For now, do not support using TSC packets - refer + * intel_pt_calc_cyc_to_tsc(). + */ + if (data->from_mtc) + return 1; timestamp = pkt_info->packet.payload | (data->timestamp & (0xffULL << 56)); if (data->from_mtc && timestamp < data->timestamp && @@ -828,6 +834,14 @@ static void intel_pt_calc_cyc_to_tsc(struct intel_pt_decoder *decoder, .cbr_cyc_to_tsc = 0, }; + /* + * For now, do not support using TSC packets for at least the reasons: + * 1) timing might have stopped + * 2) TSC packets within PSB+ can slip against CYC packets + */ + if (!from_mtc) + return; + intel_pt_pkt_lookahead(decoder, intel_pt_calc_cyc_cb, &data); } -- 2.9.4