From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id AC133C433B4 for ; Wed, 14 Apr 2021 15:53:40 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 7AF2E6120E for ; Wed, 14 Apr 2021 15:53:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1351012AbhDNPxr (ORCPT ); Wed, 14 Apr 2021 11:53:47 -0400 Received: from mga05.intel.com ([192.55.52.43]:37627 "EHLO mga05.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1352442AbhDNPxK (ORCPT ); Wed, 14 Apr 2021 11:53:10 -0400 IronPort-SDR: YgqaTmEZuTha6PLsJKYH54ZctbJlusWAn83cXt4+Kj2v1Qvy7MJQndVRBlCRpR3OEIMu68uDdH 3AMy4gGLVN9A== X-IronPort-AV: E=McAfee;i="6200,9189,9954"; a="279979739" X-IronPort-AV: E=Sophos;i="5.82,222,1613462400"; d="scan'208";a="279979739" Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Apr 2021 08:51:02 -0700 IronPort-SDR: jfbCfrwYhwQxqKkQsYOLHpPCHplyHhu3S+5x/t1gLornIi/I/MxrFhHJPPky0gB+0+EkWgCKi5 tbU7Z6IyPmjQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.82,222,1613462400"; d="scan'208";a="399226043" Received: from black.fi.intel.com (HELO black.fi.intel.com.) ([10.237.72.28]) by orsmga002.jf.intel.com with ESMTP; 14 Apr 2021 08:50:59 -0700 From: Alexander Shishkin To: Peter Zijlstra , Arnaldo Carvalho de Melo , adrian.hunter@intel.com Cc: Ingo Molnar , linux-kernel@vger.kernel.org, Jiri Olsa , Mathieu Poirier , Alexander Shishkin Subject: [PATCH v1 2/2] perf intel-pt: Use aux_watermark Date: Wed, 14 Apr 2021 18:49:55 +0300 Message-Id: <20210414154955.49603-3-alexander.shishkin@linux.intel.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20210414154955.49603-1-alexander.shishkin@linux.intel.com> References: <20210414154955.49603-1-alexander.shishkin@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Turns out, the default setting of attr.aux_watermark to half of the total buffer size is not very useful, especially with smaller buffers. The problem is that, after half of the buffer is filled up, the kernel updates ->aux_head and sets up the next "transaction", while observing that ->aux_tail is still zero (as userspace haven't had the chance to update it), meaning that the trace will have to stop at the end of this second "transaction". This means, for example, that the second PERF_RECORD_AUX in every trace comes with TRUNCATED flag set. Setting attr.aux_watermark to quarter of the buffer gives enough space for the ->aux_tail update to be observed and prevents the data loss. The obligatory before/after showcase: > # perf_before record -e intel_pt//u -m,8 uname > Linux > [ perf record: Woken up 6 times to write data ] > Warning: > AUX data lost 4 times out of 10! > > [ perf record: Captured and wrote 0.099 MB perf.data ] > # perf record -e intel_pt//u -m,8 uname > Linux > [ perf record: Woken up 4 times to write data ] > [ perf record: Captured and wrote 0.039 MB perf.data ] The effect is still visible with large workloads and large buffers, although less pronounced. Signed-off-by: Alexander Shishkin --- tools/perf/arch/x86/util/intel-pt.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index a6420c647959..6df0dc00d73a 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -776,6 +776,12 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, } } + if (!opts->auxtrace_snapshot_mode && !opts->auxtrace_sample_mode) { + u32 aux_watermark = opts->auxtrace_mmap_pages * page_size / 4; + + intel_pt_evsel->core.attr.aux_watermark = aux_watermark; + } + intel_pt_parse_terms(intel_pt_pmu->name, &intel_pt_pmu->format, "tsc", &tsc_bit); -- 2.30.2