From patchwork Tue Oct 26 22:06:12 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585783 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 136CCC433F5 for ; Tue, 26 Oct 2021 22:07:13 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id E2FD9610A0 for ; Tue, 26 Oct 2021 22:07:12 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237361AbhJZWJf convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:09:35 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:33134 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237407AbhJZWJd (ORCPT ); Tue, 26 Oct 2021 18:09:33 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-588-N8sjzlchMXas3WNbjyqc4g-1; Tue, 26 Oct 2021 18:07:05 -0400 X-MC-Unique: N8sjzlchMXas3WNbjyqc4g-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 65D9C36303; Tue, 26 Oct 2021 22:07:03 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id B48C760BF1; Tue, 26 Oct 2021 22:06:54 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 01/20] trace/osnoise: Do not follow tracing_cpumask Date: Wed, 27 Oct 2021 00:06:12 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org In preparation to support multiple instances, decouple the osnoise/timelat workload from instance-specific tracing_cpumask. Different instances can have conflicting cpumasks, making osnoise workload management needlessly complex. Osnoise already has its global cpumask. I also thought about using the first instance mask, but the "first" instance could be removed before the others. This also fixes the problem that changing the tracing_mask was not re-starting the trace. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 24 ++++++------------------ 1 file changed, 6 insertions(+), 18 deletions(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index ce053619f289..73f9609cba3d 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1553,13 +1553,9 @@ static int start_per_cpu_kthreads(struct trace_array *tr) cpus_read_lock(); /* - * Run only on CPUs in which trace and osnoise are allowed to run. + * Run only on online CPUs in which osnoise is allowed to run. */ - cpumask_and(current_mask, tr->tracing_cpumask, &osnoise_cpumask); - /* - * And the CPU is online. - */ - cpumask_and(current_mask, cpu_online_mask, current_mask); + cpumask_and(current_mask, cpu_online_mask, &osnoise_cpumask); for_each_possible_cpu(cpu) per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL; @@ -1580,10 +1576,8 @@ static int start_per_cpu_kthreads(struct trace_array *tr) #ifdef CONFIG_HOTPLUG_CPU static void osnoise_hotplug_workfn(struct work_struct *dummy) { - struct trace_array *tr = osnoise_trace; unsigned int cpu = smp_processor_id(); - mutex_lock(&trace_types_lock); if (!osnoise_busy) @@ -1595,9 +1589,6 @@ static void osnoise_hotplug_workfn(struct work_struct *dummy) if (!cpumask_test_cpu(cpu, &osnoise_cpumask)) goto out_unlock; - if (!cpumask_test_cpu(cpu, tr->tracing_cpumask)) - goto out_unlock; - start_kthread(cpu); out_unlock: @@ -1700,13 +1691,10 @@ static void osnoise_tracer_stop(struct trace_array *tr); * interface to the osnoise trace. By default, it lists all CPUs, * in this way, allowing osnoise threads to run on any online CPU * of the system. It serves to restrict the execution of osnoise to the - * set of CPUs writing via this interface. Note that osnoise also - * respects the "tracing_cpumask." Hence, osnoise threads will run only - * on the set of CPUs allowed here AND on "tracing_cpumask." Why not - * have just "tracing_cpumask?" Because the user might be interested - * in tracing what is running on other CPUs. For instance, one might - * run osnoise in one HT CPU while observing what is running on the - * sibling HT CPU. + * set of CPUs writing via this interface. Why not use "tracing_cpumask"? + * Because the user might be interested in tracing what is running on + * other CPUs. For instance, one might run osnoise in one HT CPU + * while observing what is running on the sibling HT CPU. */ static ssize_t osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, From patchwork Tue Oct 26 22:06:13 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585785 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E10A9C433EF for ; Tue, 26 Oct 2021 22:07:57 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id C1F4760F02 for ; Tue, 26 Oct 2021 22:07:57 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237390AbhJZWKV convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:21 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:29130 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237318AbhJZWKR (ORCPT ); Tue, 26 Oct 2021 18:10:17 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-47--2cc785WMvKh1kmWmATw4g-1; Tue, 26 Oct 2021 18:07:47 -0400 X-MC-Unique: -2cc785WMvKh1kmWmATw4g-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id AD978802B7A; Tue, 26 Oct 2021 22:07:45 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 02CE260BF1; Tue, 26 Oct 2021 22:07:03 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 02/20] trace/osnoise: Improve comments about barrier need for NMI callbacks Date: Wed, 27 Oct 2021 00:06:13 +0200 Message-Id: <17c9b30b50e304269f08be3cbf8ca8188d2ebf7e.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org trace_osnoise_callback_enabled is used by ftrace_nmi_enter/exit() to know when to call the NMI callback. The barrier is used to avoid having callbacks enabled before the resetting date during the start or to touch the values after stopping the tracer. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Suggested-by: Steven Rostedt Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 73f9609cba3d..f941a322ca89 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1929,8 +1929,10 @@ static int __osnoise_tracer_start(struct trace_array *tr) retval = osnoise_hook_events(); if (retval) return retval; + /* - * Make sure NMIs see reseted values. + * Make sure that ftrace_nmi_enter/exit() see reset values + * before enabling trace_osnoise_callback_enabled. */ barrier(); trace_osnoise_callback_enabled = true; @@ -1965,6 +1967,10 @@ static void osnoise_tracer_stop(struct trace_array *tr) return; trace_osnoise_callback_enabled = false; + /* + * Make sure that ftrace_nmi_enter/exit() see + * trace_osnoise_callback_enabled as false before continuing. + */ barrier(); stop_per_cpu_kthreads(); From patchwork Tue Oct 26 22:06:14 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585787 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 58FC6C433F5 for ; Tue, 26 Oct 2021 22:08:05 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 37D2B608FE for ; Tue, 26 Oct 2021 22:08:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239781AbhJZWK2 convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:28 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:22679 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230336AbhJZWK0 (ORCPT ); Tue, 26 Oct 2021 18:10:26 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-131-6_VYi_kRPae59H1WBjF_gQ-1; Tue, 26 Oct 2021 18:07:58 -0400 X-MC-Unique: 6_VYi_kRPae59H1WBjF_gQ-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 5E28A8797E2; Tue, 26 Oct 2021 22:07:56 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 21EC260243; Tue, 26 Oct 2021 22:07:45 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 03/20] trace/osnoise: Split workload start from the tracer start Date: Wed, 27 Oct 2021 00:06:14 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org In preparation from supporting multiple trace instances, create workload start/stop specific functions. No functional change. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 59 ++++++++++++++++++++++-------------- 1 file changed, 36 insertions(+), 23 deletions(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index f941a322ca89..4381a06af916 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1545,7 +1545,7 @@ static int start_kthread(unsigned int cpu) * This starts the kernel thread that will look for osnoise on many * cpus. */ -static int start_per_cpu_kthreads(struct trace_array *tr) +static int start_per_cpu_kthreads(void) { struct cpumask *current_mask = &save_cpumask; int retval = 0; @@ -1677,8 +1677,8 @@ osnoise_cpus_read(struct file *filp, char __user *ubuf, size_t count, return count; } -static void osnoise_tracer_start(struct trace_array *tr); -static void osnoise_tracer_stop(struct trace_array *tr); +static int osnoise_workload_start(void); +static void osnoise_workload_stop(void); /* * osnoise_cpus_write - Write function for "cpus" entry @@ -1700,7 +1700,6 @@ static ssize_t osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, loff_t *ppos) { - struct trace_array *tr = osnoise_trace; cpumask_var_t osnoise_cpumask_new; int running, err; char buf[256]; @@ -1725,7 +1724,7 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, mutex_lock(&trace_types_lock); running = osnoise_busy; if (running) - osnoise_tracer_stop(tr); + osnoise_workload_stop(); mutex_lock(&interface_lock); /* @@ -1739,7 +1738,7 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, mutex_unlock(&interface_lock); if (running) - osnoise_tracer_start(tr); + osnoise_workload_start(); mutex_unlock(&trace_types_lock); free_cpumask_var(osnoise_cpumask_new); @@ -1920,7 +1919,10 @@ static int osnoise_hook_events(void) return -EINVAL; } -static int __osnoise_tracer_start(struct trace_array *tr) +/* + * osnoise_workload_start - start the workload and hook to events + */ +static int osnoise_workload_start(void) { int retval; @@ -1937,7 +1939,7 @@ static int __osnoise_tracer_start(struct trace_array *tr) barrier(); trace_osnoise_callback_enabled = true; - retval = start_per_cpu_kthreads(tr); + retval = start_per_cpu_kthreads(); if (retval) { unhook_irq_events(); return retval; @@ -1948,20 +1950,10 @@ static int __osnoise_tracer_start(struct trace_array *tr) return 0; } -static void osnoise_tracer_start(struct trace_array *tr) -{ - int retval; - - if (osnoise_busy) - return; - - retval = __osnoise_tracer_start(tr); - if (retval) - pr_err(BANNER "Error starting osnoise tracer\n"); - -} - -static void osnoise_tracer_stop(struct trace_array *tr) +/* + * osnoise_workload_stop - stop the workload and unhook the events + */ +static void osnoise_workload_stop(void) { if (!osnoise_busy) return; @@ -1982,6 +1974,27 @@ static void osnoise_tracer_stop(struct trace_array *tr) osnoise_busy = false; } +static void osnoise_tracer_start(struct trace_array *tr) +{ + int retval; + + if (osnoise_busy) + return; + + retval = osnoise_workload_start(); + if (retval) + pr_err(BANNER "Error starting osnoise tracer\n"); + +} + +static void osnoise_tracer_stop(struct trace_array *tr) +{ + if (!osnoise_busy) + return; + + osnoise_workload_stop(); +} + static int osnoise_tracer_init(struct trace_array *tr) { @@ -2022,7 +2035,7 @@ static void timerlat_tracer_start(struct trace_array *tr) osnoise_data.timerlat_tracer = 1; - retval = __osnoise_tracer_start(tr); + retval = osnoise_workload_start(); if (retval) goto out_err; From patchwork Tue Oct 26 22:06:15 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585789 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 047C3C433F5 for ; Tue, 26 Oct 2021 22:08:08 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id E36E5608FE for ; Tue, 26 Oct 2021 22:08:07 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237681AbhJZWKb convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:31 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:43591 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239740AbhJZWKa (ORCPT ); Tue, 26 Oct 2021 18:10:30 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-97-R0QxaRbkM3KzzXqAIRtX7A-1; Tue, 26 Oct 2021 18:08:02 -0400 X-MC-Unique: R0QxaRbkM3KzzXqAIRtX7A-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id EC7CB8030A0; Tue, 26 Oct 2021 22:08:00 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id BCFE95FCCE; Tue, 26 Oct 2021 22:07:56 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 04/20] trace/osnoise: Use start/stop_per_cpu_kthreads() on osnoise_cpus_write() Date: Wed, 27 Oct 2021 00:06:15 +0200 Message-Id: <2b4ea45b1802844cffd98b6a208b22bc4664c795.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org When writing a new CPU mask via osnoise/cpus, if the tracer is running, the workload is restarted to follow the new cpumask. The restart is currently done using osnoise_workload_start/stop(), which disables the workload *and* the instrumentation. However, disabling the instrumentation is not necessary. Calling start/stop_per_cpu_kthreads() is enough to apply the new osnoise/cpus config. No functional change. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 4381a06af916..7453b96c548b 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1677,9 +1677,6 @@ osnoise_cpus_read(struct file *filp, char __user *ubuf, size_t count, return count; } -static int osnoise_workload_start(void); -static void osnoise_workload_stop(void); - /* * osnoise_cpus_write - Write function for "cpus" entry * @filp: The active open file structure @@ -1724,7 +1721,7 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, mutex_lock(&trace_types_lock); running = osnoise_busy; if (running) - osnoise_workload_stop(); + stop_per_cpu_kthreads(); mutex_lock(&interface_lock); /* @@ -1738,7 +1735,7 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, mutex_unlock(&interface_lock); if (running) - osnoise_workload_start(); + start_per_cpu_kthreads(); mutex_unlock(&trace_types_lock); free_cpumask_var(osnoise_cpumask_new); From patchwork Tue Oct 26 22:06:16 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585791 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 923C4C433F5 for ; Tue, 26 Oct 2021 22:08:14 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 7363160F02 for ; Tue, 26 Oct 2021 22:08:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239830AbhJZWKh convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:37 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:30176 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239804AbhJZWKe (ORCPT ); Tue, 26 Oct 2021 18:10:34 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-94-N-rtEgXXP4ucnIe4a3Tk_A-1; Tue, 26 Oct 2021 18:08:06 -0400 X-MC-Unique: N-rtEgXXP4ucnIe4a3Tk_A-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id DB6D0802B7A; Tue, 26 Oct 2021 22:08:04 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 56BB060BF1; Tue, 26 Oct 2021 22:08:01 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 05/20] trace/osnoise: Support a list of trace_array *tr Date: Wed, 27 Oct 2021 00:06:16 +0200 Message-Id: <96ce8e8e1ccb5df711e0f6e9c04392cda2782c4f.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org osnoise/timerlat were built to run a single instance, and for this, a single variable is enough to store the current struct trace_array *tr with information about the tracing instance. This is done via the *osnoise_trace variable. A trace_array represents a trace instance. In preparation to support multiple instances, replace the *osnoise_trace variable with an RCU protected list of instances. The operations that refer to an instance now propagate to all elements of the list (all instances). Also, replace the osnoise_busy variable with a check if the list has elements (busy). No functional change is expected with this patch, i.e., only one instance is allowed yet. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 261 ++++++++++++++++++++++++++--------- 1 file changed, 192 insertions(+), 69 deletions(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 7453b96c548b..3db506f49a90 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -38,8 +38,6 @@ #define CREATE_TRACE_POINTS #include -static struct trace_array *osnoise_trace; - /* * Default values. */ @@ -50,6 +48,81 @@ static struct trace_array *osnoise_trace; #define DEFAULT_TIMERLAT_PERIOD 1000 /* 1ms */ #define DEFAULT_TIMERLAT_PRIO 95 /* FIFO 95 */ +/* + * trace_array of the enabled osnoise/timerlat instances. + */ +struct osnoise_instance { + struct list_head list; + struct trace_array *tr; +}; +struct list_head osnoise_instances; + +static bool osnoise_has_registered_instances(void) +{ + return !!list_first_or_null_rcu(&osnoise_instances, + struct osnoise_instance, + list); +} + +/* + * osnoise_register_instance - register a new trace instance + * + * Register a trace_array *tr in the list of instances running + * osnoise/timerlat tracers. + */ +static int osnoise_register_instance(struct trace_array *tr) +{ + struct osnoise_instance *inst; + + /* + * register/unregister serialization is provided by trace's + * trace_types_lock. + */ + lockdep_assert_held(&trace_types_lock); + + inst = kmalloc(sizeof(*inst), GFP_KERNEL); + if (!inst) + return -ENOMEM; + + INIT_LIST_HEAD_RCU(&inst->list); + inst->tr = tr; + list_add_tail_rcu(&inst->list, &osnoise_instances); + + return 0; +} + +/* + * osnoise_unregister_instance - unregister a registered trace instance + * + * Remove the trace_array *tr from the list of instances running + * osnoise/timerlat tracers. + */ +static void osnoise_unregister_instance(struct trace_array *tr) +{ + struct osnoise_instance *inst; + int found = 0; + + /* + * register/unregister serialization is provided by trace's + * trace_types_lock. + */ + lockdep_assert_held(&trace_types_lock); + + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + if (inst->tr == tr) { + list_del_rcu(&inst->list); + found = 1; + break; + } + } + + if (!found) + return; + + synchronize_rcu(); + kfree(inst); +} + /* * NMI runtime info. */ @@ -248,11 +321,6 @@ static struct osnoise_data { #endif }; -/* - * Boolean variable used to inform that the tracer is currently sampling. - */ -static bool osnoise_busy; - #ifdef CONFIG_PREEMPT_RT /* * Print the osnoise header info. @@ -315,19 +383,24 @@ static void print_osnoise_headers(struct seq_file *s) * osnoise_taint - report an osnoise error. */ #define osnoise_taint(msg) ({ \ - struct trace_array *tr = osnoise_trace; \ + struct osnoise_instance *inst; \ + struct trace_buffer *buffer; \ \ - trace_array_printk_buf(tr->array_buffer.buffer, _THIS_IP_, msg); \ + rcu_read_lock(); \ + list_for_each_entry_rcu(inst, &osnoise_instances, list) { \ + buffer = inst->tr->array_buffer.buffer; \ + trace_array_printk_buf(buffer, _THIS_IP_, msg); \ + } \ + rcu_read_unlock(); \ osnoise_data.tainted = true; \ }) /* * Record an osnoise_sample into the tracer buffer. */ -static void trace_osnoise_sample(struct osnoise_sample *sample) +static void +__trace_osnoise_sample(struct osnoise_sample *sample, struct trace_buffer *buffer) { - struct trace_array *tr = osnoise_trace; - struct trace_buffer *buffer = tr->array_buffer.buffer; struct trace_event_call *call = &event_osnoise; struct ring_buffer_event *event; struct osnoise_entry *entry; @@ -350,6 +423,22 @@ static void trace_osnoise_sample(struct osnoise_sample *sample) trace_buffer_unlock_commit_nostack(buffer, event); } +/* + * Record an osnoise_sample on all osnoise instances. + */ +static void trace_osnoise_sample(struct osnoise_sample *sample) +{ + struct osnoise_instance *inst; + struct trace_buffer *buffer; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + buffer = inst->tr->array_buffer.buffer; + __trace_osnoise_sample(sample, buffer); + } + rcu_read_unlock(); +} + #ifdef CONFIG_TIMERLAT_TRACER /* * Print the timerlat header info. @@ -387,14 +476,10 @@ static void print_timerlat_headers(struct seq_file *s) } #endif /* CONFIG_PREEMPT_RT */ -/* - * Record an timerlat_sample into the tracer buffer. - */ -static void trace_timerlat_sample(struct timerlat_sample *sample) +static void +__trace_timerlat_sample(struct timerlat_sample *sample, struct trace_buffer *buffer) { - struct trace_array *tr = osnoise_trace; struct trace_event_call *call = &event_osnoise; - struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct timerlat_entry *entry; @@ -411,6 +496,22 @@ static void trace_timerlat_sample(struct timerlat_sample *sample) trace_buffer_unlock_commit_nostack(buffer, event); } +/* + * Record an timerlat_sample into the tracer buffer. + */ +static void trace_timerlat_sample(struct timerlat_sample *sample) +{ + struct osnoise_instance *inst; + struct trace_buffer *buffer; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + buffer = inst->tr->array_buffer.buffer; + __trace_timerlat_sample(sample, buffer); + } + rcu_read_unlock(); +} + #ifdef CONFIG_STACKTRACE #define MAX_CALLS 256 @@ -450,29 +551,18 @@ static void timerlat_save_stack(int skip) return; } -/* - * timerlat_dump_stack - dump a stack trace previously saved - * - * Dump a saved stack trace into the trace buffer. - */ -static void timerlat_dump_stack(void) + +static void +__timerlat_dump_stack(struct trace_buffer *buffer, struct trace_stack *fstack, unsigned int size) { struct trace_event_call *call = &event_osnoise; - struct trace_array *tr = osnoise_trace; - struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; - struct trace_stack *fstack; struct stack_entry *entry; - unsigned int size; - - preempt_disable_notrace(); - fstack = this_cpu_ptr(&trace_stack); - size = fstack->stack_size; event = trace_buffer_lock_reserve(buffer, TRACE_STACK, sizeof(*entry) + size, tracing_gen_ctx()); if (!event) - goto out; + return; entry = ring_buffer_event_data(event); @@ -481,8 +571,29 @@ static void timerlat_dump_stack(void) if (!call_filter_check_discard(call, entry, buffer, event)) trace_buffer_unlock_commit_nostack(buffer, event); +} -out: +/* + * timerlat_dump_stack - dump a stack trace previously saved + */ +static void timerlat_dump_stack(void) +{ + struct osnoise_instance *inst; + struct trace_buffer *buffer; + struct trace_stack *fstack; + unsigned int size; + + preempt_disable_notrace(); + fstack = this_cpu_ptr(&trace_stack); + size = fstack->stack_size; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + buffer = inst->tr->array_buffer.buffer; + __timerlat_dump_stack(buffer, fstack, size); + + } + rcu_read_unlock(); preempt_enable_notrace(); } #else @@ -1077,12 +1188,37 @@ diff_osn_sample_stats(struct osnoise_variables *osn_var, struct osnoise_sample * */ static __always_inline void osnoise_stop_tracing(void) { - struct trace_array *tr = osnoise_trace; + struct osnoise_instance *inst; + struct trace_array *tr; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + tr = inst->tr; + trace_array_printk_buf(tr->array_buffer.buffer, _THIS_IP_, + "stop tracing hit on cpu %d\n", smp_processor_id()); + + tracer_tracing_off(tr); + } + rcu_read_unlock(); +} - trace_array_printk_buf(tr->array_buffer.buffer, _THIS_IP_, - "stop tracing hit on cpu %d\n", smp_processor_id()); +/* + * notify_new_max_latency - Notify a new max latency via fsnotify interface. + */ +void notify_new_max_latency(u64 latency) +{ + struct osnoise_instance *inst; + struct trace_array *tr; - tracer_tracing_off(tr); + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + tr = inst->tr; + if (tr->max_latency < latency) { + tr->max_latency = latency; + latency_fsnotify(tr); + } + } + rcu_read_unlock(); } /* @@ -1096,7 +1232,6 @@ static __always_inline void osnoise_stop_tracing(void) static int run_osnoise(void) { struct osnoise_variables *osn_var = this_cpu_osn_var(); - struct trace_array *tr = osnoise_trace; u64 start, sample, last_sample; u64 last_int_count, int_count; s64 noise = 0, max_noise = 0; @@ -1231,11 +1366,7 @@ static int run_osnoise(void) trace_osnoise_sample(&s); - /* Keep a running maximum ever recorded osnoise "latency" */ - if (max_noise > tr->max_latency) { - tr->max_latency = max_noise; - latency_fsnotify(tr); - } + notify_new_max_latency(max_noise); if (osnoise_data.stop_tracing_total) if (s.noise > osnoise_data.stop_tracing_total) @@ -1293,7 +1424,6 @@ static int osnoise_main(void *data) static enum hrtimer_restart timerlat_irq(struct hrtimer *timer) { struct osnoise_variables *osn_var = this_cpu_osn_var(); - struct trace_array *tr = osnoise_trace; struct timerlat_variables *tlat; struct timerlat_sample s; u64 now; @@ -1364,11 +1494,7 @@ static enum hrtimer_restart timerlat_irq(struct hrtimer *timer) trace_timerlat_sample(&s); - /* Keep a running maximum ever recorded os noise "latency" */ - if (diff > tr->max_latency) { - tr->max_latency = diff; - latency_fsnotify(tr); - } + notify_new_max_latency(diff); if (osnoise_data.stop_tracing) if (time_to_us(diff) >= osnoise_data.stop_tracing) @@ -1580,7 +1706,7 @@ static void osnoise_hotplug_workfn(struct work_struct *dummy) mutex_lock(&trace_types_lock); - if (!osnoise_busy) + if (!osnoise_has_registered_instances()) goto out_unlock_trace; mutex_lock(&interface_lock); @@ -1715,11 +1841,10 @@ osnoise_cpus_write(struct file *filp, const char __user *ubuf, size_t count, goto err_free; /* - * trace_types_lock is taken to avoid concurrency on start/stop - * and osnoise_busy. + * trace_types_lock is taken to avoid concurrency on start/stop. */ mutex_lock(&trace_types_lock); - running = osnoise_busy; + running = osnoise_has_registered_instances(); if (running) stop_per_cpu_kthreads(); @@ -1942,8 +2067,6 @@ static int osnoise_workload_start(void) return retval; } - osnoise_busy = true; - return 0; } @@ -1952,7 +2075,7 @@ static int osnoise_workload_start(void) */ static void osnoise_workload_stop(void) { - if (!osnoise_busy) + if (osnoise_has_registered_instances()) return; trace_osnoise_callback_enabled = false; @@ -1967,28 +2090,28 @@ static void osnoise_workload_stop(void) unhook_irq_events(); unhook_softirq_events(); unhook_thread_events(); - - osnoise_busy = false; } static void osnoise_tracer_start(struct trace_array *tr) { int retval; - if (osnoise_busy) + if (osnoise_has_registered_instances()) return; retval = osnoise_workload_start(); if (retval) pr_err(BANNER "Error starting osnoise tracer\n"); + osnoise_register_instance(tr); } static void osnoise_tracer_stop(struct trace_array *tr) { - if (!osnoise_busy) + if (!osnoise_has_registered_instances()) return; + osnoise_unregister_instance(tr); osnoise_workload_stop(); } @@ -1996,14 +2119,12 @@ static int osnoise_tracer_init(struct trace_array *tr) { /* Only allow one instance to enable this */ - if (osnoise_busy) + if (osnoise_has_registered_instances()) return -EBUSY; - osnoise_trace = tr; tr->max_latency = 0; osnoise_tracer_start(tr); - return 0; } @@ -2027,7 +2148,7 @@ static void timerlat_tracer_start(struct trace_array *tr) { int retval; - if (osnoise_busy) + if (osnoise_has_registered_instances()) return; osnoise_data.timerlat_tracer = 1; @@ -2036,6 +2157,8 @@ static void timerlat_tracer_start(struct trace_array *tr) if (retval) goto out_err; + osnoise_register_instance(tr); + return; out_err: pr_err(BANNER "Error starting timerlat tracer\n"); @@ -2045,7 +2168,7 @@ static void timerlat_tracer_stop(struct trace_array *tr) { int cpu; - if (!osnoise_busy) + if (!osnoise_has_registered_instances()) return; for_each_online_cpu(cpu) @@ -2059,11 +2182,9 @@ static void timerlat_tracer_stop(struct trace_array *tr) static int timerlat_tracer_init(struct trace_array *tr) { /* Only allow one instance to enable this */ - if (osnoise_busy) + if (osnoise_has_registered_instances()) return -EBUSY; - osnoise_trace = tr; - tr->max_latency = 0; timerlat_tracer_start(tr); @@ -2110,6 +2231,8 @@ __init static int init_osnoise_tracer(void) #endif osnoise_init_hotplug_support(); + INIT_LIST_HEAD_RCU(&osnoise_instances); + init_tracefs(); return 0; From patchwork Tue Oct 26 22:06:17 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585793 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2E5FDC433EF for ; Tue, 26 Oct 2021 22:08:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 14671610A6 for ; Tue, 26 Oct 2021 22:08:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239824AbhJZWKn convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:43 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:51453 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239807AbhJZWKj (ORCPT ); Tue, 26 Oct 2021 18:10:39 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-127-qk73EZyfMI-eBn9zjQAL5w-1; Tue, 26 Oct 2021 18:08:11 -0400 X-MC-Unique: qk73EZyfMI-eBn9zjQAL5w-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 7475A18D6A2C; Tue, 26 Oct 2021 22:08:09 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 5495D60BF1; Tue, 26 Oct 2021 22:08:05 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 06/20] trace/osnoise: Allow multiple instances of the same tracer Date: Wed, 27 Oct 2021 00:06:17 +0200 Message-Id: <69cbbd98cce2515c84127c8827d733dc87b04823.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Currently, the user can start only one instance of timerlat/osnoise tracers and the tracers cannot run in parallel. As starting point to add more flexibility, let's allow the same tracer to run on different trace instances. The workload will start when the first trace_array (instance) is registered and stop when the last instance is unregistered. So, while this patch allows the same tracer to run in multiple instances (e.g., two instances running osnoise), it still does not allow instances of timerlat and osnoise in parallel (e.g., one timerlat and osnoise). That is because the osnoise: events have different behavior depending on which tracer is enabled (osnoise or timerlat). Enabling the parallel usage of these two tracers is in my TODO list. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- kernel/trace/trace_osnoise.c | 101 +++++++++++++++++++++++++++-------- 1 file changed, 78 insertions(+), 23 deletions(-) diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 3db506f49a90..8681ffc3817b 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -64,6 +64,24 @@ static bool osnoise_has_registered_instances(void) list); } +/* + * osnoise_instance_registered - check if a tr is already registered + */ +static int osnoise_instance_registered(struct trace_array *tr) +{ + struct osnoise_instance *inst; + int found = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(inst, &osnoise_instances, list) { + if (inst->tr == tr) + found = 1; + } + rcu_read_unlock(); + + return found; +} + /* * osnoise_register_instance - register a new trace instance * @@ -2048,6 +2066,16 @@ static int osnoise_workload_start(void) { int retval; + /* + * Instances need to be registered after calling workload + * start. Hence, if there is already an instance, the + * workload was already registered. Otherwise, this + * code is on the way to register the first instance, + * and the workload will start. + */ + if (osnoise_has_registered_instances()) + return 0; + osn_var_reset_all(); retval = osnoise_hook_events(); @@ -2075,6 +2103,13 @@ static int osnoise_workload_start(void) */ static void osnoise_workload_stop(void) { + /* + * Instances need to be unregistered before calling + * stop. Hence, if there is a registered instance, more + * than one instance is running, and the workload will not + * yet stop. Otherwise, this code is on the way to disable + * the last instance, and the workload can stop. + */ if (osnoise_has_registered_instances()) return; @@ -2096,7 +2131,11 @@ static void osnoise_tracer_start(struct trace_array *tr) { int retval; - if (osnoise_has_registered_instances()) + /* + * If the instance is already registered, there is no need to + * register it again. + */ + if (osnoise_instance_registered(tr)) return; retval = osnoise_workload_start(); @@ -2108,18 +2147,17 @@ static void osnoise_tracer_start(struct trace_array *tr) static void osnoise_tracer_stop(struct trace_array *tr) { - if (!osnoise_has_registered_instances()) - return; - osnoise_unregister_instance(tr); osnoise_workload_stop(); } static int osnoise_tracer_init(struct trace_array *tr) { - - /* Only allow one instance to enable this */ - if (osnoise_has_registered_instances()) + /* + * Only allow osnoise tracer if timerlat tracer is not running + * already. + */ + if (osnoise_data.timerlat_tracer) return -EBUSY; tr->max_latency = 0; @@ -2148,45 +2186,55 @@ static void timerlat_tracer_start(struct trace_array *tr) { int retval; - if (osnoise_has_registered_instances()) + /* + * If the instance is already registered, there is no need to + * register it again. + */ + if (osnoise_instance_registered(tr)) return; - osnoise_data.timerlat_tracer = 1; - retval = osnoise_workload_start(); if (retval) - goto out_err; + pr_err(BANNER "Error starting timerlat tracer\n"); osnoise_register_instance(tr); return; -out_err: - pr_err(BANNER "Error starting timerlat tracer\n"); } static void timerlat_tracer_stop(struct trace_array *tr) { int cpu; - if (!osnoise_has_registered_instances()) - return; - - for_each_online_cpu(cpu) - per_cpu(per_cpu_osnoise_var, cpu).sampling = 0; + osnoise_unregister_instance(tr); - osnoise_tracer_stop(tr); + /* + * Instruct the threads to stop only if this is the last instance. + */ + if (!osnoise_has_registered_instances()) { + for_each_online_cpu(cpu) + per_cpu(per_cpu_osnoise_var, cpu).sampling = 0; + } - osnoise_data.timerlat_tracer = 0; + osnoise_workload_stop(); } static int timerlat_tracer_init(struct trace_array *tr) { - /* Only allow one instance to enable this */ - if (osnoise_has_registered_instances()) + /* + * Only allow timerlat tracer if osnoise tracer is not running already. + */ + if (osnoise_has_registered_instances() && !osnoise_data.timerlat_tracer) return -EBUSY; - tr->max_latency = 0; + /* + * If this is the first instance, set timerlat_tracer to block + * osnoise tracer start. + */ + if (!osnoise_has_registered_instances()) + osnoise_data.timerlat_tracer = 1; + tr->max_latency = 0; timerlat_tracer_start(tr); return 0; @@ -2195,6 +2243,13 @@ static int timerlat_tracer_init(struct trace_array *tr) static void timerlat_tracer_reset(struct trace_array *tr) { timerlat_tracer_stop(tr); + + /* + * If this is the last instance, reset timerlat_tracer allowing + * osnoise to be started. + */ + if (!osnoise_has_registered_instances()) + osnoise_data.timerlat_tracer = 0; } static struct tracer timerlat_tracer __read_mostly = { From patchwork Tue Oct 26 22:06:18 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585795 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id BE842C4332F for ; Tue, 26 Oct 2021 22:08:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id A75A560F02 for ; Tue, 26 Oct 2021 22:08:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239799AbhJZWKo convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:10:44 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:54623 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239771AbhJZWKl (ORCPT ); Tue, 26 Oct 2021 18:10:41 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-560-D-UFq02bPAeeDYeqE4ULnA-1; Tue, 26 Oct 2021 18:08:15 -0400 X-MC-Unique: D-UFq02bPAeeDYeqE4ULnA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id EEA08A40C1; Tue, 26 Oct 2021 22:08:13 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0DAB060BF1; Tue, 26 Oct 2021 22:08:09 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 07/20] rtla: Real-Time Linux Analysis tool Date: Wed, 27 Oct 2021 00:06:18 +0200 Message-Id: <60dbd3e36ac9489b6aadbc1c3d095608e6c7e4bb.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The rtla is a meta-tool that includes a set of commands that aims to analyze the real-time properties of Linux. But instead of testing Linux as a black box, rtla leverages kernel tracing capabilities to provide precise information about the properties and root causes of unexpected results. rtla --help works and provide information about the available options. This is just the "main" and the Makefile, no function yet. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/Makefile | 76 +++++++++++++++++++++++++++++++++++ tools/tracing/rtla/src/rtla.c | 72 +++++++++++++++++++++++++++++++++ 2 files changed, 148 insertions(+) create mode 100644 tools/tracing/rtla/Makefile create mode 100644 tools/tracing/rtla/src/rtla.c diff --git a/tools/tracing/rtla/Makefile b/tools/tracing/rtla/Makefile new file mode 100644 index 000000000000..525e15b76156 --- /dev/null +++ b/tools/tracing/rtla/Makefile @@ -0,0 +1,76 @@ +NAME := rtla +VERSION := 0.2 + +# From libtracefs: +# Makefiles suck: This macro sets a default value of $(2) for the +# variable named by $(1), unless the variable has been set by +# environment or command line. This is necessary for CC and AR +# because make sets default values, so the simpler ?= approach +# won't work as expected. +define allow-override + $(if $(or $(findstring environment,$(origin $(1))),\ + $(findstring command line,$(origin $(1)))),,\ + $(eval $(1) = $(2))) +endef + +# Allow setting CC and AR, or setting CROSS_COMPILE as a prefix. +$(call allow-override,CC,$(CROSS_COMPILE)gcc) +$(call allow-override,AR,$(CROSS_COMPILE)ar) +$(call allow-override,STRIP,$(CROSS_COMPILE)strip) +$(call allow-override,PKG_CONFIG,pkg-config) +$(call allow-override,LD_SO_CONF_PATH,/etc/ld.so.conf.d/) +$(call allow-override,LDCONFIG,ldconfig) + +INSTALL = install +FOPTS := -flto=auto -ffat-lto-objects -fexceptions -fstack-protector-strong \ + -fasynchronous-unwind-tables -fstack-clash-protection +WOPTS := -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -Wno-maybe-uninitialized + +TRACEFS_HEADERS := $$($(PKG_CONFIG) --cflags libtracefs) + +CFLAGS := -O -g -DVERSION=\"$(VERSION)\" $(FOPTS) $(MOPTS) $(WOPTS) $(TRACEFS_HEADERS) +LDFLAGS := -ggdb +LIBS := -ltracefs -ltraceevent -lpthread -lprocps + +SRC := $(wildcard src/*.c) +HDR := $(wildcard src/*.h) +OBJ := $(SRC:.c=.o) +DIRS := src +FILES := Makefile +CEXT := bz2 +TARBALL := $(NAME)-$(VERSION).tar.$(CEXT) +TAROPTS := -cvjf $(TARBALL) +BINDIR := /usr/bin +DATADIR := /usr/share +DOCDIR := $(DATADIR)/doc +MANDIR := $(DATADIR)/man +LICDIR := $(DATADIR)/licenses + +.PHONY: all +all: rtla + +rtla: $(OBJ) + $(CC) -o rtla $(LDFLAGS) $(OBJ) $(LIBS) + +static: $(OBJ) + $(CC) -o rtla-static $(LDFLAGS) --static $(OBJ) $(LIBS) -lpthread -ldl + +.PHONY: install +install: + $(INSTALL) -d -m 755 $(DESTDIR)$(BINDIR) + $(INSTALL) rtla -m 755 $(DESTDIR)$(BINDIR) + $(STRIP) $(DESTDIR)$(BINDIR)/rtla + +.PHONY: clean tarball +clean: + @test ! -f rtla || rm rtla + @test ! -f rtla-static || rm rtla-static + @test ! -f src/rtla.o || rm src/rtla.o + @test ! -f $(TARBALL) || rm -f $(TARBALL) + @rm -rf *~ $(OBJ) *.tar.$(CEXT) + +tarball: clean + rm -rf $(NAME)-$(VERSION) && mkdir $(NAME)-$(VERSION) + cp -r $(DIRS) $(FILES) $(NAME)-$(VERSION) + tar $(TAROPTS) --exclude='*~' $(NAME)-$(VERSION) + rm -rf $(NAME)-$(VERSION) diff --git a/tools/tracing/rtla/src/rtla.c b/tools/tracing/rtla/src/rtla.c new file mode 100644 index 000000000000..5ae2664ed47d --- /dev/null +++ b/tools/tracing/rtla/src/rtla.c @@ -0,0 +1,72 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include + +/* + * rtla_usage - print rtla usage + */ +static void rtla_usage(void) +{ + int i; + + static const char *msg[] = { + "", + "rtla version " VERSION, + "", + " usage: rtla COMMAND ...", + "", + " commands:", + "", + NULL, + }; + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +/* + * run_command - try to run a rtla tool command + * + * It returns 0 if it fails. The tool's main will generally not + * return as they should call exit(). + */ +int run_command(int argc, char **argv, int start_position) +{ + return 0; +} + +int main(int argc, char *argv[]) +{ + int retval; + + /* is it an alias? */ + retval = run_command(argc, argv, 0); + if (retval) + exit(0); + + if (argc < 2) + goto usage; + + if (strcmp(argv[1], "-h") == 0) { + rtla_usage(); + exit(0); + } else if (strcmp(argv[1], "--help") == 0) { + rtla_usage(); + exit(0); + } + + retval = run_command(argc, argv, 1); + if (retval) + exit(0); + +usage: + rtla_usage(); + exit(1); +} From patchwork Tue Oct 26 22:06:19 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585797 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 513D0C433F5 for ; Tue, 26 Oct 2021 22:08:42 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 36AA8610A4 for ; Tue, 26 Oct 2021 22:08:42 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239907AbhJZWLE convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:04 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:28201 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239844AbhJZWKs (ORCPT ); Tue, 26 Oct 2021 18:10:48 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-423-2cYykLBdNcCn8JY3oUX_aQ-1; Tue, 26 Oct 2021 18:08:21 -0400 X-MC-Unique: 2cYykLBdNcCn8JY3oUX_aQ-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id DF65F802B7E; Tue, 26 Oct 2021 22:08:19 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 8928860BF1; Tue, 26 Oct 2021 22:08:14 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 08/20] rtla: Helper functions for rtla Date: Wed, 27 Oct 2021 00:06:19 +0200 Message-Id: <183637700a31169e5d6d1c8a827c5365599ec806.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org This is a set of utils and tracer helper functions. They are used by rtla mostly to parse config, display data and some trace operations that are not part of libtracefs (because they are only useful it for this case). Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/src/trace.c | 192 +++++++++++++++ tools/tracing/rtla/src/trace.h | 27 ++ tools/tracing/rtla/src/utils.c | 433 +++++++++++++++++++++++++++++++++ tools/tracing/rtla/src/utils.h | 56 +++++ 4 files changed, 708 insertions(+) create mode 100644 tools/tracing/rtla/src/trace.c create mode 100644 tools/tracing/rtla/src/trace.h create mode 100644 tools/tracing/rtla/src/utils.c create mode 100644 tools/tracing/rtla/src/utils.h diff --git a/tools/tracing/rtla/src/trace.c b/tools/tracing/rtla/src/trace.c new file mode 100644 index 000000000000..ab11c29cdcc2 --- /dev/null +++ b/tools/tracing/rtla/src/trace.c @@ -0,0 +1,192 @@ +// SPDX-License-Identifier: GPL-2.0 +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include + +#include "trace.h" +#include "utils.h" + +/* + * enable_tracer_by_name - enable a tracer on the given instance + */ +int enable_tracer_by_name(struct tracefs_instance *inst, const char *tracer) +{ + enum tracefs_tracers t; + int retval; + + t = TRACEFS_TRACER_CUSTOM; + + debug_msg("enabling %s tracer\n", tracer); + + retval = tracefs_tracer_set(inst, t, tracer); + if (retval < 0) { + if (errno == ENODEV) + err_msg("tracer %s not found!\n", tracer); + + err_msg("failed to enable the tracer %s\n", tracer); + return -1; + } + + return 0; +} + +/* + * disable_tracer - set nop tracer to the insta + */ +void disable_tracer(struct tracefs_instance *inst) +{ + enum tracefs_tracers t = TRACEFS_TRACER_NOP; + int retval; + + retval = tracefs_tracer_set(inst, t); + if (retval < 0) + err_msg("oops, error disabling tracer\n"); +} + +/* + * create_instance - create a trace instance with *instance_name + */ +struct tracefs_instance *create_instance(char *instance_name) +{ + return tracefs_instance_create(instance_name); +} + +/* + * destroy_instance - remove a trace instance and free the data + */ +void destroy_instance(struct tracefs_instance *inst) +{ + tracefs_instance_destroy(inst); + tracefs_instance_free(inst); +} + +/* + * save_trace_to_file - save the trace output of the instance to the file + */ +int save_trace_to_file(struct tracefs_instance *inst, const char *filename) +{ + const char *file = "trace"; + mode_t mode = 0644; + char *buffer[4096]; + int out_fd, in_fd; + int retval = -1; + + in_fd = tracefs_instance_file_open(inst, file, O_RDONLY); + if (in_fd < 0) { + err_msg("Failed to open trace file\n"); + return -1; + } + + out_fd = creat(filename, mode); + if (out_fd < 0) { + err_msg("Failed to create output file %s\n", filename); + goto out_close_in; + } + + do { + retval = read(in_fd, buffer, sizeof(buffer)); + if (retval <= 0) + goto out_close; + + retval = write(out_fd, buffer, retval); + if (retval < 0) + goto out_close; + } while (retval > 0); + + retval = 0; +out_close: + close(out_fd); +out_close_in: + close(in_fd); + return retval; +} + +/* + * collect_registered_events - call the existing callback function for the event + * + * If an event has a registered callback function, call it. + * Otherwise, ignore the event. + */ +int +collect_registered_events(struct tep_event *event, struct tep_record *record, + int cpu, void *context) +{ + struct trace_instance *trace = context; + struct trace_seq *s = trace->seq; + + if (!event->handler) + return 0; + + event->handler(s, record, event, context); + + return 0; +} + +/* + * trace_instance_destroy - destroy and free a rtla trace instance + */ +void trace_instance_destroy(struct trace_instance *trace) +{ + if (trace->inst) { + disable_tracer(trace->inst); + destroy_instance(trace->inst); + } + + if (trace->seq) + free(trace->seq); + + if (trace->tep) + tep_free(trace->tep); +} + +/* + * trace_instance_init - create an rtla trace instance + * + * It is more than the tracefs instance, as it contains other + * things required for the tracing, such as the local events and + * a seq file. + * + * Note that the trace instance is returned disabled. This allows + * the tool to apply some other configs, like setting priority + * to the kernel threads, before starting generating trace entries. + */ +int trace_instance_init(struct trace_instance *trace, char *tool_name) +{ + trace->seq = calloc(1, sizeof(*trace->seq)); + if (!trace->seq) + goto out_err; + + trace_seq_init(trace->seq); + + trace->inst = create_instance(tool_name); + if (!trace->inst) + goto out_err; + + trace->tep = tracefs_local_events(NULL); + if (!trace->tep) + goto out_err; + + /* + * Let the main enable the record after setting some other + * things such as the priority of the tracer's threads. + */ + tracefs_trace_off(trace->inst); + + return 0; + +out_err: + trace_instance_destroy(trace); + return 1; +} + +/* + * trace_instance_start - start tracing a given rtla instance + */ +int trace_instance_start(struct trace_instance *trace) +{ + return tracefs_trace_on(trace->inst); +} diff --git a/tools/tracing/rtla/src/trace.h b/tools/tracing/rtla/src/trace.h new file mode 100644 index 000000000000..7c3930909b70 --- /dev/null +++ b/tools/tracing/rtla/src/trace.h @@ -0,0 +1,27 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include + +struct trace_instance { + struct tracefs_instance *inst; + struct tep_handle *tep; + struct trace_seq *seq; +}; + +int trace_instance_init(struct trace_instance *trace, char *tool_name); +int trace_instance_start(struct trace_instance *trace); +void trace_instance_destroy(struct trace_instance *trace); + +struct trace_seq *get_trace_seq(void); +int enable_tracer_by_name(struct tracefs_instance *inst, const char *tracer); +void disable_tracer(struct tracefs_instance *inst); + +int enable_osnoise(struct trace_instance *trace); +int enable_timerlat(struct trace_instance *trace); + +struct tracefs_instance *create_instance(char *instance_name); +void destroy_instance(struct tracefs_instance *inst); + +int save_trace_to_file(struct tracefs_instance *inst, const char *filename); +int collect_registered_events(struct tep_event *tep, struct tep_record *record, + int cpu, void *context); diff --git a/tools/tracing/rtla/src/utils.c b/tools/tracing/rtla/src/utils.c new file mode 100644 index 000000000000..4a35e62b1514 --- /dev/null +++ b/tools/tracing/rtla/src/utils.c @@ -0,0 +1,433 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "utils.h" + +#define MAX_MSG_LENGTH 1024 +int config_debug; + +/* + * err_msg - print an error message to the stderr + */ +void err_msg(const char *fmt, ...) +{ + char message[MAX_MSG_LENGTH]; + va_list ap; + + va_start(ap, fmt); + vsnprintf(message, sizeof(message), fmt, ap); + va_end(ap); + + fprintf(stderr, "%s", message); +} + +/* + * debug_msg - print a debug message to stderr if debug is set + */ +void debug_msg(const char *fmt, ...) +{ + char message[MAX_MSG_LENGTH]; + va_list ap; + + if (!config_debug) + return; + + va_start(ap, fmt); + vsnprintf(message, sizeof(message), fmt, ap); + va_end(ap); + + fprintf(stderr, "%s", message); +} + +/* + * get_long_from_str - get a long int from a string + */ +long get_long_from_str(char *start) +{ + long value; + char *end; + + errno = 0; + value = strtol(start, &end, 10); + if (errno || start == end) + return -1; + + return value; +} + +/* + * get_duration - fill output with a human readable duration since start_time + */ +void get_duration(time_t start_time, char *output, int output_size) +{ + time_t now = time(NULL); + struct tm *tm_info; + time_t duration; + + duration = difftime(now, start_time); + tm_info = localtime(&duration); + + snprintf(output, output_size, "%3d %02d:%02d:%02d", + tm_info->tm_yday, + tm_info->tm_hour - 1, + tm_info->tm_min, + tm_info->tm_sec); +} + +/* + * parse_cpu_list - parse a cpu_list filling a char vector with cpus set + * + * Receives a cpu list, like 1-3,5 (cpus 1, 2, 3, 5), and then set the char + * in the monitored_cpus. + * + * XXX: convert to a bitmask. + */ +int parse_cpu_list(char *cpu_list, char **monitored_cpus) +{ + char *mon_cpus; + const char *p; + int end_cpu; + int nr_cpus; + int cpu; + int i; + + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + mon_cpus = malloc(nr_cpus * sizeof(char)); + memset(mon_cpus, 0, (nr_cpus * sizeof(char))); + + for (p = cpu_list; *p; ) { + cpu = atoi(p); + if (cpu < 0 || (!cpu && *p != '0') || cpu >= nr_cpus) + goto err; + + while (isdigit(*p)) + p++; + if (*p == '-') { + p++; + end_cpu = atoi(p); + if (end_cpu < cpu || (!end_cpu && *p != '0') || end_cpu >= nr_cpus) + goto err; + while (isdigit(*p)) + p++; + } else + end_cpu = cpu; + + if (cpu == end_cpu) { + debug_msg("cpu_list: adding cpu %d\n", cpu); + mon_cpus[cpu] = 1; + } else { + for (i = cpu; i <= end_cpu; i++) { + debug_msg("cpu_list: adding cpu %d\n", i); + mon_cpus[i] = 1; + } + } + + if (*p == ',') + p++; + } + + *monitored_cpus = mon_cpus; + + return 0; + +err: + debug_msg("Error parsing the cpu list %s", cpu_list); + return 1; +} + +/* + * parse_duration - parse duration with s/m/h/d suffix converting it to seconds + */ +long parse_seconds_duration(char *val) +{ + char *end; + long t; + + t = strtol(val, &end, 10); + + if (end) { + switch (*end) { + case 's': + case 'S': + break; + case 'm': + case 'M': + t *= 60; + break; + case 'h': + case 'H': + t *= 60 * 60; + break; + + case 'd': + case 'D': + t *= 24 * 60 * 60; + break; + } + } + + return t; +} + +/* + * parse_ns_duration - parse duration with ns/us/ms/s converting it to nanoseconds + */ +long parse_ns_duration(char *val) +{ + char *end; + long t; + + t = strtol(val, &end, 10); + + if (end) { + if (!strncmp(end, "ns", 2)) { + return t; + } else if (!strncmp(end, "us", 2)) { + t *= 1000; + return t; + } else if (!strncmp(end, "ms", 2)) { + t *= 1000 * 1000; + return t; + } else if (!strncmp(end, "s", 1)) { + t *= 1000 * 1000 * 100; + return t; + } + return -1; + } + + return t; +} + +/* + * This is a set of helper functions to use SCHED_DEADLINE. + */ +#ifdef __x86_64__ +# define __NR_sched_setattr 314 +# define __NR_sched_getattr 315 +#elif __i386__ +# define __NR_sched_setattr 351 +# define __NR_sched_getattr 352 +#elif __arm__ +# define __NR_sched_setattr 380 +# define __NR_sched_getattr 381 +#elif __aarch64__ +# define __NR_sched_setattr 274 +# define __NR_sched_getattr 275 +#elif __powerpc__ +# define __NR_sched_setattr 355 +# define __NR_sched_getattr 356 +#elif __s390x__ +# define __NR_sched_setattr 345 +# define __NR_sched_getattr 346 +#endif + +#define SCHED_DEADLINE 6 + +static inline int sched_setattr(pid_t pid, const struct sched_attr *attr, + unsigned int flags) { + return syscall(__NR_sched_setattr, pid, attr, flags); +} + +static inline int sched_getattr(pid_t pid, struct sched_attr *attr, + unsigned int size, unsigned int flags) +{ + return syscall(__NR_sched_getattr, pid, attr, size, flags); +} + +int __set_sched_attr(int pid, struct sched_attr *attr) +{ + int flags = 0; + int retval; + + retval = sched_setattr(pid, attr, flags); + if (retval < 0) { + err_msg("boost_with_deadline failed to boost pid %d: %s\n", + pid, strerror(errno)); + return 1; + } + + return 0; +} +/* + * set_comm_sched_attr - set sched params to threads starting with char *comm + * + * This function uses procps to list the currently running threads and then + * set the sched_attr *attr to the threads that start with char *comm. It is + * mainly used to set the priority to the kernel threads created by the + * tracers. + */ +int set_comm_sched_attr(const char *comm, struct sched_attr *attr) +{ + int flags = PROC_FILLCOM | PROC_FILLSTAT; + PROCTAB *ptp; + proc_t task; + int retval; + + ptp = openproc(flags); + if (!ptp) { + err_msg("error openproc()\n"); + return -ENOENT; + } + + memset(&task, 0, sizeof(task)); + + while (readproc(ptp, &task)) { + retval = strncmp(comm, task.cmd, strlen(comm)); + if (retval) + continue; + retval = __set_sched_attr(task.tid, attr); + if (retval) + goto out_err; + } + + closeproc(ptp); + return 0; + +out_err: + closeproc(ptp); + return 1; +} + +#define INVALID_VAL (~0L) +static long get_long_ns_after_colon(char *start) +{ + long val = INVALID_VAL; + + /* find the ":" */ + start = strstr(start, ":"); + if (!start) + return -1; + + /* skip ":" */ + start++; + val = parse_ns_duration(start); + + return val; +} + +static long get_long_after_colon(char *start) +{ + long val = INVALID_VAL; + + /* find the ":" */ + start = strstr(start, ":"); + if (!start) + return -1; + + /* skip ":" */ + start++; + val = get_long_from_str(start); + + return val; +} + +/* + * parse priority in the format: + * SCHED_OTHER: + * o: + * O: + * SCHED_RR: + * r: + * R: + * SCHED_FIFO: + * f: + * F: + * SCHED_DEADLINE: + * d:runtime:period + * D:runtime:period + */ +int parse_prio(char *arg, struct sched_attr *sched_param) +{ + long prio; + long runtime; + long period; + + memset(sched_param, 0, sizeof(*sched_param)); + sched_param->size = sizeof(*sched_param); + + switch (arg[0]) { + case 'd': + case 'D': + /* d:runtime:period */ + if (strlen(arg) < 4) + return -1; + + runtime = get_long_ns_after_colon(arg); + if (runtime == INVALID_VAL) + return -1; + + period = get_long_ns_after_colon(&arg[2]); + if (period == INVALID_VAL) + return -1; + + if (runtime > period) + return -1; + + sched_param->sched_policy = SCHED_DEADLINE; + sched_param->sched_runtime = runtime; + sched_param->sched_deadline = period; + sched_param->sched_period = period; + break; + case 'f': + case 'F': + /* f:prio */ + prio = get_long_after_colon(arg); + if (prio == INVALID_VAL) + return -1; + + if (prio < sched_get_priority_min(SCHED_FIFO)) + return -1; + if (prio > sched_get_priority_max(SCHED_FIFO)) + return -1; + + sched_param->sched_policy = SCHED_FIFO; + sched_param->sched_priority = prio; + break; + case 'r': + case 'R': + /* r:prio */ + prio = get_long_after_colon(arg); + if (prio == INVALID_VAL) + return -1; + + if (prio < sched_get_priority_min(SCHED_RR)) + return -1; + if (prio > sched_get_priority_max(SCHED_RR)) + return -1; + + sched_param->sched_policy = SCHED_RR; + sched_param->sched_priority = prio; + break; + case 'o': + case 'O': + /* o:prio */ + prio = get_long_after_colon(arg); + if (prio == INVALID_VAL) + return -1; + + if (prio < sched_get_priority_min(SCHED_OTHER)) + return -1; + if (prio > sched_get_priority_max(SCHED_OTHER)) + return -1; + + sched_param->sched_policy = SCHED_OTHER; + sched_param->sched_priority = prio; + break; + default: + return -1; + } + return 0; +} diff --git a/tools/tracing/rtla/src/utils.h b/tools/tracing/rtla/src/utils.h new file mode 100644 index 000000000000..2bce867387b4 --- /dev/null +++ b/tools/tracing/rtla/src/utils.h @@ -0,0 +1,56 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include + +/* + * '18446744073709551615\0' + */ +#define BUFF_U64_STR_SIZE 24 + +#define container_of(ptr, type, member)({ \ + const typeof(((type *)0)->member) *__mptr = (ptr); \ + (type *)((char *)__mptr - offsetof(type, member)) ; }) + +extern int config_debug; +void debug_msg(const char *fmt, ...); +void err_msg(const char *fmt, ...); + +long parse_seconds_duration(char *val); +void get_duration(time_t start_time, char *output, int output_size); + +int parse_cpu_list(char *cpu_list, char **monitored_cpus); +long get_long_from_str(char *start); + +static inline void +update_min(unsigned long long *a, unsigned long long *b) +{ + if (*a > *b) + *a = *b; +} + +static inline void +update_max(unsigned long long *a, unsigned long long *b) +{ + if (*a < *b) + *a = *b; +} + +static inline void +update_sum(unsigned long long *a, unsigned long long *b) +{ + *a += *b; +} + +struct sched_attr { + uint32_t size; + uint32_t sched_policy; + uint64_t sched_flags; + int32_t sched_nice; + uint32_t sched_priority; + uint64_t sched_runtime; + uint64_t sched_deadline; + uint64_t sched_period; +}; + +int parse_prio(char *arg, struct sched_attr *sched_param); +int set_comm_sched_attr(const char *comm, struct sched_attr *attr); From patchwork Tue Oct 26 22:06:20 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585803 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 80A93C433FE for ; Tue, 26 Oct 2021 22:08:53 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 64B8160F92 for ; Tue, 26 Oct 2021 22:08:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239819AbhJZWLQ convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:16 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:49870 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239890AbhJZWK7 (ORCPT ); Tue, 26 Oct 2021 18:10:59 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-511-ajUaAjrwO7ypZdAVKMVqgw-1; Tue, 26 Oct 2021 18:08:27 -0400 X-MC-Unique: ajUaAjrwO7ypZdAVKMVqgw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 6E795806697; Tue, 26 Oct 2021 22:08:25 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 7A62B60BF1; Tue, 26 Oct 2021 22:08:20 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 09/20] rtla: Add osnoise tool Date: Wed, 27 Oct 2021 00:06:20 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The osnoise tool is the interface for the osnoise tracer. The osnoise tool will have multiple "modes" with different outputs. At this point, no mode is included. The osnoise.c includes the osnoise_context abstraction. It serves to read-save-change-restore the default values from tracing/osnoise/ directory. When the context is deleted, the default values are restored. It also includes some other helper functions for managing osnoise tracer sessions. With these bits and pieces in place, we can start adding some functionality to rtla. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/Makefile | 2 + tools/tracing/rtla/src/osnoise.c | 815 +++++++++++++++++++++++++++++++ tools/tracing/rtla/src/osnoise.h | 85 ++++ tools/tracing/rtla/src/rtla.c | 10 + 4 files changed, 912 insertions(+) create mode 100644 tools/tracing/rtla/src/osnoise.c create mode 100644 tools/tracing/rtla/src/osnoise.h diff --git a/tools/tracing/rtla/Makefile b/tools/tracing/rtla/Makefile index 525e15b76156..5e93d000a821 100644 --- a/tools/tracing/rtla/Makefile +++ b/tools/tracing/rtla/Makefile @@ -60,6 +60,8 @@ install: $(INSTALL) -d -m 755 $(DESTDIR)$(BINDIR) $(INSTALL) rtla -m 755 $(DESTDIR)$(BINDIR) $(STRIP) $(DESTDIR)$(BINDIR)/rtla + @test ! -f $(DESTDIR)$(BINDIR)/osnoise || rm $(DESTDIR)$(BINDIR)/osnoise + ln -s $(DESTDIR)$(BINDIR)/rtla $(DESTDIR)$(BINDIR)/osnoise .PHONY: clean tarball clean: diff --git a/tools/tracing/rtla/src/osnoise.c b/tools/tracing/rtla/src/osnoise.c new file mode 100644 index 000000000000..fa3786c4aedc --- /dev/null +++ b/tools/tracing/rtla/src/osnoise.c @@ -0,0 +1,815 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "osnoise.h" +#include "utils.h" + +/* + * osnoise_set_cpus - configure osnoise to run on *cpus + * + * "osnoise/cpus" file is used to set the cpus in which osnoise/timerlat + * will run. This function opens this file, saves the current value, + * and set the cpus passed as argument. + */ +int osnoise_set_cpus(struct osnoise_context *context, char *cpus) +{ + char *osnoise_cpus = tracefs_get_tracing_file("osnoise/cpus"); + char curr_cpus[1024]; + int retval; + + context->cpus_fd = open(osnoise_cpus, O_RDWR); + if (!context->cpus_fd) + goto out_err; + + retval = read(context->cpus_fd, &curr_cpus, sizeof(curr_cpus)); + if (!retval) + goto out_close; + + context->orig_cpus = strdup(curr_cpus); + if (!context->orig_cpus) + goto out_err; + + retval = write(context->cpus_fd, cpus, strlen(cpus) + 1); + if (!retval) + goto out_err; + + tracefs_put_tracing_file(osnoise_cpus); + + return 0; + +out_close: + close(context->cpus_fd); + context->cpus_fd = -1; +out_err: + tracefs_put_tracing_file(osnoise_cpus); + return 1; +} + +/* + * osnoise_restore_cpus - restore the original "osnoise/cpus" + * + * osnoise_set_cpus() saves the original data for the "osnoise/cpus" + * file. This function restore the original config it was previously + * modified. + */ +void osnoise_restore_cpus(struct osnoise_context *context) +{ + int retval; + + if (!context->orig_cpus) + return; + + retval = write(context->cpus_fd, context->orig_cpus, strlen(context->orig_cpus)); + if (!retval) + err_msg("could not restore original osnoise cpus\n"); + + free(context->orig_cpus); +} + +/* + * osnoise_get_runtime - return the original "osnoise/runtime_us" value + * + * It also saves the value to be restored. + */ +unsigned long long osnoise_get_runtime(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *runtime_path; + int retval; + + if (context->runtime_us) + return context->runtime_us; + + runtime_path = tracefs_get_tracing_file("osnoise/runtime_us"); + + context->runtime_fd = open(runtime_path, O_RDWR); + if (!context->runtime_fd) + goto out_err; + + retval = read(context->runtime_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->runtime_us = get_long_from_str(buffer); + if (!context->runtime_us) + goto out_close; + + context->orig_runtime_us = context->runtime_us; + + tracefs_put_tracing_file(runtime_path); + + return context->runtime_us; + +out_close: + close(context->runtime_fd); + context->runtime_fd = -1; +out_err: + tracefs_put_tracing_file(runtime_path); + return 0; +} + +/* + * osnoise_get_period - return the original "osnoise/period_us" value + * + * It also saves the value to be restored. + */ +unsigned long long osnoise_get_period(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *period_path; + int retval; + + if (context->period_us) + return context->period_us; + + period_path = tracefs_get_tracing_file("osnoise/period_us"); + + context->period_fd = open(period_path, O_RDWR); + if (!context->period_fd) + goto out_err; + + retval = read(context->period_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->period_us = get_long_from_str(buffer); + if (!context->period_us) + goto out_close; + + context->orig_period_us = context->period_us; + + tracefs_put_tracing_file(period_path); + + return context->period_us; + +out_close: + close(context->period_fd); + context->period_fd = -1; +out_err: + tracefs_put_tracing_file(period_path); + return 0; +} + +static int __osnoise_write_runtime(struct osnoise_context *context, + unsigned long long runtime) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->runtime_fd == -1) { + err_msg("tried to write osnoise runtime before getting it\n"); + return 0; + } + + snprintf(buffer, sizeof(buffer), "%llu\n", runtime); + + retval = write(context->runtime_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->runtime_us = runtime; + return 0; +} + +static int __osnoise_write_period(struct osnoise_context *context, + unsigned long long period) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->period_fd == -1) + return 0; + + snprintf(buffer, sizeof(buffer), "%llu\n", period); + + retval = write(context->period_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->period_us = period; + return 0; +} + +/* + * osnoise_set_runtime_period - set osnoise runtime and period + * + * Osnoise's runtime and period are related as runtime <= period. + * Thus, this function saves the original values, and then tries + * to set the runtime and period if they are != 0. + */ +int osnoise_set_runtime_period(struct osnoise_context *context, + unsigned long long runtime, + unsigned long long period) +{ + unsigned long long curr_runtime_us; + unsigned long long curr_period_us; + int retval; + + if (!period && !runtime) + return 0; + + curr_runtime_us = osnoise_get_runtime(context); + curr_period_us = osnoise_get_period(context); + + if (!curr_period_us || !curr_runtime_us) + return -1; + + if (!period) { + if (runtime > curr_period_us) + return -1; + return __osnoise_write_runtime(context, runtime); + } else if (!runtime) { + if (period < curr_runtime_us) + return -1; + return __osnoise_write_period(context, period); + } + + if (runtime > curr_period_us) { + retval = __osnoise_write_period(context, period); + if (retval) + return -1; + retval = __osnoise_write_runtime(context, runtime); + if (retval) + return -1; + } else { + retval = __osnoise_write_runtime(context, runtime); + if (retval) + return -1; + retval = __osnoise_write_period(context, period); + if (retval) + return -1; + } + + return 0; +} + +/* + * osnoise_restore_runtime_period - restore the original runtime and period + */ +void osnoise_restore_runtime_period(struct osnoise_context *context) +{ + unsigned long long runtime = context->orig_runtime_us; + unsigned long long period = context->orig_period_us; + int retval; + + if (context->runtime_fd < 0 && context->period_fd < 0) + return; + + retval = osnoise_set_runtime_period(context, runtime, period); + if (retval) + err_msg("Could not restore original osnoise runtime/period\n"); +} + +/* + * osnoise_get_stop_us - read and save the original "stop_tracing_us" + */ +static long long +osnoise_get_stop_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *stop_path; + int retval; + + if (context->stop_us_fd > 0) + return context->stop_us; + + stop_path = tracefs_get_tracing_file("osnoise/stop_tracing_us"); + + context->stop_us_fd = open(stop_path, O_RDWR); + if (!context->stop_us_fd) + goto out_err; + + retval = read(context->stop_us_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->stop_us = get_long_from_str(buffer); + context->orig_stop_us = context->stop_us; + + tracefs_put_tracing_file(stop_path); + + return context->stop_us; + +out_close: + close(context->stop_us_fd); + context->stop_us_fd = -1; +out_err: + tracefs_put_tracing_file(stop_path); + return -1; +} + +/* + * osnoise_set_stop_us - set "stop_tracing_us" + */ +int osnoise_set_stop_us(struct osnoise_context *context, long long stop_us) +{ + long long curr_stop_us = osnoise_get_stop_us(context); + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (curr_stop_us < 0) + return -1; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", stop_us); + + retval = write(context->stop_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->stop_us = stop_us; + + return 0; +} + +/* + * osnoise_restore_stop_us - restore the original stop_tracing_us + */ +void osnoise_restore_stop_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->stop_us_fd < 0) + return; + + if (context->orig_stop_us == context->stop_us) + return; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", context->orig_stop_us); + + retval = write(context->stop_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + err_msg("Could not restore original osnoise stop_us\n"); +} + +/* + * osnoise_get_stop_us - read and save the original "stop_tracing_total_us" + */ +static long long +osnoise_get_stop_total_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *stop_path; + int retval; + + if (context->stop_total_us_fd > 0) + return context->stop_total_us; + + stop_path = tracefs_get_tracing_file("osnoise/stop_tracing_total_us"); + + context->stop_total_us_fd = open(stop_path, O_RDWR); + if (!context->stop_total_us_fd) + goto out_err; + + retval = read(context->stop_total_us_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->stop_total_us = get_long_from_str(buffer); + context->orig_stop_total_us = context->stop_total_us; + + tracefs_put_tracing_file(stop_path); + + return context->stop_total_us; + +out_close: + close(context->stop_total_us_fd); + context->stop_total_us_fd = -1; +out_err: + tracefs_put_tracing_file(stop_path); + return -1; +} + +/* + * osnoise_set_stop_us - set "stop_tracing_total_us" + */ +int osnoise_set_stop_total_us(struct osnoise_context *context, long long stop_total_us) +{ + long long curr_stop_total_us = osnoise_get_stop_total_us(context); + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (curr_stop_total_us < 0) + return -1; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", stop_total_us); + + retval = write(context->stop_total_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->stop_total_us = stop_total_us; + + return 0; +} + +/* + * osnoise_restore_stop_us - restore the original stop_tracing_us + */ +void osnoise_restore_stop_total_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->stop_total_us_fd < 0) + return; + + if (context->orig_stop_total_us == context->stop_total_us) + return; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", context->orig_stop_total_us); + + retval = write(context->stop_total_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + err_msg("Could not restore original osnoise stop_total_us\n"); +} + +/* + * osnoise_get_timerlat_period_us - read and save the original "timerlat_period_us" + */ +static long long +osnoise_get_timerlat_period_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *stop_path; + int retval; + + if (context->timerlat_period_us_fd > 0) + return context->timerlat_period_us; + + stop_path = tracefs_get_tracing_file("osnoise/timerlat_period_us"); + + context->timerlat_period_us_fd = open(stop_path, O_RDWR); + if (!context->timerlat_period_us_fd) + goto out_err; + + retval = read(context->timerlat_period_us_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->timerlat_period_us = get_long_from_str(buffer); + context->orig_timerlat_period_us = context->timerlat_period_us; + + tracefs_put_tracing_file(stop_path); + + return context->timerlat_period_us; + +out_close: + close(context->timerlat_period_us_fd); + context->timerlat_period_us_fd = -1; +out_err: + tracefs_put_tracing_file(stop_path); + return -1; +} + +/* + * osnoise_set_timerlat_period_us - set "timerlat_period_us" + */ +int osnoise_set_timerlat_period_us(struct osnoise_context *context, long long timerlat_period_us) +{ + long long curr_timerlat_period_us = osnoise_get_timerlat_period_us(context); + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (curr_timerlat_period_us < 0) + return -1; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", timerlat_period_us); + + retval = write(context->timerlat_period_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->timerlat_period_us = timerlat_period_us; + + return 0; +} + +/* + * osnoise_restore_timerlat_period_us - restore "timerlat_period_us" + */ +void osnoise_restore_timerlat_period_us(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->timerlat_period_us_fd < 0) + return; + + if (context->orig_timerlat_period_us == context->timerlat_period_us) + return; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", context->orig_timerlat_period_us); + + retval = write(context->timerlat_period_us_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + err_msg("Could not restore original osnoise timerlat_period_us\n"); +} + +/* + * osnoise_get_print_stack - read and save the original "print_stack" + */ +static long long +osnoise_get_print_stack(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + char *stop_path; + int retval; + + if (context->print_stack_fd > 0) + return context->print_stack; + + stop_path = tracefs_get_tracing_file("osnoise/print_stack"); + + context->print_stack_fd = open(stop_path, O_RDWR); + if (!context->print_stack_fd) + goto out_err; + + retval = read(context->print_stack_fd, &buffer, sizeof(buffer)); + if (!retval) + goto out_close; + + context->print_stack = get_long_from_str(buffer); + context->orig_print_stack = context->print_stack; + + tracefs_put_tracing_file(stop_path); + + return context->print_stack; + +out_close: + close(context->print_stack_fd); + context->print_stack_fd = -1; +out_err: + tracefs_put_tracing_file(stop_path); + return -1; +} + +/* + * osnoise_set_print_stack - set "print_stack" + */ +int osnoise_set_print_stack(struct osnoise_context *context, long long print_stack) +{ + long long curr_print_stack = osnoise_get_print_stack(context); + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (curr_print_stack < 0) + return -1; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", print_stack); + + retval = write(context->print_stack_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + return -1; + + context->print_stack = print_stack; + + return 0; +} + +/* + * osnoise_restore_print_stack - restore the original "print_stack" + */ +void osnoise_restore_print_stack(struct osnoise_context *context) +{ + char buffer[BUFF_U64_STR_SIZE]; + int retval; + + if (context->print_stack_fd < 0) + return; + + if (context->orig_print_stack == context->print_stack) + return; + + snprintf(buffer, BUFF_U64_STR_SIZE, "%lld\n", context->orig_print_stack); + + retval = write(context->print_stack_fd, buffer, strlen(buffer) + 1); + if (retval < 0) + err_msg("Could not restore original osnoise print_stack\n"); +} + +/* + * enable_osnoise - enable osnoise tracer in the trace_instance + */ +int enable_osnoise(struct trace_instance *trace) +{ + return enable_tracer_by_name(trace->inst, "osnoise"); +} + +/* + * enable_timerlat - enable timerlat tracer in the trace_instance + */ +int enable_timerlat(struct trace_instance *trace) +{ + return enable_tracer_by_name(trace->inst, "timerlat"); +} + +enum { + FLAG_CONTEXT_NEWLY_CREATED = (1 << 0), + FLAG_CONTEXT_DELETED = (1 << 1), +}; + +/* + * osnoise_get_context - increase the usage of a context and return it + */ +int osnoise_get_context(struct osnoise_context *context) +{ + int ret; + + pthread_mutex_lock(&context->lock); + if (context->flags & FLAG_CONTEXT_DELETED) { + ret = -1; + } else { + context->ref++; + ret = 0; + } + pthread_mutex_unlock(&context->lock); + + return ret; +} + +/* + * osnoise_context_alloc - alloc an osnoise_context + * + * The osnoise context contains the information of the "osnoise/" configs. + * It is used to set and restore the config. + */ +struct osnoise_context *osnoise_context_alloc(void) +{ + struct osnoise_context *context; + + context = calloc(1, sizeof(*context)); + if (!context) + goto out_err; + + if (pthread_mutex_init(&context->lock, NULL) < 0) + goto out_err; + + context->cpus_fd = -1; + context->runtime_fd = -1; + context->period_fd = -1; + context->stop_us_fd = -1; + context->stop_total_us_fd = -1; + + osnoise_get_context(context); + + return context; +out_err: + if (context) + free(context); + return NULL; +} + +/* + * osnoise_put_context - put the osnoise_put_context + * + * If there is no other user for the context, the original data + * is restored. + */ +void osnoise_put_context(struct osnoise_context *context) +{ + pthread_mutex_lock(&context->lock); + if (--context->ref < 1) + context->flags |= FLAG_CONTEXT_DELETED; + pthread_mutex_unlock(&context->lock); + + if (!(context->flags & FLAG_CONTEXT_DELETED)) + return; + + osnoise_restore_cpus(context); + if (context->cpus_fd >= 0) + close(context->cpus_fd); + + osnoise_restore_runtime_period(context); + if (context->runtime_fd >= 0) + close(context->runtime_fd); + if (context->period_fd >= 0) + close(context->period_fd); + + osnoise_restore_stop_us(context); + osnoise_restore_stop_total_us(context); + osnoise_restore_print_stack(context); + osnoise_restore_timerlat_period_us(context); + + pthread_mutex_destroy(&context->lock); + free(context); +} + +/* + * osnoise_destroy_tool - disable trace, restore configs and free data + */ +void osnoise_destroy_tool(struct osnoise_tool *top) +{ + trace_instance_destroy(&top->trace); + + if (top->context) + osnoise_put_context(top->context); + + free(top); +} + +/* + * osnoise_init_tool - init an osnoise tool + * + * It allocs data, create a context to store data and + * creates a new trace instance for the tool. + */ +struct osnoise_tool *osnoise_init_tool(char *tool_name) +{ + struct osnoise_tool *top; + int retval; + + top = calloc(1, sizeof(*top)); + if (!top) + return NULL; + + top->context = osnoise_context_alloc(); + if (!top->context) + goto out_err; + + retval = trace_instance_init(&top->trace, tool_name); + if (retval) + goto out_err; + + return top; +out_err: + osnoise_destroy_tool(top); + return NULL; +} + +/* + * osnoise_init_trace_tool - init a tracer instance to trace osnoise events + */ +struct osnoise_tool *osnoise_init_trace_tool(char *tracer) +{ + struct osnoise_tool *trace; + int retval; + + trace = osnoise_init_tool("osnoise_trace"); + if (!trace) + return NULL; + + retval = tracefs_event_enable(trace->trace.inst, "osnoise", NULL); + if (retval < 0 && !errno) { + err_msg("Could not find osnoise events\n"); + goto out_err; + } + + retval = enable_tracer_by_name(trace->trace.inst, tracer); + if (retval) { + err_msg("Could not enable osnoiser tracer for tracing\n"); + goto out_err; + } + + return trace; +out_err: + osnoise_destroy_tool(trace); + return NULL; +} + +static void osnoise_usage(void) +{ + int i; + + static const char *msg[] = { + "", + "osnoise version " VERSION, + "", + " usage: [rtla] osnoise [MODE] ...", + "", + " modes:", + "", + NULL, + }; + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +int osnoise_main(int argc, char *argv[]) +{ + if (argc <= 1) + goto usage; + + if ((strcmp(argv[1], "-h") == 0) || (strcmp(argv[1], "--help") == 0)) { + osnoise_usage(); + exit(0); + } + +usage: + osnoise_usage(); + exit(1); +} diff --git a/tools/tracing/rtla/src/osnoise.h b/tools/tracing/rtla/src/osnoise.h new file mode 100644 index 000000000000..c8db59e7c337 --- /dev/null +++ b/tools/tracing/rtla/src/osnoise.h @@ -0,0 +1,85 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "trace.h" + +/* + * osnoise_context - read, store, write, restore osnoise configs. + */ +struct osnoise_context { + int flags; + int ref; + pthread_mutex_t lock; + + int cpus_fd; + int runtime_fd; + int period_fd; + int stop_us_fd; + int stop_total_us_fd; + int timerlat_period_us_fd; + int print_stack_fd; + + char *curr_cpus; + char *orig_cpus; + + unsigned long long orig_runtime_us; + unsigned long long runtime_us; + + unsigned long long orig_period_us; + unsigned long long period_us; + + long long orig_stop_us; + long long stop_us; + + long long orig_stop_total_us; + long long stop_total_us; + + long long orig_timerlat_period_us; + long long timerlat_period_us; + + long long orig_print_stack; + long long print_stack; +}; + +struct osnoise_context *osnoise_context_alloc(void); +int osnoise_get_context(struct osnoise_context *context); +void osnoise_put_context(struct osnoise_context *context); + +int osnoise_set_cpus(struct osnoise_context *context, char *cpus); +void osnoise_restore_cpus(struct osnoise_context *context); + +int osnoise_set_runtime_period(struct osnoise_context *context, + unsigned long long runtime, + unsigned long long period); +void osnoise_restore_runtime_period(struct osnoise_context *context); + +int osnoise_set_stop_us(struct osnoise_context *context, + long long stop_us); +void osnoise_restore_stop_us(struct osnoise_context *context); + +int osnoise_set_stop_total_us(struct osnoise_context *context, + long long stop_total_us); +void osnoise_restore_stop_total_us(struct osnoise_context *context); + +int osnoise_set_timerlat_period_us(struct osnoise_context *context, + long long timerlat_period_us); +void osnoise_restore_timerlat_period_us(struct osnoise_context *context); + +void osnoise_restore_print_stack(struct osnoise_context *context); +int osnoise_set_print_stack(struct osnoise_context *context, + long long print_stack); + +/* + * osnoise_tool - osnoise based tool definition. + */ +struct osnoise_tool { + struct trace_instance trace; + struct osnoise_context *context; + void *data; + void *params; + time_t start_time; +}; + +void osnoise_destroy_tool(struct osnoise_tool *top); +struct osnoise_tool *osnoise_init_tool(char *tool_name); +struct osnoise_tool *osnoise_init_trace_tool(char *tracer); + +int osnoise_main(int argc, char **argv); diff --git a/tools/tracing/rtla/src/rtla.c b/tools/tracing/rtla/src/rtla.c index 5ae2664ed47d..669b9750b3b3 100644 --- a/tools/tracing/rtla/src/rtla.c +++ b/tools/tracing/rtla/src/rtla.c @@ -8,6 +8,8 @@ #include #include +#include "osnoise.h" + /* * rtla_usage - print rtla usage */ @@ -22,6 +24,7 @@ static void rtla_usage(void) " usage: rtla COMMAND ...", "", " commands:", + " osnoise - gives information about the operating system noise (osnoise)", "", NULL, }; @@ -39,7 +42,14 @@ static void rtla_usage(void) */ int run_command(int argc, char **argv, int start_position) { + if (strcmp(argv[start_position], "osnoise") == 0) { + osnoise_main(argc-start_position, &argv[start_position]); + goto ran; + } + return 0; +ran: + return 1; } int main(int argc, char *argv[]) From patchwork Tue Oct 26 22:06:21 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585799 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 671DAC433EF for ; Tue, 26 Oct 2021 22:08:52 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 4906860F02 for ; Tue, 26 Oct 2021 22:08:52 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239800AbhJZWLP convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:15 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:40618 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239902AbhJZWLA (ORCPT ); Tue, 26 Oct 2021 18:11:00 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-350-VBsSuHncPsGUTMrVgNQ2Gg-1; Tue, 26 Oct 2021 18:08:32 -0400 X-MC-Unique: VBsSuHncPsGUTMrVgNQ2Gg-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id BCA9E10247A6; Tue, 26 Oct 2021 22:08:30 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id D589F60BF1; Tue, 26 Oct 2021 22:08:25 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 10/20] rtla/osnoise: Add osnoise top mode Date: Wed, 27 Oct 2021 00:06:21 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The rtla osnoise tool is an interface for the osnoise tracer. The osnoise tracer dispatches a kernel thread per-cpu. These threads read the time in a loop while with preemption, softirqs and IRQs enabled, thus allowing all the sources of osnoise during its execution. The osnoise threads take note of the entry and exit point of any source of interferences, increasing a per-cpu interference counter. The osnoise tracer also saves an interference counter for each source of interference. The rtla osnoise top mode displays information about the periodic summary from the osnoise tracer. One example of rtla osnoise top output is: [root@alien ~]# rtla osnoise top -c 0-3 -d 1m -q -r 900000 -P F:1 Operating System Noise duration: 0 00:01:00 | time is in us CPU Period Runtime Noise % CPU Aval Max Noise Max Single HW NMI IRQ Softirq Thread 0 #58 52200000 1031 99.99802 91 60 0 0 52285 0 101 1 #59 53100000 5 99.99999 5 5 0 9 53122 0 18 2 #59 53100000 7 99.99998 7 7 0 8 53115 0 18 3 #59 53100000 8274 99.98441 277 23 0 9 53778 0 660 "rtla osnoise top --help" works and provide information about the available options. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/src/osnoise.c | 20 +- tools/tracing/rtla/src/osnoise.h | 1 + tools/tracing/rtla/src/osnoise_top.c | 577 +++++++++++++++++++++++++++ 3 files changed, 597 insertions(+), 1 deletion(-) create mode 100644 tools/tracing/rtla/src/osnoise_top.c diff --git a/tools/tracing/rtla/src/osnoise.c b/tools/tracing/rtla/src/osnoise.c index fa3786c4aedc..b9866dfdda66 100644 --- a/tools/tracing/rtla/src/osnoise.c +++ b/tools/tracing/rtla/src/osnoise.c @@ -790,7 +790,9 @@ static void osnoise_usage(void) " usage: [rtla] osnoise [MODE] ...", "", " modes:", + " top - prints the summary from osnoise tracer", "", + "if no MODE is given, the top mode is called, passing the arguments", NULL, }; @@ -801,12 +803,28 @@ static void osnoise_usage(void) int osnoise_main(int argc, char *argv[]) { - if (argc <= 1) + if (argc == 0) goto usage; + /* + * if osnoise was called without any argument, run the + * default cmdline. + */ + if (argc == 1) { + osnoise_top_main(argc, argv); + exit(0); + } + if ((strcmp(argv[1], "-h") == 0) || (strcmp(argv[1], "--help") == 0)) { osnoise_usage(); exit(0); + } else if (strncmp(argv[1], "-", 1) == 0) { + /* the user skipped the tool, call the default one */ + osnoise_top_main(argc, argv); + exit(0); + } else if (strcmp(argv[1], "top") == 0) { + osnoise_top_main(argc-1, &argv[1]); + exit(0); } usage: diff --git a/tools/tracing/rtla/src/osnoise.h b/tools/tracing/rtla/src/osnoise.h index c8db59e7c337..4882ee275ea0 100644 --- a/tools/tracing/rtla/src/osnoise.h +++ b/tools/tracing/rtla/src/osnoise.h @@ -82,4 +82,5 @@ void osnoise_destroy_tool(struct osnoise_tool *top); struct osnoise_tool *osnoise_init_tool(char *tool_name); struct osnoise_tool *osnoise_init_trace_tool(char *tracer); +int osnoise_top_main(int argc, char **argv); int osnoise_main(int argc, char **argv); diff --git a/tools/tracing/rtla/src/osnoise_top.c b/tools/tracing/rtla/src/osnoise_top.c new file mode 100644 index 000000000000..30f0ec087775 --- /dev/null +++ b/tools/tracing/rtla/src/osnoise_top.c @@ -0,0 +1,577 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "osnoise.h" +#include "utils.h" + +/* + * osnoise top parameters + */ +struct osnoise_top_params { + char *cpus; + char *monitored_cpus; + char *trace_output; + unsigned long long runtime; + unsigned long long period; + long long stop_us; + long long stop_total_us; + int sleep_time; + int duration; + int quiet; + int set_sched; + struct sched_attr sched_param; +}; + +struct osnoise_top_cpu { + unsigned long long sum_runtime; + unsigned long long sum_noise; + unsigned long long max_noise; + unsigned long long max_sample; + + unsigned long long hw_count; + unsigned long long nmi_count; + unsigned long long irq_count; + unsigned long long softirq_count; + unsigned long long thread_count; + + int sum_cycles; +}; + +struct osnoise_top_data { + struct osnoise_top_cpu *cpu_data; + int nr_cpus; +}; + +/* + * osnoise_free_top - free runtime data + */ +static void +osnoise_free_top(struct osnoise_top_data *data) +{ + free(data->cpu_data); + free(data); +} + +/* + * osnoise_alloc_histogram - alloc runtime data + */ +static struct osnoise_top_data *osnoise_alloc_top(int nr_cpus) +{ + struct osnoise_top_data *data; + + data = calloc(1, sizeof(*data)); + if (!data) + return NULL; + + data->nr_cpus = nr_cpus; + + /* one set of histograms per CPU */ + data->cpu_data = calloc(1, sizeof(*data->cpu_data) * nr_cpus); + if (!data->cpu_data) + goto cleanup; + + return data; + +cleanup: + osnoise_free_top(data); + return NULL; +} + +/* + * osnoise_top_handler - this is the handler for osnoise tracer events + */ +static int +osnoise_top_handler(struct trace_seq *s, struct tep_record *record, + struct tep_event *event, void *context) +{ + struct trace_instance *trace = context; + struct osnoise_tool *tool; + unsigned long long val; + struct osnoise_top_cpu *cpu_data; + struct osnoise_top_data *data; + int cpu = record->cpu; + + tool = container_of(trace, struct osnoise_tool, trace); + + data = tool->data; + cpu_data = &data->cpu_data[cpu]; + + cpu_data->sum_cycles++; + + tep_get_field_val(s, event, "runtime", record, &val, 1); + update_sum(&cpu_data->sum_runtime, &val); + + tep_get_field_val(s, event, "noise", record, &val, 1); + update_max(&cpu_data->max_noise, &val); + update_sum(&cpu_data->sum_noise, &val); + + tep_get_field_val(s, event, "max_sample", record, &val, 1); + update_max(&cpu_data->max_sample, &val); + + tep_get_field_val(s, event, "hw_count", record, &val, 1); + update_sum(&cpu_data->hw_count, &val); + + tep_get_field_val(s, event, "nmi_count", record, &val, 1); + update_sum(&cpu_data->nmi_count, &val); + + tep_get_field_val(s, event, "irq_count", record, &val, 1); + update_sum(&cpu_data->irq_count, &val); + + tep_get_field_val(s, event, "softirq_count", record, &val, 1); + update_sum(&cpu_data->softirq_count, &val); + + tep_get_field_val(s, event, "thread_count", record, &val, 1); + update_sum(&cpu_data->thread_count, &val); + + return 0; +} + +/* + * osnoise_top_header - print the header of the tool output + */ +static void osnoise_top_header(struct osnoise_tool *top) +{ + struct trace_seq *s = top->trace.seq; + char duration[26]; + + get_duration(top->start_time, duration, sizeof(duration)); + + trace_seq_printf(s, "\033[2;37;40m"); + trace_seq_printf(s, " Operating System Noise"); + trace_seq_printf(s, " "); + trace_seq_printf(s, " "); + trace_seq_printf(s, "\033[0;0;0m"); + trace_seq_printf(s, "\n"); + + trace_seq_printf(s, "duration: %9s | time is in us\n", duration); + + trace_seq_printf(s, "\033[2;30;47m"); + trace_seq_printf(s, "CPU Period Runtime "); + trace_seq_printf(s, " Noise "); + trace_seq_printf(s, " %% CPU Aval "); + trace_seq_printf(s, " Max Noise Max Single "); + trace_seq_printf(s, " HW NMI IRQ Softirq Thread"); + trace_seq_printf(s, "\033[0;0;0m"); + trace_seq_printf(s, "\n"); +} + +/* + * clear_terminal - clears the output terminal + */ +static void clear_terminal(struct trace_seq *seq) +{ + if (!config_debug) + trace_seq_printf(seq, "\033c"); +} + +/* + * osnoise_top_print - prints the output of a given CPU + */ +static void osnoise_top_print(struct osnoise_tool *tool, int cpu) +{ + struct trace_seq *s = tool->trace.seq; + struct osnoise_top_cpu *cpu_data; + struct osnoise_top_data *data; + int percentage; + int decimal; + + data = tool->data; + cpu_data = &data->cpu_data[cpu]; + + if (!cpu_data->sum_runtime) + return; + + percentage = ((cpu_data->sum_runtime - cpu_data->sum_noise) * 10000000) + / cpu_data->sum_runtime; + decimal = percentage % 100000; + percentage = percentage / 100000; + + trace_seq_printf(s, "%3d #%-6d %12llu ", cpu, cpu_data->sum_cycles, cpu_data->sum_runtime); + trace_seq_printf(s, "%12llu ", cpu_data->sum_noise); + trace_seq_printf(s, " %3d.%05d", percentage, decimal); + trace_seq_printf(s, "%12llu %12llu", cpu_data->max_noise, cpu_data->max_sample); + + trace_seq_printf(s, "%12llu ", cpu_data->hw_count); + trace_seq_printf(s, "%12llu ", cpu_data->nmi_count); + trace_seq_printf(s, "%12llu ", cpu_data->irq_count); + trace_seq_printf(s, "%12llu ", cpu_data->softirq_count); + trace_seq_printf(s, "%12llu\n", cpu_data->thread_count); +} + +/* + * osnoise_print_stats - print data for all cpus + */ +static void +osnoise_print_stats(struct osnoise_top_params *params, struct osnoise_tool *top) +{ + struct trace_instance *trace = &top->trace; + static int nr_cpus = -1; + int i; + + if (nr_cpus == -1) + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + if (!params->quiet) + clear_terminal(trace->seq); + + osnoise_top_header(top); + + for (i = 0; i < nr_cpus; i++) { + if (params->cpus && !params->monitored_cpus[i]) + continue; + osnoise_top_print(top, i); + } + + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); +} + +/* + * osnoise_top_usage - prints osnoise top usage message + */ +void osnoise_top_usage(char *usage) +{ + int i; + + static const char * const msg[] = { + " usage: rtla osnoise [top] [-h] [-q] [-p us] [-r us] [-s us] [-S us] [-T[=file]] \\", + " [-c cpu-list] -P priority", + "", + " -h/--help: print this menu", + " -p/--period us: osnoise period in us", + " -r/--runtime us: osnoise runtime in us", + " -s/--stop us: stop trace if a single sample is higher than the argument in us", + " -S/--stop-total us: stop trace if the total sample is higher than the argument in us", + " -c/--cpus cpu-list: list of cpus to run osnoise threads", + " -d/--duration time[s|m|h|d]: duration of the session", + " -t/--trace[=file]: save the stopped trace to [file|osnoise_trace.txt]", + " -q/--quiet print only a summary at the end", + " -P/--priority o:prio|r:prio|f:prio|d:runtime:period : set scheduling parameters", + " o:prio - use SCHED_OTHER with prio", + " r:prio - use SCHED_RR with prio", + " f:prio - use SCHED_FIFO with prio", + " d:runtime[us|ms|s]:period[us|ms|s] - use SCHED_DEADLINE with runtime and period", + " in nanoseconds", + NULL, + }; + + if (usage) + fprintf(stderr, "%s\n", usage); + + fprintf(stderr, "rtla osnoise top: a per-cpu summary of the OS noise (version %s)\n", + VERSION); + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +/* + * osnoise_top_parse_args - allocs, parse and fill the cmd line parameters + */ +struct osnoise_top_params *osnoise_top_parse_args(int argc, char **argv) +{ + struct osnoise_top_params *params; + int retval; + int c; + + params = calloc(1, sizeof(*params)); + if (!params) + exit(1); + + while (1) { + static struct option long_options[] = { + {"cpus", required_argument, 0, 'c'}, + {"debug", no_argument, 0, 'D'}, + {"duration", required_argument, 0, 'd'}, + {"help", no_argument, 0, 'h'}, + {"period", required_argument, 0, 'p'}, + {"priority", required_argument, 0, 'P'}, + {"quiet", no_argument, 0, 'q'}, + {"runtime", required_argument, 0, 'r'}, + {"stop", required_argument, 0, 's'}, + {"stop-total", required_argument, 0, 'S'}, + {"trace", optional_argument, 0, 't'}, + {0, 0, 0, 0} + }; + + /* getopt_long stores the option index here. */ + int option_index = 0; + + c = getopt_long(argc, argv, "c:d:Dhp:P:qr:s:S:t::", + long_options, &option_index); + + /* Detect the end of the options. */ + if (c == -1) + break; + + switch (c) { + case 'c': + retval = parse_cpu_list(optarg, ¶ms->monitored_cpus); + if (retval) + osnoise_top_usage("\nInvalid -c cpu list\n"); + params->cpus = optarg; + debug_msg("-c for %s\n", params->cpus); + break; + case 'D': + config_debug = 1; + break; + case 'd': + params->duration = parse_seconds_duration(optarg); + if (!params->duration) + osnoise_top_usage("Invalid -D duration\n"); + break; + case 'h': + case '?': + osnoise_top_usage(NULL); + break; + case 'p': + params->period = get_long_from_str(optarg); + if (params->period > 10000000) + osnoise_top_usage("Period longer than 10 s\n"); + break; + case 'P': + retval = parse_prio(optarg, ¶ms->sched_param); + if (retval == -1) + osnoise_top_usage("Invalid -P priority"); + params->set_sched = 1; + break; + case 'q': + params->quiet = 1; + break; + case 'r': + params->runtime = get_long_from_str(optarg); + if (params->runtime < 100) + osnoise_top_usage("Runtime shorter than 100 us\n"); + break; + case 's': + params->stop_us = get_long_from_str(optarg); + break; + case 'S': + params->stop_total_us = get_long_from_str(optarg); + break; + case 't': + if (optarg) + /* skip = */ + params->trace_output = &optarg[1]; + else + params->trace_output = "osnoise_trace.txt"; + break; + default: + osnoise_top_usage("Invalid option"); + } + } + + if (geteuid()) { + err_msg("osnoise needs root permission\n"); + exit(EXIT_FAILURE); + } + + return params; +} + +/* + * osnoise_top_apply_config - apply the top configs to the initialized tool + */ +static int +osnoise_top_apply_config(struct osnoise_tool *tool, struct osnoise_top_params *params) +{ + int retval; + + if (!params->sleep_time) + params->sleep_time = 1; + + if (params->cpus) { + retval = osnoise_set_cpus(tool->context, params->cpus); + if (retval) { + err_msg("Failed to apply CPUs config\n"); + goto out_err; + } + } + + if (params->runtime || params->period) { + retval = osnoise_set_runtime_period(tool->context, + params->runtime, + params->period); + if (retval) { + err_msg("Failed to set runtime and/or period\n"); + goto out_err; + } + } + + if (params->stop_us) { + retval = osnoise_set_stop_us(tool->context, params->stop_us); + if (retval) { + err_msg("Failed to set stop us\n"); + goto out_err; + } + } + + if (params->stop_total_us) { + retval = osnoise_set_stop_total_us(tool->context, params->stop_total_us); + if (retval) { + err_msg("Failed to set stop total us\n"); + goto out_err; + } + } + + return 0; + +out_err: + return -1; +} + +/* + * osnoise_init_top - initialize a osnoise top tool with parameters + */ +struct osnoise_tool *osnoise_init_top(struct osnoise_top_params *params) +{ + struct osnoise_tool *tool; + int nr_cpus; + + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + tool = osnoise_init_tool("osnoise_top"); + if (!tool) + return NULL; + + tool->data = osnoise_alloc_top(nr_cpus); + if (!tool->data) + goto out_err; + + tool->params = params; + + tep_register_event_handler(tool->trace.tep, -1, "ftrace", "osnoise", + osnoise_top_handler, NULL); + + return tool; + +out_err: + osnoise_free_top(tool->data); + osnoise_destroy_tool(tool); + return NULL; +} + +static int stop_tracing; +static void stop_top(int sig) +{ + stop_tracing = 1; +} + +/* + * osnoise_top_set_signals - handles the signal to stop the tool + */ +static void osnoise_top_set_signals(struct osnoise_top_params *params) +{ + signal(SIGINT, stop_top); + if (params->duration) { + signal(SIGALRM, stop_top); + alarm(params->duration); + } +} + +int osnoise_top_main(int argc, char **argv) +{ + struct osnoise_top_params *params; + struct trace_instance *trace; + struct osnoise_tool *record; + struct osnoise_tool *tool; + int return_value = 1; + int retval; + + params = osnoise_top_parse_args(argc, argv); + if (!params) + exit(1); + + tool = osnoise_init_top(params); + if (!tool) { + err_msg("Could not init osnoise top\n"); + goto out_exit; + } + + retval = osnoise_top_apply_config(tool, params); + if (retval) { + err_msg("Could not apply config\n"); + goto out_top; + } + + trace = &tool->trace; + + retval = enable_osnoise(trace); + if (retval) { + err_msg("Failed to enable osnoise tracer\n"); + goto out_top; + } + + if (params->set_sched) { + retval = set_comm_sched_attr("osnoise/", ¶ms->sched_param); + if (retval) + osnoise_top_usage("Failed to set sched parameters\n"); + } + + trace_instance_start(trace); + + if (params->trace_output) { + record = osnoise_init_trace_tool("osnoise"); + if (!record) { + err_msg("Failed to enable the trace instance\n"); + goto out_top; + } + trace_instance_start(&record->trace); + } + + tool->start_time = time(NULL); + osnoise_top_set_signals(params); + + do { + sleep(params->sleep_time); + + retval = tracefs_iterate_raw_events(trace->tep, + trace->inst, + NULL, + 0, + collect_registered_events, + trace); + if (retval < 0) { + err_msg("Error iterating on events\n"); + goto out_top; + } + + if (!params->quiet) + osnoise_print_stats(params, tool); + + if (!tracefs_trace_is_on(trace->inst)) + break; + + } while (!stop_tracing); + + osnoise_print_stats(params, tool); + + return_value = 0; + + if (!tracefs_trace_is_on(trace->inst)) { + printf("osnoise hit stop tracing\n"); + if (params->trace_output) { + printf(" Saving trace to %s\n", params->trace_output); + save_trace_to_file(record->trace.inst, params->trace_output); + } + } + +out_top: + osnoise_free_top(tool->data); + osnoise_destroy_tool(tool); + if (params->trace_output) + osnoise_destroy_tool(record); +out_exit: + exit(return_value); +} From patchwork Tue Oct 26 22:06:22 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585801 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9B113C4332F for ; Tue, 26 Oct 2021 22:08:53 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 8812D610E5 for ; Tue, 26 Oct 2021 22:08:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239888AbhJZWLR convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:17 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:59027 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239855AbhJZWLH (ORCPT ); Tue, 26 Oct 2021 18:11:07 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-221-LAYjVWPkMyOxmwNNisupzA-1; Tue, 26 Oct 2021 18:08:39 -0400 X-MC-Unique: LAYjVWPkMyOxmwNNisupzA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id C59A4A40D1; Tue, 26 Oct 2021 22:08:36 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 8903860BF1; Tue, 26 Oct 2021 22:08:31 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 11/20] rtla/osnoise: Add the hist mode Date: Wed, 27 Oct 2021 00:06:22 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The rtla osnoise hist tool collects all osnoise:sample_threshold occurrence in a histogram, displaying the results in a user-friendly way. The tool also allows many configurations of the osnoise tracer and the collection of the tracer output. Here is one example of the rtla osnoise hist tool output: ---------- %< ---------- [root@f34 ~]# rtla osnoise hist --bucket-size 10 --entries 100 -c 0-8 -d 1M -r 9000 -P F:1 # RTLA osnoise histogram # Time unit is microseconds (us) # Duration: 0 00:01:00 Index CPU-000 CPU-001 CPU-002 CPU-003 CPU-004 CPU-005 CPU-006 CPU-007 CPU-008 0 430 434 352 455 440 463 467 436 484 10 88 88 92 141 120 100 126 166 100 20 19 7 12 22 8 8 13 13 16 30 6 0 2 0 1 2 2 1 0 50 0 0 0 0 0 0 1 0 0 over: 0 0 0 0 0 0 0 0 0 count: 543 529 458 618 569 573 609 616 600 min: 0 0 0 0 0 0 0 0 0 avg: 0 0 0 0 0 0 0 0 0 max: 30 20 30 20 30 30 50 30 20 ---------- >% ---------- Running - rtla osnoise hist --help provides information about the available options. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/src/osnoise.c | 6 +- tools/tracing/rtla/src/osnoise.h | 1 + tools/tracing/rtla/src/osnoise_hist.c | 793 ++++++++++++++++++++++++++ 3 files changed, 799 insertions(+), 1 deletion(-) create mode 100644 tools/tracing/rtla/src/osnoise_hist.c diff --git a/tools/tracing/rtla/src/osnoise.c b/tools/tracing/rtla/src/osnoise.c index b9866dfdda66..6c6358ebf66f 100644 --- a/tools/tracing/rtla/src/osnoise.c +++ b/tools/tracing/rtla/src/osnoise.c @@ -790,7 +790,8 @@ static void osnoise_usage(void) " usage: [rtla] osnoise [MODE] ...", "", " modes:", - " top - prints the summary from osnoise tracer", + " top - prints the summary from osnoise tracer", + " hist - prints a histogram of osnoise samples", "", "if no MODE is given, the top mode is called, passing the arguments", NULL, @@ -825,6 +826,9 @@ int osnoise_main(int argc, char *argv[]) } else if (strcmp(argv[1], "top") == 0) { osnoise_top_main(argc-1, &argv[1]); exit(0); + } else if (strcmp(argv[1], "hist") == 0) { + osnoise_hist_main(argc-1, &argv[1]); + exit(0); } usage: diff --git a/tools/tracing/rtla/src/osnoise.h b/tools/tracing/rtla/src/osnoise.h index 4882ee275ea0..dbaad7a162a2 100644 --- a/tools/tracing/rtla/src/osnoise.h +++ b/tools/tracing/rtla/src/osnoise.h @@ -82,5 +82,6 @@ void osnoise_destroy_tool(struct osnoise_tool *top); struct osnoise_tool *osnoise_init_tool(char *tool_name); struct osnoise_tool *osnoise_init_trace_tool(char *tracer); +int osnoise_hist_main(int argc, char *argv[]); int osnoise_top_main(int argc, char **argv); int osnoise_main(int argc, char **argv); diff --git a/tools/tracing/rtla/src/osnoise_hist.c b/tools/tracing/rtla/src/osnoise_hist.c new file mode 100644 index 000000000000..c93f30d69795 --- /dev/null +++ b/tools/tracing/rtla/src/osnoise_hist.c @@ -0,0 +1,793 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "utils.h" +#include "osnoise.h" + +struct osnoise_hist_params { + char *cpus; + char *monitored_cpus; + char *trace_output; + unsigned long long runtime; + unsigned long long period; + long long stop_us; + long long stop_total_us; + int sleep_time; + int duration; + int set_sched; + int output_divisor; + struct sched_attr sched_param; + + char no_header; + char no_summary; + char no_index; + char with_zeros; + int bucket_size; + int entries; +}; + +struct osnoise_hist_cpu { + int *samples; + int *thread; + + int count; + int thread_count; + + unsigned long long min_sample; + unsigned long long sum_sample; + unsigned long long max_sample; + + unsigned long long min_thread; + unsigned long long sum_thread; + unsigned long long max_thread; +}; + +struct osnoise_hist_data { + struct tracefs_hist *trace_hist; + struct osnoise_hist_cpu *hist; + int entries; + int bucket_size; + int nr_cpus; +}; + +/* + * osnoise_free_histogram - free runtime data + */ +static void +osnoise_free_histogram(struct osnoise_hist_data *data) +{ + int cpu; + + /* one histogram for IRQ and one for thread, per CPU */ + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (data->hist[cpu].samples) + free(data->hist[cpu].samples); + + if (data->hist[cpu].thread) + free(data->hist[cpu].thread); + } + + /* one set of histograms per CPU */ + if (data->hist) + free(data->hist); + + free(data); +} + +/* + * osnoise_alloc_histogram - alloc runtime data + */ +static struct osnoise_hist_data +*osnoise_alloc_histogram(int nr_cpus, int entries, int bucket_size) +{ + struct osnoise_hist_data *data; + int cpu; + + data = calloc(1, sizeof(*data)); + if (!data) + return NULL; + + data->entries = entries; + data->bucket_size = bucket_size; + data->nr_cpus = nr_cpus; + + /* one set of histograms per CPU */ + data->hist = calloc(1, sizeof(*data->hist) * nr_cpus); + if (!data->hist) + goto cleanup; + + /* one histogram for IRQ and one for thread, per cpu */ + for (cpu = 0; cpu < nr_cpus; cpu++) { + data->hist[cpu].samples = calloc(1, sizeof(*data->hist) * (entries + 1)); + if (!data->hist[cpu].samples) + goto cleanup; + data->hist[cpu].thread = calloc(1, sizeof(*data->hist) * (entries + 1)); + if (!data->hist[cpu].thread) + goto cleanup; + } + + /* set the min to max */ + for (cpu = 0; cpu < nr_cpus; cpu++) { + data->hist[cpu].min_sample = ~0; + data->hist[cpu].min_thread = ~0; + } + + return data; + +cleanup: + osnoise_free_histogram(data); + return NULL; +} + +static void osnoise_hist_update_multiple(struct osnoise_tool *tool, int cpu, + unsigned long long duration, int count) +{ + struct osnoise_hist_params *params = tool->params; + struct osnoise_hist_data *data = tool->data; + int entries = data->entries; + int bucket; + int *hist; + + if (params->output_divisor) + duration = duration / params->output_divisor; + + if (data->bucket_size) + bucket = duration / data->bucket_size; + + hist = data->hist[cpu].samples; + data->hist[cpu].count += count; + update_min(&data->hist[cpu].min_sample, &duration); + update_sum(&data->hist[cpu].sum_sample, &duration); + update_max(&data->hist[cpu].max_sample, &duration); + + if (bucket < entries) + hist[bucket] += count; + else + hist[entries] += count; +} + +/* + * osnoise_destroy_trace_hist - disable events used to collect histogram + */ +static void osnoise_destroy_trace_hist(struct osnoise_tool *tool) +{ + struct osnoise_hist_data *data = tool->data; + + tracefs_hist_pause(tool->trace.inst, data->trace_hist); + tracefs_hist_destroy(tool->trace.inst, data->trace_hist); +} + +/* + * osnoise_init_trace_hist - enable events used to collect histogram + */ +static int osnoise_init_trace_hist(struct osnoise_tool *tool) +{ + struct osnoise_hist_params *params = tool->params; + struct osnoise_hist_data *data = tool->data; + int bucket_size; + char buff[128]; + int retval = 0; + + /* + * Set the size of the bucket. + */ + bucket_size = params->output_divisor * params->bucket_size; + snprintf(buff, sizeof(buff), "duration.buckets=%d", bucket_size); + + data->trace_hist = tracefs_hist_alloc(tool->trace.tep, "osnoise", "sample_threshold", + buff, TRACEFS_HIST_KEY_NORMAL); + if (!data->trace_hist) + return 1; + + retval = tracefs_hist_add_key(data->trace_hist, "cpu", 0); + if (retval) + goto out_err; + + retval = tracefs_hist_start(tool->trace.inst, data->trace_hist); + if (retval) + goto out_err; + + return 0; + +out_err: + osnoise_destroy_trace_hist(tool); + return 1; +} + +/* + * osnoise_read_trace_hist - parse histogram file and file osnoise histogram + */ +static void osnoise_read_trace_hist(struct osnoise_tool *tool) +{ + struct osnoise_hist_data *data = tool->data; + long cpu, counter, duration; + char *content, *position; + + tracefs_hist_pause(tool->trace.inst, data->trace_hist); + + content = tracefs_event_file_read(tool->trace.inst, "osnoise", + "sample_threshold", + "hist", NULL); + if (!content) + return; + + position = content; + while (true) { + position = strstr(position, "duration: ~"); + if (!position) + break; + position += strlen("duration: ~"); + duration = get_long_from_str(position); + if (duration == -1) + err_msg("error reading duration from histogram\n"); + + position = strstr(position, "cpu:"); + if (!position) + break; + position += strlen("cpu: "); + cpu = get_long_from_str(position); + if (cpu == -1) + err_msg("error reading cpu from histogram\n"); + + position = strstr(position, "hitcount:"); + if (!position) + break; + position += strlen("hitcount: "); + counter = get_long_from_str(position); + if (counter == -1) + err_msg("error reading counter from histogram\n"); + + osnoise_hist_update_multiple(tool, cpu, duration, counter); + } + free(content); +} + +/* + * osnoise_hist_header - print the header of the tracer to the output + */ +static void osnoise_hist_header(struct osnoise_tool *tool) +{ + struct osnoise_hist_params *params = tool->params; + struct osnoise_hist_data *data = tool->data; + struct trace_seq *s = tool->trace.seq; + char duration[26]; + int cpu; + + if (params->no_header) + return; + + get_duration(tool->start_time, duration, sizeof(duration)); + trace_seq_printf(s, "# RTLA osnoise histogram\n"); + trace_seq_printf(s, "# Time unit is %s (%s)\n", + params->output_divisor == 1 ? "nanoseconds" : "microseconds", + params->output_divisor == 1 ? "ns" : "us"); + + trace_seq_printf(s, "# Duration: %s\n", duration); + + if (!params->no_index) + trace_seq_printf(s, "Index"); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + trace_seq_printf(s, " CPU-%03d", cpu); + } + trace_seq_printf(s, "\n"); + + trace_seq_do_printf(s); + trace_seq_reset(s); +} + +/* + * osnoise_print_summary - print the summary of the hist data to the output + */ +static void +osnoise_print_summary(struct osnoise_hist_params *params, + struct trace_instance *trace, + struct osnoise_hist_data *data) +{ + int cpu; + + if (params->no_summary) + return; + + if (!params->no_index) + trace_seq_printf(trace->seq, "count:"); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + trace_seq_printf(trace->seq, "%9d ", data->hist[cpu].count); + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "min: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + trace_seq_printf(trace->seq, "%9llu ", data->hist[cpu].min_sample); + + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "avg: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (data->hist[cpu].count) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].sum_sample / data->hist[cpu].count); + else + trace_seq_printf(trace->seq, " - "); + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "max: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + trace_seq_printf(trace->seq, "%9llu ", data->hist[cpu].max_sample); + + } + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); +} + +/* + * osnoise_print_stats - print data for all CPUs + */ +static void +osnoise_print_stats(struct osnoise_hist_params *params, struct osnoise_tool *tool) +{ + struct osnoise_hist_data *data = tool->data; + struct trace_instance *trace = &tool->trace; + int bucket, cpu; + int total; + + osnoise_hist_header(tool); + + for (bucket = 0; bucket < data->entries; bucket++) { + total = 0; + + if (!params->no_index) + trace_seq_printf(trace->seq, "%-6d", + bucket * data->bucket_size); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + total += data->hist[cpu].samples[bucket]; + trace_seq_printf(trace->seq, "%9d ", data->hist[cpu].samples[bucket]); + } + + if (total == 0 && !params->with_zeros) { + trace_seq_reset(trace->seq); + continue; + } + + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); + } + + if (!params->no_index) + trace_seq_printf(trace->seq, "over: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].samples[data->entries]); + } + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); + + osnoise_print_summary(params, trace, data); +} + +/* + * osnoise_hist_usage - prints osnoise hist usage message + */ +static void osnoise_hist_usage(char *usage) +{ + int i; + + static const char * const msg[] = { + "", + " usage: rtla osnoise hist [-h] [-p us] [-r us] [-s us] [-S us] [-T[=file]] \\", + " [-c cpu-list] [-P priority] [-b N] [-e N] [--no-header] [--no-summary] \\", + " [--no-index] [--with-zeros]", + "", + " -h/--help: print this menu", + " -p/--period us: osnoise period in us", + " -r/--runtime us: osnoise runtime in us", + " -s/--stop us: stop trace if a single sample is higher than the argument in us", + " -S/--stop-total us: stop trace if the total sample is higher than the argument in us", + " -c/--cpus cpu-list: list of cpus to run osnoise threads", + " -d/--duration time[s|m|h|d]: duration of the session", + " -t/--trace[=file]: save the stopped trace to [file|osnoise_trace.txt]", + " -b/--bucket-size N: set the histogram bucket size (default 1)", + " -e/--entries N: set the number of entries of the histogram (default 256)", + " --no-header: do not print header", + " --no-summary: do not print summary", + " --no-index: do not print index", + " --with-zeros: print zero only entries", + " -P/--priority o:prio|r:prio|f:prio|d:runtime:period: set scheduling parameters", + " o:prio - use SCHED_OTHER with prio", + " r:prio - use SCHED_RR with prio", + " f:prio - use SCHED_FIFO with prio", + " d:runtime[us|ms|s]:period[us|ms|s] - use SCHED_DEADLINE with runtime and period", + " in nanoseconds", + NULL, + }; + + if (usage) + fprintf(stderr, "%s\n", usage); + + fprintf(stderr, "rtla osnoise hist: a per-cpu histogram of the OS noise (version %s)\n", + VERSION); + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +/* + * osnoise_hist_parse_args - allocs, parse and fill the cmd line parameters + */ +static struct osnoise_hist_params +*osnoise_hist_parse_args(int argc, char *argv[]) +{ + struct osnoise_hist_params *params; + int retval; + int c; + + params = calloc(1, sizeof(*params)); + if (!params) + exit(1); + + /* display data in microseconds */ + params->output_divisor = 1000; + params->bucket_size = 1; + params->entries = 256; + + while (1) { + static struct option long_options[] = { + {"bucket-size", required_argument, 0, 'b'}, + {"entries", required_argument, 0, 'e'}, + {"cpus", required_argument, 0, 'c'}, + {"debug", no_argument, 0, 'D'}, + {"duration", required_argument, 0, 'd'}, + {"help", no_argument, 0, 'h'}, + {"period", required_argument, 0, 'p'}, + {"priority", required_argument, 0, 'P'}, + {"runtime", required_argument, 0, 'r'}, + {"stop", required_argument, 0, 's'}, + {"stop-total", required_argument, 0, 'S'}, + {"trace", optional_argument, 0, 't'}, + {"no-header", no_argument, 0, '0'}, + {"no-summary", no_argument, 0, '1'}, + {"no-index", no_argument, 0, '2'}, + {"with-zeros", no_argument, 0, '3'}, + {0, 0, 0, 0} + }; + + /* getopt_long stores the option index here. */ + int option_index = 0; + + c = getopt_long(argc, argv, "c:b:d:e:Dhp:P:r:s:S:t::0123", + long_options, &option_index); + + /* detect the end of the options. */ + if (c == -1) + break; + + switch (c) { + case 'b': + params->bucket_size = get_long_from_str(optarg); + if ((params->bucket_size == 0) || (params->bucket_size >= 1000000)) + osnoise_hist_usage("Bucket size needs to be > 0 and <= 1000000\n"); + break; + case 'c': + retval = parse_cpu_list(optarg, ¶ms->monitored_cpus); + if (retval) + osnoise_hist_usage("\nInvalid -c cpu list\n"); + params->cpus = optarg; + debug_msg("-c for %s\n", params->cpus); + break; + case 'D': + config_debug = 1; + break; + case 'd': + params->duration = parse_seconds_duration(optarg); + if (!params->duration) + osnoise_hist_usage("Invalid -D duration\n"); + break; + case 'e': + params->entries = get_long_from_str(optarg); + if ((params->entries < 10) || (params->entries > 9999999)) + osnoise_hist_usage("Entries must be > 10 and < 9999999\n"); + break; + case 'h': + case '?': + osnoise_hist_usage(NULL); + break; + case 'p': + params->period = get_long_from_str(optarg); + if (params->period > 10000000) + osnoise_hist_usage("Period longer than 10 s\n"); + break; + case 'P': + retval = parse_prio(optarg, ¶ms->sched_param); + if (retval == -1) + osnoise_hist_usage("Invalid -P priority"); + params->set_sched = 1; + break; + case 'r': + params->runtime = get_long_from_str(optarg); + if (params->runtime < 100) + osnoise_hist_usage("Runtime shorter than 100 us\n"); + break; + case 's': + params->stop_us = get_long_from_str(optarg); + break; + case 'S': + params->stop_total_us = get_long_from_str(optarg); + break; + case 't': + if (optarg) + /* skip = */ + params->trace_output = &optarg[1]; + else + params->trace_output = "osnoise_trace.txt"; + break; + case '0': /* no header */ + params->no_header = 1; + break; + case '1': /* no summary */ + params->no_summary = 1; + break; + case '2': /* no index */ + params->no_index = 1; + break; + case '3': /* with zeros */ + params->with_zeros = 1; + break; + default: + osnoise_hist_usage("Invalid option"); + } + } + + if (geteuid()) { + err_msg("rtla needs root permission\n"); + exit(EXIT_FAILURE); + } + + if (params->no_index && !params->with_zeros) + osnoise_hist_usage("no-index set and with-zeros not set - it does not make sense"); + + return params; +} + +/* + * osnoise_hist_apply_config - apply the hist configs to the initialized tool + */ +static int +osnoise_hist_apply_config(struct osnoise_tool *tool, struct osnoise_hist_params *params) +{ + int retval; + + if (!params->sleep_time) + params->sleep_time = 1; + + if (params->cpus) { + retval = osnoise_set_cpus(tool->context, params->cpus); + if (retval) { + err_msg("Failed to apply CPUs config\n"); + goto out_err; + } + } + + if (params->runtime || params->period) { + retval = osnoise_set_runtime_period(tool->context, + params->runtime, + params->period); + if (retval) { + err_msg("Failed to set runtime and/or period\n"); + goto out_err; + } + } + + if (params->stop_us) { + retval = osnoise_set_stop_us(tool->context, params->stop_us); + if (retval) { + err_msg("Failed to set stop us\n"); + goto out_err; + } + } + + if (params->stop_total_us) { + retval = osnoise_set_stop_total_us(tool->context, params->stop_total_us); + if (retval) { + err_msg("Failed to set stop total us\n"); + goto out_err; + } + } + + return 0; + +out_err: + return -1; +} + +/* + * osnoise_init_hist - initialize a osnoise hist tool with parameters + */ +static struct osnoise_tool +*osnoise_init_hist(struct osnoise_hist_params *params) +{ + struct osnoise_tool *tool; + int nr_cpus; + + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + tool = osnoise_init_tool("osnoise_hist"); + if (!tool) + return NULL; + + tool->data = osnoise_alloc_histogram(nr_cpus, params->entries, params->bucket_size); + if (!tool->data) + goto out_err; + + tool->params = params; + + return tool; + +out_err: + osnoise_destroy_tool(tool); + return NULL; +} + +static int stop_tracing; +static void stop_hist(int sig) +{ + stop_tracing = 1; +} + +/* + * osnoise_hist_set_signals - handles the signal to stop the tool + */ +static void +osnoise_hist_set_signals(struct osnoise_hist_params *params) +{ + signal(SIGINT, stop_hist); + if (params->duration) { + signal(SIGALRM, stop_hist); + alarm(params->duration); + } +} + +int osnoise_hist_main(int argc, char *argv[]) +{ + struct osnoise_hist_params *params; + struct trace_instance *trace; + struct osnoise_tool *record; + struct osnoise_tool *tool; + int return_value = 1; + int retval; + + params = osnoise_hist_parse_args(argc, argv); + if (!params) + exit(1); + + tool = osnoise_init_hist(params); + if (!tool) { + err_msg("Could not init osnoise hist\n"); + goto out_exit; + } + + retval = osnoise_hist_apply_config(tool, params); + if (retval) { + err_msg("Could not apply config\n"); + goto out_destroy; + } + + trace = &tool->trace; + + retval = enable_osnoise(trace); + if (retval) { + err_msg("Failed to enable osnoise tracer\n"); + goto out_destroy; + } + + retval = osnoise_init_trace_hist(tool); + if (retval) + goto out_destroy; + + if (params->set_sched) { + retval = set_comm_sched_attr("osnoise/", ¶ms->sched_param); + if (retval) + osnoise_hist_usage("Failed to set sched parameters\n"); + } + + trace_instance_start(trace); + + if (params->trace_output) { + record = osnoise_init_trace_tool("osnoise"); + if (!record) { + err_msg("Failed to enable the trace instance\n"); + goto out_hist; + } + trace_instance_start(&record->trace); + } + + tool->start_time = time(NULL); + osnoise_hist_set_signals(params); + + while (!stop_tracing) { + sleep(params->sleep_time); + + retval = tracefs_iterate_raw_events(trace->tep, + trace->inst, + NULL, + 0, + collect_registered_events, + trace); + if (retval < 0) { + err_msg("Error iterating on events\n"); + goto out_hist; + } + + if (!tracefs_trace_is_on(trace->inst)) + break; + }; + + osnoise_read_trace_hist(tool); + + osnoise_print_stats(params, tool); + + return_value = 0; + + if (!tracefs_trace_is_on(trace->inst)) { + printf("rtla timelat hit stop tracing\n"); + if (params->trace_output) { + printf(" Saving trace to %s\n", params->trace_output); + save_trace_to_file(record->trace.inst, params->trace_output); + } + } + +out_hist: + osnoise_free_histogram(tool->data); +out_destroy: + osnoise_destroy_tool(tool); + if (params->trace_output) + osnoise_destroy_tool(record); + free(params); +out_exit: + exit(return_value); +} From patchwork Tue Oct 26 22:06:23 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585805 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9571BC433EF for ; Tue, 26 Oct 2021 22:08:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 81B07608FE for ; Tue, 26 Oct 2021 22:08:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239904AbhJZWLU convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:20 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:26162 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239865AbhJZWLM (ORCPT ); Tue, 26 Oct 2021 18:11:12 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-203-6alPer-KNGu1LNmScVdNbw-1; Tue, 26 Oct 2021 18:08:44 -0400 X-MC-Unique: 6alPer-KNGu1LNmScVdNbw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 542E2112A6A0; Tue, 26 Oct 2021 22:08:42 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 60E0560BF1; Tue, 26 Oct 2021 22:08:36 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 12/20] rtla: Add timerlat tool and timelart top mode Date: Wed, 27 Oct 2021 00:06:23 +0200 Message-Id: <3cf784d2e0eb826481935255eba3bceefb638cf1.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The rtla timerlat tool is an interface for the timerlat tracer. The timerlat tracer dispatches a kernel thread per-cpu. These threads set a periodic timer to wake themselves up and go back to sleep. After the wakeup, they collect and generate useful information for the debugging of operating system timer latency. The timerlat tracer outputs information in two ways. It periodically prints the timer latency at the timer IRQ handler and the Thread handler. It also provides information for each noise via the osnoise tracepoints. The rtla timerlat top mode displays a summary of the periodic output from the timerlat tracer. Here is one example of the rtla timerlat tool output: ---------- %< ---------- [root@alien ~]# rtla timerlat top -c 0-3 -d 1m Timer Latency 0 00:01:00 | IRQ Timer Latency (us) | Thread Timer Latency (us) CPU COUNT | cur min avg max | cur min avg max 0 #60001 | 0 0 0 3 | 1 1 1 6 1 #60001 | 0 0 0 3 | 2 1 1 5 2 #60001 | 0 0 1 6 | 1 1 2 7 3 #60001 | 0 0 0 7 | 1 1 1 11 ---------- >% ---------- Running: # rtla timerlat --help # rtla timerlat top --help provides information about the available options. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/Makefile | 2 + tools/tracing/rtla/src/rtla.c | 5 + tools/tracing/rtla/src/timerlat.c | 68 +++ tools/tracing/rtla/src/timerlat.h | 4 + tools/tracing/rtla/src/timerlat_top.c | 603 ++++++++++++++++++++++++++ 5 files changed, 682 insertions(+) create mode 100644 tools/tracing/rtla/src/timerlat.c create mode 100644 tools/tracing/rtla/src/timerlat.h create mode 100644 tools/tracing/rtla/src/timerlat_top.c diff --git a/tools/tracing/rtla/Makefile b/tools/tracing/rtla/Makefile index 5e93d000a821..3f8d5421ccc1 100644 --- a/tools/tracing/rtla/Makefile +++ b/tools/tracing/rtla/Makefile @@ -62,6 +62,8 @@ install: $(STRIP) $(DESTDIR)$(BINDIR)/rtla @test ! -f $(DESTDIR)$(BINDIR)/osnoise || rm $(DESTDIR)$(BINDIR)/osnoise ln -s $(DESTDIR)$(BINDIR)/rtla $(DESTDIR)$(BINDIR)/osnoise + @test ! -f $(DESTDIR)$(BINDIR)/timerlat || rm $(DESTDIR)$(BINDIR)/timerlat + ln -s $(DESTDIR)$(BINDIR)/rtla $(DESTDIR)$(BINDIR)/timerlat .PHONY: clean tarball clean: diff --git a/tools/tracing/rtla/src/rtla.c b/tools/tracing/rtla/src/rtla.c index 669b9750b3b3..09bd21b8af81 100644 --- a/tools/tracing/rtla/src/rtla.c +++ b/tools/tracing/rtla/src/rtla.c @@ -9,6 +9,7 @@ #include #include "osnoise.h" +#include "timerlat.h" /* * rtla_usage - print rtla usage @@ -25,6 +26,7 @@ static void rtla_usage(void) "", " commands:", " osnoise - gives information about the operating system noise (osnoise)", + " timerlat - measures the timer irq and thread latency", "", NULL, }; @@ -45,6 +47,9 @@ int run_command(int argc, char **argv, int start_position) if (strcmp(argv[start_position], "osnoise") == 0) { osnoise_main(argc-start_position, &argv[start_position]); goto ran; + } else if (strcmp(argv[start_position], "timerlat") == 0) { + timerlat_main(argc-start_position, &argv[start_position]); + goto ran; } return 0; diff --git a/tools/tracing/rtla/src/timerlat.c b/tools/tracing/rtla/src/timerlat.c new file mode 100644 index 000000000000..29a51c01b084 --- /dev/null +++ b/tools/tracing/rtla/src/timerlat.c @@ -0,0 +1,68 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "timerlat.h" + +static void timerlat_usage(void) +{ + int i; + + static const char * const msg[] = { + "", + "timerlat version " VERSION, + "", + " usage: [rtla] timerlat [MODE] ...", + "", + " modes:", + " top - prints the summary from timerlat tracer", + "", + "if no MODE is given, the top mode is called, passing the arguments", + NULL, + }; + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +int timerlat_main(int argc, char *argv[]) +{ + if (argc == 0) + goto usage; + + /* + * if timerlat was called without any argument, run the + * default cmdline. + */ + if (argc == 1) { + timerlat_top_main(argc, argv); + exit(0); + } + + if ((strcmp(argv[1], "-h") == 0) || (strcmp(argv[1], "--help") == 0)) { + timerlat_usage(); + exit(0); + } else if (strncmp(argv[1], "-", 1) == 0) { + /* the user skipped the tool, call the default one */ + timerlat_top_main(argc, argv); + exit(0); + } else if (strcmp(argv[1], "top") == 0) { + timerlat_top_main(argc-1, &argv[1]); + exit(0); + } + +usage: + timerlat_usage(); + exit(1); +} diff --git a/tools/tracing/rtla/src/timerlat.h b/tools/tracing/rtla/src/timerlat.h new file mode 100644 index 000000000000..6f9d3925acd0 --- /dev/null +++ b/tools/tracing/rtla/src/timerlat.h @@ -0,0 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 + +int timerlat_top_main(int argc, char *argv[]); +int timerlat_main(int argc, char *argv[]); diff --git a/tools/tracing/rtla/src/timerlat_top.c b/tools/tracing/rtla/src/timerlat_top.c new file mode 100644 index 000000000000..239f8cf3c221 --- /dev/null +++ b/tools/tracing/rtla/src/timerlat_top.c @@ -0,0 +1,603 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "utils.h" +#include "osnoise.h" +#include "timerlat.h" + +struct timerlat_top_params { + char *cpus; + char *monitored_cpus; + char *trace_output; + unsigned long long runtime; + long long stop_us; + long long stop_total_us; + long long timerlat_period_us; + long long print_stack; + int sleep_time; + int output_divisor; + int duration; + int quiet; + int set_sched; + struct sched_attr sched_param; +}; + +struct timerlat_top_cpu { + int irq_count; + int thread_count; + + unsigned long long cur_irq; + unsigned long long min_irq; + unsigned long long sum_irq; + unsigned long long max_irq; + + unsigned long long cur_thread; + unsigned long long min_thread; + unsigned long long sum_thread; + unsigned long long max_thread; +}; + +struct timerlat_top_data { + struct timerlat_top_cpu *cpu_data; + int nr_cpus; +}; + +/* + * timerlat_free_top - free runtime data + */ +static void +timerlat_free_top(struct timerlat_top_data *data) +{ + free(data->cpu_data); + free(data); +} + +/* + * timerlat_alloc_histogram - alloc runtime data + */ +static struct timerlat_top_data *timerlat_alloc_top(int nr_cpus) +{ + struct timerlat_top_data *data; + int cpu; + + data = calloc(1, sizeof(*data)); + if (!data) + return NULL; + + data->nr_cpus = nr_cpus; + + /* one set of histograms per CPU */ + data->cpu_data = calloc(1, sizeof(*data->cpu_data) * nr_cpus); + if (!data->cpu_data) + goto cleanup; + + /* set the min to max */ + for (cpu = 0; cpu < nr_cpus; cpu++) { + data->cpu_data[cpu].min_irq = ~0; + data->cpu_data[cpu].min_thread = ~0; + } + + return data; + +cleanup: + timerlat_free_top(data); + return NULL; +} + +/* + * timerlat_hist_update - record a new timerlat occurent on cpu, updating data + */ +static void +timerlat_top_update(struct osnoise_tool *tool, int cpu, + unsigned long long thread, + unsigned long long latency) +{ + struct timerlat_top_data *data = tool->data; + struct timerlat_top_cpu *cpu_data = &data->cpu_data[cpu]; + + if (!thread) { + cpu_data->irq_count++; + cpu_data->cur_irq = latency; + update_min(&cpu_data->min_irq, &latency); + update_sum(&cpu_data->sum_irq, &latency); + update_max(&cpu_data->max_irq, &latency); + } else { + cpu_data->thread_count++; + cpu_data->cur_thread = latency; + update_min(&cpu_data->min_thread, &latency); + update_sum(&cpu_data->sum_thread, &latency); + update_max(&cpu_data->max_thread, &latency); + } +} + +/* + * timerlat_top_handler - this is the handler for timerlat tracer events + */ +static int +timerlat_top_handler(struct trace_seq *s, struct tep_record *record, + struct tep_event *event, void *context) +{ + struct trace_instance *trace = context; + unsigned long long latency, thread; + struct osnoise_tool *top; + int cpu = record->cpu; + + top = container_of(trace, struct osnoise_tool, trace); + + tep_get_field_val(s, event, "context", record, &thread, 1); + tep_get_field_val(s, event, "timer_latency", record, &latency, 1); + + timerlat_top_update(top, cpu, thread, latency); + + return 0; +} + +/* + * timerlat_top_header - print the header of the tool output + */ +static void timerlat_top_header(struct osnoise_tool *top) +{ + struct timerlat_top_params *params = top->params; + struct trace_seq *s = top->trace.seq; + char duration[26]; + + get_duration(top->start_time, duration, sizeof(duration)); + + trace_seq_printf(s, "\033[2;37;40m"); + trace_seq_printf(s, " Timer Latency "); + trace_seq_printf(s, "\033[0;0;0m"); + trace_seq_printf(s, "\n"); + + trace_seq_printf(s, "%-6s | IRQ Timer Latency (%s) | Thread Timer Latency (%s)\n", duration, + params->output_divisor == 1 ? "ns" : "us", + params->output_divisor == 1 ? "ns" : "us"); + + trace_seq_printf(s, "\033[2;30;47m"); + trace_seq_printf(s, "CPU COUNT | cur min avg max | cur min avg max"); + trace_seq_printf(s, "\033[0;0;0m"); + trace_seq_printf(s, "\n"); +} + +/* + * timerlat_top_print - prints the output of a given CPU + */ +static void timerlat_top_print(struct osnoise_tool *top, int cpu) +{ + + struct timerlat_top_params *params = top->params; + struct timerlat_top_data *data = top->data; + struct timerlat_top_cpu *cpu_data = &data->cpu_data[cpu]; + int divisor = params->output_divisor; + struct trace_seq *s = top->trace.seq; + + if (divisor == 0) + return; + + trace_seq_printf(s, "%3d #%-9d |", cpu, cpu_data->irq_count); + + trace_seq_printf(s, "%9llu ", cpu_data->cur_irq / params->output_divisor); + trace_seq_printf(s, "%9llu ", cpu_data->min_irq / params->output_divisor); + + if (!cpu_data->irq_count) + trace_seq_printf(s, " "); + else + trace_seq_printf(s, "%9llu ", (cpu_data->sum_irq / cpu_data->irq_count) / divisor); + + trace_seq_printf(s, "%9llu |", cpu_data->max_irq / divisor); + + trace_seq_printf(s, "%9llu ", cpu_data->cur_thread / divisor); + trace_seq_printf(s, "%9llu ", cpu_data->min_thread / divisor); + + if (!cpu_data->thread_count) { + trace_seq_printf(s, " "); + } else { + trace_seq_printf(s, "%9llu ", + (cpu_data->sum_thread / cpu_data->thread_count) / divisor); + } + + trace_seq_printf(s, "%9llu\n", cpu_data->max_thread / divisor); +} + +/* + * clear_terminal - clears the output terminal + */ +static void clear_terminal(struct trace_seq *seq) +{ + if (!config_debug) + trace_seq_printf(seq, "\033c"); +} + +/* + * timerlat_print_stats - print data for all cpus + */ +static void +timerlat_print_stats(struct timerlat_top_params *params, struct osnoise_tool *top) +{ + struct trace_instance *trace = &top->trace; + static int nr_cpus = -1; + int i; + + if (nr_cpus == -1) + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + if (!params->quiet) + clear_terminal(trace->seq); + + timerlat_top_header(top); + + for (i = 0; i < nr_cpus; i++) { + if (params->cpus && !params->monitored_cpus[i]) + continue; + timerlat_top_print(top, i); + } + + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); +} + +/* + * timerlat_top_usage - prints timerlat top usage message + */ +static void timerlat_top_usage(char *usage) +{ + int i; + + static const char *const msg[] = { + "", + " usage: rtla timerlat [top] [-h] [-q] [-p us] [-i us] [-t us] [-s us] [-T[=file]] \\", + " [-c cpu-list] [-P priority]", + "", + " -h/--help: print this menu", + " -p/--period us: timerlat period in us", + " -i/--irq us: stop trace if the irq latency is higher than the argument in us", + " -T/--thread us: stop trace if the thread latency is higher than the argument in us", + " -s/--stack us: save the stack trace at the IRQ if a thread latency is higher than the argument in us", + " -c/--cpus cpus: run the tracer only on the given cpus -d/--duration time[m|h|d]: duration of the session in seconds", + " -t/--trace[=file]: save the stopped trace to [file|timerlat_trace.txt]", + " -n/--nano: display data in nanoseconds", + " -q/--quiet print only a summary at the end", + " -P/--priority o:prio|r:prio|f:prio|d:runtime:period : set scheduling parameters", + " o:prio - use SCHED_OTHER with prio", + " r:prio - use SCHED_RR with prio", + " f:prio - use SCHED_FIFO with prio", + " d:runtime[us|ms|s]:period[us|ms|s] - use SCHED_DEADLINE with runtime and period", + " in nanoseconds", + NULL, + }; + + if (usage) + fprintf(stderr, "%s\n", usage); + + fprintf(stderr, "rtla timerlat top: a per-cpu summary of the timer latency (version %s)\n", + VERSION); + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +/* + * timerlat_top_parse_args - allocs, parse and fill the cmd line parameters + */ +static struct timerlat_top_params +*timerlat_top_parse_args(int argc, char **argv) +{ + struct timerlat_top_params *params; + int retval; + int c; + + params = calloc(1, sizeof(*params)); + if (!params) + exit(1); + + /* display data in microseconds */ + params->output_divisor = 1000; + + while (1) { + static struct option long_options[] = { + {"cpus", required_argument, 0, 'c'}, + {"debug", no_argument, 0, 'D'}, + {"duration", required_argument, 0, 'd'}, + {"help", no_argument, 0, 'h'}, + {"irq", required_argument, 0, 'i'}, + {"nano", no_argument, 0, 'n'}, + {"period", required_argument, 0, 'p'}, + {"priority", required_argument, 0, 'P'}, + {"quiet", no_argument, 0, 'q'}, + {"stack", required_argument, 0, 's'}, + {"thread", required_argument, 0, 'T'}, + {"trace", optional_argument, 0, 't'}, + {0, 0, 0, 0} + }; + + /* getopt_long stores the option index here. */ + int option_index = 0; + + c = getopt_long(argc, argv, "c:d:Dhi:np:P:qs:t::T:", + long_options, &option_index); + + /* detect the end of the options. */ + if (c == -1) + break; + + switch (c) { + case 'c': + retval = parse_cpu_list(optarg, ¶ms->monitored_cpus); + if (retval) + timerlat_top_usage("\nInvalid -c cpu list\n"); + params->cpus = optarg; + debug_msg("-c for %s\n", params->cpus); + break; + case 'D': + config_debug = 1; + break; + case 'd': + params->duration = parse_seconds_duration(optarg); + if (!params->duration) + timerlat_top_usage("Invalid -D duration\n"); + break; + case 'h': + case '?': + timerlat_top_usage(NULL); + break; + case 'i': + params->stop_us = get_long_from_str(optarg); + break; + case 'n': + params->output_divisor = 1; + break; + case 'p': + params->timerlat_period_us = get_long_from_str(optarg); + if (params->timerlat_period_us > 1000000) + timerlat_top_usage("Period longer than 1 s\n"); + break; + case 'P': + retval = parse_prio(optarg, ¶ms->sched_param); + if (retval == -1) + timerlat_top_usage("Invalid -P priority"); + params->set_sched = 1; + break; + case 'q': + params->quiet = 1; + break; + case 's': + params->print_stack = get_long_from_str(optarg); + break; + case 'T': + params->stop_total_us = get_long_from_str(optarg); + break; + case 't': + if (optarg) + /* skip = */ + params->trace_output = &optarg[1]; + else + params->trace_output = "timerlat_trace.txt"; + break; + default: + timerlat_top_usage("Invalid option"); + } + } + + if (geteuid()) { + err_msg("rtla needs root permission\n"); + exit(EXIT_FAILURE); + } + + return params; +} + +/* + * timerlat_top_apply_config - apply the top configs to the initialized tool + */ +static int +timerlat_top_apply_config(struct osnoise_tool *top, struct timerlat_top_params *params) +{ + int retval; + + if (!params->sleep_time) + params->sleep_time = 1; + + if (params->cpus) { + retval = osnoise_set_cpus(top->context, params->cpus); + if (retval) { + err_msg("Failed to apply CPUs config\n"); + goto out_err; + } + } + + if (params->stop_us) { + retval = osnoise_set_stop_us(top->context, params->stop_us); + if (retval) { + err_msg("Failed to set stop us\n"); + goto out_err; + } + } + + if (params->stop_total_us) { + retval = osnoise_set_stop_total_us(top->context, params->stop_total_us); + if (retval) { + err_msg("Failed to set stop total us\n"); + goto out_err; + } + } + + + if (params->timerlat_period_us) { + retval = osnoise_set_timerlat_period_us(top->context, params->timerlat_period_us); + if (retval) { + err_msg("Failed to set timerlat period\n"); + goto out_err; + } + } + + + if (params->print_stack) { + retval = osnoise_set_print_stack(top->context, params->print_stack); + if (retval) { + err_msg("Failed to set print stack\n"); + goto out_err; + } + } + + return 0; + +out_err: + return -1; +} + +/* + * timerlat_init_top - initialize a timerlat top tool with parameters + */ +static struct osnoise_tool +*timerlat_init_top(struct timerlat_top_params *params) +{ + struct osnoise_tool *top; + int nr_cpus; + + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + top = osnoise_init_tool("timerlat_top"); + if (!top) + return NULL; + + top->data = timerlat_alloc_top(nr_cpus); + if (!top->data) + goto out_err; + + top->params = params; + + tep_register_event_handler(top->trace.tep, -1, "ftrace", "timerlat", + timerlat_top_handler, top); + + return top; + +out_err: + osnoise_destroy_tool(top); + return NULL; +} + +static int stop_tracing; +static void stop_top(int sig) +{ + stop_tracing = 1; +} + +/* + * timerlat_top_set_signals - handles the signal to stop the tool + */ +static void +timerlat_top_set_signals(struct timerlat_top_params *params) +{ + signal(SIGINT, stop_top); + if (params->duration) { + signal(SIGALRM, stop_top); + alarm(params->duration); + } +} + +int timerlat_top_main(int argc, char *argv[]) +{ + struct timerlat_top_params *params; + struct trace_instance *trace; + struct osnoise_tool *record; + struct osnoise_tool *top; + int return_value = 1; + int retval; + + params = timerlat_top_parse_args(argc, argv); + if (!params) + exit(1); + + top = timerlat_init_top(params); + if (!top) { + err_msg("Could not init osnoise top\n"); + goto out_exit; + } + + retval = timerlat_top_apply_config(top, params); + if (retval) { + err_msg("Could not apply config\n"); + goto out_top; + } + + trace = &top->trace; + + retval = enable_timerlat(trace); + if (retval) { + err_msg("Failed to enable timerlat tracer\n"); + goto out_top; + } + + if (params->set_sched) { + retval = set_comm_sched_attr("timerlat/", ¶ms->sched_param); + if (retval) + timerlat_top_usage("Failed to set sched parameters\n"); + } + + trace_instance_start(trace); + + if (params->trace_output) { + record = osnoise_init_trace_tool("timerlat"); + if (!record) { + err_msg("Failed to enable the trace instance\n"); + goto out_top; + } + trace_instance_start(&record->trace); + } + + top->start_time = time(NULL); + timerlat_top_set_signals(params); + + while (!stop_tracing) { + sleep(params->sleep_time); + + retval = tracefs_iterate_raw_events(trace->tep, + trace->inst, + NULL, + 0, + collect_registered_events, + trace); + if (retval < 0) { + err_msg("Error iterating on events\n"); + goto out_top; + } + + if (!params->quiet) + timerlat_print_stats(params, top); + + if (!tracefs_trace_is_on(trace->inst)) + break; + + }; + + timerlat_print_stats(params, top); + + return_value = 0; + + if (!tracefs_trace_is_on(trace->inst)) { + printf("rtla timelat hit stop tracing\n"); + if (params->trace_output) { + printf(" Saving trace to %s\n", params->trace_output); + save_trace_to_file(record->trace.inst, params->trace_output); + } + } + +out_top: + timerlat_free_top(top->data); + osnoise_destroy_tool(top); + if (params->trace_output) + osnoise_destroy_tool(record); + free(params); +out_exit: + exit(return_value); +} From patchwork Tue Oct 26 22:06:24 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585807 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 05947C433F5 for ; Tue, 26 Oct 2021 22:09:03 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id E3D0C60F92 for ; Tue, 26 Oct 2021 22:09:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239929AbhJZWL0 convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:26 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:41739 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239887AbhJZWLR (ORCPT ); Tue, 26 Oct 2021 18:11:17 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-211-BzCiv6tDONmmJLC1V27UMA-1; Tue, 26 Oct 2021 18:08:49 -0400 X-MC-Unique: BzCiv6tDONmmJLC1V27UMA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id BBD9510247A6; Tue, 26 Oct 2021 22:08:47 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id E238260BF1; Tue, 26 Oct 2021 22:08:42 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 13/20] rtla/timerlat: Add timerlat hist mode Date: Wed, 27 Oct 2021 00:06:24 +0200 Message-Id: <70e3b0f73ea94b0acb53af2188d187bc3f60590a.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org The rtla hist hist mode displays a histogram of each tracer event occurrence, both for IRQ and timer latencies. The tool also allows many configurations of the timerlat tracer and the collection of the tracer output. Here is one example of the rtla timerlat hist mode output: ---------- %< ---------- [root@alien ~]# rtla timerlat hist -c 0-3 -d 1M # RTLA timerlat histogram # Time unit is microseconds (us) # Duration: 0 00:01:00 Index IRQ-000 Thr-000 IRQ-001 Thr-001 IRQ-002 Thr-002 IRQ-003 Thr-003 0 58572 0 59373 0 58691 0 58895 0 1 1422 57021 628 57241 1310 56160 1102 56805 2 6 2931 0 2695 0 3567 4 3031 3 1 40 0 53 0 260 0 142 4 0 7 0 5 0 6 0 17 5 0 2 0 5 0 7 0 4 6 0 0 0 2 0 1 0 1 8 0 0 0 0 0 0 0 1 over: 0 0 0 0 0 0 0 0 count: 60001 60001 60001 60001 60001 60001 60001 60001 min: 0 1 0 1 0 1 0 1 avg: 0 1 0 1 0 1 0 1 max: 3 5 1 6 1 6 2 8 ---------- >% ---------- Running - rtla timerlat hist --help provides information about the available options. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- tools/tracing/rtla/src/timerlat.c | 6 +- tools/tracing/rtla/src/timerlat.h | 2 +- tools/tracing/rtla/src/timerlat_hist.c | 800 +++++++++++++++++++++++++ 3 files changed, 806 insertions(+), 2 deletions(-) create mode 100644 tools/tracing/rtla/src/timerlat_hist.c diff --git a/tools/tracing/rtla/src/timerlat.c b/tools/tracing/rtla/src/timerlat.c index 29a51c01b084..97abbf494fee 100644 --- a/tools/tracing/rtla/src/timerlat.c +++ b/tools/tracing/rtla/src/timerlat.c @@ -25,7 +25,8 @@ static void timerlat_usage(void) " usage: [rtla] timerlat [MODE] ...", "", " modes:", - " top - prints the summary from timerlat tracer", + " top - prints the summary from timerlat tracer", + " hist - prints a histogram of timer latencies", "", "if no MODE is given, the top mode is called, passing the arguments", NULL, @@ -60,6 +61,9 @@ int timerlat_main(int argc, char *argv[]) } else if (strcmp(argv[1], "top") == 0) { timerlat_top_main(argc-1, &argv[1]); exit(0); + } else if (strcmp(argv[1], "hist") == 0) { + timerlat_hist_main(argc-1, &argv[1]); + exit(0); } usage: diff --git a/tools/tracing/rtla/src/timerlat.h b/tools/tracing/rtla/src/timerlat.h index 6f9d3925acd0..88561bfd14f3 100644 --- a/tools/tracing/rtla/src/timerlat.h +++ b/tools/tracing/rtla/src/timerlat.h @@ -1,4 +1,4 @@ // SPDX-License-Identifier: GPL-2.0 - +int timerlat_hist_main(int argc, char *argv[]); int timerlat_top_main(int argc, char *argv[]); int timerlat_main(int argc, char *argv[]); diff --git a/tools/tracing/rtla/src/timerlat_hist.c b/tools/tracing/rtla/src/timerlat_hist.c new file mode 100644 index 000000000000..76d7b46f7ef0 --- /dev/null +++ b/tools/tracing/rtla/src/timerlat_hist.c @@ -0,0 +1,800 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021 Red Hat Inc, Daniel Bristot de Oliveira + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "utils.h" +#include "osnoise.h" +#include "timerlat.h" + +struct timerlat_hist_params { + char *cpus; + char *monitored_cpus; + char *trace_output; + unsigned long long runtime; + long long stop_us; + long long stop_total_us; + long long timerlat_period_us; + long long print_stack; + int sleep_time; + int output_divisor; + int duration; + int set_sched; + struct sched_attr sched_param; + + char no_irq; + char no_thread; + char no_header; + char no_summary; + char no_index; + char with_zeros; + int bucket_size; + int entries; +}; + +struct timerlat_hist_cpu { + int *irq; + int *thread; + + int irq_count; + int thread_count; + + unsigned long long min_irq; + unsigned long long sum_irq; + unsigned long long max_irq; + + unsigned long long min_thread; + unsigned long long sum_thread; + unsigned long long max_thread; +}; + +struct timerlat_hist_data { + struct timerlat_hist_cpu *hist; + int entries; + int bucket_size; + int nr_cpus; +}; + +/* + * timerlat_free_histogram - free runtime data + */ +static void +timerlat_free_histogram(struct timerlat_hist_data *data) +{ + int cpu; + + /* one histogram for IRQ and one for thread, per CPU */ + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (data->hist[cpu].irq) + free(data->hist[cpu].irq); + + if (data->hist[cpu].thread) + free(data->hist[cpu].thread); + } + + /* one set of histograms per CPU */ + if (data->hist) + free(data->hist); + + free(data); +} + +/* + * timerlat_alloc_histogram - alloc runtime data + */ +static struct timerlat_hist_data +*timerlat_alloc_histogram(int nr_cpus, int entries, int bucket_size) +{ + struct timerlat_hist_data *data; + int cpu; + + data = calloc(1, sizeof(*data)); + if (!data) + return NULL; + + data->entries = entries; + data->bucket_size = bucket_size; + data->nr_cpus = nr_cpus; + + /* one set of histograms per CPU */ + data->hist = calloc(1, sizeof(*data->hist) * nr_cpus); + if (!data->hist) + goto cleanup; + + /* one histogram for IRQ and one for thread, per cpu */ + for (cpu = 0; cpu < nr_cpus; cpu++) { + data->hist[cpu].irq = calloc(1, sizeof(*data->hist) * (entries + 1)); + if (!data->hist[cpu].irq) + goto cleanup; + data->hist[cpu].thread = calloc(1, sizeof(*data->hist) * (entries + 1)); + if (!data->hist[cpu].thread) + goto cleanup; + } + + /* set the min to max */ + for (cpu = 0; cpu < nr_cpus; cpu++) { + data->hist[cpu].min_irq = ~0; + data->hist[cpu].min_thread = ~0; + } + + return data; + +cleanup: + timerlat_free_histogram(data); + return NULL; +} + +/* + * timerlat_hist_update - record a new timerlat occurent on cpu, updating data + */ +static void +timerlat_hist_update(struct osnoise_tool *tool, int cpu, + unsigned long long thread, + unsigned long long latency) +{ + struct timerlat_hist_params *params = tool->params; + struct timerlat_hist_data *data = tool->data; + int entries = data->entries; + int bucket; + int *hist; + + if (params->output_divisor) + latency = latency / params->output_divisor; + + if (data->bucket_size) + bucket = latency / data->bucket_size; + + if (!thread) { + hist = data->hist[cpu].irq; + data->hist[cpu].irq_count++; + update_min(&data->hist[cpu].min_irq, &latency); + update_sum(&data->hist[cpu].sum_irq, &latency); + update_max(&data->hist[cpu].max_irq, &latency); + } else { + hist = data->hist[cpu].thread; + data->hist[cpu].thread_count++; + update_min(&data->hist[cpu].min_thread, &latency); + update_sum(&data->hist[cpu].sum_thread, &latency); + update_max(&data->hist[cpu].max_thread, &latency); + } + + if (bucket < entries) + hist[bucket]++; + else + hist[entries]++; +} + +/* + * timerlat_hist_handler - this is the handler for timerlat tracer events + */ +static int +timerlat_hist_handler(struct trace_seq *s, struct tep_record *record, + struct tep_event *event, void *data) +{ + struct trace_instance *trace = data; + unsigned long long thread, latency; + struct osnoise_tool *tool; + int cpu = record->cpu; + + tool = container_of(trace, struct osnoise_tool, trace); + + tep_get_field_val(s, event, "context", record, &thread, 1); + tep_get_field_val(s, event, "timer_latency", record, &latency, 1); + + timerlat_hist_update(tool, cpu, thread, latency); + + return 0; +} + +/* + * timerlat_hist_header - print the header of the tracer to the output + */ +static void timerlat_hist_header(struct osnoise_tool *tool) +{ + struct timerlat_hist_params *params = tool->params; + struct timerlat_hist_data *data = tool->data; + struct trace_seq *s = tool->trace.seq; + char duration[26]; + int cpu; + + if (params->no_header) + return; + + get_duration(tool->start_time, duration, sizeof(duration)); + trace_seq_printf(s, "# RTLA timerlat histogram\n"); + trace_seq_printf(s, "# Time unit is %s (%s)\n", + params->output_divisor == 1 ? "nanoseconds" : "microseconds", + params->output_divisor == 1 ? "ns" : "us"); + + trace_seq_printf(s, "# Duration: %s\n", duration); + + if (!params->no_index) + trace_seq_printf(s, "Index"); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) + trace_seq_printf(s, " IRQ-%03d", cpu); + + if (!params->no_thread) + trace_seq_printf(s, " Thr-%03d", cpu); + } + trace_seq_printf(s, "\n"); + + + trace_seq_do_printf(s); + trace_seq_reset(s); +} + +/* + * timerlat_print_summary - print the summary of the hist data to the output + */ +static void +timerlat_print_summary(struct timerlat_hist_params *params, + struct trace_instance *trace, + struct timerlat_hist_data *data) +{ + int cpu; + + if (params->no_summary) + return; + + if (!params->no_index) + trace_seq_printf(trace->seq, "count:"); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].irq_count); + + if (!params->no_thread) + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].thread_count); + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "min: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].min_irq); + + if (!params->no_thread) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].min_thread); + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "avg: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) { + if (data->hist[cpu].irq_count) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].sum_irq / data->hist[cpu].irq_count); + else + trace_seq_printf(trace->seq, " - "); + } + + if (!params->no_thread) { + if (data->hist[cpu].thread_count) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].sum_thread / data->hist[cpu].thread_count); + else + trace_seq_printf(trace->seq, " - "); + } + } + trace_seq_printf(trace->seq, "\n"); + + if (!params->no_index) + trace_seq_printf(trace->seq, "max: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].max_irq); + + if (!params->no_thread) + trace_seq_printf(trace->seq, "%9llu ", + data->hist[cpu].max_thread); + } + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); +} + +/* + * timerlat_print_stats - print data for all CPUs + */ +static void +timerlat_print_stats(struct timerlat_hist_params *params, struct osnoise_tool *tool) +{ + struct timerlat_hist_data *data = tool->data; + struct trace_instance *trace = &tool->trace; + int bucket, cpu; + int total; + + timerlat_hist_header(tool); + + for (bucket = 0; bucket < data->entries; bucket++) { + total = 0; + + if (!params->no_index) + trace_seq_printf(trace->seq, "%-6d", + bucket * data->bucket_size); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + + if (!params->no_irq) { + total += data->hist[cpu].irq[bucket]; + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].irq[bucket]); + } + + if (!params->no_thread) { + total += data->hist[cpu].thread[bucket]; + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].thread[bucket]); + } + + } + + if (total == 0 && !params->with_zeros) { + trace_seq_reset(trace->seq); + continue; + } + + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); + } + + if (!params->no_index) + trace_seq_printf(trace->seq, "over: "); + + for (cpu = 0; cpu < data->nr_cpus; cpu++) { + if (params->cpus && !params->monitored_cpus[cpu]) + continue; + + if (!params->no_irq) + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].irq[data->entries]); + + if (!params->no_thread) + trace_seq_printf(trace->seq, "%9d ", + data->hist[cpu].thread[data->entries]); + } + trace_seq_printf(trace->seq, "\n"); + trace_seq_do_printf(trace->seq); + trace_seq_reset(trace->seq); + + timerlat_print_summary(params, trace, data); +} + +/* + * timerlat_hist_usage - prints timerlat top usage message + */ +static void timerlat_hist_usage(char *usage) +{ + int i; + + char *msg[] = { + "", + " usage: [rtla] timerlat hist [-h] [-q] [-p us] [-i us] [-t us] [-s us] [-T[=file]] \\", + " [-c cpu-list] [-P priority] [--no-irq] [--no-thread] [--no-header] [--no-summary] \\", + " [--no-index] [--with-zeros]", + "", + " -h/--help: print this menu", + " -p/--period us: timerlat period in us", + " -i/--irq us: stop trace if the irq latency is higher than the argument in us", + " -T/--thread us: stop trace if the thread latency is higher than the argument in us", + " -s/--stack us: save the stack trace at the IRQ if a thread latency is higher than the argument in us", + " -c/--cpus cpus: run the tracer only on the given cpus", + " -d/--duration time[m|h|d]: duration of the session in seconds", + " -T/--trace[=file]: save the stopped trace to [file|timerlat_trace.txt]", + " -n/--nano: display data in nanoseconds", + " -b/--bucket-size N: set the histogram bucket size (default 1)", + " -e/--entries N: set the number of entries of the histogram (default 256)", + " --no-irq: ignore IRQ latencies", + " --no-thread: ignore thread latencies", + " --no-header: do not print header", + " --no-summary: do not print summary", + " --no-index: do not print index", + " --with-zeros: print zero only entries", + " -P/--priority o:prio|r:prio|f:prio|d:runtime:period : set scheduling parameters", + " o:prio - use SCHED_OTHER with prio", + " r:prio - use SCHED_RR with prio", + " f:prio - use SCHED_FIFO with prio", + " d:runtime[us|ms|s]:period[us|ms|s] - use SCHED_DEADLINE with runtime and period", + " in nanoseconds", + NULL, + }; + + if (usage) + fprintf(stderr, "%s\n", usage); + + fprintf(stderr, "rtla timerlat hist: a per-cpu histogram of the timer latency (version %s)\n", + VERSION); + + for (i = 0; msg[i]; i++) + fprintf(stderr, "%s\n", msg[i]); + exit(1); +} + +/* + * timerlat_hist_parse_args - allocs, parse and fill the cmd line parameters + */ +static struct timerlat_hist_params +*timerlat_hist_parse_args(int argc, char *argv[]) +{ + struct timerlat_hist_params *params; + int retval; + int c; + + params = calloc(1, sizeof(*params)); + if (!params) + exit(1); + + /* display data in microseconds */ + params->output_divisor = 1000; + params->bucket_size = 1; + params->entries = 256; + + while (1) { + static struct option long_options[] = { + {"cpus", required_argument, 0, 'c'}, + {"bucket-size", required_argument, 0, 'b'}, + {"debug", no_argument, 0, 'D'}, + {"entries", required_argument, 0, 'e'}, + {"duration", required_argument, 0, 'd'}, + {"help", no_argument, 0, 'h'}, + {"irq", required_argument, 0, 'i'}, + {"nano", no_argument, 0, 'n'}, + {"period", required_argument, 0, 'p'}, + {"priority", required_argument, 0, 'P'}, + {"stack", required_argument, 0, 's'}, + {"thread", required_argument, 0, 'T'}, + {"trace", optional_argument, 0, 't'}, + {"no-irq", no_argument, 0, '0'}, + {"no-thread", no_argument, 0, '1'}, + {"no-header", no_argument, 0, '2'}, + {"no-summary", no_argument, 0, '3'}, + {"no-index", no_argument, 0, '4'}, + {"with-zeros", no_argument, 0, '5'}, + {0, 0, 0, 0} + }; + + /* getopt_long stores the option index here. */ + int option_index = 0; + + c = getopt_long(argc, argv, "c:b:d:e:Dhi:np:P:s:t::T:012345", + long_options, &option_index); + + /* detect the end of the options. */ + if (c == -1) + break; + + switch (c) { + case 'c': + retval = parse_cpu_list(optarg, ¶ms->monitored_cpus); + if (retval) + timerlat_hist_usage("\nInvalid -c cpu list\n"); + params->cpus = optarg; + debug_msg("-c for %s\n", params->cpus); + break; + case 'b': + params->bucket_size = get_long_from_str(optarg); + if ((params->bucket_size == 0) || (params->bucket_size >= 1000000)) + timerlat_hist_usage("Bucket size needs to be > 0 and <= 1000000\n"); + break; + case 'D': + config_debug = 1; + break; + case 'd': + params->duration = parse_seconds_duration(optarg); + if (!params->duration) + timerlat_hist_usage("Invalid -D duration\n"); + break; + case 'e': + params->entries = get_long_from_str(optarg); + if ((params->entries < 10) || (params->entries > 9999999)) + timerlat_hist_usage("Entries must be > 10 and < 9999999\n"); + break; + case 'h': + case '?': + timerlat_hist_usage(NULL); + break; + case 'i': + params->stop_us = get_long_from_str(optarg); + break; + case 'n': + params->output_divisor = 1; + break; + case 'p': + params->timerlat_period_us = get_long_from_str(optarg); + if (params->timerlat_period_us > 1000000) + timerlat_hist_usage("Period longer than 1 s\n"); + break; + case 'P': + retval = parse_prio(optarg, ¶ms->sched_param); + if (retval == -1) + timerlat_hist_usage("Invalid -P priority"); + params->set_sched = 1; + break; + case 's': + params->print_stack = get_long_from_str(optarg); + break; + case 'T': + params->stop_total_us = get_long_from_str(optarg); + break; + case 't': + if (optarg) + /* skip = */ + params->trace_output = &optarg[1]; + else + params->trace_output = "timerlat_trace.txt"; + break; + case '0': /* no irq */ + params->no_irq = 1; + break; + case '1': /* no thread */ + params->no_thread = 1; + break; + case '2': /* no header */ + params->no_header = 1; + break; + case '3': /* no summary */ + params->no_summary = 1; + break; + case '4': /* no index */ + params->no_index = 1; + break; + case '5': /* with zeros */ + params->with_zeros = 1; + break; + default: + timerlat_hist_usage("Invalid option"); + } + } + + if (geteuid()) { + err_msg("rtla needs root permission\n"); + exit(EXIT_FAILURE); + } + + if (params->no_irq && params->no_thread) + timerlat_hist_usage("no-irq and no-thread set, there is nothing to do here"); + + if (params->no_index && !params->with_zeros) + timerlat_hist_usage("no-index set with with-zeros is not set - it does not make sense"); + + return params; +} + +/* + * timerlat_hist_apply_config - apply the hist configs to the initialized tool + */ +static int +timerlat_hist_apply_config(struct osnoise_tool *tool, struct timerlat_hist_params *params) +{ + int retval; + + if (!params->sleep_time) + params->sleep_time = 1; + + if (params->cpus) { + retval = osnoise_set_cpus(tool->context, params->cpus); + if (retval) { + err_msg("Failed to apply CPUs config\n"); + goto out_err; + } + } + + if (params->stop_us) { + retval = osnoise_set_stop_us(tool->context, params->stop_us); + if (retval) { + err_msg("Failed to set stop us\n"); + goto out_err; + } + } + + if (params->stop_total_us) { + retval = osnoise_set_stop_total_us(tool->context, params->stop_total_us); + if (retval) { + err_msg("Failed to set stop total us\n"); + goto out_err; + } + } + + if (params->timerlat_period_us) { + retval = osnoise_set_timerlat_period_us(tool->context, params->timerlat_period_us); + if (retval) { + err_msg("Failed to set timerlat period\n"); + goto out_err; + } + } + + if (params->print_stack) { + retval = osnoise_set_print_stack(tool->context, params->print_stack); + if (retval) { + err_msg("Failed to set print stack\n"); + goto out_err; + } + } + + return 0; + +out_err: + return -1; +} + +/* + * timerlat_init_hist - initialize a timerlat hist tool with parameters + */ +static struct osnoise_tool +*timerlat_init_hist(struct timerlat_hist_params *params) +{ + struct osnoise_tool *tool; + int nr_cpus; + + nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + + tool = osnoise_init_tool("timerlat_hist"); + if (!tool) + return NULL; + + tool->data = timerlat_alloc_histogram(nr_cpus, params->entries, params->bucket_size); + if (!tool->data) + goto out_err; + + tool->params = params; + + tep_register_event_handler(tool->trace.tep, -1, "ftrace", "timerlat", + timerlat_hist_handler, tool); + + return tool; + +out_err: + osnoise_destroy_tool(tool); + return NULL; +} + +static int stop_tracing; +static void stop_hist(int sig) +{ + stop_tracing = 1; +} + +/* + * timerlat_hist_set_signals - handles the signal to stop the tool + */ +static void +timerlat_hist_set_signals(struct timerlat_hist_params *params) +{ + signal(SIGINT, stop_hist); + if (params->duration) { + signal(SIGALRM, stop_hist); + alarm(params->duration); + } +} + +int timerlat_hist_main(int argc, char *argv[]) +{ + struct timerlat_hist_params *params; + struct trace_instance *trace; + struct osnoise_tool *record; + struct osnoise_tool *tool; + int return_value = 1; + int retval; + + params = timerlat_hist_parse_args(argc, argv); + if (!params) + exit(1); + + tool = timerlat_init_hist(params); + if (!tool) { + err_msg("Could not init osnoise hist\n"); + goto out_exit; + } + + retval = timerlat_hist_apply_config(tool, params); + if (retval) { + err_msg("Could not apply config\n"); + goto out_hist; + } + + trace = &tool->trace; + + retval = enable_timerlat(trace); + if (retval) { + err_msg("Failed to enable timerlat tracer\n"); + goto out_hist; + } + + if (params->set_sched) { + retval = set_comm_sched_attr("timerlat/", ¶ms->sched_param); + if (retval) + timerlat_hist_usage("Failed to set sched parameters\n"); + } + + trace_instance_start(trace); + + if (params->trace_output) { + record = osnoise_init_trace_tool("timerlat"); + if (!record) { + err_msg("Failed to enable the trace instance\n"); + goto out_hist; + } + trace_instance_start(&record->trace); + } + + tool->start_time = time(NULL); + timerlat_hist_set_signals(params); + + while (!stop_tracing) { + sleep(params->sleep_time); + + retval = tracefs_iterate_raw_events(trace->tep, + trace->inst, + NULL, + 0, + collect_registered_events, + trace); + if (retval < 0) { + err_msg("Error iterating on events\n"); + goto out_hist; + } + + if (!tracefs_trace_is_on(trace->inst)) + break; + }; + + timerlat_print_stats(params, tool); + + return_value = 0; + + if (!tracefs_trace_is_on(trace->inst)) { + printf("rtla timelat hit stop tracing\n"); + if (params->trace_output) { + printf(" Saving trace to %s\n", params->trace_output); + save_trace_to_file(record->trace.inst, params->trace_output); + } + } + +out_hist: + timerlat_free_histogram(tool->data); + osnoise_destroy_tool(tool); + if (params->trace_output) + osnoise_destroy_tool(record); + free(params); +out_exit: + exit(return_value); +} From patchwork Tue Oct 26 22:06:25 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585809 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1DD9BC433FE for ; Tue, 26 Oct 2021 22:09:18 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 0525D61100 for ; Tue, 26 Oct 2021 22:09:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240001AbhJZWLl convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:41 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:31315 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239867AbhJZWLW (ORCPT ); Tue, 26 Oct 2021 18:11:22 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-572-lw9KFpy_MJqfJpnaaYepbw-1; Tue, 26 Oct 2021 18:08:54 -0400 X-MC-Unique: lw9KFpy_MJqfJpnaaYepbw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 730A410247A6; Tue, 26 Oct 2021 22:08:52 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 2573760BF1; Tue, 26 Oct 2021 22:08:47 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org, Jonathan Corbet Subject: [PATCH V6 14/20] rtla: Add Documentation Date: Wed, 27 Oct 2021 00:06:25 +0200 Message-Id: <66e35d65e1ce22a4084f5c6778a986924cdf57ab.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Adds the basis for rtla documentation. This patch also includes the rtla(1) man page. As suggested by Jonathan Corbet, we are placing these man pages at Documentation/tools/rtla, using rst format. It is not linked to the official documentation, though. The Makefile is based on bpftool's Documentation one. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Suggested-by: Steven Rostedt Suggested-by: Jonathan Corbet Signed-off-by: Daniel Bristot de Oliveira --- Documentation/tools/rtla/Makefile | 41 +++++++++++++++++ Documentation/tools/rtla/common_appendix.rst | 12 +++++ Documentation/tools/rtla/rtla.rst | 48 ++++++++++++++++++++ tools/tracing/rtla/Makefile | 30 ++++++++++-- 4 files changed, 127 insertions(+), 4 deletions(-) create mode 100644 Documentation/tools/rtla/Makefile create mode 100644 Documentation/tools/rtla/common_appendix.rst create mode 100644 Documentation/tools/rtla/rtla.rst diff --git a/Documentation/tools/rtla/Makefile b/Documentation/tools/rtla/Makefile new file mode 100644 index 000000000000..9f2b84af1a6c --- /dev/null +++ b/Documentation/tools/rtla/Makefile @@ -0,0 +1,41 @@ +# SPDX-License-Identifier: GPL-2.0-only +# Based on bpftool's Documentation Makefile + +INSTALL ?= install +RM ?= rm -f +RMDIR ?= rmdir --ignore-fail-on-non-empty + +PREFIX ?= /usr/share +MANDIR ?= $(PREFIX)/man +MAN1DIR = $(MANDIR)/man1 + +MAN1_RST = $(wildcard rtla*.rst) + +_DOC_MAN1 = $(patsubst %.rst,%.1,$(MAN1_RST)) +DOC_MAN1 = $(addprefix $(OUTPUT),$(_DOC_MAN1)) + +RST2MAN_DEP := $(shell command -v rst2man 2>/dev/null) +RST2MAN_OPTS += --verbose + +$(OUTPUT)%.1: %.rst +ifndef RST2MAN_DEP + $(error "rst2man not found, but required to generate man pages") +endif + rst2man $(RST2MAN_OPTS) $< > $@ + +man1: $(DOC_MAN1) +man: man1 + +clean: + $(RM) $(DOC_MAN1) + +install: man + $(INSTALL) -d -m 755 $(DESTDIR)$(MAN1DIR) + $(INSTALL) -m 644 $(DOC_MAN1) $(DESTDIR)$(MAN1DIR) + +uninstall: + $(RM) $(addprefix $(DESTDIR)$(MAN1DIR)/,$(_DOC_MAN1)) + $(RMDIR) $(DESTDIR)$(MAN1DIR) + +.PHONY: man man1 clean install uninstall +.DEFAULT_GOAL := man diff --git a/Documentation/tools/rtla/common_appendix.rst b/Documentation/tools/rtla/common_appendix.rst new file mode 100644 index 000000000000..b494084acccd --- /dev/null +++ b/Documentation/tools/rtla/common_appendix.rst @@ -0,0 +1,12 @@ +REPORTING BUGS +============== +Report bugs to + +LICENSE +======= +**rtla** is Free Software licensed under the GNU GPLv2 + +COPYING +======= +Copyright \(C) 2021 Red Hat, Inc. Free use of this software is granted under +the terms of the GNU Public License (GPL). diff --git a/Documentation/tools/rtla/rtla.rst b/Documentation/tools/rtla/rtla.rst new file mode 100644 index 000000000000..fc0d233efcd5 --- /dev/null +++ b/Documentation/tools/rtla/rtla.rst @@ -0,0 +1,48 @@ +========= +rtla +========= +-------------------------------- +Real-time Linux Analysis tool +-------------------------------- + +:Manual section: 1 + +SYNOPSIS +======== +**rtla** *COMMAND* [*OPTIONS*] + +DESCRIPTION +=========== +The **rtla** is a meta-tool that includes a set of commands that aims to +analyze the real-time properties of Linux. But instead of testing Linux +as a black box, **rtla** leverages kernel tracing capabilities to provide +precise information about the properties and root causes of unexpected +results. + +COMMANDS +======== +**osnoise** + + Gives information about the operating system noise (osnoise). + +**timerlat** + + Measures the IRQ and thread timer latency. + +OPTIONS +======= +**-h**, **--help** + + Display the help text. + +For other options, see the man page for the corresponding command. + +SEE ALSO +======== +**rtla-osnoise**\(1), **rtla-timerlat**\(1) + +AUTHOR +====== +Daniel Bristot de Oliveira + +.. include:: common_appendix.rst diff --git a/tools/tracing/rtla/Makefile b/tools/tracing/rtla/Makefile index 3f8d5421ccc1..4d1d158f006f 100644 --- a/tools/tracing/rtla/Makefile +++ b/tools/tracing/rtla/Makefile @@ -45,18 +45,28 @@ DATADIR := /usr/share DOCDIR := $(DATADIR)/doc MANDIR := $(DATADIR)/man LICDIR := $(DATADIR)/licenses +SRCTREE := $(if $(BUILD_SRC),$(BUILD_SRC),$(CURDIR)) + +# If running from the tarball, man pages are stored in the Documentation +# dir. If running from the kernel source, man pages are stored in +# Documentation/tools/rtla/. +ifneq ($(wildcard Documentation/.*),) +DOCSRC = Documentation/ +else +DOCSRC = $(SRCTREE)/../../../Documentation/tools/rtla/ +endif .PHONY: all all: rtla -rtla: $(OBJ) +rtla: $(OBJ) doc $(CC) -o rtla $(LDFLAGS) $(OBJ) $(LIBS) static: $(OBJ) $(CC) -o rtla-static $(LDFLAGS) --static $(OBJ) $(LIBS) -lpthread -ldl .PHONY: install -install: +install: doc_install $(INSTALL) -d -m 755 $(DESTDIR)$(BINDIR) $(INSTALL) rtla -m 755 $(DESTDIR)$(BINDIR) $(STRIP) $(DESTDIR)$(BINDIR)/rtla @@ -66,15 +76,27 @@ install: ln -s $(DESTDIR)$(BINDIR)/rtla $(DESTDIR)$(BINDIR)/timerlat .PHONY: clean tarball -clean: +clean: doc_clean @test ! -f rtla || rm rtla @test ! -f rtla-static || rm rtla-static @test ! -f src/rtla.o || rm src/rtla.o @test ! -f $(TARBALL) || rm -f $(TARBALL) @rm -rf *~ $(OBJ) *.tar.$(CEXT) -tarball: clean +tarball: clean rm -rf $(NAME)-$(VERSION) && mkdir $(NAME)-$(VERSION) cp -r $(DIRS) $(FILES) $(NAME)-$(VERSION) + mkdir $(NAME)-$(VERSION)/Documentation/ + cp -rp $(SRCTREE)/../../../Documentation/tools/rtla/* $(NAME)-$(VERSION)/Documentation/ tar $(TAROPTS) --exclude='*~' $(NAME)-$(VERSION) rm -rf $(NAME)-$(VERSION) + +.PHONY: doc doc_clean doc_install +doc: + $(MAKE) -C $(DOCSRC) + +doc_clean: + $(MAKE) -C $(DOCSRC) clean + +doc_install: + $(MAKE) -C $(DOCSRC) install From patchwork Tue Oct 26 22:06:26 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585811 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4ED85C433F5 for ; Tue, 26 Oct 2021 22:09:28 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 3962260F92 for ; Tue, 26 Oct 2021 22:09:28 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239921AbhJZWLv convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:11:51 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:37385 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239895AbhJZWL1 (ORCPT ); Tue, 26 Oct 2021 18:11:27 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-302-JNp1OmZHNGykqJUpDWKjrw-1; Tue, 26 Oct 2021 18:08:59 -0400 X-MC-Unique: JNp1OmZHNGykqJUpDWKjrw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 6437E81424A; Tue, 26 Oct 2021 22:08:57 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id D2ADE60BF1; Tue, 26 Oct 2021 22:08:52 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 15/20] rtla: Add rtla osnoise man page Date: Wed, 27 Oct 2021 00:06:26 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla osnoise command. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- .../tools/rtla/common_osnoise_description.rst | 8 +++ Documentation/tools/rtla/rtla-osnoise.rst | 59 +++++++++++++++++++ 2 files changed, 67 insertions(+) create mode 100644 Documentation/tools/rtla/common_osnoise_description.rst create mode 100644 Documentation/tools/rtla/rtla-osnoise.rst diff --git a/Documentation/tools/rtla/common_osnoise_description.rst b/Documentation/tools/rtla/common_osnoise_description.rst new file mode 100644 index 000000000000..8973c5df888f --- /dev/null +++ b/Documentation/tools/rtla/common_osnoise_description.rst @@ -0,0 +1,8 @@ +The **rtla osnoise** tool is an interface for the *osnoise* tracer. The +*osnoise* tracer dispatches a kernel thread per-cpu. These threads read the +time in a loop while with preemption, softirq and IRQs enabled, thus +allowing all the sources of operating systme noise during its execution. +The *osnoise*'s tracer threads take note of the delta between each time +read, along with an interference counter of all sources of interference. +At the end of each period, the *osnoise* tracer displays a summary of +the results. diff --git a/Documentation/tools/rtla/rtla-osnoise.rst b/Documentation/tools/rtla/rtla-osnoise.rst new file mode 100644 index 000000000000..c129b206ce34 --- /dev/null +++ b/Documentation/tools/rtla/rtla-osnoise.rst @@ -0,0 +1,59 @@ +=============== +rtla-osnoise +=============== +------------------------------------------------------------------ +Measure the operating system noise +------------------------------------------------------------------ + +:Manual section: 1 + +SYNOPSIS +======== +**rtla osnoise** [*MODE*] ... + +DESCRIPTION +=========== + +.. include:: common_osnoise_description.rst + +The *osnoise* tracer outputs information in two ways. It periodically prints +a summary of the noise of the operating system, including the counters of +the occurrence of the source of interference. It also provides information +for each noise via the **osnoise:** tracepoints. The **rtla osnoise top** +mode displays information about the periodic summary from the *osnoise* tracer. +The **rtla osnoise hist** mode displays information about the noise using +the **osnoise:** tracepoints. For further details, please refer to the +respective man page. + +MODES +===== +**top** + + Prints the summary from osnoise tracer. + +**hist** + + Prints a histogram of osnoise samples. + +If no MODE is given, the top mode is called, passing the arguments. + +OPTIONS +======= + +**-h**, **--help** + + Display the help text. + +For other options, see the man page for the corresponding mode. + +SEE ALSO +======== +**rtla-osnoise-top**\(1), **rtla-osnoise-hist**\(1) + +Osnoise tracer documentation: + +AUTHOR +====== +Written by Daniel Bristot de Oliveira + +.. include:: common_appendix.rst From patchwork Tue Oct 26 22:06:27 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585813 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8096FC433EF for ; Tue, 26 Oct 2021 22:09:43 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 6B47260F92 for ; Tue, 26 Oct 2021 22:09:43 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239957AbhJZWMH convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:12:07 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:54686 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239971AbhJZWLh (ORCPT ); Tue, 26 Oct 2021 18:11:37 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-178-53qjkjPsNCuTjeQqzm0zCw-1; Tue, 26 Oct 2021 18:09:03 -0400 X-MC-Unique: 53qjkjPsNCuTjeQqzm0zCw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 0BED418D6A25; Tue, 26 Oct 2021 22:09:02 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id BF00060BF1; Tue, 26 Oct 2021 22:08:57 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 16/20] rtla: Add rtla osnoise top documentation Date: Wed, 27 Oct 2021 00:06:27 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla osnoise top mode. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- Documentation/tools/rtla/common_options.rst | 24 ++++++++ .../tools/rtla/common_osnoise_options.rst | 17 ++++++ .../tools/rtla/common_top_options.rst | 3 + Documentation/tools/rtla/rtla-osnoise-top.rst | 61 +++++++++++++++++++ 4 files changed, 105 insertions(+) create mode 100644 Documentation/tools/rtla/common_options.rst create mode 100644 Documentation/tools/rtla/common_osnoise_options.rst create mode 100644 Documentation/tools/rtla/common_top_options.rst create mode 100644 Documentation/tools/rtla/rtla-osnoise-top.rst diff --git a/Documentation/tools/rtla/common_options.rst b/Documentation/tools/rtla/common_options.rst new file mode 100644 index 000000000000..ea6529667e7c --- /dev/null +++ b/Documentation/tools/rtla/common_options.rst @@ -0,0 +1,24 @@ +**-c**, **--cpus** *cpu-list* + + Set the osnoise tracer to run the sample threads in the cpu-list. + +**-d**, **--duration** *time[s|m|h|d]* + + Set the duration of the session. + +**-t**, **--trace**\[*=file*] + + Save the stopped trace to [*file|osnoise_trace.txt*]. + +**-P**, **--priority** *o:prio|r:prio|f:prio|d:runtime:period* + + Set scheduling parameters to the osnoise tracer threads, the format to set the priority are: + + - *o:prio* - use SCHED_OTHER with *prio*; + - *r:prio* - use SCHED_RR with *prio*; + - *f:prio* - use SCHED_FIFO with *prio*; + - *d:runtime[us|ms|s]:period[us|ms|s]* - use SCHED_DEADLINE with *runtime* and *period* in nanoseconds. + +**-h**, **--help** + + Print help menu. diff --git a/Documentation/tools/rtla/common_osnoise_options.rst b/Documentation/tools/rtla/common_osnoise_options.rst new file mode 100644 index 000000000000..d556883e4e26 --- /dev/null +++ b/Documentation/tools/rtla/common_osnoise_options.rst @@ -0,0 +1,17 @@ +**-p**, **--period** *us* + + Set the *osnoise* tracer period in microseconds. + +**-r**, **--runtime** *us* + + Set the *osnoise* tracer runtime in microseconds. + +**-s**, **--stop** *us* + + Stop the trace if a single sample is higher than the argument in microseconds. + If **-T** is set, it will also save the trace to the output. + +**-S**, **--stop-total** *us* + + Stop the trace if the total sample is higher than the argument in microseconds. + If **-T** is set, it will also save the trace to the output. diff --git a/Documentation/tools/rtla/common_top_options.rst b/Documentation/tools/rtla/common_top_options.rst new file mode 100644 index 000000000000..f48878938f84 --- /dev/null +++ b/Documentation/tools/rtla/common_top_options.rst @@ -0,0 +1,3 @@ +**-q**, **--quiet** + + Print only a summary at the end of the session. diff --git a/Documentation/tools/rtla/rtla-osnoise-top.rst b/Documentation/tools/rtla/rtla-osnoise-top.rst new file mode 100644 index 000000000000..5d75d1394516 --- /dev/null +++ b/Documentation/tools/rtla/rtla-osnoise-top.rst @@ -0,0 +1,61 @@ +=================== +rtla-osnoise-top +=================== +----------------------------------------------- +Display a summary of the operating system noise +----------------------------------------------- + +:Manual section: 1 + +SYNOPSIS +======== +**rtla osnoise top** [*OPTIONS*] + +DESCRIPTION +=========== +.. include:: common_osnoise_description.rst + +**rtla osnoise top** collects the periodic summary from the *osnoise* tracer, +including the counters of the occurrence of the interference source, +displaying the results in a user-friendly format. + +The tool also allows many configurations of the *osnoise* tracer and the +collection of the tracer output. + +OPTIONS +======= +.. include:: common_osnoise_options.rst + +.. include:: common_top_options.rst + +.. include:: common_options.rst + +EXAMPLE +======= +In the example below, the **rtla osnoise top** tool is set to run with a +real-time priority *FIFO:1*, on CPUs *0-3*, for *900ms* at each period +(*1s* by default). The reason for reducing the runtime is to avoid starving +the rtla tool. The tool is also set to run for *one minute* and to display +a summary of the report at the end of the session:: + + [root@f34 ~]# rtla osnoise top -P F:1 -c 0-3 -r 900000 -d 1M -q + Operating System Noise + duration: 0 00:01:00 | time is in us + CPU Period Runtime Noise % CPU Aval Max Noise Max Single HW NMI IRQ Softirq Thread + 0 #59 53100000 304896 99.42580 6978 56 549 0 53111 1590 13 + 1 #59 53100000 338339 99.36282 8092 24 399 0 53130 1448 31 + 2 #59 53100000 290842 99.45227 6582 39 855 0 53110 1406 12 + 3 #59 53100000 204935 99.61405 6251 33 290 0 53156 1460 12 + +SEE ALSO +======== + +**rtla-osnoise**\(1), **rtla-osnoise-hist**\(1) + +Osnoise tracer documentation: + +AUTHOR +====== +Written by Daniel Bristot de Oliveira + +.. include:: common_appendix.rst From patchwork Tue Oct 26 22:06:28 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585815 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E7BCBC433EF for ; Tue, 26 Oct 2021 22:09:53 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id D17C160F92 for ; Tue, 26 Oct 2021 22:09:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236202AbhJZWMM convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:12:12 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:24552 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239987AbhJZWLi (ORCPT ); Tue, 26 Oct 2021 18:11:38 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-255-5an5Lcf1NC63eHoYkHvh8w-1; Tue, 26 Oct 2021 18:09:08 -0400 X-MC-Unique: 5an5Lcf1NC63eHoYkHvh8w-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 94AAE10247A7; Tue, 26 Oct 2021 22:09:06 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 8F50A60BF1; Tue, 26 Oct 2021 22:09:02 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 17/20] rtla: Add rtla osnoise hist documentation Date: Wed, 27 Oct 2021 00:06:28 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla osnoise hist mode. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- .../tools/rtla/common_hist_options.rst | 23 +++++++ .../tools/rtla/rtla-osnoise-hist.rst | 66 +++++++++++++++++++ 2 files changed, 89 insertions(+) create mode 100644 Documentation/tools/rtla/common_hist_options.rst create mode 100644 Documentation/tools/rtla/rtla-osnoise-hist.rst diff --git a/Documentation/tools/rtla/common_hist_options.rst b/Documentation/tools/rtla/common_hist_options.rst new file mode 100644 index 000000000000..0266cd08a6c9 --- /dev/null +++ b/Documentation/tools/rtla/common_hist_options.rst @@ -0,0 +1,23 @@ +**-b**, **--bucket-size** *N* + + Set the histogram bucket size (default *1*). + +**-e**, **--entries** *N* + + Set the number of entries of the histogram (default 256). + +**--no-header** + + Do not print header. + +**--no-summary** + + Do not print summary. + +**--no-index** + + Do not print index. + +**--with-zeros** + + Print zero only entries. diff --git a/Documentation/tools/rtla/rtla-osnoise-hist.rst b/Documentation/tools/rtla/rtla-osnoise-hist.rst new file mode 100644 index 000000000000..52298ddd8701 --- /dev/null +++ b/Documentation/tools/rtla/rtla-osnoise-hist.rst @@ -0,0 +1,66 @@ +=================== +rtla-osnoise-hist +=================== +------------------------------------------------------ +Display a histogram of the osnoise tracer samples +------------------------------------------------------ + +:Manual section: 1 + +SYNOPSIS +======== +**rtla osnoise hist** [*OPTIONS*] + +DESCRIPTION +=========== +.. include:: common_osnoise_description.rst + +The **rtla osnoise hist** tool collects all **osnoise:sample_threshold** +occurrence in a histogram, displaying the results in a user-friendly way. +The tool also allows many configurations of the *osnoise* tracer and the +collection of the tracer output. + +OPTIONS +======= +.. include:: common_osnoise_options.rst + +.. include:: common_hist_options.rst + +.. include:: common_options.rst + +EXAMPLE +======= +In the example below, *osnoise* tracer threads are set to run with real-time +priority *FIFO:1*, on CPUs *0-11*, for *900ms* at each period (*1s* by +default). The reason for reducing the runtime is to avoid starving the +**rtla** tool. The tool is also set to run for *one minute*. The output +histogram is set to group outputs in buckets of *10us* and *25* entries:: + + [root@f34 ~/]# rtla osnoise hist -P F:1 -c 0-11 -r 900000 -d 1M -b 10 -e 25 + # RTLA osnoise histogram + # Time unit is microseconds (us) + # Duration: 0 00:01:00 + Index CPU-000 CPU-001 CPU-002 CPU-003 CPU-004 CPU-005 CPU-006 CPU-007 CPU-008 CPU-009 CPU-010 CPU-011 + 0 42982 46287 51779 53740 52024 44817 49898 36500 50408 50128 49523 52377 + 10 12224 8356 2912 878 2667 10155 4573 18894 4214 4836 5708 2413 + 20 8 5 12 2 13 24 20 41 29 53 39 39 + 30 1 1 0 0 10 3 6 19 15 31 30 38 + 40 0 0 0 0 0 4 2 7 2 3 8 11 + 50 0 0 0 0 0 0 0 0 0 1 1 2 + over: 0 0 0 0 0 0 0 0 0 0 0 0 + count: 55215 54649 54703 54620 54714 55003 54499 55461 54668 55052 55309 54880 + min: 0 0 0 0 0 0 0 0 0 0 0 0 + avg: 0 0 0 0 0 0 0 0 0 0 0 0 + max: 30 30 20 20 30 40 40 40 40 50 50 50 + +SEE ALSO +======== +**rtla-osnoise**\(1), **rtla-osnoise-top**\(1) + +*osnoise* tracer documentation: + +AUTHOR +====== +Written by Daniel Bristot de Oliveira + +.. include:: common_appendix.rst From patchwork Tue Oct 26 22:06:29 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585817 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id B8C64C433F5 for ; Tue, 26 Oct 2021 22:10:11 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id A47D760F02 for ; Tue, 26 Oct 2021 22:10:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240023AbhJZWMd convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:12:33 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:44372 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240028AbhJZWLw (ORCPT ); Tue, 26 Oct 2021 18:11:52 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-223-ykweK9dxNBGCp72_ijyg7Q-1; Tue, 26 Oct 2021 18:09:24 -0400 X-MC-Unique: ykweK9dxNBGCp72_ijyg7Q-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id D92C5CF98F; Tue, 26 Oct 2021 22:09:22 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 300BC60BF1; Tue, 26 Oct 2021 22:09:06 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 18/20] rtla: Add rtla timerlat documentation Date: Wed, 27 Oct 2021 00:06:29 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla timerlat tool. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- .../rtla/common_timerlat_description.rst | 10 ++++ Documentation/tools/rtla/rtla-timerlat.rst | 57 +++++++++++++++++++ 2 files changed, 67 insertions(+) create mode 100644 Documentation/tools/rtla/common_timerlat_description.rst create mode 100644 Documentation/tools/rtla/rtla-timerlat.rst diff --git a/Documentation/tools/rtla/common_timerlat_description.rst b/Documentation/tools/rtla/common_timerlat_description.rst new file mode 100644 index 000000000000..321201cb8597 --- /dev/null +++ b/Documentation/tools/rtla/common_timerlat_description.rst @@ -0,0 +1,10 @@ +The **rtla timerlat** tool is an interface for the *timerlat* tracer. The +*timerlat* tracer dispatches a kernel thread per-cpu. These threads +set a periodic timer to wake themselves up and go back to sleep. After +the wakeup, they collect and generate useful information for the +debugging of operating system timer latency. + +The *timerlat* tracer outputs information in two ways. It periodically +prints the timer latency at the timer *IRQ* handler and the *Thread* +handler. It also enable the trace of the most relevant information via +**osnoise:** tracepoints. diff --git a/Documentation/tools/rtla/rtla-timerlat.rst b/Documentation/tools/rtla/rtla-timerlat.rst new file mode 100644 index 000000000000..44a49e6f302b --- /dev/null +++ b/Documentation/tools/rtla/rtla-timerlat.rst @@ -0,0 +1,57 @@ +================ +rtla-timerlat +================ +------------------------------------------- +Measures the operating system timer latency +------------------------------------------- + +:Manual section: 1 + +SYNOPSIS +======== +**rtla timerlat** [*MODE*] ... + +DESCRIPTION +=========== + +.. include:: common_timerlat_description.rst + +The *timerlat* tracer outputs information in two ways. It periodically +prints the timer latency at the timer *IRQ* handler and the *Thread* handler. +It also provides information for each noise via the **osnoise:** tracepoints. +The **rtla timerlat top** mode displays a summary of the periodic output +from the *timerlat* tracer. The **rtla hist hist** mode displays a histogram +of each tracer event occurrence. For further details, please refer to the +respective man page. + +MODES +===== +**top** + + Prints the summary from *timerlat* tracer. + +**hist** + + Prints a histogram of timerlat samples. + +If no *MODE* is given, the top mode is called, passing the arguments. + +OPTIONS +======= +**-h**, **--help** + + Display the help text. + +For other options, see the man page for the corresponding mode. + +SEE ALSO +======== +**rtla-timerlat-top**\(1), **rtla-timerlat-hist**\(1) + +*timerlat* tracer documentation: + +AUTHOR +====== +Written by Daniel Bristot de Oliveira + +.. include:: common_appendix.rst From patchwork Tue Oct 26 22:06:30 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585819 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8456FC433EF for ; Tue, 26 Oct 2021 22:10:21 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 6B384610A2 for ; Tue, 26 Oct 2021 22:10:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240031AbhJZWMm convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:12:42 -0400 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:55568 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240035AbhJZWL7 (ORCPT ); Tue, 26 Oct 2021 18:11:59 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-2-5tmm8Q0qOoaH1krXjEt3sA-1; Tue, 26 Oct 2021 18:09:30 -0400 X-MC-Unique: 5tmm8Q0qOoaH1krXjEt3sA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 3A8B4801A85; Tue, 26 Oct 2021 22:09:28 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9245C60BF1; Tue, 26 Oct 2021 22:09:23 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 19/20] rtla: Add rtla timerlat top documentation Date: Wed, 27 Oct 2021 00:06:30 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla timerlat top mode. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- .../tools/rtla/common_timerlat_options.rst | 16 ++ .../tools/rtla/rtla-timerlat-top.rst | 145 ++++++++++++++++++ 2 files changed, 161 insertions(+) create mode 100644 Documentation/tools/rtla/common_timerlat_options.rst create mode 100644 Documentation/tools/rtla/rtla-timerlat-top.rst diff --git a/Documentation/tools/rtla/common_timerlat_options.rst b/Documentation/tools/rtla/common_timerlat_options.rst new file mode 100644 index 000000000000..e9c1bfd55d48 --- /dev/null +++ b/Documentation/tools/rtla/common_timerlat_options.rst @@ -0,0 +1,16 @@ +**-p**, **--period** *us* + + Set the *timerlat* tracer period in microseconds. + +**-i**, **--irq** *us* + + Stop trace if the *IRQ* latency is higher than the argument in us. + +**-T**, **--thread** *us* + + Stop trace if the *Thread* latency is higher than the argument in us. + +**-s**, **--stack** *us* + + Save the stack trace at the *IRQ* if a *Thread* latency is higher than the + argument in us. diff --git a/Documentation/tools/rtla/rtla-timerlat-top.rst b/Documentation/tools/rtla/rtla-timerlat-top.rst new file mode 100644 index 000000000000..1c321de1c171 --- /dev/null +++ b/Documentation/tools/rtla/rtla-timerlat-top.rst @@ -0,0 +1,145 @@ +==================== +rtla-timerlat-top +==================== +------------------------------------------- +Measures the operating system timer latency +------------------------------------------- + +:Manual section: 1 + +SYNOPSIS +======== +**rtla timerlat top** [*OPTIONS*] ... + +DESCRIPTION +=========== + +.. include:: common_timerlat_description.rst + +The **rtla timerlat top** displays a summary of the periodic output +from the *timerlat* tracer. It also provides information for each +operating system noise via the **osnoise:** tracepoints that can be +seem with the option **-T**. + +OPTIONS +======= + +.. include:: common_timerlat_options.rst + +.. include:: common_top_options.rst + +.. include:: common_options.rst + +EXAMPLE +======= + +In the example below, the *timerlat* tracer is set to capture the stack trace at +the IRQ handler, printing it to the buffer if the *Thread* timer latency is +higher than *30 us*. It is also set to stop the session if a *Thread* timer +latency higher than *30 us* is hit. Finally, it is set to save the trace +buffer if the stop condition is hit:: + + [root@alien ~]# rtla timerlat top -s 30 -t 30 -T + Timer Latency + 0 00:00:59 | IRQ Timer Latency (us) | Thread Timer Latency (us) + CPU COUNT | cur min avg max | cur min avg max + 0 #58634 | 1 0 1 10 | 11 2 10 23 + 1 #58634 | 1 0 1 9 | 12 2 9 23 + 2 #58634 | 0 0 1 11 | 10 2 9 23 + 3 #58634 | 1 0 1 11 | 11 2 9 24 + 4 #58634 | 1 0 1 10 | 11 2 9 26 + 5 #58634 | 1 0 1 8 | 10 2 9 25 + 6 #58634 | 12 0 1 12 | 30 2 10 30 <--- CPU with spike + 7 #58634 | 1 0 1 9 | 11 2 9 23 + 8 #58633 | 1 0 1 9 | 11 2 9 26 + 9 #58633 | 1 0 1 9 | 10 2 9 26 + 10 #58633 | 1 0 1 13 | 11 2 9 28 + 11 #58633 | 1 0 1 13 | 12 2 9 24 + 12 #58633 | 1 0 1 8 | 10 2 9 23 + 13 #58633 | 1 0 1 10 | 10 2 9 22 + 14 #58633 | 1 0 1 18 | 12 2 9 27 + 15 #58633 | 1 0 1 10 | 11 2 9 28 + 16 #58633 | 0 0 1 11 | 7 2 9 26 + 17 #58633 | 1 0 1 13 | 10 2 9 24 + 18 #58633 | 1 0 1 9 | 13 2 9 22 + 19 #58633 | 1 0 1 10 | 11 2 9 23 + 20 #58633 | 1 0 1 12 | 11 2 9 28 + 21 #58633 | 1 0 1 14 | 11 2 9 24 + 22 #58633 | 1 0 1 8 | 11 2 9 22 + 23 #58633 | 1 0 1 10 | 11 2 9 27 + timerlat hit stop tracing + saving trace to timerlat_trace.txt + [root@alien bristot]# tail -60 timerlat_trace.txt + [...] + timerlat/5-79755 [005] ....... 426.271226: #58634 context thread timer_latency 10823 ns + sh-109404 [006] dnLh213 426.271247: #58634 context irq timer_latency 12505 ns + sh-109404 [006] dNLh313 426.271258: irq_noise: local_timer:236 start 426.271245463 duration 12553 ns + sh-109404 [006] d...313 426.271263: thread_noise: sh:109404 start 426.271245853 duration 4769 ns + timerlat/6-79756 [006] ....... 426.271264: #58634 context thread timer_latency 30328 ns + timerlat/6-79756 [006] ....1.. 426.271265: + => timerlat_irq + => __hrtimer_run_queues + => hrtimer_interrupt + => __sysvec_apic_timer_interrupt + => sysvec_apic_timer_interrupt + => asm_sysvec_apic_timer_interrupt + => _raw_spin_unlock_irqrestore <---- spinlock that disabled interrupt. + => try_to_wake_up + => autoremove_wake_function + => __wake_up_common + => __wake_up_common_lock + => ep_poll_callback + => __wake_up_common + => __wake_up_common_lock + => fsnotify_add_event + => inotify_handle_inode_event + => fsnotify + => __fsnotify_parent + => __fput + => task_work_run + => exit_to_user_mode_prepare + => syscall_exit_to_user_mode + => do_syscall_64 + => entry_SYSCALL_64_after_hwframe + => 0x7265000001378c + => 0x10000cea7 + => 0x25a00000204a + => 0x12e302d00000000 + => 0x19b51010901b6 + => 0x283ce00726500 + => 0x61ea308872 + => 0x00000fe3 + bash-109109 [007] d..h... 426.271265: #58634 context irq timer_latency 1211 ns + timerlat/6-79756 [006] ....... 426.271267: timerlat_main: stop tracing hit on cpu 6 + +In the trace, it is possible the notice that the *IRQ* timer latency was +already high, accounting *12505 ns*. The IRQ delay was caused by the +*bash-109109* process that disabled IRQs in the wake-up path +(*_try_to_wake_up()* function). The duration of the IRQ handler that woke +up the timerlat thread, informed with the **osnoise:irq_noise** event, was +also high and added more *12553 ns* to the Thread latency. Finally, the +**osnoise:thread_noise** added by the currently running thread (including +the scheduling overhead) added more *4769 ns*. Summing up these values, +the *Thread* timer latency accounted for *30328 ns*. + +The primary reason for this high value is the wake-up path that was hit +twice during this case: when the *bash-109109* was waking up a thread +and then when the *timerlat* thread was awakened. This information can +then be used as the starting point of a more fine-grained analysis. + +Note that **rtla timerlat** was dispatched without changing *timerlat* tracer +threads' priority. That is generally not needed because these threads hava +priority *FIFO:95* by default, which is a common priority used by real-time +kernel developers to analyze scheduling delays. + +SEE ALSO +-------- +**rtla-timerlat**\(1), **rtla-timerlat-hist**\(1) + +*timerlat* tracer documentation: + +AUTHOR +------ +Written by Daniel Bristot de Oliveira + +.. include:: common_appendix.rst From patchwork Tue Oct 26 22:06:31 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Bristot de Oliveira X-Patchwork-Id: 12585821 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id D3998C433F5 for ; Tue, 26 Oct 2021 22:10:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id B4E4A608FE for ; Tue, 26 Oct 2021 22:10:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239949AbhJZWND convert rfc822-to-8bit (ORCPT ); Tue, 26 Oct 2021 18:13:03 -0400 Received: from us-smtp-delivery-44.mimecast.com ([207.211.30.44]:37737 "EHLO us-smtp-delivery-44.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239956AbhJZWMG (ORCPT ); Tue, 26 Oct 2021 18:12:06 -0400 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-561-O3WKBLXnMEyl7PqH1H_rJQ-1; Tue, 26 Oct 2021 18:09:38 -0400 X-MC-Unique: O3WKBLXnMEyl7PqH1H_rJQ-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 1EB92100C62D; Tue, 26 Oct 2021 22:09:37 +0000 (UTC) Received: from x1.com (unknown [10.22.32.3]) by smtp.corp.redhat.com (Postfix) with ESMTP id 93F9160BF1; Tue, 26 Oct 2021 22:09:28 +0000 (UTC) From: Daniel Bristot de Oliveira To: Steven Rostedt Cc: Tao Zhou , Daniel Bristot de Oliveira , Ingo Molnar , Tom Zanussi , Masami Hiramatsu , Juri Lelli , Clark Williams , John Kacur , Peter Zijlstra , Thomas Gleixner , Sebastian Andrzej Siewior , linux-rt-users@vger.kernel.org, linux-trace-devel@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH V6 20/20] rtla: Add rtla timerlat hist documentation Date: Wed, 27 Oct 2021 00:06:31 +0200 Message-Id: <373043cb663619d03737db8b8da33acf40a0fc4e.1635284863.git.bristot@kernel.org> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=bristot@kernel.org X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-trace-devel@vger.kernel.org Man page for rtla timerlat hist mode. Cc: Steven Rostedt Cc: Ingo Molnar Cc: Tom Zanussi Cc: Masami Hiramatsu Cc: Juri Lelli Cc: Clark Williams Cc: John Kacur Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Sebastian Andrzej Siewior Cc: Daniel Bristot de Oliveira Cc: linux-rt-users@vger.kernel.org Cc: linux-trace-devel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Daniel Bristot de Oliveira --- .../tools/rtla/rtla-timerlat-hist.rst | 106 ++++++++++++++++++ 1 file changed, 106 insertions(+) create mode 100644 Documentation/tools/rtla/rtla-timerlat-hist.rst diff --git a/Documentation/tools/rtla/rtla-timerlat-hist.rst b/Documentation/tools/rtla/rtla-timerlat-hist.rst new file mode 100644 index 000000000000..e12eae1f3301 --- /dev/null +++ b/Documentation/tools/rtla/rtla-timerlat-hist.rst @@ -0,0 +1,106 @@ +===================== +rtla-timerlat-hist +===================== +------------------------------------------------ +Histograms of the operating system timer latency +------------------------------------------------ + +:Manual section: 1 + +SYNOPSIS +======== +**rtla timerlat hist** [*OPTIONS*] ... + +DESCRIPTION +=========== + +.. include:: common_timerlat_description.rst + +The **rtla timerlat hist** displays a histogram of each tracer event +occurrence. This tool uses the periodic information, and the +**osnoise:** tracepoints are enabled when using the **-T** option. + +OPTIONS +======= + +.. include:: common_timerlat_options.rst + +.. include:: common_hist_options.rst + +.. include:: common_options.rst + +EXAMPLE +======= +In the example below, **rtla timerlat hist** is set to run for *10* minutes, +in the cpus *0-4*, *skipping zero* only lines. Moreover, **rtla timerlat +hist** will change the priority of the *timelat* threads to run under +*SCHED_DEADLINE* priority, with a *10us* runtime every *1ms* period. The +*1ms* period is also passed to the *timerlat* tracer:: + + [root@alien ~]# timerlat hist -d 10m -c 0-4 -P d:100us:1ms -p 1ms + # RTLA timerlat histogram + # Time unit is microseconds (us) + # Duration: 0 00:10:00 + Index IRQ-000 Thr-000 IRQ-001 Thr-001 IRQ-002 Thr-002 IRQ-003 Thr-003 IRQ-004 Thr-004 + 0 276489 0 206089 0 466018 0 481102 0 205546 0 + 1 318327 35487 388149 30024 94531 48382 83082 71078 388026 55730 + 2 3282 122584 4019 126527 28231 109012 23311 89309 4568 98739 + 3 940 11815 837 9863 6209 16227 6895 17196 910 9780 + 4 444 17287 424 11574 2097 38443 2169 36736 462 13476 + 5 206 43291 255 25581 1223 101908 1304 101137 236 28913 + 6 132 101501 96 64584 635 213774 757 215471 99 73453 + 7 74 169347 65 124758 350 57466 441 53639 69 148573 + 8 53 85183 31 156751 229 9052 306 9026 39 139907 + 9 22 10387 12 42762 161 2554 225 2689 19 26192 + 10 13 1898 8 5770 114 1247 128 1405 13 3772 + 11 9 560 9 924 71 686 76 765 8 713 + 12 4 256 2 360 50 411 64 474 3 278 + 13 2 167 2 172 43 256 53 350 4 180 + 14 1 88 1 116 15 198 42 223 0 115 + 15 2 63 3 94 11 139 20 150 0 58 + 16 2 37 0 56 5 78 10 102 0 39 + 17 0 18 0 28 4 57 8 80 0 15 + 18 0 8 0 17 2 50 6 56 0 12 + 19 0 9 0 5 0 19 0 48 0 18 + 20 0 4 0 8 0 11 2 27 0 4 + 21 0 2 0 3 1 9 1 18 0 6 + 22 0 1 0 3 1 7 0 3 0 5 + 23 0 2 0 4 0 2 0 7 0 2 + 24 0 2 0 2 1 3 0 3 0 5 + 25 0 0 0 1 0 1 0 1 0 3 + 26 0 1 0 0 0 2 0 2 0 0 + 27 0 0 0 3 0 1 0 0 0 1 + 28 0 0 0 3 0 0 0 1 0 0 + 29 0 0 0 2 0 2 0 1 0 3 + 30 0 1 0 0 0 0 0 0 0 0 + 31 0 1 0 0 0 0 0 2 0 2 + 32 0 0 0 1 0 2 0 0 0 0 + 33 0 0 0 2 0 0 0 0 0 1 + 34 0 0 0 0 0 0 0 0 0 2 + 35 0 1 0 1 0 0 0 0 0 1 + 36 0 1 0 0 0 1 0 1 0 0 + 37 0 0 0 1 0 0 0 0 0 0 + 40 0 0 0 0 0 1 0 1 0 0 + 41 0 0 0 0 0 0 0 0 0 1 + 42 0 0 0 0 0 0 0 0 0 1 + 44 0 0 0 0 0 1 0 0 0 0 + 46 0 0 0 0 0 0 0 1 0 0 + 47 0 0 0 0 0 0 0 0 0 1 + 50 0 0 0 0 0 0 0 0 0 1 + 54 0 0 0 1 0 0 0 0 0 0 + 58 0 0 0 1 0 0 0 0 0 0 + over: 0 0 0 0 0 0 0 0 0 0 + count: 600002 600002 600002 600002 600002 600002 600002 600002 600002 600002 + min: 0 1 0 1 0 1 0 1 0 1 + avg: 0 5 0 5 0 4 0 4 0 5 + max: 16 36 15 58 24 44 21 46 13 50 + +SEE ALSO +======== +**rtla-timerlat**\(1), **rtla-timerlat-top**\(1) + +*timerlat* tracer documentation: + +AUTHOR +====== +Written by Daniel Bristot de Oliveira