From patchwork Sun Oct 18 18:24:39 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mathieu Poirier X-Patchwork-Id: 55204 Return-Path: X-Original-To: linaro@patches.linaro.org Delivered-To: linaro@patches.linaro.org Received: from mail-lb0-f199.google.com (mail-lb0-f199.google.com [209.85.217.199]) by patches.linaro.org (Postfix) with ESMTPS id 5DE9922FFA for ; Sun, 18 Oct 2015 18:28:41 +0000 (UTC) Received: by lbbms9 with SMTP id ms9sf37789040lbb.3 for ; Sun, 18 Oct 2015 11:28:40 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:delivered-to:from:to:cc:subject :date:message-id:in-reply-to:references:sender:precedence:list-id :x-original-sender:x-original-authentication-results:mailing-list :list-post:list-help:list-archive:list-unsubscribe; bh=HFozbMR6CDXVTvmbnyyAsbZM8LgRYA1hyfrCWFvcuis=; b=EgEEWsFShtQL0wQU3ZF4uoUgrry0u89fOwCH2JfGxMQeIHyGqIVl8aR5aIxm1Zso0Z SBiSFaE99TBVa6V8KWe6DucIl/3C4+SNYNt3+zvinJO7tuYIKCiF2PLhaNWrycrMmYRP uqs/mzIZBLHtNE5Wy4UY/EMsWVJBvqv+wninkGTzRscJV5WkYqVYGdA2Pd0rLizVeabL FwCDtA26zv1v0uHDg9Dudb+8Mm7FeJk+8xvKR4NcORbEut30bOtBW+Tz1uNcIXnaMcT8 1R0g3LZjdraiBwjywP5cinASMfm3xoFew1EEuu2TvCl//tmwEhBKAYRBmiRyTyUigIEM TeeQ== X-Gm-Message-State: ALoCoQk5xe+heDtP0ol7S7pWbCCca5jpIC5URpMee1KHER+RwtqQv4UKqfriviu8t6SK/1RkDllG X-Received: by 10.112.184.170 with SMTP id ev10mr6107014lbc.4.1445192920196; Sun, 18 Oct 2015 11:28:40 -0700 (PDT) MIME-Version: 1.0 X-BeenThere: patchwork-forward@linaro.org Received: by 10.25.15.147 with SMTP id 19ls496146lfp.5.gmail; Sun, 18 Oct 2015 11:28:40 -0700 (PDT) X-Received: by 10.25.23.208 with SMTP id 77mr368981lfx.44.1445192920024; Sun, 18 Oct 2015 11:28:40 -0700 (PDT) Received: from mail-lb0-f178.google.com (mail-lb0-f178.google.com. [209.85.217.178]) by mx.google.com with ESMTPS id jn7si20099802lbc.125.2015.10.18.11.28.39 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sun, 18 Oct 2015 11:28:39 -0700 (PDT) Received-SPF: pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 209.85.217.178 as permitted sender) client-ip=209.85.217.178; Received: by lbcao8 with SMTP id ao8so127465264lbc.3 for ; Sun, 18 Oct 2015 11:28:39 -0700 (PDT) X-Received: by 10.112.168.228 with SMTP id zz4mr12813812lbb.73.1445192919856; Sun, 18 Oct 2015 11:28:39 -0700 (PDT) X-Forwarded-To: patchwork-forward@linaro.org X-Forwarded-For: patch@linaro.org patchwork-forward@linaro.org Delivered-To: patch@linaro.org Received: by 10.112.59.35 with SMTP id w3csp1078586lbq; Sun, 18 Oct 2015 11:28:38 -0700 (PDT) X-Received: by 10.66.61.205 with SMTP id s13mr29061788par.96.1445192918265; Sun, 18 Oct 2015 11:28:38 -0700 (PDT) Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id xp4si46454562pbc.180.2015.10.18.11.28.37; Sun, 18 Oct 2015 11:28:38 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753020AbbJRS2f (ORCPT + 28 others); Sun, 18 Oct 2015 14:28:35 -0400 Received: from mail-pa0-f50.google.com ([209.85.220.50]:35663 "EHLO mail-pa0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752489AbbJRS0K (ORCPT ); Sun, 18 Oct 2015 14:26:10 -0400 Received: by pasz6 with SMTP id z6so6768634pas.2 for ; Sun, 18 Oct 2015 11:26:10 -0700 (PDT) X-Received: by 10.66.219.195 with SMTP id pq3mr29323575pac.98.1445192770273; Sun, 18 Oct 2015 11:26:10 -0700 (PDT) Received: from t430.cg.shawcable.net (S0106002369de4dac.cg.shawcable.net. [184.64.168.246]) by smtp.gmail.com with ESMTPSA id hq1sm20402076pbb.43.2015.10.18.11.26.08 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Sun, 18 Oct 2015 11:26:09 -0700 (PDT) From: Mathieu Poirier To: gregkh@linuxfoundation.org, a.p.zijlstra@chello.nl, alexander.shishkin@linux.intel.com, acme@kernel.org, mingo@redhat.com, corbet@lwn.net, nicolas.pitre@linaro.org Cc: adrian.hunter@intel.com, zhang.chunyan@linaro.org, mike.leach@arm.com, tor@ti.com, al.grant@arm.com, pawel.moll@arm.com, linux-arm-kernel@lists.infradead.org, linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, mathieu.poirier@linaro.org Subject: [PATCH V2 22/30] coresight: etm-perf: new PMU driver for ETM tracers Date: Sun, 18 Oct 2015 12:24:39 -0600 Message-Id: <1445192687-24112-23-git-send-email-mathieu.poirier@linaro.org> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1445192687-24112-1-git-send-email-mathieu.poirier@linaro.org> References: <1445192687-24112-1-git-send-email-mathieu.poirier@linaro.org> Sender: linux-kernel-owner@vger.kernel.org Precedence: list List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-Removed-Original-Auth: Dkim didn't pass. X-Original-Sender: mathieu.poirier@linaro.org X-Original-Authentication-Results: mx.google.com; spf=pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 209.85.217.178 as permitted sender) smtp.mailfrom=patch+caf_=patchwork-forward=linaro.org@linaro.org Mailing-list: list patchwork-forward@linaro.org; contact patchwork-forward+owners@linaro.org X-Google-Group-Id: 836684582541 List-Post: , List-Help: , List-Archive: List-Unsubscribe: , Perf is a well known and used tool for performance monitoring and much more. A such it is an ideal condaditate for integration with coresight based HW tracing. This patch introduces a PMU that represent a coresight tracer to the Perf core. Signed-off-by: Mathieu Poirier --- drivers/hwtracing/coresight/Makefile | 3 +- drivers/hwtracing/coresight/coresight-etm-perf.c | 533 +++++++++++++++++++++++ drivers/hwtracing/coresight/coresight-etm-perf.h | 27 ++ drivers/hwtracing/coresight/coresight-etm3x.c | 7 + 4 files changed, 569 insertions(+), 1 deletion(-) create mode 100644 drivers/hwtracing/coresight/coresight-etm-perf.c create mode 100644 drivers/hwtracing/coresight/coresight-etm-perf.h diff --git a/drivers/hwtracing/coresight/Makefile b/drivers/hwtracing/coresight/Makefile index 233d66cf22d3..cf8c6d689747 100644 --- a/drivers/hwtracing/coresight/Makefile +++ b/drivers/hwtracing/coresight/Makefile @@ -9,6 +9,7 @@ obj-$(CONFIG_CORESIGHT_SINK_ETBV10) += coresight-etb10.o obj-$(CONFIG_CORESIGHT_LINKS_AND_SINKS) += coresight-funnel.o \ coresight-replicator.o obj-$(CONFIG_CORESIGHT_SOURCE_ETM3X) += coresight-etm3x.o coresight-etm-cp14.o \ - coresight-etm3x-sysfs.o + coresight-etm3x-sysfs.o \ + coresight-etm-perf.o obj-$(CONFIG_CORESIGHT_SOURCE_ETM4X) += coresight-etm4x.o obj-$(CONFIG_CORESIGHT_QCOM_REPLICATOR) += coresight-replicator-qcom.o diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.c b/drivers/hwtracing/coresight/coresight-etm-perf.c new file mode 100644 index 000000000000..dbd02277fcda --- /dev/null +++ b/drivers/hwtracing/coresight/coresight-etm-perf.c @@ -0,0 +1,533 @@ +/* + * Copyright(C) 2015 Linaro Limited. All rights reserved. + * Author: Mathieu Poirier + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "coresight-priv.h" + +#define CORESIGHT_ETM_PMU_NAME "cs_etm" + +static struct pmu etm_pmu; +static bool etm_perf_up; + +/** + * struct etm_event_data - Coresight specifics associated to an event + * @mask: hold the CPU(s) this event was set for. + * @source_config: per CPU tracer configuration associated to a + * trace session. + * @sink_config: per CPU AUX configuration associated to a + * trace session. + * @sink: sink associated to a CPU. + */ +struct etm_event_data { + cpumask_t mask; + void **source_config; + void **sink_config; + void **sink; +}; + +static DEFINE_PER_CPU(struct perf_output_handle, ctx_handle); +static DEFINE_PER_CPU(struct coresight_device *, csdev_src); + +/* ETMCR is 'config' */ +PMU_FORMAT_ATTR(cycacc, "config:12"); +PMU_FORMAT_ATTR(timestamp, "config:28"); + +static struct attribute *etm_config_formats_attr[] = { + &format_attr_cycacc.attr, + &format_attr_timestamp.attr, + NULL, +}; + +static struct attribute_group etm_pmu_format_group = { + .name = "format", + .attrs = etm_config_formats_attr, +}; + +static const struct attribute_group *etm_pmu_attr_groups[] = { + &etm_pmu_format_group, + NULL, +}; + +static void etm_event_read(struct perf_event *event) {} + +/** + * etm_event_build_path() - setup a path between source and sink + * @cpu: The CPU the tracer is associated to. + * @build: Whether the path should be setup or thorned down. + * + * Return: The _first_ sink buffer discovered during the walkthrough. + */ +static struct coresight_device *etm_event_build_path(int cpu, bool build) +{ + int ret = 0; + LIST_HEAD(path); + LIST_HEAD(sinks); + struct coresight_device *csdev_source; + struct coresight_device *csdev_sink = NULL; + + csdev_source = per_cpu(csdev_src, cpu); + + if (!csdev_source) + return ERR_PTR(-EINVAL); + + if (csdev_source->type != CORESIGHT_DEV_TYPE_SOURCE) + return ERR_PTR(-EINVAL); + + if (build) { + ret = coresight_build_paths(csdev_source, &path, &sinks, build); + if (ret) { + dev_dbg(&csdev_source->dev, + "creating path(s) failed\n"); + goto out; + } + + /* Everything is good, record first enabled sink buffer */ + csdev_sink = list_first_entry(&sinks, + struct coresight_device, sinks); + } else { + ret = coresight_build_paths(csdev_source, &path, NULL, build); + if (ret) + dev_dbg(&csdev_source->dev, + "releasing path(s) failed\n"); + } + +out: + return csdev_sink; +} + +static int etm_event_pmu_start(struct perf_event *event) +{ + int cpu, ret; + cpumask_t mask; + struct coresight_device *csdev; + + cpumask_clear(&mask); + if (event->cpu != -1) + cpumask_set_cpu(event->cpu, &mask); + else + cpumask_copy(&mask, cpu_online_mask); + + for_each_cpu(cpu, &mask) { + csdev = per_cpu(csdev_src, cpu); + + if (!source_ops(csdev)->perf_start) + continue; + + ret = source_ops(csdev)->perf_start(csdev); + if (ret) + goto err; + } + +out: + return ret; +err: + for_each_cpu(cpu, &mask) { + csdev = per_cpu(csdev_src, cpu); + + if (!source_ops(csdev)->perf_stop) + continue; + source_ops(csdev)->perf_stop(csdev); + } + + goto out; +} + +static void etm_event_destroy(struct perf_event *event) +{ + int cpu; + cpumask_t mask; + struct coresight_device *csdev; + + cpumask_clear(&mask); + if (event->cpu != -1) + cpumask_set_cpu(event->cpu, &mask); + else + cpumask_copy(&mask, cpu_online_mask); + + for_each_cpu(cpu, &mask) { + csdev = per_cpu(csdev_src, cpu); + etm_event_build_path(cpu, false); + + if (!source_ops(csdev)->perf_stop) + continue; + source_ops(csdev)->perf_stop(csdev); + } +} + +static int etm_event_init(struct perf_event *event) +{ + int ret; + + if (event->attr.type != etm_pmu.type) + return -ENOENT; + + if (event->cpu >= nr_cpu_ids) + return -EINVAL; + + ret = etm_event_pmu_start(event); + if (ret) + return ret; + + event->destroy = etm_event_destroy; + + return 0; +} + +static void *alloc_event_data(int cpu) +{ + int size; + struct etm_event_data *event_data; + void *source_config, *sink_config, *sink; + + event_data = kzalloc(sizeof(struct etm_event_data), GFP_KERNEL); + if (!event_data) + return NULL; + + if (cpu != -1) + size = 1; + else + size = num_online_cpus(); + + source_config = kcalloc(size, sizeof(void *), GFP_KERNEL); + if (!source_config) + goto source_config_err; + + sink_config = kcalloc(size, sizeof(void *), GFP_KERNEL); + if (!sink_config) + goto sink_config_err; + + sink = kcalloc(size, sizeof(void *), GFP_KERNEL); + if (!sink) + goto sink_err; + + cpumask_clear(&event_data->mask); + event_data->source_config = source_config; + event_data->sink_config = sink_config; + event_data->sink = sink; + +out: + return event_data; + +sink_err: + kfree(sink_config); +sink_config_err: + kfree(source_config); +source_config_err: + kfree(event_data); + event_data = NULL; + goto out; +} + +static void free_event_data(struct etm_event_data *event_data) +{ + int cpu; + cpumask_t *mask = &event_data->mask; + + for_each_cpu(cpu, mask) { + kfree(event_data->source_config[cpu]); + kfree(event_data->sink_config[cpu]); + kfree(event_data->sink[cpu]); + } + + kfree(event_data->source_config); + kfree(event_data->sink_config); + kfree(event_data->sink); + kfree(event_data); +} + +static void *etm_setup_aux(struct perf_event *event, void **pages, + int nr_pages, bool overwrite) +{ + int cpu; + cpumask_t *mask; + struct etm_event_data *event_data = NULL; + struct coresight_device *csdev; + + event_data = alloc_event_data(event->cpu); + if (!event_data) + return NULL; + + mask = &event_data->mask; + + if (event->cpu != -1) + cpumask_set_cpu(event->cpu, mask); + else + cpumask_copy(mask, cpu_online_mask); + + for_each_cpu(cpu, mask) { + struct coresight_device *sink; + + csdev = per_cpu(csdev_src, cpu); + if (!csdev) + goto err; + + /* Get the tracer's config from perf */ + if (!source_ops(csdev)->perf_get_config) + goto err; + + event_data->source_config[cpu] = + source_ops(csdev)->perf_get_config(csdev, event); + + if (!event_data->source_config[cpu]) + goto err; + + /* + * Get a handle on the sink buffer associated + * with this tracer. + */ + event_data->sink[cpu] = (void *)etm_event_build_path(cpu, true); + + if (!event_data->sink[cpu]) + goto err; + + sink = event_data->sink[cpu]; + + if (!sink_ops(sink)->setup_aux) + goto err; + + /* Finally get the AUX specific data from the sink buffer */ + event_data->sink_config[cpu] = + sink_ops(sink)->setup_aux(sink, cpu, pages, + nr_pages, overwrite); + if (!event_data->sink_config[cpu]) + goto err; + } + +out: + return event_data; + +err: + for_each_cpu(cpu, mask) { + etm_event_build_path(cpu, false); + } + + free_event_data(event_data); + event_data = NULL; + goto out; +} + +static void etm_free_aux(void *data) +{ + free_event_data(data); +} + +static void etm_event_stop(struct perf_event *event, int mode) +{ + int cpu = smp_processor_id(); + struct coresight_device *csdev = per_cpu(csdev_src, cpu); + + if (event->hw.state == PERF_HES_STOPPED) + return; + + if (!csdev) + return; + + /* stop tracer */ + if (!source_ops(csdev)->perf_disable) + return; + + if (source_ops(csdev)->perf_disable(csdev)) + return; + + /* tell the core */ + event->hw.state = PERF_HES_STOPPED; + + + if (mode & PERF_EF_UPDATE) { + struct coresight_device *sink; + struct perf_output_handle *handle = this_cpu_ptr(&ctx_handle); + struct etm_event_data *event_data = perf_get_aux(handle); + + if (WARN_ON_ONCE(handle->event != event)) + return; + + if (WARN_ON_ONCE(!event_data)) + return; + + sink = event_data->sink[cpu]; + if (WARN_ON_ONCE(!sink)) + return; + + /* update trace information */ + if (!sink_ops(sink)->update_buffer) + return; + + sink_ops(sink)->update_buffer(sink, handle, + event_data->sink_config[cpu]); + } +} + +static void etm_event_start(struct perf_event *event, int flags) +{ + int cpu = smp_processor_id(); + struct coresight_device *csdev = per_cpu(csdev_src, cpu); + + if (!csdev) + goto fail; + + /* tell the perf core the event is alive */ + event->hw.state = 0; + + if (!source_ops(csdev)->perf_enable) + goto fail; + + if (source_ops(csdev)->perf_enable(csdev)) + goto fail; + + return; + +fail: + event->hw.state = PERF_HES_STOPPED; +} + +static void etm_event_del(struct perf_event *event, int mode) +{ + int cpu = smp_processor_id(); + struct coresight_device *sink; + struct perf_output_handle *handle = this_cpu_ptr(&ctx_handle); + struct etm_event_data *event_data = perf_get_aux(handle); + + if (WARN_ON_ONCE(!event_data)) + return; + + sink = event_data->sink[cpu]; + if (!sink) + return; + + etm_event_stop(event, PERF_EF_UPDATE); + + if (!sink_ops(sink)->reset_buffer) + return; + + sink_ops(sink)->reset_buffer(sink, handle, + event_data->sink_config[cpu]); +} + +static int etm_event_add(struct perf_event *event, int mode) +{ + + int ret = -EBUSY, cpu = smp_processor_id(); + struct etm_event_data *event_data; + struct perf_output_handle *handle = this_cpu_ptr(&ctx_handle); + struct hw_perf_event *hwc = &event->hw; + struct coresight_device *csdev = per_cpu(csdev_src, cpu); + struct coresight_device *sink; + + if (handle->event) + goto out; + + event_data = perf_aux_output_begin(handle, event); + ret = -EINVAL; + if (WARN_ON_ONCE(!event_data)) + goto fail_stop; + + sink = event_data->sink[cpu]; + if (!sink) + goto fail_end_stop; + + if (!sink_ops(sink)->set_buffer) + goto fail_end_stop; + + ret = sink_ops(sink)->set_buffer(sink, handle, + event_data->sink_config[cpu]); + if (ret) + goto fail_end_stop; + + if (!source_ops(csdev)->perf_set_config) { + ret = -EINVAL; + goto fail_end_stop; + } + + source_ops(csdev)->perf_set_config(csdev, + event_data->source_config[cpu]); + + if (mode & PERF_EF_START) { + etm_event_start(event, 0); + if (hwc->state & PERF_HES_STOPPED) { + etm_event_del(event, 0); + return -EBUSY; + } + } + +out: + return ret; + +fail_end_stop: + perf_aux_output_end(handle, 0, true); +fail_stop: + hwc->state = PERF_HES_STOPPED; + goto out; +} + +int etm_perf_symlink(struct coresight_device *csdev, bool link) +{ + char entry[sizeof("cpu9999999")]; + int ret = 0, cpu = source_ops(csdev)->cpu_id(csdev); + struct device *pmu_dev = etm_pmu.dev; + struct device *cs_dev = &csdev->dev; + + sprintf(entry, "cpu%d", cpu); + + if (!etm_perf_up) + return -EPROBE_DEFER; + + if (link) { + ret = sysfs_create_link(&pmu_dev->kobj, &cs_dev->kobj, entry); + if (ret) + return ret; + per_cpu(csdev_src, cpu) = csdev; + } else { + sysfs_remove_link(&pmu_dev->kobj, entry); + per_cpu(csdev_src, cpu) = NULL; + } + + return 0; +} + +static int __init etm_perf_init(void) +{ + int ret; + + etm_pmu.capabilities = PERF_PMU_CAP_EXCLUSIVE; + + etm_pmu.attr_groups = etm_pmu_attr_groups; + etm_pmu.task_ctx_nr = perf_sw_context; + etm_pmu.read = etm_event_read; + etm_pmu.event_init = etm_event_init; + etm_pmu.setup_aux = etm_setup_aux; + etm_pmu.free_aux = etm_free_aux; + etm_pmu.stop = etm_event_stop; + etm_pmu.start = etm_event_start; + etm_pmu.del = etm_event_del; + etm_pmu.add = etm_event_add; + + ret = perf_pmu_register(&etm_pmu, CORESIGHT_ETM_PMU_NAME, -1); + if (ret == 0) + etm_perf_up = true; + + return ret; +} +module_init(etm_perf_init); diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.h b/drivers/hwtracing/coresight/coresight-etm-perf.h new file mode 100644 index 000000000000..4dd900f2362a --- /dev/null +++ b/drivers/hwtracing/coresight/coresight-etm-perf.h @@ -0,0 +1,27 @@ +/* Copyright (c) 2014-2015, The Linux Foundation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 and + * only version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _CORESIGHT_ETM_PERF_H +#define _CORESIGHT_ETM_PERF_H + +struct coresight_device; + +#ifdef CONFIG_CORESIGHT +int etm_perf_symlink(struct coresight_device *csdev, bool link); + +#else +static inline int etm_perf_symlink(struct coresight_device *csdev, bool link) +{ return -EINVAL; } + +#endif /* CONFIG_CORESIGHT */ + +#endif diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c index 9b4c0359ca29..7407c7ecf668 100644 --- a/drivers/hwtracing/coresight/coresight-etm3x.c +++ b/drivers/hwtracing/coresight/coresight-etm3x.c @@ -35,6 +35,7 @@ #include #include "coresight-etm.h" +#include "coresight-etm-perf.h" static int boot_enable; module_param_named(boot_enable, boot_enable, int, S_IRUGO); @@ -850,6 +851,12 @@ static int etm_probe(struct amba_device *adev, const struct amba_id *id) goto err_arch_supported; } + ret = etm_perf_symlink(drvdata->csdev, true); + if (ret) { + coresight_unregister(drvdata->csdev); + goto err_arch_supported; + } + pm_runtime_put(&adev->dev); dev_info(dev, "%s initialized\n", (char *)id->data);