From patchwork Mon Jun 8 07:41:58 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chen Yu X-Patchwork-Id: 11592687 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id E129692A for ; Mon, 8 Jun 2020 07:41:14 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id CF7582082F for ; Mon, 8 Jun 2020 07:41:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729071AbgFHHlO (ORCPT ); Mon, 8 Jun 2020 03:41:14 -0400 Received: from mga12.intel.com ([192.55.52.136]:7078 "EHLO mga12.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729046AbgFHHlN (ORCPT ); Mon, 8 Jun 2020 03:41:13 -0400 IronPort-SDR: 3Gg1FWpqdHZ/jkl2vgxRsOqja8svUsIKF0wam3WstZ8IsYGjLSuzp2IjPZj3OyVO7v/btyKpu8 hVl2gZ2ubJjw== X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 08 Jun 2020 00:41:09 -0700 IronPort-SDR: +cZS7Uo1JfltpvAh8vpLKL6o/4HjE9wSx6oKD/5ZWMnBDOFREMOURCMXx9wuFGAUD0b5U3yt9j bjB2QRT3O/Vg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.73,487,1583222400"; d="scan'208";a="274134916" Received: from chenyu-office.sh.intel.com ([10.239.158.173]) by orsmga006.jf.intel.com with ESMTP; 08 Jun 2020 00:41:07 -0700 From: Chen Yu To: "Rafael J. Wysocki" , Len Brown , Greg Kroah-Hartman , Michal Miroslaw Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, Chen Yu Subject: [PATCH 1/2][RFC v2] PM-runtime: Move all runtime usage related function to runtime.c Date: Mon, 8 Jun 2020 15:41:58 +0800 Message-Id: X-Mailer: git-send-email 2.17.1 In-Reply-To: References: Sender: linux-pm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pm@vger.kernel.org In order to track all the runtime usage count change, move the code related to runtime usage count change from pm_runtime.h to runtime.c, so that in runtime.c we can leverage trace event to do the tracking. Meanwhile export pm_runtime_get_noresume() and pm_runtime_put_noidle() so the module can use them. No functional change. Reported-by: kernel test robot Signed-off-by: Chen Yu --- drivers/base/power/runtime.c | 12 ++++++++++++ include/linux/pm_runtime.h | 12 ++---------- 2 files changed, 14 insertions(+), 10 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 9f62790f644c..85a248e196ca 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -1738,6 +1738,18 @@ void pm_runtime_drop_link(struct device *dev) spin_unlock_irq(&dev->power.lock); } +void pm_runtime_get_noresume(struct device *dev) +{ + atomic_inc(&dev->power.usage_count); +} +EXPORT_SYMBOL_GPL(pm_runtime_get_noresume); + +void pm_runtime_put_noidle(struct device *dev) +{ + atomic_add_unless(&dev->power.usage_count, -1, 0); +} +EXPORT_SYMBOL_GPL(pm_runtime_put_noidle); + static bool pm_runtime_need_not_resume(struct device *dev) { return atomic_read(&dev->power.usage_count) <= 1 && diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h index 3dbc207bff53..26510fef2acd 100644 --- a/include/linux/pm_runtime.h +++ b/include/linux/pm_runtime.h @@ -59,6 +59,8 @@ extern void pm_runtime_get_suppliers(struct device *dev); extern void pm_runtime_put_suppliers(struct device *dev); extern void pm_runtime_new_link(struct device *dev); extern void pm_runtime_drop_link(struct device *dev); +extern void pm_runtime_get_noresume(struct device *dev); +extern void pm_runtime_put_noidle(struct device *dev); static inline int pm_runtime_get_if_in_use(struct device *dev) { @@ -70,16 +72,6 @@ static inline void pm_suspend_ignore_children(struct device *dev, bool enable) dev->power.ignore_children = enable; } -static inline void pm_runtime_get_noresume(struct device *dev) -{ - atomic_inc(&dev->power.usage_count); -} - -static inline void pm_runtime_put_noidle(struct device *dev) -{ - atomic_add_unless(&dev->power.usage_count, -1, 0); -} - static inline bool pm_runtime_suspended(struct device *dev) { return dev->power.runtime_status == RPM_SUSPENDED From patchwork Mon Jun 8 07:42:15 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Chen Yu X-Patchwork-Id: 11592689 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 9A2D8618 for ; Mon, 8 Jun 2020 07:41:21 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 8A5AE2076A for ; Mon, 8 Jun 2020 07:41:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729046AbgFHHlU (ORCPT ); Mon, 8 Jun 2020 03:41:20 -0400 Received: from mga07.intel.com ([134.134.136.100]:65231 "EHLO mga07.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728974AbgFHHlU (ORCPT ); Mon, 8 Jun 2020 03:41:20 -0400 IronPort-SDR: CQmoUCxqpyVY144gDNcmQ9HjHsisomzOjrz0DHv1s9xO6t+B/HdceY015PPPcC2dzM/TJ79oZr xv9ySoJNkojQ== X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga006.jf.intel.com ([10.7.209.51]) by orsmga105.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 08 Jun 2020 00:41:19 -0700 IronPort-SDR: 2wrP6yDSENKSDdYN7WA/ndnPhqnAaLsVJevuZ3NP6eQqlIbc1xnCMLuG61+81uKX+Aw6+cDS0x jIFLB9JLqAVQ== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.73,487,1583222400"; d="scan'208";a="274134975" Received: from chenyu-office.sh.intel.com ([10.239.158.173]) by orsmga006.jf.intel.com with ESMTP; 08 Jun 2020 00:41:17 -0700 From: Chen Yu To: "Rafael J. Wysocki" , Len Brown , Greg Kroah-Hartman , Michal Miroslaw Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, Chen Yu Subject: [PATCH 2/2][RFC v2] PM-runtime: change the tracepoints to cover all usage_count Date: Mon, 8 Jun 2020 15:42:15 +0800 Message-Id: <0a3ae7be4c467912327005995b1defb4fefd8101.1591600914.git.yu.c.chen@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: References: Sender: linux-pm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pm@vger.kernel.org Commit d229290689ae ("PM-runtime: add tracepoints for usage_count changes") has added some tracepoints to monitor the change of runtime usage, and there is something to improve: 1. There are some places that adjust the usage count have not been traced yet. For example, pm_runtime_get_noresume() and pm_runtime_put_noidle() 2. The change of the usage count will not be tracked if decreased from 1 to 0. This patch intends to adjust the logic to be consistent with the change of usage_counter, that is to say, only after the counter has been possibly modified, we record it. Besides, all usage changes will be shown using rpm_usage even if included by other trace points. And these changes has helped track down the e1000e runtime issue. Signed-off-by: Chen Yu Reviewed-by: Michał Mirosław --- v2: According to Michal's suggestion, adjust the commit log to better describe the meaning of this patch. -- drivers/base/power/runtime.c | 38 +++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 14 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 85a248e196ca..5789d2624513 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -1004,10 +1004,11 @@ int __pm_runtime_idle(struct device *dev, int rpmflags) int retval; if (rpmflags & RPM_GET_PUT) { - if (!atomic_dec_and_test(&dev->power.usage_count)) { - trace_rpm_usage_rcuidle(dev, rpmflags); + bool non_zero = !atomic_dec_and_test(&dev->power.usage_count); + + trace_rpm_usage_rcuidle(dev, rpmflags); + if (non_zero) return 0; - } } might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); @@ -1038,10 +1039,12 @@ int __pm_runtime_suspend(struct device *dev, int rpmflags) int retval; if (rpmflags & RPM_GET_PUT) { - if (!atomic_dec_and_test(&dev->power.usage_count)) { - trace_rpm_usage_rcuidle(dev, rpmflags); + bool non_zero = !atomic_dec_and_test(&dev->power.usage_count); + + trace_rpm_usage_rcuidle(dev, rpmflags); + if (non_zero) return 0; - } + } might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); @@ -1073,8 +1076,10 @@ int __pm_runtime_resume(struct device *dev, int rpmflags) might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe && dev->power.runtime_status != RPM_ACTIVE); - if (rpmflags & RPM_GET_PUT) + if (rpmflags & RPM_GET_PUT) { atomic_inc(&dev->power.usage_count); + trace_rpm_usage_rcuidle(dev, rpmflags); + } spin_lock_irqsave(&dev->power.lock, flags); retval = rpm_resume(dev, rpmflags); @@ -1433,6 +1438,7 @@ void pm_runtime_forbid(struct device *dev) dev->power.runtime_auto = false; atomic_inc(&dev->power.usage_count); + trace_rpm_usage_rcuidle(dev, 0); rpm_resume(dev, 0); out: @@ -1448,16 +1454,17 @@ EXPORT_SYMBOL_GPL(pm_runtime_forbid); */ void pm_runtime_allow(struct device *dev) { + bool is_zero; + spin_lock_irq(&dev->power.lock); if (dev->power.runtime_auto) goto out; dev->power.runtime_auto = true; - if (atomic_dec_and_test(&dev->power.usage_count)) + is_zero = atomic_dec_and_test(&dev->power.usage_count); + trace_rpm_usage_rcuidle(dev, RPM_AUTO | RPM_ASYNC); + if (is_zero) rpm_idle(dev, RPM_AUTO | RPM_ASYNC); - else - trace_rpm_usage_rcuidle(dev, RPM_AUTO | RPM_ASYNC); - out: spin_unlock_irq(&dev->power.lock); } @@ -1523,9 +1530,8 @@ static void update_autosuspend(struct device *dev, int old_delay, int old_use) /* If it used to be allowed then prevent it. */ if (!old_use || old_delay >= 0) { atomic_inc(&dev->power.usage_count); - rpm_resume(dev, 0); - } else { trace_rpm_usage_rcuidle(dev, 0); + rpm_resume(dev, 0); } } @@ -1533,8 +1539,10 @@ static void update_autosuspend(struct device *dev, int old_delay, int old_use) else { /* If it used to be prevented then allow it. */ - if (old_use && old_delay < 0) + if (old_use && old_delay < 0) { atomic_dec(&dev->power.usage_count); + trace_rpm_usage_rcuidle(dev, 0); + } /* Maybe we can autosuspend now. */ rpm_idle(dev, RPM_AUTO); @@ -1741,12 +1749,14 @@ void pm_runtime_drop_link(struct device *dev) void pm_runtime_get_noresume(struct device *dev) { atomic_inc(&dev->power.usage_count); + trace_rpm_usage_rcuidle(dev, 0); } EXPORT_SYMBOL_GPL(pm_runtime_get_noresume); void pm_runtime_put_noidle(struct device *dev) { atomic_add_unless(&dev->power.usage_count, -1, 0); + trace_rpm_usage_rcuidle(dev, 0); } EXPORT_SYMBOL_GPL(pm_runtime_put_noidle);