From nobody Fri Sep 19 02:17:27 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id F26F3C433FE for ; Tue, 29 Nov 2022 23:34:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229971AbiK2Xen (ORCPT ); Tue, 29 Nov 2022 18:34:43 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35858 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229595AbiK2Xeg (ORCPT ); Tue, 29 Nov 2022 18:34:36 -0500 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 3678CF020; Tue, 29 Nov 2022 15:34:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1669764876; x=1701300876; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Z2p0ZEt70m3lASgdlGcNqNlgamI3K0iU68fdPui8BqQ=; b=F54it81DKS6gQC3b9YRtjxWg/ifh5hBhGWupmXTUjZtKFVhSqmQ+uRgX RZOP2sU/0m5AZdOuFkyE/6dxrXBZbCYci3WZKup8mL+wzcMpaQWNrmbiM Ix2PTOaWJFXJGTM2zQKH6ZvyUOP+9utWAhmckWSLQs3C7lGCcoQigC6Uj cCVGF3566Z57pA+UjnAG3j2TpdFXM3deHTsJ/1gsWebv9jbYHJq+1r3Ax wGh5FGTGFlJ430NpVUBPQo02rY1fz96h/hWko1FQYTuFnXvppjGbj5twF FAu5tQTJ+zI4Kta1A++b2geIITMXHJx3QL5lMJT4ivTv8siSBOYh+eyhk g==; X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="342178167" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="342178167" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 Nov 2022 15:34:35 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="768617600" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="768617600" Received: from spandruv-desk.jf.intel.com ([10.54.75.8]) by orsmga004.jf.intel.com with ESMTP; 29 Nov 2022 15:34:35 -0800 From: Srinivas Pandruvada To: rafael@kernel.org Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, daniel.lezcano@linaro.org, rui.zhang@intel.com, amitk@kernel.org, Srinivas Pandruvada Subject: [PATCH v2 1/4] powercap: idle_inject: Export symbols Date: Tue, 29 Nov 2022 15:34:16 -0800 Message-Id: <20221129233419.4022830-2-srinivas.pandruvada@linux.intel.com> X-Mailer: git-send-email 2.31.1 In-Reply-To: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> References: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Export symbols for external interfaces, so that they can be used in other loadable modules. Export is done under name space IDLE_INJECT. Signed-off-by: Srinivas Pandruvada Acked-by: Rafael J. Wysocki Reported-by: kernel test robot --- v2: No change drivers/powercap/idle_inject.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/powercap/idle_inject.c b/drivers/powercap/idle_inject.c index fe86a09e3b67..dfa989182e71 100644 --- a/drivers/powercap/idle_inject.c +++ b/drivers/powercap/idle_inject.c @@ -160,6 +160,7 @@ void idle_inject_set_duration(struct idle_inject_device= *ii_dev, WRITE_ONCE(ii_dev->idle_duration_us, idle_duration_us); } } +EXPORT_SYMBOL_NS_GPL(idle_inject_set_duration, IDLE_INJECT); =20 /** * idle_inject_get_duration - idle and run duration retrieval helper @@ -174,6 +175,7 @@ void idle_inject_get_duration(struct idle_inject_device= *ii_dev, *run_duration_us =3D READ_ONCE(ii_dev->run_duration_us); *idle_duration_us =3D READ_ONCE(ii_dev->idle_duration_us); } +EXPORT_SYMBOL_NS_GPL(idle_inject_get_duration, IDLE_INJECT); =20 /** * idle_inject_set_latency - set the maximum latency allowed @@ -185,6 +187,7 @@ void idle_inject_set_latency(struct idle_inject_device = *ii_dev, { WRITE_ONCE(ii_dev->latency_us, latency_us); } +EXPORT_SYMBOL_NS_GPL(idle_inject_set_latency, IDLE_INJECT); =20 /** * idle_inject_start - start idle injections @@ -216,6 +219,7 @@ int idle_inject_start(struct idle_inject_device *ii_dev) =20 return 0; } +EXPORT_SYMBOL_NS_GPL(idle_inject_start, IDLE_INJECT); =20 /** * idle_inject_stop - stops idle injections @@ -262,6 +266,7 @@ void idle_inject_stop(struct idle_inject_device *ii_dev) =20 cpu_hotplug_enable(); } +EXPORT_SYMBOL_NS_GPL(idle_inject_stop, IDLE_INJECT); =20 /** * idle_inject_setup - prepare the current task for idle injection @@ -337,6 +342,7 @@ struct idle_inject_device *idle_inject_register(struct = cpumask *cpumask) =20 return NULL; } +EXPORT_SYMBOL_NS_GPL(idle_inject_register, IDLE_INJECT); =20 /** * idle_inject_unregister - unregister idle injection control device @@ -357,6 +363,7 @@ void idle_inject_unregister(struct idle_inject_device *= ii_dev) =20 kfree(ii_dev); } +EXPORT_SYMBOL_NS_GPL(idle_inject_unregister, IDLE_INJECT); =20 static struct smp_hotplug_thread idle_inject_threads =3D { .store =3D &idle_inject_thread.tsk, --=20 2.31.1 From nobody Fri Sep 19 02:17:27 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B8462C433FE for ; Tue, 29 Nov 2022 23:34:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229877AbiK2Xep (ORCPT ); Tue, 29 Nov 2022 18:34:45 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35870 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229679AbiK2Xeh (ORCPT ); Tue, 29 Nov 2022 18:34:37 -0500 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C351A13D46; Tue, 29 Nov 2022 15:34:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1669764876; x=1701300876; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Hvisxqquf2yYaCVhVJZxHzOaZVsvlPXCmm2MsGkCZx8=; b=YOK9OlxLztQdsHakZ2JkVlgV+INQoIbtgZLvJ9CgrCmIxOh/YYCOh0Yl 0R6JxO9IQUSrO4x+Uz5fmlU7PceH2/LfU6/n9uBwXCxfxf5A69RMZu2dW UD2Y9fqxiMAKvT1BaY3m1oLFzUyrjR/Xpkw8zqpxGc1IaxVYOKsQgCm92 DxBSO278mnVDaNwpzRXZy7JXox7NTHwG6GNOOPL3hVJL7n+3fHwrOTka5 dh+HjaaQte3hyo6UYXo1aq6e0cKOKr7mpSAJo0KIG/KWx31kLKBnzzHMm nEreD7NdDlp3imd1l42pYm/nxJVjyl4X2+pb12Tbvy7SUGChnU1Qpbm8m Q==; X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="342178170" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="342178170" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 Nov 2022 15:34:35 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="768617604" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="768617604" Received: from spandruv-desk.jf.intel.com ([10.54.75.8]) by orsmga004.jf.intel.com with ESMTP; 29 Nov 2022 15:34:35 -0800 From: Srinivas Pandruvada To: rafael@kernel.org Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, daniel.lezcano@linaro.org, rui.zhang@intel.com, amitk@kernel.org, Srinivas Pandruvada Subject: [PATCH v2 2/4] powercap: idle_inject: Add prepare/complete callbacks Date: Tue, 29 Nov 2022 15:34:17 -0800 Message-Id: <20221129233419.4022830-3-srinivas.pandruvada@linux.intel.com> X-Mailer: git-send-email 2.31.1 In-Reply-To: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> References: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The actual idle percentage can be less than the desired because of interrupts. Since the objective for CPU Idle injection is for thermal control, there should be some way to compensate for lost idle percentage. Some architectures provide interface to get actual idle percent observed by the hardware. So, the idle percent can be adjusted using the hardware feedback. For example, Intel CPUs provides package idle counters, which is currently used by intel powerclamp driver to adjust idle time. The only way this can be done currently is by monitoring hardware idle percent from a different software thread. This can be avoided by adding callbacks. Add a capability to register a prepare and complete callback during idle inject registry. Add a new register function idle_inject_register_full() which also allows to register callbacks. If they are not NULL, then prepare callback is called before calling play_idle_precise() and complete callback is called after calling play_idle_precise(). If prepare callback is present and returns non 0 value then play_idle_precise() is not called to avoid over compensation. Signed-off-by: Srinivas Pandruvada Reported-by: kernel test robot --- v2 - Replace begin/end with prepare/complete - Add new interface idle_inject_register_full with callbacks - Update kernel doc - Update commit description drivers/powercap/idle_inject.c | 62 +++++++++++++++++++++++++++++++--- include/linux/idle_inject.h | 4 +++ 2 files changed, 62 insertions(+), 4 deletions(-) diff --git a/drivers/powercap/idle_inject.c b/drivers/powercap/idle_inject.c index dfa989182e71..f48e71501429 100644 --- a/drivers/powercap/idle_inject.c +++ b/drivers/powercap/idle_inject.c @@ -63,13 +63,31 @@ struct idle_inject_thread { * @idle_duration_us: duration of CPU idle time to inject * @run_duration_us: duration of CPU run time to allow * @latency_us: max allowed latency + * @prepare: Callback function which is called before calling + * play_idle_precise() + * @complete: Callback function which is called after calling + * play_idle_precise() * @cpumask: mask of CPUs affected by idle injection + * + * This structure is used to define per instance idle inject device data. = Each + * instance has an idle duration, a run duration and mask of CPUs to inject + * idle. + * Actual idle is injected by calling kernel scheduler interface + * play_idle_precise(). There are two optional callbacks which the caller = can + * register by calling idle_inject_register_full(): + * prepare() - This callback is called just before calling play_idle_preci= se() + * If this callback returns non zero value then + * play_idle_precise() is not called. This means skip injecting + * idle during this period. + * complete() - This callback is called after calling play_idle_precise(). */ struct idle_inject_device { struct hrtimer timer; unsigned int idle_duration_us; unsigned int run_duration_us; unsigned int latency_us; + int (*prepare)(unsigned int cpu); + void (*complete)(unsigned int cpu); unsigned long cpumask[]; }; =20 @@ -132,6 +150,7 @@ static void idle_inject_fn(unsigned int cpu) { struct idle_inject_device *ii_dev; struct idle_inject_thread *iit; + int ret; =20 ii_dev =3D per_cpu(idle_inject_device, cpu); iit =3D per_cpu_ptr(&idle_inject_thread, cpu); @@ -141,8 +160,18 @@ static void idle_inject_fn(unsigned int cpu) */ iit->should_run =3D 0; =20 + if (ii_dev->prepare) { + ret =3D ii_dev->prepare(cpu); + if (ret) + goto skip; + } + play_idle_precise(READ_ONCE(ii_dev->idle_duration_us) * NSEC_PER_USEC, READ_ONCE(ii_dev->latency_us) * NSEC_PER_USEC); + +skip: + if (ii_dev->complete) + ii_dev->complete(cpu); } =20 /** @@ -295,17 +324,23 @@ static int idle_inject_should_run(unsigned int cpu) } =20 /** - * idle_inject_register - initialize idle injection on a set of CPUs + * idle_inject_register_full - initialize idle injection on a set of CPUs * @cpumask: CPUs to be affected by idle injection + * @prepare: callback called before calling play_idle_precise() + * @complete: callback called after calling play_idle_precise() * * This function creates an idle injection control device structure for the - * given set of CPUs and initializes the timer associated with it. It doe= s not - * start any injection cycles. + * given set of CPUs and initializes the timer associated with it. This + * function also allows to register prepare() and complete() callbacks. + * It does not start any injection cycles. * * Return: NULL if memory allocation fails, idle injection control device * pointer on success. */ -struct idle_inject_device *idle_inject_register(struct cpumask *cpumask) + +struct idle_inject_device *idle_inject_register_full(struct cpumask *cpuma= sk, + int (*prepare)(unsigned int cpu), + void (*complete)(unsigned int cpu)) { struct idle_inject_device *ii_dev; int cpu, cpu_rb; @@ -318,6 +353,8 @@ struct idle_inject_device *idle_inject_register(struct = cpumask *cpumask) hrtimer_init(&ii_dev->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); ii_dev->timer.function =3D idle_inject_timer_fn; ii_dev->latency_us =3D UINT_MAX; + ii_dev->prepare =3D prepare; + ii_dev->complete =3D complete; =20 for_each_cpu(cpu, to_cpumask(ii_dev->cpumask)) { =20 @@ -342,6 +379,23 @@ struct idle_inject_device *idle_inject_register(struct= cpumask *cpumask) =20 return NULL; } +EXPORT_SYMBOL_NS_GPL(idle_inject_register_full, IDLE_INJECT); + +/** + * idle_inject_register - initialize idle injection on a set of CPUs + * @cpumask: CPUs to be affected by idle injection + * + * This function creates an idle injection control device structure for the + * given set of CPUs and initializes the timer associated with it. It doe= s not + * start any injection cycles. + * + * Return: NULL if memory allocation fails, idle injection control device + * pointer on success. + */ +struct idle_inject_device *idle_inject_register(struct cpumask *cpumask) +{ + return idle_inject_register_full(cpumask, NULL, NULL); +} EXPORT_SYMBOL_NS_GPL(idle_inject_register, IDLE_INJECT); =20 /** diff --git a/include/linux/idle_inject.h b/include/linux/idle_inject.h index fb88e23a99d3..e18d89793490 100644 --- a/include/linux/idle_inject.h +++ b/include/linux/idle_inject.h @@ -13,6 +13,10 @@ struct idle_inject_device; =20 struct idle_inject_device *idle_inject_register(struct cpumask *cpumask); =20 +struct idle_inject_device *idle_inject_register_full(struct cpumask *cpuma= sk, + int (*prepare)(unsigned int cpu), + void (*complete)(unsigned int cpu)); + void idle_inject_unregister(struct idle_inject_device *ii_dev); =20 int idle_inject_start(struct idle_inject_device *ii_dev); --=20 2.31.1 From nobody Fri Sep 19 02:17:27 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0A855C433FE for ; Tue, 29 Nov 2022 23:34:52 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230124AbiK2Xeu (ORCPT ); Tue, 29 Nov 2022 18:34:50 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35876 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229642AbiK2Xei (ORCPT ); Tue, 29 Nov 2022 18:34:38 -0500 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id D14B013F5A; Tue, 29 Nov 2022 15:34:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1669764876; x=1701300876; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=JdAL99X3U/wRyQMrUSkXLIkWotT0+se/Lxk0222z2TI=; b=Ovn1PSfee7L/bwSh6OqkU9BcesoTdC5k1nTz6jbEizdeFNnys25g3Hxm vJobmVb/M8hS2FTFMhmX0XZRj20qzJpArbg3qxj8ZkvPsbjNhkDci9J6s PBP5vr4cs13vlY4towX6J5cNUPjU4QI8qvtOPgTdMAwd7ONWXIjpV1Nhb 6HU+Ko66tiIzWcOi5tkMemIpXKFJ2LSmYscqmjvRGNqrD4aRm516r6ZeW c1GHpfga1OQsubR7IR+6baFhMwEfTCbcwd/9ACaEN6JQcKy5pLesc5oBD XxGI410XPAOxFWGvzIw3oJDFT1KjsLARucAKJ+RTdXPF1ENG58UCL3eGf Q==; X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="342178173" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="342178173" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 Nov 2022 15:34:35 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="768617608" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="768617608" Received: from spandruv-desk.jf.intel.com ([10.54.75.8]) by orsmga004.jf.intel.com with ESMTP; 29 Nov 2022 15:34:35 -0800 From: Srinivas Pandruvada To: rafael@kernel.org Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, daniel.lezcano@linaro.org, rui.zhang@intel.com, amitk@kernel.org, Srinivas Pandruvada , kernel test robot Subject: [PATCH v2 3/4] thermal/drivers/intel_powerclamp: Use powercap idle-inject framework Date: Tue, 29 Nov 2022 15:34:18 -0800 Message-Id: <20221129233419.4022830-4-srinivas.pandruvada@linux.intel.com> X-Mailer: git-send-email 2.31.1 In-Reply-To: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> References: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" There are two idle injection implementation in the Linux kernel. One via intel_powerclamp and the other using powercap/idle_inject. Both implementation end up in calling play_idle* function from a FIFO priority thread. Both can't be used at the same time. Currently per core idle injection (cpuidle_cooling) is using powercap/idle_inject, which is not used in platforms where intel_powerclamp is used for system wide idle injection. So there is no conflict. But there are some use cases where per core idle injection is beneficial on the same system where system wide idle injection is also used via intel_powerclamp. To avoid conflict only one of the idle injection type must be in use at a time. This require a common framework which both per core and system wide idle injection can use. Here powercap/idle_inject can be used for both per-core and for system wide idle injection. This framework has a well defined interface which allow registry for per-core or for all CPUs (system wide). If particular CPU is already participating in idle injection, the call to registry fails. Here the registry can be done when user space changes the current cooling device state. Also one framework for idle injection is better as there is one loop calling play_idle*, instead of multiple for better maintenance. So, reuse powercap/idle_inject calls in intel_powerclamp. This simplifies the code as all per CPU kthreads which calls play_idle* can be removed. The changes include: - Remove unneeded include files - Remove per CPU kthread workers: balancing_work and idle_injection_work - Reuse the compensation related code by moving from previous worker thread to idle_injection callbacks - Adjust the idle_duration and runtime by using powercap/idle_inject interface - Remove all variables, which are not required once powercap/idle_inject is used - Add mutex to avoid race during removal of idle injection during module unload and user action to change idle inject percent - Use READ_ONCE and WRITE_ONCE for data accessed from multiple CPUs Signed-off-by: Srinivas Pandruvada Reported-by: kernel test robot --- v2: - Use idle_inject_register_full instead of idle_inject_register - Also fix dependency issue with POWERCAP config Reported-by: kernel test robot drivers/thermal/intel/Kconfig | 2 + drivers/thermal/intel/intel_powerclamp.c | 292 ++++++++++------------- 2 files changed, 126 insertions(+), 168 deletions(-) diff --git a/drivers/thermal/intel/Kconfig b/drivers/thermal/intel/Kconfig index f0c845679250..6c2a95f41c81 100644 --- a/drivers/thermal/intel/Kconfig +++ b/drivers/thermal/intel/Kconfig @@ -3,6 +3,8 @@ config INTEL_POWERCLAMP tristate "Intel PowerClamp idle injection driver" depends on X86 depends on CPU_SUP_INTEL + select POWERCAP + select IDLE_INJECT help Enable this to enable Intel PowerClamp idle injection driver. This enforce idle time which results in more package C-state residency. The diff --git a/drivers/thermal/intel/intel_powerclamp.c b/drivers/thermal/int= el/intel_powerclamp.c index b80e25ec1261..3f2b20ae8f68 100644 --- a/drivers/thermal/intel/intel_powerclamp.c +++ b/drivers/thermal/intel/intel_powerclamp.c @@ -2,7 +2,7 @@ /* * intel_powerclamp.c - package c-state idle injection * - * Copyright (c) 2012, Intel Corporation. + * Copyright (c) 2022, Intel Corporation. * * Authors: * Arjan van de Ven @@ -27,21 +27,15 @@ #include #include #include -#include #include #include -#include -#include #include #include -#include -#include +#include =20 -#include #include #include #include -#include =20 #define MAX_TARGET_RATIO (50U) /* For each undisturbed clamping period (no extra wake ups during idle tim= e), @@ -60,6 +54,7 @@ static struct dentry *debug_dir; =20 /* user selected target */ static unsigned int set_target_ratio; +static bool target_ratio_updated; static unsigned int current_ratio; static bool should_skip; =20 @@ -67,26 +62,20 @@ static unsigned int control_cpu; /* The cpu assigned to= collect stat and update * control parameters. default to BSP but BSP * can be offlined. */ -static bool clamping; - -struct powerclamp_worker_data { - struct kthread_worker *worker; - struct kthread_work balancing_work; - struct kthread_delayed_work idle_injection_work; +struct powerclamp_data { unsigned int cpu; unsigned int count; unsigned int guard; unsigned int window_size_now; unsigned int target_ratio; - unsigned int duration_jiffies; bool clamping; }; =20 -static struct powerclamp_worker_data __percpu *worker_data; +static struct powerclamp_data powerclamp_data; + static struct thermal_cooling_device *cooling_dev; -static unsigned long *cpu_clamping_mask; /* bit map for tracking per cpu - * clamping kthread worker - */ + +static DEFINE_MUTEX(powerclamp_lock); =20 static unsigned int duration; static unsigned int pkg_cstate_ratio_cur; @@ -344,79 +333,33 @@ static bool powerclamp_adjust_controls(unsigned int t= arget_ratio, return set_target_ratio + guard <=3D current_ratio; } =20 -static void clamp_balancing_func(struct kthread_work *work) +static unsigned int get_run_time(void) { - struct powerclamp_worker_data *w_data; - int sleeptime; - unsigned long target_jiffies; unsigned int compensated_ratio; - int interval; /* jiffies to sleep for each attempt */ - - w_data =3D container_of(work, struct powerclamp_worker_data, - balancing_work); + unsigned int runtime; =20 /* * make sure user selected ratio does not take effect until * the next round. adjust target_ratio if user has changed * target such that we can converge quickly. */ - w_data->target_ratio =3D READ_ONCE(set_target_ratio); - w_data->guard =3D 1 + w_data->target_ratio / 20; - w_data->window_size_now =3D window_size; - w_data->duration_jiffies =3D msecs_to_jiffies(duration); - w_data->count++; + powerclamp_data.target_ratio =3D READ_ONCE(set_target_ratio); + powerclamp_data.guard =3D 1 + powerclamp_data.target_ratio / 20; + powerclamp_data.window_size_now =3D window_size; =20 /* * systems may have different ability to enter package level * c-states, thus we need to compensate the injected idle ratio * to achieve the actual target reported by the HW. */ - compensated_ratio =3D w_data->target_ratio + - get_compensation(w_data->target_ratio); + compensated_ratio =3D powerclamp_data.target_ratio + + get_compensation(powerclamp_data.target_ratio); if (compensated_ratio <=3D 0) compensated_ratio =3D 1; - interval =3D w_data->duration_jiffies * 100 / compensated_ratio; - - /* align idle time */ - target_jiffies =3D roundup(jiffies, interval); - sleeptime =3D target_jiffies - jiffies; - if (sleeptime <=3D 0) - sleeptime =3D 1; - - if (clamping && w_data->clamping && cpu_online(w_data->cpu)) - kthread_queue_delayed_work(w_data->worker, - &w_data->idle_injection_work, - sleeptime); -} =20 -static void clamp_idle_injection_func(struct kthread_work *work) -{ - struct powerclamp_worker_data *w_data; - - w_data =3D container_of(work, struct powerclamp_worker_data, - idle_injection_work.work); + runtime =3D duration * 100 / compensated_ratio - duration; =20 - /* - * only elected controlling cpu can collect stats and update - * control parameters. - */ - if (w_data->cpu =3D=3D control_cpu && - !(w_data->count % w_data->window_size_now)) { - should_skip =3D - powerclamp_adjust_controls(w_data->target_ratio, - w_data->guard, - w_data->window_size_now); - smp_mb(); - } - - if (should_skip) - goto balance; - - play_idle(jiffies_to_usecs(w_data->duration_jiffies)); - -balance: - if (clamping && w_data->clamping && cpu_online(w_data->cpu)) - kthread_queue_work(w_data->worker, &w_data->balancing_work); + return runtime; } =20 /* @@ -452,104 +395,127 @@ static void poll_pkg_cstate(struct work_struct *dum= my) msr_last =3D msr_now; tsc_last =3D tsc_now; =20 - if (true =3D=3D clamping) + if (powerclamp_data.clamping) schedule_delayed_work(&poll_pkg_cstate_work, HZ); } =20 -static void start_power_clamp_worker(unsigned long cpu) +static struct idle_inject_device *ii_dev; + +static int idle_inject_begin(unsigned int cpu) { - struct powerclamp_worker_data *w_data =3D per_cpu_ptr(worker_data, cpu); - struct kthread_worker *worker; + /* + * only elected controlling cpu can collect stats and update + * control parameters. + */ + if (cpu =3D=3D control_cpu) { + bool update =3D READ_ONCE(target_ratio_updated); + + if (!(powerclamp_data.count % powerclamp_data.window_size_now)) { + bool skip =3D powerclamp_adjust_controls(powerclamp_data.target_ratio, + powerclamp_data.guard, + powerclamp_data.window_size_now); + WRITE_ONCE(should_skip, skip); + update =3D true; + } =20 - worker =3D kthread_create_worker_on_cpu(cpu, 0, "kidle_inj/%ld", cpu); - if (IS_ERR(worker)) - return; + if (update) { + unsigned int runtime; + + runtime =3D get_run_time(); + idle_inject_set_duration(ii_dev, runtime, duration); + WRITE_ONCE(target_ratio_updated, false); + } + powerclamp_data.count++; + } + + if (READ_ONCE(should_skip)) + return -EAGAIN; =20 - w_data->worker =3D worker; - w_data->count =3D 0; - w_data->cpu =3D cpu; - w_data->clamping =3D true; - set_bit(cpu, cpu_clamping_mask); - sched_set_fifo(worker->task); - kthread_init_work(&w_data->balancing_work, clamp_balancing_func); - kthread_init_delayed_work(&w_data->idle_injection_work, - clamp_idle_injection_func); - kthread_queue_work(w_data->worker, &w_data->balancing_work); + return 0; } =20 -static void stop_power_clamp_worker(unsigned long cpu) +static void trigger_idle_injection(void) { - struct powerclamp_worker_data *w_data =3D per_cpu_ptr(worker_data, cpu); + unsigned int runtime =3D get_run_time(); =20 - if (!w_data->worker) - return; + idle_inject_set_duration(ii_dev, runtime, duration); + idle_inject_start(ii_dev); + powerclamp_data.clamping =3D true; +} + +static int powerclamp_idle_injection_register(void) +{ + static cpumask_t idle_injection_cpu_mask; + unsigned long cpu; =20 - w_data->clamping =3D false; - /* - * Make sure that all works that get queued after this point see - * the clamping disabled. The counter part is not needed because - * there is an implicit memory barrier when the queued work - * is proceed. - */ - smp_wmb(); - kthread_cancel_work_sync(&w_data->balancing_work); - kthread_cancel_delayed_work_sync(&w_data->idle_injection_work); /* - * The balancing work still might be queued here because - * the handling of the "clapming" variable, cancel, and queue - * operations are not synchronized via a lock. But it is not - * a big deal. The balancing work is fast and destroy kthread - * will wait for it. + * The idle inject core will only inject for online CPUs, + * So we can register for all present CPUs. In this way + * if some CPU goes online/offline while idle inject + * is registered, nothing additional calls are required. + * The same runtime and idle time is applicable for + * newly onlined CPUs if any. */ - clear_bit(w_data->cpu, cpu_clamping_mask); - kthread_destroy_worker(w_data->worker); + for_each_present_cpu(cpu) { + cpumask_set_cpu(cpu, &idle_injection_cpu_mask); + } + + ii_dev =3D idle_inject_register_full(&idle_injection_cpu_mask, + idle_inject_begin, NULL); + if (!ii_dev) { + pr_err("powerclamp: idle_inject_register failed\n"); + return -EAGAIN; + } =20 - w_data->worker =3D NULL; + idle_inject_set_duration(ii_dev, TICK_USEC, duration); + idle_inject_set_latency(ii_dev, UINT_MAX); + + return 0; +} + +static void remove_idle_injection(void) +{ + if (!powerclamp_data.clamping) + return; + + powerclamp_data.clamping =3D false; + idle_inject_stop(ii_dev); } =20 static int start_power_clamp(void) { - unsigned long cpu; + int ret; =20 - set_target_ratio =3D clamp(set_target_ratio, 0U, MAX_TARGET_RATIO - 1); /* prevent cpu hotplug */ cpus_read_lock(); =20 /* prefer BSP */ control_cpu =3D cpumask_first(cpu_online_mask); =20 - clamping =3D true; - schedule_delayed_work(&poll_pkg_cstate_work, 0); - - /* start one kthread worker per online cpu */ - for_each_online_cpu(cpu) { - start_power_clamp_worker(cpu); + ret =3D powerclamp_idle_injection_register(); + if (!ret) { + trigger_idle_injection(); + schedule_delayed_work(&poll_pkg_cstate_work, 0); } + cpus_read_unlock(); =20 - return 0; + return ret; } =20 static void end_power_clamp(void) { - int i; - - /* - * Block requeuing in all the kthread workers. They will flush and - * stop faster. - */ - clamping =3D false; - for_each_set_bit(i, cpu_clamping_mask, num_possible_cpus()) { - pr_debug("clamping worker for cpu %d alive, destroy\n", i); - stop_power_clamp_worker(i); + if (powerclamp_data.clamping) { + remove_idle_injection(); + idle_inject_unregister(ii_dev); } } =20 static int powerclamp_cpu_online(unsigned int cpu) { - if (clamping =3D=3D false) + if (!powerclamp_data.clamping) return 0; - start_power_clamp_worker(cpu); + /* prefer BSP as controlling CPU */ if (cpu =3D=3D 0) { control_cpu =3D 0; @@ -560,10 +526,6 @@ static int powerclamp_cpu_online(unsigned int cpu) =20 static int powerclamp_cpu_predown(unsigned int cpu) { - if (clamping =3D=3D false) - return 0; - - stop_power_clamp_worker(cpu); if (cpu !=3D control_cpu) return 0; =20 @@ -585,7 +547,7 @@ static int powerclamp_get_max_state(struct thermal_cool= ing_device *cdev, static int powerclamp_get_cur_state(struct thermal_cooling_device *cdev, unsigned long *state) { - if (true =3D=3D clamping) + if (powerclamp_data.clamping) *state =3D pkg_cstate_ratio_cur; else /* to save power, do not poll idle ratio while not clamping */ @@ -599,24 +561,30 @@ static int powerclamp_set_cur_state(struct thermal_co= oling_device *cdev, { int ret =3D 0; =20 + mutex_lock(&powerclamp_lock); + new_target_ratio =3D clamp(new_target_ratio, 0UL, - (unsigned long) (MAX_TARGET_RATIO-1)); - if (set_target_ratio =3D=3D 0 && new_target_ratio > 0) { + (unsigned long) (MAX_TARGET_RATIO - 1)); + if (READ_ONCE(set_target_ratio) =3D=3D 0 && new_target_ratio > 0) { pr_info("Start idle injection to reduce power\n"); - set_target_ratio =3D new_target_ratio; + WRITE_ONCE(set_target_ratio, new_target_ratio); ret =3D start_power_clamp(); + if (ret) + WRITE_ONCE(set_target_ratio, 0); goto exit_set; - } else if (set_target_ratio > 0 && new_target_ratio =3D=3D 0) { + } else if (READ_ONCE(set_target_ratio) > 0 && new_target_ratio =3D=3D 0) { pr_info("Stop forced idle injection\n"); end_power_clamp(); - set_target_ratio =3D 0; + WRITE_ONCE(set_target_ratio, 0); + WRITE_ONCE(target_ratio_updated, false); } else /* adjust currently running */ { - set_target_ratio =3D new_target_ratio; - /* make new set_target_ratio visible to other cpus */ - smp_mb(); + WRITE_ONCE(set_target_ratio, new_target_ratio); + WRITE_ONCE(target_ratio_updated, true); } =20 exit_set: + mutex_unlock(&powerclamp_lock); + return ret; } =20 @@ -686,14 +654,10 @@ static int __init powerclamp_init(void) { int retval; =20 - cpu_clamping_mask =3D bitmap_zalloc(num_possible_cpus(), GFP_KERNEL); - if (!cpu_clamping_mask) - return -ENOMEM; - /* probe cpu features and ids here */ retval =3D powerclamp_probe(); if (retval) - goto exit_free; + return retval; =20 /* set default limit, maybe adjusted during runtime based on feedback */ window_size =3D 2; @@ -702,53 +666,45 @@ static int __init powerclamp_init(void) powerclamp_cpu_online, powerclamp_cpu_predown); if (retval < 0) - goto exit_free; + return retval; =20 hp_state =3D retval; =20 - worker_data =3D alloc_percpu(struct powerclamp_worker_data); - if (!worker_data) { - retval =3D -ENOMEM; - goto exit_unregister; - } - cooling_dev =3D thermal_cooling_device_register("intel_powerclamp", NULL, - &powerclamp_cooling_ops); + &powerclamp_cooling_ops); if (IS_ERR(cooling_dev)) { retval =3D -ENODEV; - goto exit_free_thread; + goto exit_unregister; } =20 if (!duration) - duration =3D jiffies_to_msecs(DEFAULT_DURATION_JIFFIES); + duration =3D jiffies_to_usecs(DEFAULT_DURATION_JIFFIES); =20 powerclamp_create_debug_files(); =20 return 0; =20 -exit_free_thread: - free_percpu(worker_data); exit_unregister: cpuhp_remove_state_nocalls(hp_state); -exit_free: - bitmap_free(cpu_clamping_mask); return retval; } module_init(powerclamp_init); =20 static void __exit powerclamp_exit(void) { + mutex_lock(&powerclamp_lock); end_power_clamp(); + mutex_unlock(&powerclamp_lock); cpuhp_remove_state_nocalls(hp_state); - free_percpu(worker_data); thermal_cooling_device_unregister(cooling_dev); - bitmap_free(cpu_clamping_mask); =20 cancel_delayed_work_sync(&poll_pkg_cstate_work); debugfs_remove_recursive(debug_dir); } module_exit(powerclamp_exit); =20 +MODULE_IMPORT_NS(IDLE_INJECT); + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Arjan van de Ven "); MODULE_AUTHOR("Jacob Pan "); --=20 2.31.1 From nobody Fri Sep 19 02:17:27 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5B65DC433FE for ; Tue, 29 Nov 2022 23:34:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230159AbiK2Xe4 (ORCPT ); Tue, 29 Nov 2022 18:34:56 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35880 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229751AbiK2Xei (ORCPT ); Tue, 29 Nov 2022 18:34:38 -0500 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7D3D4F020; Tue, 29 Nov 2022 15:34:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1669764877; x=1701300877; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=UB+AsI11L5AiZWNtd3SOlQlASLZ7XjEIFfhbV5ckI+E=; b=HuYim+0Qa99/TKxsuDUMKGp6YI+43Yfi0/3UZ0N4okpSye7zuJrYfaDW 7xyyGTZTQBoFcroF10lbPzMkIBOs2PyPiZIyJk/r7OkVrRgnawfh13NGj E1b3zQN7Pjcg14KER4kmMaRZqezAMxrcqxkMQeBPFS7r533VuD6hMaZT1 kY7tW7r4QSsZ6vu/GH6l/3LrpAiw+LrG1TNQc0OvauTsF9ag+nIkmvfRb OntBwKkLMVHerBU8d/PQ5LXpOvN9AGiLRNfiGYd6ZeBY6nOwCaY8a8kPc ZsDEE2cJB2kmcjBe1rNogSWCeGwtVMUVvgpFm0OVLDB/y8CZjs6qR2Ig9 w==; X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="342178175" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="342178175" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 29 Nov 2022 15:34:36 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6500,9779,10546"; a="768617611" X-IronPort-AV: E=Sophos;i="5.96,204,1665471600"; d="scan'208";a="768617611" Received: from spandruv-desk.jf.intel.com ([10.54.75.8]) by orsmga004.jf.intel.com with ESMTP; 29 Nov 2022 15:34:35 -0800 From: Srinivas Pandruvada To: rafael@kernel.org Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, daniel.lezcano@linaro.org, rui.zhang@intel.com, amitk@kernel.org, Srinivas Pandruvada Subject: [PATCH v2 4/4] thermal/drivers/intel_cpu_idle_cooling: Introduce Intel cpu idle cooling driver Date: Tue, 29 Nov 2022 15:34:19 -0800 Message-Id: <20221129233419.4022830-5-srinivas.pandruvada@linux.intel.com> X-Mailer: git-send-email 2.31.1 In-Reply-To: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> References: <20221129233419.4022830-1-srinivas.pandruvada@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The cpu idle cooling is used to cool down a CPU by injecting idle cycles at runtime. The objective is similar to intel_powerclamp driver, which is used for system wide cooling by injecting idle on each CPU. This driver is modeled after drivers/thermal/cpuidle_cooling.c by reusing powercap/idle_inject framework. On each CPU online a thermal cooling device is registered. The minimum state of the cooling device is 0 and maximum is 100. When user space changes the current state to non zero, then register with idle inject framework and start idle inject. The default idle duration is 24 milli seconds, matching intel_powerclamp, which doesn't change based on the current state of cooling device. The runtime is changed based on the current state. Signed-off-by: Srinivas Pandruvada Reported-by: kernel test robot --- v2: - Removed callback arguments for idle_inject_register drivers/thermal/intel/Kconfig | 10 + drivers/thermal/intel/Makefile | 1 + .../thermal/intel/intel_cpu_idle_cooling.c | 261 ++++++++++++++++++ 3 files changed, 272 insertions(+) create mode 100644 drivers/thermal/intel/intel_cpu_idle_cooling.c diff --git a/drivers/thermal/intel/Kconfig b/drivers/thermal/intel/Kconfig index 6c2a95f41c81..8c88d6e18414 100644 --- a/drivers/thermal/intel/Kconfig +++ b/drivers/thermal/intel/Kconfig @@ -115,3 +115,13 @@ config INTEL_HFI_THERMAL These capabilities may change as a result of changes in the operating conditions of the system such power and thermal limits. If selected, the kernel relays updates in CPUs' capabilities to userspace. + +config INTEL_CPU_IDLE_COOLING + tristate "Intel CPU idle cooling device" + depends on IDLE_INJECT + help + This implements the CPU cooling mechanism through + idle injection. This will throttle the CPU by injecting + idle cycle. + Unlike Intel Power clamp driver, this driver provides + idle injection for each CPU. diff --git a/drivers/thermal/intel/Makefile b/drivers/thermal/intel/Makefile index 9a8d8054f316..8d5f7b5cf9b7 100644 --- a/drivers/thermal/intel/Makefile +++ b/drivers/thermal/intel/Makefile @@ -14,3 +14,4 @@ obj-$(CONFIG_INTEL_TCC_COOLING) +=3D intel_tcc_cooling.o obj-$(CONFIG_X86_THERMAL_VECTOR) +=3D therm_throt.o obj-$(CONFIG_INTEL_MENLOW) +=3D intel_menlow.o obj-$(CONFIG_INTEL_HFI_THERMAL) +=3D intel_hfi.o +obj-$(CONFIG_INTEL_CPU_IDLE_COOLING) +=3D intel_cpu_idle_cooling.o diff --git a/drivers/thermal/intel/intel_cpu_idle_cooling.c b/drivers/therm= al/intel/intel_cpu_idle_cooling.c new file mode 100644 index 000000000000..cdd62756cc3d --- /dev/null +++ b/drivers/thermal/intel/intel_cpu_idle_cooling.c @@ -0,0 +1,261 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Per CPU Idle injection cooling device implementation + * + * Copyright (c) 2022, Intel Corporation. + * All rights reserved. + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +/* Duration match with intel_powerclamp driver */ +#define IDLE_DURATION 24000 +#define IDLE_LATENCY UINT_MAX + +static int idle_duration_us =3D IDLE_DURATION; +static int idle_latency_us =3D IDLE_LATENCY; + +module_param(idle_duration_us, int, 0644); +MODULE_PARM_DESC(idle_duration_us, + "Idle duration in us."); + +module_param(idle_latency_us, int, 0644); +MODULE_PARM_DESC(idle_latency_us, + "Idle latency in us."); + +/** + * struct cpuidle_cooling - Per instance data for cooling device + * @cpu: CPU number for this cooling device + * @ii_dev: Idle inject core instance pointer + * @cdev: Thermal core cooling device instance + * @state: Current cooling device state + * + * Stores per instance cooling device state. + */ +struct cpuidle_cooling { + int cpu; + struct idle_inject_device *ii_dev; + struct thermal_cooling_device *cdev; + unsigned long state; +}; + +static DEFINE_PER_CPU(struct cpuidle_cooling, cooling_devs); +static cpumask_t cpuidle_cpu_mask; + +/* Used for module unload protection with idle injection operations */ +static DEFINE_MUTEX(idle_cooling_lock); + +static unsigned int cpuidle_cooling_runtime(unsigned int idle_duration_us, + unsigned long state) +{ + if (!state) + return 0; + + return ((idle_duration_us * 100) / state) - idle_duration_us; +} + +static int cpuidle_idle_injection_register(struct cpuidle_cooling *cooling= _dev) +{ + struct idle_inject_device *ii_dev; + + ii_dev =3D idle_inject_register((struct cpumask *)cpumask_of(cooling_dev-= >cpu)); + if (!ii_dev) { + /* + * It is busy as some other device claimed idle injection for this CPU + * Also it is possible that memory allocation failure. + */ + pr_err("idle_inject_register failed for cpu:%d\n", cooling_dev->cpu); + return -EAGAIN; + } + + idle_inject_set_duration(ii_dev, TICK_USEC, idle_duration_us); + idle_inject_set_latency(ii_dev, idle_latency_us); + + cooling_dev->ii_dev =3D ii_dev; + + return 0; +} + +static void cpuidle_idle_injection_unregister(struct cpuidle_cooling *cool= ing_dev) +{ + idle_inject_unregister(cooling_dev->ii_dev); +} + +static int cpuidle_cooling_get_max_state(struct thermal_cooling_device *cd= ev, + unsigned long *state) +{ + *state =3D 100; + + return 0; +} + +static int cpuidle_cooling_get_cur_state(struct thermal_cooling_device *cd= ev, + unsigned long *state) +{ + struct cpuidle_cooling *cooling_dev =3D cdev->devdata; + + *state =3D READ_ONCE(cooling_dev->state); + + return 0; +} + +static int cpuidle_cooling_set_cur_state(struct thermal_cooling_device *cd= ev, + unsigned long state) +{ + struct cpuidle_cooling *cooling_dev =3D cdev->devdata; + unsigned int runtime_us; + unsigned long curr_state; + int ret =3D 0; + + mutex_lock(&idle_cooling_lock); + + curr_state =3D READ_ONCE(cooling_dev->state); + + if (!curr_state && state > 0) { + /* + * This is the first time to start cooling, so register with + * idle injection framework. + */ + if (!cooling_dev->ii_dev) { + ret =3D cpuidle_idle_injection_register(cooling_dev); + if (ret) + goto unlock_set_state; + } + + runtime_us =3D cpuidle_cooling_runtime(idle_duration_us, state); + + idle_inject_set_duration(cooling_dev->ii_dev, runtime_us, idle_duration_= us); + idle_inject_start(cooling_dev->ii_dev); + } else if (curr_state > 0 && state) { + /* Simply update runtime */ + runtime_us =3D cpuidle_cooling_runtime(idle_duration_us, state); + idle_inject_set_duration(cooling_dev->ii_dev, runtime_us, idle_duration_= us); + } else if (curr_state > 0 && !state) { + idle_inject_stop(cooling_dev->ii_dev); + cpuidle_idle_injection_unregister(cooling_dev); + cooling_dev->ii_dev =3D NULL; + } + + WRITE_ONCE(cooling_dev->state, state); + +unlock_set_state: + mutex_unlock(&idle_cooling_lock); + + return ret; +} + +/** + * cpuidle_cooling_ops - thermal cooling device ops + */ +static struct thermal_cooling_device_ops cpuidle_cooling_ops =3D { + .get_max_state =3D cpuidle_cooling_get_max_state, + .get_cur_state =3D cpuidle_cooling_get_cur_state, + .set_cur_state =3D cpuidle_cooling_set_cur_state, +}; + +static int cpuidle_cooling_register(int cpu) +{ + struct cpuidle_cooling *cooling_dev =3D &per_cpu(cooling_devs, cpu); + struct thermal_cooling_device *cdev; + char name[14]; /* storage for cpuidle-XXXX */ + int ret =3D 0; + + mutex_lock(&idle_cooling_lock); + + snprintf(name, sizeof(name), "cpuidle-%d", cpu); + cdev =3D thermal_cooling_device_register(name, cooling_dev, &cpuidle_cool= ing_ops); + if (IS_ERR(cdev)) { + ret =3D PTR_ERR(cdev); + goto unlock_register; + } + + cooling_dev->cdev =3D cdev; + cpumask_set_cpu(cpu, &cpuidle_cpu_mask); + cooling_dev->cpu =3D cpu; + +unlock_register: + mutex_unlock(&idle_cooling_lock); + + return ret; +} + +static void cpuidle_cooling_unregister(int cpu) +{ + struct cpuidle_cooling *cooling_dev =3D &per_cpu(cooling_devs, cpu); + + mutex_lock(&idle_cooling_lock); + + if (cooling_dev->state) { + idle_inject_stop(cooling_dev->ii_dev); + cpuidle_idle_injection_unregister(cooling_dev); + } + + thermal_cooling_device_unregister(cooling_dev->cdev); + cooling_dev->state =3D 0; + + mutex_unlock(&idle_cooling_lock); +} + +static int cpuidle_cooling_cpu_online(unsigned int cpu) +{ + cpuidle_cooling_register(cpu); + + return 0; +} + +static int cpuidle_cooling_cpu_offline(unsigned int cpu) +{ + cpuidle_cooling_unregister(cpu); + + return 0; +} + +static enum cpuhp_state cpuidle_cooling_hp_state __read_mostly; + +static const struct x86_cpu_id intel_cpuidle_cooling_ids[] __initconst =3D= { + X86_MATCH_VENDOR_FEATURE(INTEL, X86_FEATURE_MWAIT, NULL), + {} +}; +MODULE_DEVICE_TABLE(x86cpu, intel_cpuidle_cooling_ids); + +static int __init cpuidle_cooling_init(void) +{ + int ret; + + if (!x86_match_cpu(intel_cpuidle_cooling_ids)) + return -ENODEV; + + ret =3D cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, + "thermal/cpuidle_cooling:online", + cpuidle_cooling_cpu_online, + cpuidle_cooling_cpu_offline); + if (ret < 0) + return ret; + + cpuidle_cooling_hp_state =3D ret; + + return 0; +} +module_init(cpuidle_cooling_init) + +static void __exit cpuidle_cooling_exit(void) +{ + cpuhp_remove_state(cpuidle_cooling_hp_state); +} +module_exit(cpuidle_cooling_exit) + +MODULE_IMPORT_NS(IDLE_INJECT); + +MODULE_LICENSE("GPL"); --=20 2.31.1