From patchwork Wed Oct 22 13:57:49 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Lezcano X-Patchwork-Id: 39303 Return-Path: X-Original-To: linaro@patches.linaro.org Delivered-To: linaro@patches.linaro.org Received: from mail-la0-f69.google.com (mail-la0-f69.google.com [209.85.215.69]) by ip-10-151-82-157.ec2.internal (Postfix) with ESMTPS id 95E43202DB for ; Wed, 22 Oct 2014 13:58:12 +0000 (UTC) Received: by mail-la0-f69.google.com with SMTP id q1sf2056588lam.0 for ; Wed, 22 Oct 2014 06:58:11 -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:x-original-sender :x-original-authentication-results:precedence:mailing-list:list-id :list-post:list-help:list-archive:list-unsubscribe; bh=62OjMvCEIjzrU4BXdGBBGSLgbFDNNO4wXy0mpLq+pgM=; b=Jeog7RoWQXquB08h7sN85fkM1Juy4WRxexGQst20REWBsxvy9F2IKPcuF3pwFHoFnu xr/ka182LVRMbUJ1nzpOq0BobV0rGF+SCuoxmQVfljLNPwXBcnU4BWwucG2uPytu9/Fx n85jagXtXMpuBP1a1i4Yz8kZ9nw6UmFBdbDImwT1IqsoYytqZPuPd1jtYr9kfwh31kAq wFEP7s2G0KVKmYsho9W+IcVpmTD2n9WClsKVfMyAD6Qg4YNQQgvST3jikc7z0LKUotLo fxjfMLhUm1hjxfGORbBn5p0I5oJRM7hgLimmCC4hXzER54QTFnXlDrf1jYET2IQpHFkT osow== X-Gm-Message-State: ALoCoQlUhJ7OTCgmic350ItcOur0DNtgvGoBZ5DMyUk+L3HIelQ4gqEYSdnw9P0CC27IFN3nEuja X-Received: by 10.152.29.134 with SMTP id k6mr20603lah.9.1413986291118; Wed, 22 Oct 2014 06:58:11 -0700 (PDT) MIME-Version: 1.0 X-BeenThere: patchwork-forward@linaro.org Received: by 10.152.27.129 with SMTP id t1ls131035lag.53.gmail; Wed, 22 Oct 2014 06:58:11 -0700 (PDT) X-Received: by 10.152.179.164 with SMTP id dh4mr41896523lac.55.1413986290692; Wed, 22 Oct 2014 06:58:10 -0700 (PDT) Received: from mail-lb0-f171.google.com (mail-lb0-f171.google.com. [209.85.217.171]) by mx.google.com with ESMTPS id ua7si23406128lac.103.2014.10.22.06.58.10 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 22 Oct 2014 06:58:10 -0700 (PDT) Received-SPF: pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 209.85.217.171 as permitted sender) client-ip=209.85.217.171; Received: by mail-lb0-f171.google.com with SMTP id z12so2891595lbi.16 for ; Wed, 22 Oct 2014 06:58:10 -0700 (PDT) X-Received: by 10.112.221.197 with SMTP id qg5mr41492472lbc.32.1413986290481; Wed, 22 Oct 2014 06:58:10 -0700 (PDT) X-Forwarded-To: patchwork-forward@linaro.org X-Forwarded-For: patch@linaro.org patchwork-forward@linaro.org Delivered-To: patches@linaro.org Received: by 10.112.84.229 with SMTP id c5csp76299lbz; Wed, 22 Oct 2014 06:58:10 -0700 (PDT) X-Received: by 10.180.98.170 with SMTP id ej10mr5799016wib.17.1413986289818; Wed, 22 Oct 2014 06:58:09 -0700 (PDT) Received: from mail-wi0-f169.google.com (mail-wi0-f169.google.com. [209.85.212.169]) by mx.google.com with ESMTPS id i2si1868301wiy.55.2014.10.22.06.58.09 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 22 Oct 2014 06:58:09 -0700 (PDT) Received-SPF: pass (google.com: domain of daniel.lezcano@linaro.org designates 209.85.212.169 as permitted sender) client-ip=209.85.212.169; Received: by mail-wi0-f169.google.com with SMTP id r20so1354715wiv.0 for ; Wed, 22 Oct 2014 06:58:09 -0700 (PDT) X-Received: by 10.180.103.233 with SMTP id fz9mr36843412wib.80.1413986289581; Wed, 22 Oct 2014 06:58:09 -0700 (PDT) Received: from localhost.localdomain (AToulouse-656-1-959-39.w90-50.abo.wanadoo.fr. [90.50.216.39]) by mx.google.com with ESMTPSA id f7sm2030217wiz.13.2014.10.22.06.58.07 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 22 Oct 2014 06:58:08 -0700 (PDT) From: Daniel Lezcano To: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org Cc: axboe@kernel.dk, rafael.j.wysocki@intel.com, mingo@kernel.org, peterz@infradead.org, preeti@linux.vnet.ibm.com, Morten.Rasmussen@arm.com, mturquette@linaro.org, tuukka.tikkanen@linaro.org, nicolas.pitre@linaro.org, patches@linaro.org Subject: [RFD PATCH 06/10] sched: idle: Add io latency information for the next event Date: Wed, 22 Oct 2014 15:57:49 +0200 Message-Id: <1413986273-28522-7-git-send-email-daniel.lezcano@linaro.org> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1413986273-28522-1-git-send-email-daniel.lezcano@linaro.org> References: <1413986273-28522-1-git-send-email-daniel.lezcano@linaro.org> X-Removed-Original-Auth: Dkim didn't pass. X-Original-Sender: daniel.lezcano@linaro.org X-Original-Authentication-Results: mx.google.com; spf=pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 209.85.217.171 as permitted sender) smtp.mail=patch+caf_=patchwork-forward=linaro.org@linaro.org Precedence: list Mailing-list: list patchwork-forward@linaro.org; contact patchwork-forward+owners@linaro.org List-ID: X-Google-Group-Id: 836684582541 List-Post: , List-Help: , List-Archive: List-Unsubscribe: , As we want to improve the sleep duration estimation, the IO latency expected duration is passed to the cpuidle framework. The governors will have to deal with if they are interested in this information. Signed-off-by: Daniel Lezcano --- drivers/cpuidle/cpuidle.c | 4 ++-- drivers/cpuidle/governors/ladder.c | 3 ++- drivers/cpuidle/governors/menu.c | 5 +++-- include/linux/cpuidle.h | 12 +++++++++--- kernel/sched/idle.c | 27 ++++++++++++++++----------- 5 files changed, 32 insertions(+), 19 deletions(-) diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 1abd5a0..bf42e17 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -155,7 +155,7 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, * Returns the index of the idle state. */ int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event) + struct cpuidle_times *times) { if (off || !initialized) return -ENODEV; @@ -166,7 +166,7 @@ int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, if (unlikely(use_deepest_state)) return cpuidle_find_deepest_state(drv, dev); - return cpuidle_curr_governor->select(drv, dev, latency_req, next_event); + return cpuidle_curr_governor->select(drv, dev, times); } /** diff --git a/drivers/cpuidle/governors/ladder.c b/drivers/cpuidle/governors/ladder.c index 17381c3..a84993c 100644 --- a/drivers/cpuidle/governors/ladder.c +++ b/drivers/cpuidle/governors/ladder.c @@ -65,10 +65,11 @@ static inline void ladder_do_selection(struct ladder_device *ldev, */ static int ladder_select_state(struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event) + struct cpuidle_times *times) { struct ladder_device *ldev = &__get_cpu_var(ladder_devices); struct ladder_device_state *last_state; + int latency_req = times->latency_req; int last_residency, last_idx = ldev->last_state_idx; last_state = &ldev->states[last_idx]; diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index 9da11ce..88382d5 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -284,9 +284,10 @@ again: * @dev: the CPU */ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event) + struct cpuidle_times *times) { struct menu_device *data = &__get_cpu_var(menu_devices); + int latency_req = times->latency_req; int i; unsigned int interactivity_req; unsigned long nr_iowaiters, cpu_load; @@ -299,7 +300,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, data->last_state_idx = CPUIDLE_DRIVER_STATE_START - 1; /* determine the expected residency time, round up */ - data->next_timer_us = next_event; + data->next_timer_us = times->next_timer_event; get_iowait_load(&nr_iowaiters, &cpu_load); data->bucket = which_bucket(data->next_timer_us, nr_iowaiters); diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index b379ae5..e99823f 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -83,6 +83,12 @@ struct cpuidle_device { #endif }; +struct cpuidle_times { + unsigned int latency_req; + unsigned int next_timer_event; + unsigned int next_io_event; +}; + DECLARE_PER_CPU(struct cpuidle_device *, cpuidle_devices); DECLARE_PER_CPU(struct cpuidle_device, cpuidle_dev); @@ -123,7 +129,7 @@ extern void disable_cpuidle(void); extern int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event); + struct cpuidle_times *times); extern int cpuidle_enter(struct cpuidle_driver *drv, struct cpuidle_device *dev, int index); extern void cpuidle_reflect(struct cpuidle_device *dev, int index); @@ -152,7 +158,7 @@ extern struct cpuidle_driver *cpuidle_get_cpu_driver(struct cpuidle_device *dev) static inline void disable_cpuidle(void) { } static inline int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event) + struct cpuidle_times *times) {return -ENODEV; } static inline int cpuidle_enter(struct cpuidle_driver *drv, struct cpuidle_device *dev, int index) @@ -208,7 +214,7 @@ struct cpuidle_governor { int (*select) (struct cpuidle_driver *drv, struct cpuidle_device *dev, - int latency_req, int next_event); + struct cpuidle_times *times); void (*reflect) (struct cpuidle_device *dev, int index); struct module *owner; diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 7b6a148..6057020 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -4,7 +4,7 @@ #include #include #include -#include +#include #include #include #include @@ -14,6 +14,7 @@ #include #include "sched.h" +#include "io_latency.h" static int __read_mostly cpu_idle_force_poll; @@ -79,9 +80,9 @@ static void cpuidle_idle_call(void) { struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); - struct timespec t; - int latency_req = pm_qos_request(PM_QOS_CPU_DMA_LATENCY); - unsigned int broadcast; + struct cpuidle_times times; + int next_state, entered_state; + bool broadcast; /* * Check if the idle task must be rescheduled. If it is the @@ -105,25 +106,29 @@ static void cpuidle_idle_call(void) */ rcu_idle_enter(); + times.latency_req = pm_qos_request(PM_QOS_CPU_DMA_LATENCY); /* * The latency requirement does not allow any latency, jump to - * the default idle function + * the default idle function without entering the cpuidle code */ - if (latency_req == 0) + if (times.latency_req == 0) goto use_default; - t = ktime_to_timespec(tick_nohz_get_sleep_length()); + /* + * Retrieve the next timer event + */ + times.next_timer_event = ktime_to_us(tick_nohz_get_sleep_length()); - /* - * The next timer event for this in us + /* + * Retrieve the next IO guessed event */ - next_event = t.tv_sec * USEC_PER_SEC + t.tv_nsec / NSEC_PER_USEC; + times.next_io_event = io_latency_get_sleep_length(this_rq()); /* * Ask the cpuidle framework to choose a convenient idle state. * Fall back to the default arch idle method on errors. */ - next_state = cpuidle_select(drv, dev, latency_req, next_event); + next_state = cpuidle_select(drv, dev, ×); if (next_state < 0) { use_default: /*