From: "Rafael J. Wysocki" <rjw@rjwysocki.net> To: Peter Zijlstra <peterz@infradead.org>, Linux PM <linux-pm@vger.kernel.org> Cc: Frederic Weisbecker <fweisbec@gmail.com>, Thomas Gleixner <tglx@linutronix.de>, Paul McKenney <paulmck@linux.vnet.ibm.com>, Thomas Ilsche <thomas.ilsche@tu-dresden.de>, Doug Smythies <dsmythies@telus.net>, Rik van Riel <riel@surriel.com>, Aubrey Li <aubrey.li@linux.intel.com>, Mike Galbraith <mgalbraith@suse.de>, LKML <linux-kernel@vger.kernel.org> Subject: [RFT][PATCH v7 2/8] sched: idle: Do not stop the tick upfront in the idle loop Date: Tue, 20 Mar 2018 16:15:24 +0100 Message-ID: <2802114.rPO26tdOrH@aspire.rjw.lan> (raw) In-Reply-To: <2390019.oHdSGtR3EE@aspire.rjw.lan> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Push the decision whether or not to stop the tick somewhat deeper into the idle loop. Stopping the tick upfront leads to unpleasant outcomes in case the idle governor doesn't agree with the nohz code on the duration of the upcoming idle period. Specifically, if the tick has been stopped and the idle governor predicts short idle, the situation is bad regardless of whether or not the prediction is accurate. If it is accurate, the tick has been stopped unnecessarily which means excessive overhead. If it is not accurate, the CPU is likely to spend too much time in the (shallow, because short idle has been predicted) idle state selected by the governor [1]. As the first step towards addressing this problem, change the code to make the tick stopping decision inside of the loop in do_idle(). In particular, do not stop the tick in the cpu_idle_poll() code path. Also don't do that in tick_nohz_irq_exit() which doesn't really have enough information on whether or not to stop the tick. Link: https://marc.info/?l=linux-pm&m=150116085925208&w=2 # [1] Link: https://tu-dresden.de/zih/forschung/ressourcen/dateien/projekte/haec/powernightmares.pdf Suggested-by: Frederic Weisbecker <frederic@kernel.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> --- include/linux/tick.h | 2 ++ kernel/sched/idle.c | 9 ++++++--- kernel/time/tick-sched.c | 26 ++++++++++++++++++-------- 3 files changed, 26 insertions(+), 11 deletions(-) Index: linux-pm/kernel/sched/idle.c =================================================================== --- linux-pm.orig/kernel/sched/idle.c +++ linux-pm/kernel/sched/idle.c @@ -221,13 +221,13 @@ static void do_idle(void) __current_set_polling(); tick_nohz_idle_enter(); - tick_nohz_idle_stop_tick_protected(); while (!need_resched()) { check_pgt_cache(); rmb(); if (cpu_is_offline(cpu)) { + tick_nohz_idle_stop_tick_protected(); cpuhp_report_idle_dead(); arch_cpu_idle_dead(); } @@ -241,10 +241,13 @@ static void do_idle(void) * broadcast device expired for us, we don't want to go deep * idle as we know that the IPI is going to arrive right away. */ - if (cpu_idle_force_poll || tick_check_broadcast_expired()) + if (cpu_idle_force_poll || tick_check_broadcast_expired()) { + tick_nohz_idle_restart_tick(); cpu_idle_poll(); - else + } else { + tick_nohz_idle_stop_tick(); cpuidle_idle_call(); + } arch_cpu_idle_exit(); } Index: linux-pm/kernel/time/tick-sched.c =================================================================== --- linux-pm.orig/kernel/time/tick-sched.c +++ linux-pm/kernel/time/tick-sched.c @@ -984,12 +984,10 @@ void tick_nohz_irq_exit(void) { struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched); - if (ts->inidle) { + if (ts->inidle) tick_nohz_start_idle(ts); - __tick_nohz_idle_stop_tick(ts); - } else { + else tick_nohz_full_update_tick(ts); - } } /** @@ -1050,6 +1048,20 @@ static void tick_nohz_account_idle_ticks #endif } +static void __tick_nohz_idle_restart_tick(struct tick_sched *ts, ktime_t now) +{ + tick_nohz_restart_sched_tick(ts, now); + tick_nohz_account_idle_ticks(ts); +} + +void tick_nohz_idle_restart_tick(void) +{ + struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched); + + if (ts->tick_stopped) + __tick_nohz_idle_restart_tick(ts, ktime_get()); +} + /** * tick_nohz_idle_exit - restart the idle tick from the idle task * @@ -1074,10 +1086,8 @@ void tick_nohz_idle_exit(void) if (ts->idle_active) tick_nohz_stop_idle(ts, now); - if (ts->tick_stopped) { - tick_nohz_restart_sched_tick(ts, now); - tick_nohz_account_idle_ticks(ts); - } + if (ts->tick_stopped) + __tick_nohz_idle_restart_tick(ts, now); local_irq_enable(); } Index: linux-pm/include/linux/tick.h =================================================================== --- linux-pm.orig/include/linux/tick.h +++ linux-pm/include/linux/tick.h @@ -115,6 +115,7 @@ enum tick_dep_bits { extern bool tick_nohz_enabled; extern int tick_nohz_tick_stopped(void); extern void tick_nohz_idle_stop_tick(void); +extern void tick_nohz_idle_restart_tick(void); extern void tick_nohz_idle_enter(void); extern void tick_nohz_idle_exit(void); extern void tick_nohz_irq_exit(void); @@ -135,6 +136,7 @@ static inline void tick_nohz_idle_stop_t #define tick_nohz_enabled (0) static inline int tick_nohz_tick_stopped(void) { return 0; } static inline void tick_nohz_idle_stop_tick(void) { } +static inline void tick_nohz_idle_restart_tick(void) { } static inline void tick_nohz_idle_enter(void) { } static inline void tick_nohz_idle_exit(void) { }
next prev parent reply index Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-03-20 15:12 [RFT][PATCH v7 0/8] sched/cpuidle: Idle loop rework Rafael J. Wysocki 2018-03-20 15:13 ` [RFT][PATCH v7 1/8] time: tick-sched: Reorganize idle tick management code Rafael J. Wysocki 2018-03-20 15:15 ` Rafael J. Wysocki [this message] 2018-03-20 15:15 ` [RFT][PATCH v7 3/8] sched: idle: Do not stop the tick upfront in the idle loop Rafael J. Wysocki 2018-03-20 18:00 ` [Correction][RFT][PATCH v7 3/8] sched: idle: Do not stop the tick before cpuidle_idle_call() Rafael J. Wysocki 2018-03-20 15:16 ` [RFT][PATCH v7 4/8] jiffies: Introduce USER_TICK_USEC and redefine TICK_USEC Rafael J. Wysocki 2018-03-20 15:45 ` [RFT][PATCH v7 5/8] cpuidle: Return nohz hint from cpuidle_select() Rafael J. Wysocki 2018-03-21 6:48 ` [RFT][PATCH v7.1 " Rafael J. Wysocki 2018-03-21 11:52 ` Rafael J. Wysocki 2018-03-21 13:03 ` [RFT][PATCH v7.2 " Rafael J. Wysocki 2018-03-21 14:36 ` [RFT][PATCH v7 " Rafael J. Wysocki 2018-03-21 17:59 ` Thomas Ilsche 2018-03-21 22:15 ` Rafael J. Wysocki 2018-03-22 13:18 ` Thomas Ilsche 2018-03-22 17:23 ` Rafael J. Wysocki 2018-03-22 6:24 ` Doug Smythies 2018-03-22 15:41 ` Doug Smythies 2018-03-22 17:21 ` Rafael J. Wysocki 2018-03-21 18:23 ` Doug Smythies 2018-03-22 17:40 ` [RFT][PATCH v7.3 " Rafael J. Wysocki 2018-03-28 9:14 ` Thomas Ilsche 2018-03-30 9:39 ` Rafael J. Wysocki 2018-04-10 15:22 ` Thomas Ilsche 2018-03-22 20:46 ` Doug Smythies 2018-03-20 15:45 ` [RFT][PATCH v7 6/8] sched: idle: Select idle state before stopping the tick Rafael J. Wysocki 2018-03-27 21:50 ` Thomas Ilsche 2018-03-27 22:10 ` Rafael J. Wysocki 2018-03-28 8:13 ` Rafael J. Wysocki 2018-03-28 8:38 ` Thomas Ilsche 2018-03-28 10:37 ` Rafael J. Wysocki 2018-03-28 10:56 ` Rafael J. Wysocki 2018-03-28 15:15 ` Thomas Ilsche 2018-03-28 20:41 ` Doug Smythies 2018-03-28 23:11 ` Rafael J. Wysocki 2018-03-20 15:46 ` [RFT][PATCH v7 7/8] cpuidle: menu: Refine idle state selection for running tick Rafael J. Wysocki 2018-03-20 15:47 ` [RFT][PATCH v7 8/8] cpuidle: menu: Avoid selecting shallow states with stopped tick Rafael J. Wysocki 2018-03-20 17:52 ` [RFT][PATCH v7 3/8] sched: idle: Do not stop the tick upfront in the idle loop Doug Smythies 2018-03-20 18:01 ` Rafael J. Wysocki 2018-03-21 12:31 ` [RFT][PATCH v7 0/8] sched/cpuidle: Idle loop rework Rik van Riel 2018-03-21 13:55 ` Rafael J. Wysocki 2018-03-21 14:53 ` Rik van Riel
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=2802114.rPO26tdOrH@aspire.rjw.lan \ --to=rjw@rjwysocki.net \ --cc=aubrey.li@linux.intel.com \ --cc=dsmythies@telus.net \ --cc=fweisbec@gmail.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-pm@vger.kernel.org \ --cc=mgalbraith@suse.de \ --cc=paulmck@linux.vnet.ibm.com \ --cc=peterz@infradead.org \ --cc=riel@surriel.com \ --cc=tglx@linutronix.de \ --cc=thomas.ilsche@tu-dresden.de \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
LKML Archive on lore.kernel.org Archives are clonable: git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git git clone --mirror https://lore.kernel.org/lkml/9 lkml/git/9.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \ linux-kernel@vger.kernel.org public-inbox-index lkml Example config snippet for mirrors Newsgroup available over NNTP: nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel AGPL code for this site: git clone https://public-inbox.org/public-inbox.git