Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755630Ab1BRMrR (ORCPT ); Fri, 18 Feb 2011 07:47:17 -0500 Received: from e28smtp04.in.ibm.com ([122.248.162.4]:45461 "EHLO e28smtp04.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752475Ab1BRMrO (ORCPT ); Fri, 18 Feb 2011 07:47:14 -0500 Message-ID: <4D5E6A54.4090901@linux.vnet.ibm.com> Date: Fri, 18 Feb 2011 18:17:16 +0530 From: Ciju Rajan K User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.13) Gecko/20101209 Fedora/3.1.7-0.35.b3pre.fc14 Thunderbird/3.1.7 MIME-Version: 1.0 To: linux kernel mailing list CC: Ciju Rajan K , Peter Zijlstra , Bharata B Rao , Ingo Molnar , Srivatsa Vaddagiri , Satoru Takeuchi Subject: [PATCH 2/2 v3.0]sched: Updating the sched-stat documentation References: <4D3F3595.4020607@linux.vnet.ibm.com> <4D5E6970.1000105@linux.vnet.ibm.com> In-Reply-To: <4D5E6970.1000105@linux.vnet.ibm.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 9275 Lines: 193 From: Ciju Rajan K Date: Fri, 18 Feb 2011 16:29:14 +0530 Subject: [PATCH 2/2 v3.0] sched: Updating the sched-stat documentation Some of the unused fields are removed from /proc/schedstat. This is the documentation changes reflecting the same. Signed-off-by: Ciju Rajan K --- Documentation/scheduler/sched-stats.txt | 144 ++++++++++++------------------- 1 files changed, 55 insertions(+), 89 deletions(-) diff --git a/Documentation/scheduler/sched-stats.txt b/Documentation/scheduler/sched-stats.txt index 1cd5d51..de47562 100644 --- a/Documentation/scheduler/sched-stats.txt +++ b/Documentation/scheduler/sched-stats.txt @@ -1,3 +1,4 @@ +Version 16 of schedstats removed some of the unused fields. Version 15 of schedstats dropped counters for some sched_yield: yld_exp_empty, yld_act_empty and yld_both_empty. Otherwise, it is identical to version 14. @@ -30,112 +31,77 @@ Note that any such script will necessarily be version-specific, as the main reason to change versions is changes in the output format. For those wishing to write their own scripts, the fields are described here. +The first two fields of /proc/schedstat indicates the version (current +version is 16) and jiffies values. The following values are from +cpu & domain statistics. + CPU statistics -------------- -cpu 1 2 3 4 5 6 7 8 9 - -First field is a sched_yield() statistic: - 1) # of times sched_yield() was called - -Next three are schedule() statistics: - 2) # of times we switched to the expired queue and reused it - 3) # of times schedule() was called - 4) # of times schedule() left the processor idle - -Next two are try_to_wake_up() statistics: - 5) # of times try_to_wake_up() was called - 6) # of times try_to_wake_up() was called to wake up the local cpu - -Next three are statistics describing scheduling latency: - 7) sum of all time spent running by tasks on this processor (in jiffies) - 8) sum of all time spent waiting to run by tasks on this processor (in - jiffies) - 9) # of timeslices run on this cpu - +The format is like this: + +cpu 1 2 3 4 5 6 7 8 + + 1) # of times sched_yield() was called on this CPU + 2) # of times scheduler runs on this CPU + 3) # of times scheduler picks idle task as next task on this CPU + 4) # of times try_to_wake_up() is run on this CPU + (Number of times task wakeup is attempted from this CPU) + 5) # of times try_to_wake_up() wakes up a task on the same CPU + (local wakeup) + 6) Time(ns) for which tasks have run on this CPU + 7) Time(ns) for which tasks on this CPU's runqueue have waited + before getting to run on the CPU + 8) # of tasks that have run on this CPU Domain statistics ----------------- -One of these is produced per domain for each cpu described. (Note that if -CONFIG_SMP is not defined, *no* domains are utilized and these lines -will not appear in the output.) +One of these is produced per domain for each cpu described. +(Note that if CONFIG_SMP is not defined, *no* domains are utilized +and these lines will not appear in the output.) -domain 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 +domain 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 The first field is a bit mask indicating what cpus this domain operates over. The next 24 are a variety of load_balance() statistics in grouped into types of idleness (idle, busy, and newly idle): - 1) # of times in this domain load_balance() was called when the - cpu was idle - 2) # of times in this domain load_balance() checked but found - the load did not require balancing when the cpu was idle - 3) # of times in this domain load_balance() tried to move one or - more tasks and failed, when the cpu was idle - 4) sum of imbalances discovered (if any) with each call to - load_balance() in this domain when the cpu was idle - 5) # of times in this domain pull_task() was called when the cpu - was idle - 6) # of times in this domain pull_task() was called even though - the target task was cache-hot when idle - 7) # of times in this domain load_balance() was called but did - not find a busier queue while the cpu was idle - 8) # of times in this domain a busier queue was found while the - cpu was idle but no busier group was found - - 9) # of times in this domain load_balance() was called when the - cpu was busy - 10) # of times in this domain load_balance() checked but found the - load did not require balancing when busy - 11) # of times in this domain load_balance() tried to move one or - more tasks and failed, when the cpu was busy - 12) sum of imbalances discovered (if any) with each call to - load_balance() in this domain when the cpu was busy - 13) # of times in this domain pull_task() was called when busy - 14) # of times in this domain pull_task() was called even though the - target task was cache-hot when busy - 15) # of times in this domain load_balance() was called but did not - find a busier queue while the cpu was busy - 16) # of times in this domain a busier queue was found while the cpu - was busy but no busier group was found - - 17) # of times in this domain load_balance() was called when the - cpu was just becoming idle - 18) # of times in this domain load_balance() checked but found the - load did not require balancing when the cpu was just becoming idle - 19) # of times in this domain load_balance() tried to move one or more - tasks and failed, when the cpu was just becoming idle - 20) sum of imbalances discovered (if any) with each call to - load_balance() in this domain when the cpu was just becoming idle - 21) # of times in this domain pull_task() was called when newly idle - 22) # of times in this domain pull_task() was called even though the - target task was cache-hot when just becoming idle - 23) # of times in this domain load_balance() was called but did not - find a busier queue while the cpu was just becoming idle - 24) # of times in this domain a busier queue was found while the cpu - was just becoming idle but no busier group was found - +CPU_NOT_IDLE: Load balancer is being run on a CPU when it is + not in IDLE state (busy times) +CPU_NEWLY_IDLE: Load balancer is being run on a CPU which is + about to enter IDLE state + +There are eight stats available for each of the above three states: + - # of times in this domain load_balance() was called + - # of times in this domain load_balance() checked but found + the load did not require balancing + - # of times in this domain load_balance() tried to move one or + more tasks and failed + - sum of imbalances discovered (if any) with each call to + load_balance() in this domain + - # of times in this domain pull_task() was called + - # of times in this domain pull_task() was called even though + the target task was cache-hot + - # of times in this domain load_balance() was called but did + not find a busier queue + - # of times in this domain a busier queue was found but no + busier group was found + + The first 1-8) fields are the stats when cpu was idle (CPU_IDLE), + the next 9-15) fields are the stats when cpu was busy (CPU_NOT_IDLE), + and the next 16-24) fields are the stats when cpu was just + becoming idle (CPU_NEWLY_IDLE) + Next three are active_load_balance() statistics: 25) # of times active_load_balance() was called 26) # of times active_load_balance() tried to move a task and failed 27) # of times active_load_balance() successfully moved a task - Next three are sched_balance_exec() statistics: - 28) sbe_cnt is not used - 29) sbe_balanced is not used - 30) sbe_pushed is not used - - Next three are sched_balance_fork() statistics: - 31) sbf_cnt is not used - 32) sbf_balanced is not used - 33) sbf_pushed is not used - - Next three are try_to_wake_up() statistics: - 34) # of times in this domain try_to_wake_up() awoke a task that - last ran on a different cpu in this domain - 35) # of times in this domain try_to_wake_up() moved a task to the - waking cpu because it was cache-cold on its own cpu anyway - 36) # of times in this domain try_to_wake_up() started passive balancing + Next two are try_to_wake_up() statistics: + 28) # of times in this domain try_to_wake_up() awoke a task that + last ran on a different cpu in this domain + 29) # of times in this domain try_to_wake_up() moved a task to the + waking cpu because it was cache-cold on its own cpu anyway /proc//schedstat ---------------- -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/