linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
To: linux-kernel@vger.kernel.org
Cc: Fernando Luis Vazquez Cao <fernando_b1@lab.ntt.co.jp>,
	Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Arjan van de Ven <arjan@linux.intel.com>,
	Oleg Nesterov <oleg@redhat.com>,
	Preeti U Murthy <preeti@linux.vnet.ibm.com>,
	Denys Vlasenko <vda.linux@googlemail.com>
Subject: [PATCH 5/8] cputime, ia64: update iowait accounting
Date: Thu, 26 Jun 2014 18:13:06 +0900	[thread overview]
Message-ID: <53ABE422.1000000@jp.fujitsu.com> (raw)
In-Reply-To: <53ABE28F.6010402@jp.fujitsu.com>

Using VIRT_CPU_ACCOUNTING, ia64 utilize "timestamp at end of iowait"
like s390.

Not-Tested-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
---
 arch/ia64/include/asm/cputime.h |    2 +
 arch/ia64/kernel/time.c         |   43 ++++++++++++++++++++++++++++++++++++++-
 2 files changed, 44 insertions(+), 1 deletions(-)

diff --git a/arch/ia64/include/asm/cputime.h b/arch/ia64/include/asm/cputime.h
index e2d3f5b..6089cbd 100644
--- a/arch/ia64/include/asm/cputime.h
+++ b/arch/ia64/include/asm/cputime.h
@@ -24,6 +24,8 @@
 # include <asm/processor.h>
 # include <asm-generic/cputime_nsecs.h>
 extern void arch_vtime_task_switch(struct task_struct *tsk);
+extern void ia64_record_iowait_exit(int cpu);
+# define arch_record_iowait_exit(cpu) ia64_record_iowait_exit(cpu)
 #endif /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
 
 #endif /* __IA64_CPUTIME_H */
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 71c52bc..37c907d 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -138,9 +138,50 @@ void vtime_account_system(struct task_struct *tsk)
 }
 EXPORT_SYMBOL_GPL(vtime_account_system);
 
+/*
+ * Nasty trick to account iowait time as defined
+ */
+DEFINE_PER_CPU(u64, ia64_iowait_exit);
+
+void ia64_record_iowait_exit(int cpu)
+{
+	/* FIXME: ITC might not synchronized between cpus/sockets */
+        per_cpu(ia64_iowait_exit, cpu) = ia64_get_itc();
+}
+
 void vtime_account_idle(struct task_struct *tsk)
 {
-	account_idle_time(vtime_delta(tsk));
+	struct thread_info *ti = task_thread_info(tsk);
+	cputime_t delta_stime, iowait_time;
+	__u64 now;
+
+	WARN_ON_ONCE(!irqs_disabled());
+
+	now = ia64_get_itc();
+
+	/*
+	 * tsk must be idle process: its runtime is accumulated as system time
+	 * (ac_utime must be 0) so convert it to idle time for accounting.
+	 *
+	 * in cycles:
+	 *		              iowait_exit
+	 *  (idle_enter)              |   ac_stamp   now(=idle_exit)
+	 * -------+-------------------*------+------->
+	 *         <------- ac_stime ------->
+	 *
+	 * in cputime:
+	 *       (0)
+	 * -------+-------------------*------+------->
+	 *         <----------- delta_stime --------->
+	 *         <-- iowait_time -->
+	 */
+	delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp));
+	iowait_time = cycle_to_cputime(ti->ac_stime +
+			__ia64_per_cpu_var(ia64_iowait_exit) - ti->ac_stamp);
+	ti->ac_stime = 0;
+	ti->ac_stamp = now;
+
+	account_idle_and_iowait(0, iowait_time, delta_stime);
 }
 
 #endif /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
-- 
1.7.1



  parent reply	other threads:[~2014-06-26  9:13 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-26  9:06 [RFC PATCH 0/8] rework iowait accounting Hidetoshi Seto
2014-06-26  9:08 ` [PATCH 1/8] cputime, sched: record last_iowait Hidetoshi Seto
2014-06-26  9:09 ` [PATCH 2/8] cputime, nohz: handle last_iowait for nohz Hidetoshi Seto
2014-06-26  9:10 ` [PATCH 3/8] cputime: introduce account_idle_and_iowait Hidetoshi Seto
2014-06-26  9:12 ` [PATCH 4/8] cputime, s390: introduce s390_get_idle_and_iowait Hidetoshi Seto
2014-06-26  9:13 ` Hidetoshi Seto [this message]
2014-06-26  9:14 ` [PATCH 6/8] cputime, ppc: update iowait accounting Hidetoshi Seto
2014-06-26  9:16 ` [PATCH 7/8] cputime: generic iowait accounting for VIRT_CPU_ACCOUNTING Hidetoshi Seto
2014-06-26  9:17 ` [PATCH 8/8] cputime: iowait aware idle tick accounting Hidetoshi Seto
2014-07-07  9:30 ` [RFC PATCH 0/8] rework iowait accounting Peter Zijlstra

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=53ABE422.1000000@jp.fujitsu.com \
    --to=seto.hidetoshi@jp.fujitsu.com \
    --cc=akpm@linux-foundation.org \
    --cc=arjan@linux.intel.com \
    --cc=fernando_b1@lab.ntt.co.jp \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=oleg@redhat.com \
    --cc=penguin-kernel@I-love.SAKURA.ne.jp \
    --cc=peterz@infradead.org \
    --cc=preeti@linux.vnet.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=vda.linux@googlemail.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).