Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1536610

[PATCH 03/10] cputime: Allow accounting system time using cpustat index

Show all headers | View raw

In order to prepare for CONFIG_VIRT_CPU_ACCOUNTING_NATIVE to delay
cputime accounting to the tick, let's provide APIs to account system
time to precise contexts: hardirq, softirq, pure system, ...

Inspired-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 include/linux/kernel_stat.h |  4 ++++
 kernel/sched/cputime.c      | 19 ++++++++++++++-----
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index 00f7768..544aa86 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -80,10 +80,14 @@ static inline unsigned int kstat_cpu_irqs_sum(unsigned int cpu)
 
 extern void account_user_time(struct task_struct *, cputime_t);
 extern void account_system_time(struct task_struct *, int, cputime_t);
+extern void account_system_index_time(struct task_struct *, cputime_t,
+				      enum cpu_usage_stat);
 extern void account_steal_time(cputime_t);
 extern void account_idle_time(cputime_t);
 
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+extern void account_system_index_scaled(struct task_struct *, cputime_t,
+					cputime_t, enum cpu_usage_stat);
 static inline void account_process_tick(struct task_struct *tsk, int user)
 {
 	vtime_account_user(tsk);
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index 7700a9c..0e6ea60 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -176,8 +176,8 @@ static void account_guest_time(struct task_struct *p, cputime_t cputime)
  * @cputime: the cpu time spent in kernel space since the last update
  * @index: pointer to cpustat field that has to be updated
  */
-static inline
-void __account_system_time(struct task_struct *p, cputime_t cputime, int index)
+void account_system_index_time(struct task_struct *p,
+			       cputime_t cputime, enum cpu_usage_stat index)
 {
 	/* Add system time to process. */
 	p->stime += cputime;
@@ -190,6 +190,15 @@ void __account_system_time(struct task_struct *p, cputime_t cputime, int index)
 	acct_account_cputime(p);
 }
 
+#ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME
+void account_system_index_scaled(struct task_struct *p, cputime_t cputime,
+				 cputime_t scaled, enum cpu_usage_stat index)
+{
+	p->stimescaled += scaled;
+	account_system_index_time(p, cputime, index);
+}
+#endif
+
 /*
  * Account system cpu time to a process.
  * @p: the process that the cpu time gets accounted to
@@ -213,7 +222,7 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 	else
 		index = CPUTIME_SYSTEM;
 
-	__account_system_time(p, cputime, index);
+	account_system_index_time(p, cputime, index);
 }
 
 /*
@@ -400,7 +409,7 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick,
 		 * So, we have to handle it separately here.
 		 * Also, p->stime needs to be updated for ksoftirqd.
 		 */
-		__account_system_time(p, cputime, CPUTIME_SOFTIRQ);
+		account_system_index_time(p, cputime, CPUTIME_SOFTIRQ);
 	} else if (user_tick) {
 		account_user_time(p, cputime);
 	} else if (p == rq->idle) {
@@ -408,7 +417,7 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick,
 	} else if (p->flags & PF_VCPU) { /* System time or guest time */
 		account_guest_time(p, cputime);
 	} else {
-		__account_system_time(p, cputime, CPUTIME_SYSTEM);
+		account_system_index_time(p, cputime, CPUTIME_SYSTEM);
 	}
 }
 
-- 
2.7.4

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 00/10] vtime: Delay cputime accounting to tick Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [FIX][PATCH 02/10] ia64: Fix wrong start cputime assignment on task switch Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 09/10] s390/cputime: delayed accounting of system time Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
    Re: [PATCH 09/10] s390/cputime: delayed accounting of system time Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-10 02:50 +0100
  [PATCH 06/10] powerpc: Migrate stolen_time field to accounting structure Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 03/10] cputime: Allow accounting system time using cpustat index Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 10/10] vtime: Rename vtime_account_user() to vtime_flush() Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 07/10] powerpc/vtime: Accumulate cputime and account only on tick/task switch Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 08/10] ia64: Accumulate cputime and account only on tick/task switch Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  [PATCH 04/10] cputime: Export account_guest_time Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 03:40 +0100
  Re: [PATCH 00/10] vtime: Delay cputime accounting to tick Paul Mackerras <paulus@ozlabs.org> - 2016-12-06 05:30 +0100
    Re: [PATCH 00/10] vtime: Delay cputime accounting to tick Martin Schwidefsky <schwidefsky@de.ibm.com> - 2016-12-06 08:10 +0100
    Re: [PATCH 00/10] vtime: Delay cputime accounting to tick Frederic Weisbecker <fweisbec@gmail.com> - 2016-12-06 15:40 +0100
  Re: [PATCH 00/10] vtime: Delay cputime accounting to tick Christian Borntraeger <borntraeger@de.ibm.com> - 2016-12-06 09:50 +0100

csiph-web