mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH v2 1/2] try to fix /proc/stat scalability of irq num per cpu
@ 2010-10-21  7:01 KAMEZAWA Hiroyuki
  2010-10-21  7:03 ` [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu KAMEZAWA Hiroyuki
  0 siblings, 1 reply; 4+ messages in thread
From: KAMEZAWA Hiroyuki @ 2010-10-21  7:01 UTC (permalink / raw)
  To: linux-kernel; +Cc: akpm, steiner, yinghai, mingo

Posting an updated version against mmotm because I noticed rebase is necessary.
(kernel/irq/handle.c is modified.)
I'll post this again after merge window.
==
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

/proc/stat shows the total number of all interrupts to each cpu. But when
the number of IRQs are very large, it take very long time and 'cat /proc/stat'
takes more than 10 secs. This is because sum of all irq events are counted
when /proc/stat is read. This patch adds "sum of all irq" counter percpu
and reduce read costs.

The cost of reading /proc/stat is important because it's used by major
applications as 'top', 'ps', 'w', etc....

A test on a mechin (4096cpu, 256 nodes, 4592 irqs) shows

 %time cat /proc/stat > /dev/null
 Before Patch:  12.627 sec
 After  Patch:  2.459 sec

Tested-by: Jack Steiner <steiner@sgi.com>
Acked-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
---
 fs/proc/stat.c              |    4 +---
 include/linux/kernel_stat.h |   14 ++++++++++++--
 2 files changed, 13 insertions(+), 5 deletions(-)

Index: mmotm-1020/fs/proc/stat.c
===================================================================
--- mmotm-1020.orig/fs/proc/stat.c
+++ mmotm-1020/fs/proc/stat.c
@@ -52,9 +52,7 @@ static int show_stat(struct seq_file *p,
 		guest = cputime64_add(guest, kstat_cpu(i).cpustat.guest);
 		guest_nice = cputime64_add(guest_nice,
 			kstat_cpu(i).cpustat.guest_nice);
-		for_each_irq_nr(j) {
-			sum += kstat_irqs_cpu(j, i);
-		}
+		sum += kstat_cpu_irqs_sum(i);
 		sum += arch_irq_stat_cpu(i);
 
 		for (j = 0; j < NR_SOFTIRQS; j++) {
Index: mmotm-1020/include/linux/kernel_stat.h
===================================================================
--- mmotm-1020.orig/include/linux/kernel_stat.h
+++ mmotm-1020/include/linux/kernel_stat.h
@@ -33,6 +33,7 @@ struct kernel_stat {
 #ifndef CONFIG_GENERIC_HARDIRQS
        unsigned int irqs[NR_IRQS];
 #endif
+	unsigned long irqs_sum;
 	unsigned int softirqs[NR_SOFTIRQS];
 };
 
@@ -54,6 +55,7 @@ static inline void kstat_incr_irqs_this_
 					    struct irq_desc *desc)
 {
 	kstat_this_cpu.irqs[irq]++;
+	kstat_this_cpu.irqs_sum++;
 }
 
 static inline unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
@@ -65,8 +67,9 @@ static inline unsigned int kstat_irqs_cp
 extern unsigned int kstat_irqs_cpu(unsigned int irq, int cpu);
 #define kstat_irqs_this_cpu(DESC) \
 	((DESC)->kstat_irqs[smp_processor_id()])
-#define kstat_incr_irqs_this_cpu(irqno, DESC) \
-	((DESC)->kstat_irqs[smp_processor_id()]++)
+#define kstat_incr_irqs_this_cpu(irqno, DESC) do {\
+	((DESC)->kstat_irqs[smp_processor_id()]++);\
+	kstat_this_cpu.irqs_sum++; } while (0)
 
 #endif
 
@@ -94,6 +97,13 @@ static inline unsigned int kstat_irqs(un
 	return sum;
 }
 
+/*
+ * Number of interrupts per cpu, since bootup
+ */
+static inline unsigned int kstat_cpu_irqs_sum(unsigned int cpu)
+{
+	return kstat_cpu(cpu).irqs_sum;
+}
 
 /*
  * Lock/unlock the current runqueue - to extract task statistics:


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu
  2010-10-21  7:01 [PATCH v2 1/2] try to fix /proc/stat scalability of irq num per cpu KAMEZAWA Hiroyuki
@ 2010-10-21  7:03 ` KAMEZAWA Hiroyuki
  2010-10-21 19:58   ` Andrew Morton
  0 siblings, 1 reply; 4+ messages in thread
From: KAMEZAWA Hiroyuki @ 2010-10-21  7:03 UTC (permalink / raw)
  To: KAMEZAWA Hiroyuki; +Cc: linux-kernel, akpm, steiner, yinghai, mingo

From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

In /proc/stat, the number of per-IRQ event is shown by making a sum
each irq's events on all cpus. But we can make use of kstat_irqs().

kstat_irqs() do the same calculation, If !CONFIG_GENERIC_HARDIRQ,
it's not a big cost. (Both of the number of cpus and irqs are small.)

If a system is very big and CONFIG_GENERIC_HARDIRQ, it does

	for_each_irq()
		for_each_cpu()
			- look up a radix tree
			- read desc->irq_stat[cpu]
This seems not efficient. This patch adds kstat_irqs() for
CONFIG_GENRIC_HARDIRQ and change the calculation as

	for_each_irq()
		look up radix tree
		for_each_cpu()
			- read desc->irq_stat[cpu]

This reduces cost.

A test on (4096cpusp, 256 nodes, 4592 irqs) host (by Jack Steiner)

%time cat /proc/stat > /dev/null

Before Patch:	 2.459 sec
After Patch :	  .561 sec

Changelog:
 - rebased onto mmotm-1020 (kernel/irq/handle.c is modified)

Tested-by: Jack Steiner <steiner@sgi.com>
Acked-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
---
 fs/proc/stat.c              |    9 ++-------
 include/linux/kernel_stat.h |    4 ++++
 kernel/irq/irqdesc.c        |   16 ++++++++++++++++
 3 files changed, 22 insertions(+), 7 deletions(-)

Index: mmotm-1020/fs/proc/stat.c
===================================================================
--- mmotm-1020.orig/fs/proc/stat.c
+++ mmotm-1020/fs/proc/stat.c
@@ -108,13 +108,8 @@ static int show_stat(struct seq_file *p,
 	seq_printf(p, "intr %llu", (unsigned long long)sum);
 
 	/* sum again ? it could be updated? */
-	for_each_irq_nr(j) {
-		per_irq_sum = 0;
-		for_each_possible_cpu(i)
-			per_irq_sum += kstat_irqs_cpu(j, i);
-
-		seq_printf(p, " %u", per_irq_sum);
-	}
+	for_each_irq_nr(j)
+		seq_printf(p, " %u", kstat_irqs(j));
 
 	seq_printf(p,
 		"\nctxt %llu\n"
Index: mmotm-1020/include/linux/kernel_stat.h
===================================================================
--- mmotm-1020.orig/include/linux/kernel_stat.h
+++ mmotm-1020/include/linux/kernel_stat.h
@@ -86,6 +86,7 @@ static inline unsigned int kstat_softirq
 /*
  * Number of interrupts per specific IRQ source, since bootup
  */
+#ifndef CONFIG_GENERIC_HARDIRQS
 static inline unsigned int kstat_irqs(unsigned int irq)
 {
 	unsigned int sum = 0;
@@ -96,6 +97,9 @@ static inline unsigned int kstat_irqs(un
 
 	return sum;
 }
+#else
+extern unsigned int kstat_irqs(unsigned int irq);
+#endif
 
 /*
  * Number of interrupts per cpu, since bootup
Index: mmotm-1020/kernel/irq/irqdesc.c
===================================================================
--- mmotm-1020.orig/kernel/irq/irqdesc.c
+++ mmotm-1020/kernel/irq/irqdesc.c
@@ -393,3 +393,19 @@ unsigned int kstat_irqs_cpu(unsigned int
 	struct irq_desc *desc = irq_to_desc(irq);
 	return desc ? desc->kstat_irqs[cpu] : 0;
 }
+
+#ifdef CONFIG_GENERIC_HARDIRQS
+unsigned int  kstat_irqs(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	int cpu;
+	int sum = 0;
+
+	if (!desc)
+		return 0;
+	for_each_possible_cpu(cpu)
+		sum += desc->kstat_irqs[cpu];
+	return sum;
+}
+EXPORT_SYMBOL_GPL(kstat_irqs);
+#endif /*CONFIG_GENERIC_HARDIRQS*/


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu
  2010-10-21  7:03 ` [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu KAMEZAWA Hiroyuki
@ 2010-10-21 19:58   ` Andrew Morton
  2010-10-22  3:21     ` KAMEZAWA Hiroyuki
  0 siblings, 1 reply; 4+ messages in thread
From: Andrew Morton @ 2010-10-21 19:58 UTC (permalink / raw)
  To: KAMEZAWA Hiroyuki; +Cc: linux-kernel, steiner, yinghai, mingo, Thomas Gleixner

On Thu, 21 Oct 2010 16:03:43 +0900
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:

> From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> 
> In /proc/stat, the number of per-IRQ event is shown by making a sum
> each irq's events on all cpus. But we can make use of kstat_irqs().
> 
> kstat_irqs() do the same calculation, If !CONFIG_GENERIC_HARDIRQ,
> it's not a big cost. (Both of the number of cpus and irqs are small.)
> 
> If a system is very big and CONFIG_GENERIC_HARDIRQ, it does
> 
> 	for_each_irq()
> 		for_each_cpu()
> 			- look up a radix tree
> 			- read desc->irq_stat[cpu]
> This seems not efficient. This patch adds kstat_irqs() for
> CONFIG_GENRIC_HARDIRQ and change the calculation as
> 
> 	for_each_irq()
> 		look up radix tree
> 		for_each_cpu()
> 			- read desc->irq_stat[cpu]
> 
> This reduces cost.
> 
> A test on (4096cpusp, 256 nodes, 4592 irqs) host (by Jack Steiner)
> 
> %time cat /proc/stat > /dev/null
> 
> Before Patch:	 2.459 sec
> After Patch :	  .561 sec
> 
> Changelog:
>  - rebased onto mmotm-1020 (kernel/irq/handle.c is modified)
> 
> Tested-by: Jack Steiner <steiner@sgi.com>
> Acked-by: Jack Steiner <steiner@sgi.com>
> Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> ---
>  fs/proc/stat.c              |    9 ++-------
>  include/linux/kernel_stat.h |    4 ++++
>  kernel/irq/irqdesc.c        |   16 ++++++++++++++++
>  3 files changed, 22 insertions(+), 7 deletions(-)
> 
> Index: mmotm-1020/fs/proc/stat.c
> ===================================================================
> --- mmotm-1020.orig/fs/proc/stat.c
> +++ mmotm-1020/fs/proc/stat.c
> @@ -108,13 +108,8 @@ static int show_stat(struct seq_file *p,
>  	seq_printf(p, "intr %llu", (unsigned long long)sum);
>  
>  	/* sum again ? it could be updated? */
> -	for_each_irq_nr(j) {
> -		per_irq_sum = 0;
> -		for_each_possible_cpu(i)
> -			per_irq_sum += kstat_irqs_cpu(j, i);
> -
> -		seq_printf(p, " %u", per_irq_sum);
> -	}
> +	for_each_irq_nr(j)
> +		seq_printf(p, " %u", kstat_irqs(j));
>  
>  	seq_printf(p,
>  		"\nctxt %llu\n"
> Index: mmotm-1020/include/linux/kernel_stat.h
> ===================================================================
> --- mmotm-1020.orig/include/linux/kernel_stat.h
> +++ mmotm-1020/include/linux/kernel_stat.h
> @@ -86,6 +86,7 @@ static inline unsigned int kstat_softirq
>  /*
>   * Number of interrupts per specific IRQ source, since bootup
>   */
> +#ifndef CONFIG_GENERIC_HARDIRQS
>  static inline unsigned int kstat_irqs(unsigned int irq)
>  {
>  	unsigned int sum = 0;
> @@ -96,6 +97,9 @@ static inline unsigned int kstat_irqs(un
>  
>  	return sum;
>  }
> +#else
> +extern unsigned int kstat_irqs(unsigned int irq);
> +#endif

hrm, why on earth was that inlined.

>  /*
>   * Number of interrupts per cpu, since bootup
> Index: mmotm-1020/kernel/irq/irqdesc.c
> ===================================================================
> --- mmotm-1020.orig/kernel/irq/irqdesc.c
> +++ mmotm-1020/kernel/irq/irqdesc.c
> @@ -393,3 +393,19 @@ unsigned int kstat_irqs_cpu(unsigned int
>  	struct irq_desc *desc = irq_to_desc(irq);
>  	return desc ? desc->kstat_irqs[cpu] : 0;
>  }
> +
> +#ifdef CONFIG_GENERIC_HARDIRQS
> +unsigned int  kstat_irqs(unsigned int irq)
> +{
> +	struct irq_desc *desc = irq_to_desc(irq);
> +	int cpu;
> +	int sum = 0;
> +
> +	if (!desc)
> +		return 0;
> +	for_each_possible_cpu(cpu)
> +		sum += desc->kstat_irqs[cpu];
> +	return sum;
> +}
> +EXPORT_SYMBOL_GPL(kstat_irqs);

kstat_irqs() needs to be exported to modules because of some silliness
in drivers/isdn/hisax/config.c.  But in linux-next that silliness got
deleted and the kstat_irqs export was removed.

--- a/kernel/irq/irqdesc.c~proc-stat-fix-scalability-of-irq-sum-of-all-cpu-fix
+++ a/kernel/irq/irqdesc.c
@@ -395,7 +395,7 @@ unsigned int kstat_irqs_cpu(unsigned int
 }
 
 #ifdef CONFIG_GENERIC_HARDIRQS
-unsigned int  kstat_irqs(unsigned int irq)
+unsigned int kstat_irqs(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	int cpu;
@@ -407,5 +407,4 @@ unsigned int  kstat_irqs(unsigned int ir
 		sum += desc->kstat_irqs[cpu];
 	return sum;
 }
-EXPORT_SYMBOL_GPL(kstat_irqs);
-#endif /*CONFIG_GENERIC_HARDIRQS*/
+#endif /* CONFIG_GENERIC_HARDIRQS */
_


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu
  2010-10-21 19:58   ` Andrew Morton
@ 2010-10-22  3:21     ` KAMEZAWA Hiroyuki
  0 siblings, 0 replies; 4+ messages in thread
From: KAMEZAWA Hiroyuki @ 2010-10-22  3:21 UTC (permalink / raw)
  To: Andrew Morton; +Cc: linux-kernel, steiner, yinghai, mingo, Thomas Gleixner

On Thu, 21 Oct 2010 12:58:16 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:

> >  /*
> >   * Number of interrupts per cpu, since bootup
> > Index: mmotm-1020/kernel/irq/irqdesc.c
> > ===================================================================
> > --- mmotm-1020.orig/kernel/irq/irqdesc.c
> > +++ mmotm-1020/kernel/irq/irqdesc.c
> > @@ -393,3 +393,19 @@ unsigned int kstat_irqs_cpu(unsigned int
> >  	struct irq_desc *desc = irq_to_desc(irq);
> >  	return desc ? desc->kstat_irqs[cpu] : 0;
> >  }
> > +
> > +#ifdef CONFIG_GENERIC_HARDIRQS
> > +unsigned int  kstat_irqs(unsigned int irq)
> > +{
> > +	struct irq_desc *desc = irq_to_desc(irq);
> > +	int cpu;
> > +	int sum = 0;
> > +
> > +	if (!desc)
> > +		return 0;
> > +	for_each_possible_cpu(cpu)
> > +		sum += desc->kstat_irqs[cpu];
> > +	return sum;
> > +}
> > +EXPORT_SYMBOL_GPL(kstat_irqs);
> 
> kstat_irqs() needs to be exported to modules because of some silliness
> in drivers/isdn/hisax/config.c.  But in linux-next that silliness got
> deleted and the kstat_irqs export was removed.
> 
Okay, thank you for fixing.

-Kame


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2010-10-22  3:27 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-10-21  7:01 [PATCH v2 1/2] try to fix /proc/stat scalability of irq num per cpu KAMEZAWA Hiroyuki
2010-10-21  7:03 ` [PATCH v2 2/2] try to fix /proc/stat scalability of irq sum of all cpu KAMEZAWA Hiroyuki
2010-10-21 19:58   ` Andrew Morton
2010-10-22  3:21     ` KAMEZAWA Hiroyuki

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome