[PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[Heiko Carstens]

From: Heiko Carstens <heiko.carstens@de.ibm.com>

When performing cpu hotplug tests the kernel printk log buffer gets flooded
with pointless "Switched to NOHz mode..." messages. Especially when afterwards
analyzing a dump this might have removed more interesting stuff out of the
buffer.
Assuming that switching to NOHz mode simply works just remove the printk.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 kernel/time/tick-sched.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index d5097c4..8b3b014 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -640,8 +640,6 @@ static void tick_nohz_switch_to_nohz(void)
 		next = ktime_add(next, tick_period);
 	}
 	local_irq_enable();
-
-	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
 }
 
 /*
@@ -793,10 +791,8 @@ void tick_setup_sched_timer(void)
 	}
 
 #ifdef CONFIG_NO_HZ
-	if (tick_nohz_enabled) {
+	if (tick_nohz_enabled)
 		ts->nohz_mode = NOHZ_MODE_HIGHRES;
-		printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
-	}
 #endif
 }
 #endif /* HIGH_RES_TIMERS */

Re: [PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[Joe Perches]

On Tue, 2011-08-23 at 13:20 +0200, Heiko Carstens wrote:
> From: Heiko Carstens <heiko.carstens@de.ibm.com>
> 
> When performing cpu hotplug tests the kernel printk log buffer gets flooded
> with pointless "Switched to NOHz mode..." messages. Especially when afterwards
> analyzing a dump this might have removed more interesting stuff out of the
> buffer.
> Assuming that switching to NOHz mode simply works just remove the printk.

> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
[]
> -
> -	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());

Though I don't know if these are actually particularly
useful messages, perhaps pr_info_ratelimited() might
be appropriate instead.

> @@ -793,10 +791,8 @@ void tick_setup_sched_timer(void)
> -		printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());

and here.

Re: [PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[Stephen Boyd]

On 08/23/2011 12:12 PM, Joe Perches wrote:
> On Tue, 2011-08-23 at 13:20 +0200, Heiko Carstens wrote:
>> From: Heiko Carstens <heiko.carstens@de.ibm.com>
>>
>> When performing cpu hotplug tests the kernel printk log buffer gets flooded
>> with pointless "Switched to NOHz mode..." messages. Especially when afterwards
>> analyzing a dump this might have removed more interesting stuff out of the
>> buffer.
>> Assuming that switching to NOHz mode simply works just remove the printk.
>> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
> []
>> -
>> -	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
> Though I don't know if these are actually particularly
> useful messages, perhaps pr_info_ratelimited() might
> be appropriate instead.

Or perhaps a printk_once() would be good. Presumably the second switch
to NOHz mode wouldn't be as dangerous as the first.

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: [PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[Heiko Carstens]

On Tue, Aug 23, 2011 at 12:15:51PM -0700, Stephen Boyd wrote:
> On 08/23/2011 12:12 PM, Joe Perches wrote:
> > On Tue, 2011-08-23 at 13:20 +0200, Heiko Carstens wrote:
> >> From: Heiko Carstens <heiko.carstens@de.ibm.com>
> >>
> >> When performing cpu hotplug tests the kernel printk log buffer gets flooded
> >> with pointless "Switched to NOHz mode..." messages. Especially when afterwards
> >> analyzing a dump this might have removed more interesting stuff out of the
> >> buffer.
> >> Assuming that switching to NOHz mode simply works just remove the printk.
> >> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
> > []
> >> -
> >> -	printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id());
> > Though I don't know if these are actually particularly
> > useful messages, perhaps pr_info_ratelimited() might
> > be appropriate instead.
> 
> Or perhaps a printk_once() would be good. Presumably the second switch
> to NOHz mode wouldn't be as dangerous as the first.

Let's just remove them.

Re: [PATCH] nohz: remove "Switched to NOHz mode" debugging messages

[David Miller]

From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 23 Aug 2011 13:20:46 +0200

> From: Heiko Carstens <heiko.carstens@de.ibm.com>
> 
> When performing cpu hotplug tests the kernel printk log buffer gets flooded
> with pointless "Switched to NOHz mode..." messages. Especially when afterwards
> analyzing a dump this might have removed more interesting stuff out of the
> buffer.
> Assuming that switching to NOHz mode simply works just remove the printk.
> 
> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>

These drive me nuts during bootup on my 128+ cpu machines :-/  Thanks
for taking care of this.

Acked-by: David S. Miller <davem@davemloft.net>