[PATCH] Fix printk format vs argument warning

[PATCH] Fix printk format vs argument warning

[David Howells]

The attached patch makes the argument to this printk in
calibrate_migration_costs() always long to match the format string.

Signed-Off-By: David Howells <dhowells@redhat.com>
---
warthog>diffstat -p1 format-arg-size-2612mm1-10.diff 
 kernel/sched.c |    2 +-
 1 files changed, 1 insertion(+), 1 deletion(-)

diff -uNrp linux-2.6.12-mm1/kernel/sched.c linux-2.6.12-mm1-cachefs-wander/kernel/sched.c
--- linux-2.6.12-mm1/kernel/sched.c	2005-06-22 13:54:08.000000000 +0100
+++ linux-2.6.12-mm1-cachefs-wander/kernel/sched.c	2005-06-22 14:10:57.000000000 +0100
@@ -5572,7 +5572,7 @@ void __devinit calibrate_migration_costs
 	printk("| migration cost matrix (max_cache_size: %d, cpu: %ld MHz):\n",
 			max_cache_size,
 #ifdef CONFIG_X86
-			cpu_khz/1000
+			cpu_khz/1000L
 #else
 			-1L
 #endif

Re: [PATCH] Fix printk format vs argument warning

[Andrew Morton]

David Howells <dhowells@redhat.com> wrote:
>
> The attached patch makes the argument to this printk in
>  calibrate_migration_costs() always long to match the format string.

That darn printk again.  You wouldn't believe...

calibrate_migration_costs() causes a storm of boot-time output and I think
all those printks should be removed before this code goes up to Linus. 
Maybe split out into a separate -mm-only patch?

>  Signed-Off-By: David Howells <dhowells@redhat.com>
>  ---
>  warthog>diffstat -p1 format-arg-size-2612mm1-10.diff 
>   kernel/sched.c |    2 +-
>   1 files changed, 1 insertion(+), 1 deletion(-)
> 
>  diff -uNrp linux-2.6.12-mm1/kernel/sched.c linux-2.6.12-mm1-cachefs-wander/kernel/sched.c
>  --- linux-2.6.12-mm1/kernel/sched.c	2005-06-22 13:54:08.000000000 +0100
>  +++ linux-2.6.12-mm1-cachefs-wander/kernel/sched.c	2005-06-22 14:10:57.000000000 +0100
>  @@ -5572,7 +5572,7 @@ void __devinit calibrate_migration_costs
>   	printk("| migration cost matrix (max_cache_size: %d, cpu: %ld MHz):\n",
>   			max_cache_size,
>   #ifdef CONFIG_X86
>  -			cpu_khz/1000
>  +			cpu_khz/1000L
>   #else

It's currently %d in my tree.  This is why I converted x86's cpu_khz from
unsigned long to unsigned, to match x86_64's.

Re: [PATCH] Fix printk format vs argument warning

[Ingo Molnar]

* Andrew Morton <akpm@osdl.org> wrote:

> calibrate_migration_costs() causes a storm of boot-time output and I 
> think all those printks should be removed before this code goes up to 
> Linus.  Maybe split out into a separate -mm-only patch?

it was only a debugging measure. Patch below turns off.

------
turn off migration debugging.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

 kernel/sched.c |    2 +-
 1 files changed, 1 insertion(+), 1 deletion(-)

Index: linux-sched-curr/kernel/sched.c
===================================================================
--- linux-sched-curr.orig/kernel/sched.c
+++ linux-sched-curr/kernel/sched.c
@@ -5043,7 +5043,7 @@ __init static unsigned long domain_dista
 	return distance;
 }
 
-static __initdata unsigned int migration_debug = 1;
+static __initdata unsigned int migration_debug = 0;
 
 static int __init setup_migration_debug(char *str)
 {