[PATCH] x86: remove printk compiling warning

[PATCH] x86: remove printk compiling warning

[Shen Feng]

Since size is unsigned long, the format string is changed
from zd to lu.

Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
---
 arch/x86/kernel/setup.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index cc6f5eb..c0c68c1 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -95,7 +95,7 @@ void __init setup_per_cpu_areas(void)

 	/* Copy section for each CPU (we discard the original) */
 	size = PERCPU_ENOUGH_ROOM;
-	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
+	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
 			  size);

 	for_each_possible_cpu(i) {
-- 
1.5.4.1

Re: [PATCH] x86: remove printk compiling warning

[Li Zefan]

CC: Randy Dunlap <rdunlap@xenotime.net>

Shen Feng wrote:
> Since size is unsigned long, the format string is changed
> from zd to lu.
> 
> Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
> ---
>  arch/x86/kernel/setup.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
> index cc6f5eb..c0c68c1 100644
> --- a/arch/x86/kernel/setup.c
> +++ b/arch/x86/kernel/setup.c
> @@ -95,7 +95,7 @@ void __init setup_per_cpu_areas(void)
> 
>  	/* Copy section for each CPU (we discard the original) */
>  	size = PERCPU_ENOUGH_ROOM;
> -	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
> +	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
>  			  size);
> 
>  	for_each_possible_cpu(i) {

Let me figure out this.

Commit 0fc5035fc66d26776bebd45c508a9bce3611f388 in linux-next changed
'size' from unsigned long to ssize_t, causing a compile warning, and
then Randy posted a patch to fix it:
	http://marc.info/?l=linux-kernel&m=120958902125650&w=2

But this patch was wrongly merged into mainline, causing the compile warning
you described here.

So we should revert Randy's patch and re-apply it to linux-next ?

Re: [PATCH] x86: remove printk compiling warning

[Randy Dunlap]

On Wed, 07 May 2008 16:27:11 +0800 Li Zefan wrote:

> CC: Randy Dunlap <rdunlap@xenotime.net>
> 
> Shen Feng wrote:
> > Since size is unsigned long, the format string is changed
> > from zd to lu.
> > 
> > Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
> > ---
> >  arch/x86/kernel/setup.c |    2 +-
> >  1 files changed, 1 insertions(+), 1 deletions(-)
> > 
> > diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
> > index cc6f5eb..c0c68c1 100644
> > --- a/arch/x86/kernel/setup.c
> > +++ b/arch/x86/kernel/setup.c
> > @@ -95,7 +95,7 @@ void __init setup_per_cpu_areas(void)
> > 
> >  	/* Copy section for each CPU (we discard the original) */
> >  	size = PERCPU_ENOUGH_ROOM;
> > -	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
> > +	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
> >  			  size);
> > 
> >  	for_each_possible_cpu(i) {
> 
> Let me figure out this.
> 
> Commit 0fc5035fc66d26776bebd45c508a9bce3611f388 in linux-next changed
> 'size' from unsigned long to ssize_t, causing a compile warning, and
> then Randy posted a patch to fix it:
> 	http://marc.info/?l=linux-kernel&m=120958902125650&w=2
> 
> But this patch was wrongly merged into mainline, causing the compile warning
> you described here.
> 
> So we should revert Randy's patch and re-apply it to linux-next ?

Hi,
Thanks for the detective work.  I hadn't looked into this yet.

Yes, my patch was sent for linux-next.  How are we supposed to
keep this straightened out?

---
~Randy

Re: [PATCH] x86: remove printk compiling warning

[Thomas Gleixner]

On Wed, 7 May 2008, Li Zefan wrote:
> CC: Randy Dunlap <rdunlap@xenotime.net>
> 
> Shen Feng wrote:
> > Since size is unsigned long, the format string is changed
> > from zd to lu.
> > 
> > Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
> > ---
> >  arch/x86/kernel/setup.c |    2 +-
> >  1 files changed, 1 insertions(+), 1 deletions(-)
> > 
> > diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
> > index cc6f5eb..c0c68c1 100644
> > --- a/arch/x86/kernel/setup.c
> > +++ b/arch/x86/kernel/setup.c
> > @@ -95,7 +95,7 @@ void __init setup_per_cpu_areas(void)
> > 
> >  	/* Copy section for each CPU (we discard the original) */
> >  	size = PERCPU_ENOUGH_ROOM;
> > -	printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
> > +	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
> >  			  size);
> > 
> >  	for_each_possible_cpu(i) {
> 
> Let me figure out this.
> 
> Commit 0fc5035fc66d26776bebd45c508a9bce3611f388 in linux-next changed
> 'size' from unsigned long to ssize_t, causing a compile warning, and
> then Randy posted a patch to fix it:
> 	http://marc.info/?l=linux-kernel&m=120958902125650&w=2
> 
> But this patch was wrongly merged into mainline, causing the compile warning
> you described here.
> 
> So we should revert Randy's patch and re-apply it to linux-next ?

Yes, this came via Andrew and we did not notice that it was for
linux-next.

Thanks,

	tglx