SLUB: simplify re on_each_cpu()

SLUB: simplify re on_each_cpu()

[Alexey Dobriyan]

on_each_cpu() expands to function call on UP, too.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
---

 mm/slub.c |    8 --------
 1 file changed, 8 deletions(-)

--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1495,15 +1495,7 @@ static void flush_cpu_slab(void *d)
 
 static void flush_all(struct kmem_cache *s)
 {
-#ifdef CONFIG_SMP
 	on_each_cpu(flush_cpu_slab, s, 1);
-#else
-	unsigned long flags;
-
-	local_irq_save(flags);
-	flush_cpu_slab(s);
-	local_irq_restore(flags);
-#endif
 }
 
 /*

Re: SLUB: simplify re on_each_cpu()

[Christoph Lameter]

Acked-by: Christoph Lameter <cl@linux-foundation.org>

Re: SLUB: simplify re on_each_cpu()

[Pekka Enberg]

On Wed, Jul 16, 2008 at 8:29 PM, Alexey Dobriyan <adobriyan@gmail.com> wrote:
> on_each_cpu() expands to function call on UP, too.
>
> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>

Applied, thanks!

Re: SLUB: simplify re on_each_cpu()

[Robin Holt]

On Wed, Jul 16, 2008 at 09:29:02PM +0400, Alexey Dobriyan wrote:
> on_each_cpu() expands to function call on UP, too.

It looks to me like on_each_cpu is a #define for !SMP.  Are you certain
of the need for this change?  Are you looking at a different tree than
Linus'?

Thanks,
Robin

> 
> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
> ---
> 
>  mm/slub.c |    8 --------
>  1 file changed, 8 deletions(-)
> 
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -1495,15 +1495,7 @@ static void flush_cpu_slab(void *d)
>  
>  static void flush_all(struct kmem_cache *s)
>  {
> -#ifdef CONFIG_SMP
>  	on_each_cpu(flush_cpu_slab, s, 1);
> -#else
> -	unsigned long flags;
> -
> -	local_irq_save(flags);
> -	flush_cpu_slab(s);
> -	local_irq_restore(flags);
> -#endif
>  }
>  
>  /*
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: SLUB: simplify re on_each_cpu()

[Pekka Enberg]

Hi Robin,

On Wed, Jul 16, 2008 at 09:29:02PM +0400, Alexey Dobriyan wrote:
>> on_each_cpu() expands to function call on UP, too.

On Thu, Jul 17, 2008 at 2:38 PM, Robin Holt <holt@sgi.com> wrote:
> It looks to me like on_each_cpu is a #define for !SMP.  Are you certain
> of the need for this change?  Are you looking at a different tree than
> Linus'?

I think it's actually you who is looking at an older tree.

However, while double-checking this, I noticed that for UP,
on_each_cpu() enables interrupts unconditionally with
local_irq_enable() whereas the call while the old code in SLUB did a
local_irq_restore() that preserves the irq mask. Looking at the
callers of flush_all() in mm/slub.c, the change looks okay, but it's
still bit strange on_each_cpu() works that way on UP.

                          Pekka

Re: SLUB: simplify re on_each_cpu()

[Robin Holt]

On Thu, Jul 17, 2008 at 03:26:46PM +0300, Pekka Enberg wrote:
> Hi Robin,
> 
> On Wed, Jul 16, 2008 at 09:29:02PM +0400, Alexey Dobriyan wrote:
> >> on_each_cpu() expands to function call on UP, too.
> 
> On Thu, Jul 17, 2008 at 2:38 PM, Robin Holt <holt@sgi.com> wrote:
> > It looks to me like on_each_cpu is a #define for !SMP.  Are you certain
> > of the need for this change?  Are you looking at a different tree than
> > Linus'?
> 
> I think it's actually you who is looking at an older tree.

I did a git-checkout of
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
which got me commit id 33af79d12e0fa25545d49e86afc67ea8ad5f2f40
and looked at include/linux/smp.h line 123.  It has:

#define on_each_cpu(func,info,wait)             \
        ({                                      \
                local_irq_disable();            \
                func(info);                     \
                local_irq_enable();             \
                0;                              \
        })

This is the !CONFIG_SMP case (line 111).  When I look at
kernel/softirq.c, I notice immediately prior to the comment the #ifdef
CONFIG_SMP.

They appear equivalent.  What am I missing?

Thanks,
Robin

Re: SLUB: simplify re on_each_cpu()

[Vegard Nossum]

On Thu, Jul 17, 2008 at 2:39 PM, Robin Holt <holt@sgi.com> wrote:
> On Thu, Jul 17, 2008 at 03:26:46PM +0300, Pekka Enberg wrote:
>> Hi Robin,
>>
>> On Wed, Jul 16, 2008 at 09:29:02PM +0400, Alexey Dobriyan wrote:
>> >> on_each_cpu() expands to function call on UP, too.
>>
>> On Thu, Jul 17, 2008 at 2:38 PM, Robin Holt <holt@sgi.com> wrote:
>> > It looks to me like on_each_cpu is a #define for !SMP.  Are you certain
>> > of the need for this change?  Are you looking at a different tree than
>> > Linus'?
>>
>> I think it's actually you who is looking at an older tree.
>
> I did a git-checkout of
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
> which got me commit id 33af79d12e0fa25545d49e86afc67ea8ad5f2f40
> and looked at include/linux/smp.h line 123.  It has:
>
> #define on_each_cpu(func,info,wait)             \
>        ({                                      \
>                local_irq_disable();            \
>                func(info);                     \

That's the function call right there :-)


Vegard

-- 
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
	-- E. W. Dijkstra, EWD1036