[PATCH] mm/slab: Fix typo _RET_IP -> _RET_IP_

[PATCH] mm/slab: Fix typo _RET_IP -> _RET_IP_

[Ezequiel Garcia]

The bug was introduced by commit 7c0cb9c64f83
"mm, slab: Replace 'caller' type, void* -> unsigned long".

Cc: Pekka Enberg <penberg@kernel.org>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
---
 mm/slab.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index d011030..ca3849f 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3869,7 +3869,7 @@ void *kmem_cache_alloc_node_trace(struct kmem_cache *cachep,
 {
 	void *ret;
 
-	ret = slab_alloc_node(cachep, flags, nodeid, _RET_IP);
+	ret = slab_alloc_node(cachep, flags, nodeid, _RET_IP_);
 
 	trace_kmalloc_node(_RET_IP_, ret,
 			   size, cachep->size,
-- 
1.7.8.6

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Ezequiel Garcia]

The bug was introduced in commit 4052147c0afa
"mm, slab: Match SLAB and SLUB kmem_cache_alloc_xxx_trace() prototype".

Cc: Pekka Enberg <penberg@kernel.org>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
---
 mm/slab.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index ca3849f..3409ead 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3862,10 +3862,10 @@ void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 EXPORT_SYMBOL(kmem_cache_alloc_node);
 
 #ifdef CONFIG_TRACING
-void *kmem_cache_alloc_node_trace(struct kmem_cache *cachep,
+void *kmem_cache_alloc_node_trace(size_t size,
+				  struct kmem_cache *cachep,
 				  gfp_t flags,
-				  int nodeid,
-				  size_t size)
+				  int nodeid)
 {
 	void *ret;
 
@@ -3887,7 +3887,7 @@ __do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller)
 	cachep = kmem_find_general_cachep(size, flags);
 	if (unlikely(ZERO_OR_NULL_PTR(cachep)))
 		return cachep;
-	return kmem_cache_alloc_node_trace(cachep, flags, node, size);
+	return kmem_cache_alloc_node_trace(size, cachep, flags, node);
 }
 
 #if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_TRACING)
-- 
1.7.8.6

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[David Rientjes]

On Tue, 25 Sep 2012, Ezequiel Garcia wrote:

> The bug was introduced in commit 4052147c0afa
> "mm, slab: Match SLAB and SLUB kmem_cache_alloc_xxx_trace() prototype".
> 

This isn't a candidate for kernel-janitors@vger.kernel.org, these are 
patches that are one of Pekka's branches and would never make it to Linus' 
tree in this form.

> Cc: Pekka Enberg <penberg@kernel.org>
> Reported-by: Fengguang Wu <fengguang.wu@intel.com>
> Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>

Acked-by: David Rientjes <rientjes@google.com>

So now we have this for SLAB:

extern void *kmem_cache_alloc_node_trace(size_t size,
					 struct kmem_cache *cachep,
					 gfp_t flags,
					 int nodeid);

and this for SLUB:

extern void *kmem_cache_alloc_node_trace(struct kmem_cache *s,
					 gfp_t gfpflags,
					 int node, size_t size);

Would you like to send a follow-up patch to make these the same?  (My 
opinion is that the SLUB variant is the correct order.)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Dan Carpenter]

On Tue, Sep 25, 2012 at 09:18:02PM -0700, David Rientjes wrote:
> On Tue, 25 Sep 2012, Ezequiel Garcia wrote:
> 
> > The bug was introduced in commit 4052147c0afa
> > "mm, slab: Match SLAB and SLUB kmem_cache_alloc_xxx_trace() prototype".
> > 
> 
> This isn't a candidate for kernel-janitors@vger.kernel.org, these are 
> patches that are one of Pekka's branches and would never make it to Linus' 
> tree in this form.

kernel-janitors got CC'd because it was a compile problem.  It stops
us from sending duplicate messages to people.  It's surprising how
annoyed people get about duplicates instead of just ignoring the
second messages like sane individuals would.

regards,
dan carpenter

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Ezequiel Garcia]

Hi David,

On Wed, Sep 26, 2012 at 1:18 AM, David Rientjes <rientjes@google.com> wrote:
> On Tue, 25 Sep 2012, Ezequiel Garcia wrote:
>
>> The bug was introduced in commit 4052147c0afa
>> "mm, slab: Match SLAB and SLUB kmem_cache_alloc_xxx_trace() prototype".
>>
>
> This isn't a candidate for kernel-janitors@vger.kernel.org, these are
> patches that are one of Pekka's branches and would never make it to Linus'
> tree in this form.
>
>> Cc: Pekka Enberg <penberg@kernel.org>
>> Reported-by: Fengguang Wu <fengguang.wu@intel.com>
>> Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
>
> Acked-by: David Rientjes <rientjes@google.com>
>
> So now we have this for SLAB:
>
> extern void *kmem_cache_alloc_node_trace(size_t size,
>                                          struct kmem_cache *cachep,
>                                          gfp_t flags,
>                                          int nodeid);
>
> and this for SLUB:
>
> extern void *kmem_cache_alloc_node_trace(struct kmem_cache *s,
>                                          gfp_t gfpflags,
>                                          int node, size_t size);
>
> Would you like to send a follow-up patch to make these the same?  (My
> opinion is that the SLUB variant is the correct order.)

Yes. I just asked Pekka to revert this patch altogether.
The original patch was meant to match SLAB and SLUB, and this
fix should maintain that. But instead I fix it the wrong way.

I'll send another one.

Sorry for the mess,
Ezequiel.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Pekka Enberg]

On Wed, Sep 26, 2012 at 1:09 PM, Ezequiel Garcia <elezegarcia@gmail.com> wrote:
> Yes. I just asked Pekka to revert this patch altogether.
> The original patch was meant to match SLAB and SLUB, and this
> fix should maintain that. But instead I fix it the wrong way.
>
> I'll send another one.

Okay, I'm now confused and somewhat unhappy. What commits do you want
me to nuke exactly?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Ezequiel Garcia]

On Wed, Sep 26, 2012 at 7:42 AM, Pekka Enberg <penberg@kernel.org> wrote:
> On Wed, Sep 26, 2012 at 1:09 PM, Ezequiel Garcia <elezegarcia@gmail.com> wrote:
>> Yes. I just asked Pekka to revert this patch altogether.
>> The original patch was meant to match SLAB and SLUB, and this
>> fix should maintain that. But instead I fix it the wrong way.
>>
>> I'll send another one.
>
> Okay, I'm now confused and somewhat unhappy. What commits do you want
> me to nuke exactly?

>From your slab/tracing topic branch:

1e5965bf1f018cc30a4659fa3f1a40146e4276f6
mm/slab: Fix kmem_cache_alloc_node_trace() declaration

Thanks,
Ezequiel.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix kmem_cache_alloc_node_trace() declaration

[Ezequiel Garcia]

Pekka,

On Tue, Sep 25, 2012 at 8:07 AM, Ezequiel Garcia <elezegarcia@gmail.com> wrote:
> The bug was introduced in commit 4052147c0afa
> "mm, slab: Match SLAB and SLUB kmem_cache_alloc_xxx_trace() prototype".
>
> Cc: Pekka Enberg <penberg@kernel.org>
> Reported-by: Fengguang Wu <fengguang.wu@intel.com>
> Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
> ---
>  mm/slab.c |    8 ++++----
>  1 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/mm/slab.c b/mm/slab.c
> index ca3849f..3409ead 100644
> --- a/mm/slab.c
> +++ b/mm/slab.c
> @@ -3862,10 +3862,10 @@ void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid)
>  EXPORT_SYMBOL(kmem_cache_alloc_node);
>
>  #ifdef CONFIG_TRACING
> -void *kmem_cache_alloc_node_trace(struct kmem_cache *cachep,
> +void *kmem_cache_alloc_node_trace(size_t size,
> +                                 struct kmem_cache *cachep,
>                                   gfp_t flags,
> -                                 int nodeid,
> -                                 size_t size)
> +                                 int nodeid)
>  {
>         void *ret;
>
> @@ -3887,7 +3887,7 @@ __do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller)
>         cachep = kmem_find_general_cachep(size, flags);
>         if (unlikely(ZERO_OR_NULL_PTR(cachep)))
>                 return cachep;
> -       return kmem_cache_alloc_node_trace(cachep, flags, node, size);
> +       return kmem_cache_alloc_node_trace(size, cachep, flags, node);
>  }
>
>  #if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_TRACING)
> --

Please revert this patch. This fix is wrong, I'll send a proper one.

Sorry for the mess,
Ezequiel.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm/slab: Fix typo _RET_IP -> _RET_IP_

[David Rientjes]

On Tue, 25 Sep 2012, Ezequiel Garcia wrote:

> The bug was introduced by commit 7c0cb9c64f83
> "mm, slab: Replace 'caller' type, void* -> unsigned long".
> 

That commit SHA1 may not remain consistent, so it's better to just mention 
the name of the patch.  I also didn't see this in linux-next and had to 
look at Pekka's slab/next tree to find it since it wasn't indicated in the 
patch.

> Cc: Pekka Enberg <penberg@kernel.org>
> Reported-by: Fengguang Wu <fengguang.wu@intel.com>
> Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>

Acked-by: David Rientjes <rientjes@google.com>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>