[PATCH v2] cpumask: add cpumask_var_t documentation

[PATCH v2] cpumask: add cpumask_var_t documentation

[KOSAKI Motohiro]

cpumask_var_t has one nortable difference against cpumask_t.
This patch adds the explanation.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
 include/linux/cpumask.h |   14 ++++++++++++++
 1 files changed, 14 insertions(+), 0 deletions(-)

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 1e40dd0..471c98a 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
  *	  ... use 'tmpmask' like a normal struct cpumask * ...
  *
  *	free_cpumask_var(tmpmask);
+ *
+ *
+ * However, one notable exception is there. cpumask_var_t is allocated
+ * only nr_cpu_ids bits (in the other hand, real cpumask_t always has
+ * NR_CPUS bits). therefore You don't have to dereference cpumask_var_t.
+ *
+ *	cpumask_var_t tmpmask;
+ *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ *		return -ENOMEM;
+ *
+ *	var = *tmpmask;
+ *
+ * This code makes NR_CPUS length memcopy and bring memroy corruption.
+ * You have to use cpumask_copy() instead.
  */
 #ifdef CONFIG_CPUMASK_OFFSTACK
 typedef struct cpumask *cpumask_var_t;
-- 
1.7.3.1

Re: [PATCH v2] cpumask: add cpumask_var_t documentation

[Thiago Farina]

On Thu, Apr 28, 2011 at 12:27 PM, KOSAKI Motohiro
<kosaki.motohiro@jp.fujitsu.com> wrote:
> cpumask_var_t has one nortable difference against cpumask_t.
> This patch adds the explanation.
>
> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> ---
>  include/linux/cpumask.h |   14 ++++++++++++++
>  1 files changed, 14 insertions(+), 0 deletions(-)
>
> diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
> index 1e40dd0..471c98a 100644
> --- a/include/linux/cpumask.h
> +++ b/include/linux/cpumask.h
> @@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
>  *       ... use 'tmpmask' like a normal struct cpumask * ...
>  *
>  *     free_cpumask_var(tmpmask);
> + *
> + *
> + * However, one notable exception is there. cpumask_var_t is allocated
> + * only nr_cpu_ids bits

Maybe, instead of "is allocated only", "allocates only" ?


> (in the other hand, real cpumask_t always has
> + * NR_CPUS bits). therefore You don't have to dereference cpumask_var_t.

s/therefore You/Therefore you/ ?

> + *
> + *     cpumask_var_t tmpmask;
> + *     if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
> + *             return -ENOMEM;
> + *
> + *     var = *tmpmask;
> + *
> + * This code makes NR_CPUS length memcopy and bring memroy corruption.

/s/memroy/memory

You are saying that I should not use this code? I'm confused, could
you explain a little bit?

> + * You have to use cpumask_copy() instead.
>  */

I don't get this. :(


>  #ifdef CONFIG_CPUMASK_OFFSTACK
>  typedef struct cpumask *cpumask_var_t;
> --
> 1.7.3.1
>
>
>
> --
> 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/
>

[PATCH v3] cpumask: add cpumask_var_t documentation

[KOSAKI Motohiro]

> On Thu, Apr 28, 2011 at 12:27 PM, KOSAKI Motohiro
> <kosaki.motohiro@jp.fujitsu.com> wrote:
> > cpumask_var_t has one nortable difference against cpumask_t.
> > This patch adds the explanation.
> >
> > Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> > ---
> >  include/linux/cpumask.h |   14 ++++++++++++++
> >  1 files changed, 14 insertions(+), 0 deletions(-)
> >
> > diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
> > index 1e40dd0..471c98a 100644
> > --- a/include/linux/cpumask.h
> > +++ b/include/linux/cpumask.h
> > @@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
> >  *       ... use 'tmpmask' like a normal struct cpumask * ...
> >  *
> >  *     free_cpumask_var(tmpmask);
> > + *
> > + *
> > + * However, one notable exception is there. cpumask_var_t is allocated
> > + * only nr_cpu_ids bits
> 
> Maybe, instead of "is allocated only", "allocates only" ?

Instead, "alloc_cpumask_var() allocates only nr_cpu_ids bits" is more clear? :)


> > (in the other hand, real cpumask_t always has
> > + * NR_CPUS bits). therefore You don't have to dereference cpumask_var_t.
> 
> s/therefore You/Therefore you/ ?
> 

Thx.


> > + *
> > + *     cpumask_var_t tmpmask;
> > + *     if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
> > + *             return -ENOMEM;
> > + *
> > + *     var = *tmpmask;
> > + *
> > + * This code makes NR_CPUS length memcopy and bring memroy corruption.
> 
> /s/memroy/memory
> 
> You are saying that I should not use this code? I'm confused, could
> you explain a little bit?

Yes. you aren't confused.


> > + * You have to use cpumask_copy() instead.
> >  */
> 
> I don't get this. :(

"cpumask_copy() privide safe copy functionality." is clear?


>From d2ea9d4846e46bcc8a82b9a641ede3a10aca346c Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Tue, 26 Apr 2011 19:58:39 +0900
Subject: [PATCH v3] cpumask: add cpumask_var_t documentation

cpumask_var_t has one nortable difference against cpumask_t.
This patch adds the explanation.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
 include/linux/cpumask.h |   14 ++++++++++++++
 1 files changed, 14 insertions(+), 0 deletions(-)

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 1e40dd0..e2b9032 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
  *	  ... use 'tmpmask' like a normal struct cpumask * ...
  *
  *	free_cpumask_var(tmpmask);
+ *
+ *
+ * However, one notable exception is there. alloc_cpumask_var() allocates
+ * only nr_cpumask_bits bits (in the other hand, real cpumask_t always has
+ * NR_CPUS bits). Therefore you don't have to dereference cpumask_var_t.
+ *
+ *	cpumask_var_t tmpmask;
+ *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ *		return -ENOMEM;
+ *
+ *	var = *tmpmask;
+ *
+ * This code makes NR_CPUS length memcopy and bring memory corruption.
+ * cpumask_copy() privide safe copy functionality.
  */
 #ifdef CONFIG_CPUMASK_OFFSTACK
 typedef struct cpumask *cpumask_var_t;
-- 
1.7.3.1

Re: [PATCH v3] cpumask: add cpumask_var_t documentation

[Thiago Farina]

On Thu, Apr 28, 2011 at 12:43 PM, KOSAKI Motohiro
<kosaki.motohiro@jp.fujitsu.com> wrote:
>> On Thu, Apr 28, 2011 at 12:27 PM, KOSAKI Motohiro
>> <kosaki.motohiro@jp.fujitsu.com> wrote:
>> > cpumask_var_t has one nortable difference against cpumask_t.
>> > This patch adds the explanation.
>> >
>> > Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
>> > ---
>> >  include/linux/cpumask.h |   14 ++++++++++++++
>> >  1 files changed, 14 insertions(+), 0 deletions(-)
>> >
>> > diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
>> > index 1e40dd0..471c98a 100644
>> > --- a/include/linux/cpumask.h
>> > +++ b/include/linux/cpumask.h
>> > @@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
>> >  *       ... use 'tmpmask' like a normal struct cpumask * ...
>> >  *
>> >  *     free_cpumask_var(tmpmask);
>> > + *
>> > + *
>> > + * However, one notable exception is there. cpumask_var_t is allocated
>> > + * only nr_cpu_ids bits
>>
>> Maybe, instead of "is allocated only", "allocates only" ?
>
> Instead, "alloc_cpumask_var() allocates only nr_cpu_ids bits" is more clear? :)
>
Yup. Thx.

>
>> > (in the other hand, real cpumask_t always has
>> > + * NR_CPUS bits). therefore You don't have to dereference cpumask_var_t.
>>
>> s/therefore You/Therefore you/ ?
>>
>
> Thx.
>
>
>> > + *
>> > + *     cpumask_var_t tmpmask;
>> > + *     if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
>> > + *             return -ENOMEM;
>> > + *
>> > + *     var = *tmpmask;
>> > + *
>> > + * This code makes NR_CPUS length memcopy and bring memroy corruption.
>>
>> /s/memroy/memory
>>
>> You are saying that I should not use this code? I'm confused, could
>> you explain a little bit?
>
> Yes. you aren't confused.
>
>
>> > + * You have to use cpumask_copy() instead.
>> >  */
>>
>> I don't get this. :(
>
> "cpumask_copy() privide safe copy functionality." is clear?
>
Yes. ;)

>
> From d2ea9d4846e46bcc8a82b9a641ede3a10aca346c Mon Sep 17 00:00:00 2001
> From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> Date: Tue, 26 Apr 2011 19:58:39 +0900
> Subject: [PATCH v3] cpumask: add cpumask_var_t documentation
>
> cpumask_var_t has one nortable difference against cpumask_t.
> This patch adds the explanation.
>
> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> ---
>  include/linux/cpumask.h |   14 ++++++++++++++
>  1 files changed, 14 insertions(+), 0 deletions(-)
>
> diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
> index 1e40dd0..e2b9032 100644
> --- a/include/linux/cpumask.h
> +++ b/include/linux/cpumask.h
> @@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
>  *       ... use 'tmpmask' like a normal struct cpumask * ...
>  *
>  *     free_cpumask_var(tmpmask);
> + *
> + *
> + * However, one notable exception is there. alloc_cpumask_var() allocates
> + * only nr_cpumask_bits bits (in the other hand, real cpumask_t always has
> + * NR_CPUS bits). Therefore you don't have to dereference cpumask_var_t.
> + *
> + *     cpumask_var_t tmpmask;
> + *     if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
> + *             return -ENOMEM;
> + *
> + *     var = *tmpmask;
> + *
> + * This code makes NR_CPUS length memcopy and bring memory corruption.

Maybe:

s/bring memory/brings to a memory/

?

> + * cpumask_copy() privide safe copy functionality.
s/privide/provides ?

>  */
>  #ifdef CONFIG_CPUMASK_OFFSTACK
>  typedef struct cpumask *cpumask_var_t;
> --
> 1.7.3.1
>
>
>
>

[PATCH v4] cpumask: add cpumask_var_t documentation

[KOSAKI Motohiro]

> > + * This code makes NR_CPUS length memcopy and bring memory corruption.
> 
> Maybe:
> 
> s/bring memory/brings to a memory/
> 
> ?
> 
> > + * cpumask_copy() privide safe copy functionality.
> s/privide/provides ?


Thank you.



>From 28549f1a51072a58bd661a409e378e36a97072a3 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Tue, 26 Apr 2011 19:58:39 +0900
Subject: [PATCH v4] cpumask: add cpumask_var_t documentation

cpumask_var_t has one nortable difference against cpumask_t.
This patch adds the explanation.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
 include/linux/cpumask.h |   14 ++++++++++++++
 1 files changed, 14 insertions(+), 0 deletions(-)

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 1e40dd0..8fcd82c 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -617,6 +617,20 @@ static inline size_t cpumask_size(void)
  *	  ... use 'tmpmask' like a normal struct cpumask * ...
  *
  *	free_cpumask_var(tmpmask);
+ *
+ *
+ * However, one notable exception is there. alloc_cpumask_var() allocates
+ * only nr_cpumask_bits bits (in the other hand, real cpumask_t always has
+ * NR_CPUS bits). Therefore you don't have to dereference cpumask_var_t.
+ *
+ *	cpumask_var_t tmpmask;
+ *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ *		return -ENOMEM;
+ *
+ *	var = *tmpmask;
+ *
+ * This code makes NR_CPUS length memcopy and brings to a memory corruption.
+ * cpumask_copy() provide safe copy functionality.
  */
 #ifdef CONFIG_CPUMASK_OFFSTACK
 typedef struct cpumask *cpumask_var_t;
-- 
1.7.3.1