[PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[SF Markus Elfring]

From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 20 Aug 2016 10:10:12 +0200

Reuse existing functionality from memdup_user() instead of keeping
duplicate source code.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/net/wan/cosa.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
index b87fe0a..02f5809 100644
--- a/drivers/net/wan/cosa.c
+++ b/drivers/net/wan/cosa.c
@@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
 	if (count > COSA_MTU)
 		count = COSA_MTU;
 	
-	/* Allocate the buffer */
-	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
-	if (kbuf == NULL) {
+	kbuf = memdup_user(buf, count);
+	if (IS_ERR(kbuf)) {
 		up(&chan->wsem);
-		return -ENOMEM;
-	}
-	if (copy_from_user(kbuf, buf, count)) {
-		up(&chan->wsem);
-		kfree(kbuf);
-		return -EFAULT;
+		return PTR_ERR(kbuf);
 	}
 	chan->tx_status=0;
 	cosa_start_tx(chan, kbuf, count);
-- 
2.9.3

[PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[SF Markus Elfring]

From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 20 Aug 2016 10:10:12 +0200

Reuse existing functionality from memdup_user() instead of keeping
duplicate source code.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/net/wan/cosa.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
index b87fe0a..02f5809 100644
--- a/drivers/net/wan/cosa.c
+++ b/drivers/net/wan/cosa.c
@@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
 	if (count > COSA_MTU)
 		count = COSA_MTU;
 	
-	/* Allocate the buffer */
-	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
-	if (kbuf = NULL) {
+	kbuf = memdup_user(buf, count);
+	if (IS_ERR(kbuf)) {
 		up(&chan->wsem);
-		return -ENOMEM;
-	}
-	if (copy_from_user(kbuf, buf, count)) {
-		up(&chan->wsem);
-		kfree(kbuf);
-		return -EFAULT;
+		return PTR_ERR(kbuf);
 	}
 	chan->tx_status=0;
 	cosa_start_tx(chan, kbuf, count);
-- 
2.9.3

Re: [PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[Christophe JAILLET]

Le 20/08/2016 à 10:25, SF Markus Elfring a écrit :
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Sat, 20 Aug 2016 10:10:12 +0200
>
> Reuse existing functionality from memdup_user() instead of keeping
> duplicate source code.
>
> This issue was detected by using the Coccinelle software.
>
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
>   drivers/net/wan/cosa.c | 12 +++---------
>   1 file changed, 3 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
> index b87fe0a..02f5809 100644
> --- a/drivers/net/wan/cosa.c
> +++ b/drivers/net/wan/cosa.c
> @@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
>   	if (count > COSA_MTU)
>   		count = COSA_MTU;
>   	
> -	/* Allocate the buffer */
> -	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
In this case, 'memdup_user()' has a different meaning, as GFP_DMA will 
no more be used for this memory allocation.

> -	if (kbuf == NULL) {
> +	kbuf = memdup_user(buf, count);
> +	if (IS_ERR(kbuf)) {
>   		up(&chan->wsem);
> -		return -ENOMEM;
> -	}
> -	if (copy_from_user(kbuf, buf, count)) {
> -		up(&chan->wsem);
> -		kfree(kbuf);
> -		return -EFAULT;
> +		return PTR_ERR(kbuf);
>   	}
>   	chan->tx_status=0;
>   	cosa_start_tx(chan, kbuf, count);



---
L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
https://www.avast.com/antivirus

Re: [PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[David Miller]

From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sat, 20 Aug 2016 10:42:54 +0200

> Le 20/08/2016 à 10:25, SF Markus Elfring a écrit :
>> @@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
>>   	if (count > COSA_MTU)
>>   		count = COSA_MTU;
>>   	
>> -	/* Allocate the buffer */
>> -	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
> In this case, 'memdup_user()' has a different meaning, as GFP_DMA will
> no more be used for this memory allocation.

Agreed, this transformation is not equivalent.

Markus, please do not make these changes so blindly.

Re: [PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[David Miller]

From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sat, 20 Aug 2016 10:42:54 +0200

> Le 20/08/2016 à 10:25, SF Markus Elfring a écrit :
>> @@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
>>   	if (count > COSA_MTU)
>>   		count = COSA_MTU;
>>   	
>> -	/* Allocate the buffer */
>> -	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
> In this case, 'memdup_user()' has a different meaning, as GFP_DMA will
> no more be used for this memory allocation.

Agreed, this transformation is not equivalent.

Markus, please do not make these changes so blindly.
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[Jan Kasprzak]

	Hello,

SF Markus Elfring wrote:
: From: Markus Elfring <elfring@users.sourceforge.net>
: Date: Sat, 20 Aug 2016 10:10:12 +0200
: 
: Reuse existing functionality from memdup_user() instead of keeping
: duplicate source code.
: 
: This issue was detected by using the Coccinelle software.

	What about the GFP_DMA attribute, which your patch deletes?
The buffer in question has to be ISA DMA-able.

	Anyway, COSA is an ancient ISA device, which is out of production
for at least 10 years. I don't have a box with ISA bus anymore, so I can't
even test the modifications of the wan/cosa.c (I do have a COSA board
itself, though :-).

	I suggest keeping the driver intact if possible, in case somebody
still uses it.

	Sincerely,

-Yenya

: Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
: ---
:  drivers/net/wan/cosa.c | 12 +++---------
:  1 file changed, 3 insertions(+), 9 deletions(-)
: 
: diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
: index b87fe0a..02f5809 100644
: --- a/drivers/net/wan/cosa.c
: +++ b/drivers/net/wan/cosa.c
: @@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
:  	if (count > COSA_MTU)
:  		count = COSA_MTU;
:  	
: -	/* Allocate the buffer */
: -	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
: -	if (kbuf == NULL) {
: +	kbuf = memdup_user(buf, count);
: +	if (IS_ERR(kbuf)) {
:  		up(&chan->wsem);
: -		return -ENOMEM;
: -	}
: -	if (copy_from_user(kbuf, buf, count)) {
: -		up(&chan->wsem);
: -		kfree(kbuf);
: -		return -EFAULT;
: +		return PTR_ERR(kbuf);
:  	}
:  	chan->tx_status=0;
:  	cosa_start_tx(chan, kbuf, count);
: -- 
: 2.9.3

-- 
| Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> |
| http://www.fi.muni.cz/~kas/                         GPG: 4096R/A45477D5 |
         Like most things in Windows, on the surface it looks great.
                         -- Jeremy Allison, A Tale of Two Standards

Re: [PATCH] wan-cosa: Use memdup_user() rather than duplicating its implementation

[Jan Kasprzak]

	Hello,

SF Markus Elfring wrote:
: From: Markus Elfring <elfring@users.sourceforge.net>
: Date: Sat, 20 Aug 2016 10:10:12 +0200
: 
: Reuse existing functionality from memdup_user() instead of keeping
: duplicate source code.
: 
: This issue was detected by using the Coccinelle software.

	What about the GFP_DMA attribute, which your patch deletes?
The buffer in question has to be ISA DMA-able.

	Anyway, COSA is an ancient ISA device, which is out of production
for at least 10 years. I don't have a box with ISA bus anymore, so I can't
even test the modifications of the wan/cosa.c (I do have a COSA board
itself, though :-).

	I suggest keeping the driver intact if possible, in case somebody
still uses it.

	Sincerely,

-Yenya

: Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
: ---
:  drivers/net/wan/cosa.c | 12 +++---------
:  1 file changed, 3 insertions(+), 9 deletions(-)
: 
: diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
: index b87fe0a..02f5809 100644
: --- a/drivers/net/wan/cosa.c
: +++ b/drivers/net/wan/cosa.c
: @@ -875,16 +875,10 @@ static ssize_t cosa_write(struct file *file,
:  	if (count > COSA_MTU)
:  		count = COSA_MTU;
:  	
: -	/* Allocate the buffer */
: -	kbuf = kmalloc(count, GFP_KERNEL|GFP_DMA);
: -	if (kbuf = NULL) {
: +	kbuf = memdup_user(buf, count);
: +	if (IS_ERR(kbuf)) {
:  		up(&chan->wsem);
: -		return -ENOMEM;
: -	}
: -	if (copy_from_user(kbuf, buf, count)) {
: -		up(&chan->wsem);
: -		kfree(kbuf);
: -		return -EFAULT;
: +		return PTR_ERR(kbuf);
:  	}
:  	chan->tx_status=0;
:  	cosa_start_tx(chan, kbuf, count);
: -- 
: 2.9.3

-- 
| Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> |
| http://www.fi.muni.cz/~kas/                         GPG: 4096R/A45477D5 |
         Like most things in Windows, on the surface it looks great.
                         -- Jeremy Allison, A Tale of Two Standards

Re: wan-cosa: Use memdup_user() rather than duplicating its implementation

[SF Markus Elfring]

> 	What about the GFP_DMA attribute, which your patch deletes?
> The buffer in question has to be ISA DMA-able.

Thanks for your constructive feedback.

Would you be interested in using a variant of the function "memdup_…"
with which the corresponding memory allocation option can be preserved?

Regards,
Markus

Re: wan-cosa: Use memdup_user() rather than duplicating its implementation

[SF Markus Elfring]

> 	What about the GFP_DMA attribute, which your patch deletes?
> The buffer in question has to be ISA DMA-able.

Thanks for your constructive feedback.

Would you be interested in using a variant of the function "memdup_…"
with which the corresponding memory allocation option can be preserved?

Regards,
Markus
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: wan-cosa: Use memdup_user() rather than duplicating its implementation

[Jan Kasprzak]

SF Markus Elfring wrote:
: > 	What about the GFP_DMA attribute, which your patch deletes?
: > The buffer in question has to be ISA DMA-able.
: 
: Thanks for your constructive feedback.
: 
: Would you be interested in using a variant of the function "memdup_…"
: with which the corresponding memory allocation option can be preserved?

	I am not sure that extending an in-kernel API just for one
legacy driver is what we want. As I said, I would prefer the driver
unchanged, if possible.

	Maybe it is the time for gradually phasing out ISA DMA support and
all the legacy drivers which use it?

	Sincerely,

-Yenya

-- 
| Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> |
| http://www.fi.muni.cz/~kas/                         GPG: 4096R/A45477D5 |
         Like most things in Windows, on the surface it looks great.
                         -- Jeremy Allison, A Tale of Two Standards

Re: wan-cosa: Use memdup_user() rather than duplicating its implementation

[Jan Kasprzak]

SF Markus Elfring wrote:
: > 	What about the GFP_DMA attribute, which your patch deletes?
: > The buffer in question has to be ISA DMA-able.
: 
: Thanks for your constructive feedback.
: 
: Would you be interested in using a variant of the function "memdup_…"
: with which the corresponding memory allocation option can be preserved?

	I am not sure that extending an in-kernel API just for one
legacy driver is what we want. As I said, I would prefer the driver
unchanged, if possible.

	Maybe it is the time for gradually phasing out ISA DMA support and
all the legacy drivers which use it?

	Sincerely,

-Yenya

-- 
| Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> |
| http://www.fi.muni.cz/~kas/                         GPG: 4096R/A45477D5 |
         Like most things in Windows, on the surface it looks great.
                         -- Jeremy Allison, A Tale of Two Standards
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html