[PATCH] block: replace kmalloc and then memcpy with kmemdup

[PATCH] block: replace kmalloc and then memcpy with kmemdup

[Mihnea Dobrescu-Balaur]

Signed-off-by: Mihnea Dobrescu-Balaur <mihneadb@gmail.com>
---
 drivers/block/xen-blkfront.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index c3dae2e..9620644 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -1203,11 +1203,10 @@ static int blkif_recover(struct blkfront_info *info)
 	int j;
 
 	/* Stage 1: Make a safe copy of the shadow state. */
-	copy = kmalloc(sizeof(info->shadow),
+	copy = kmemdup(info->shadow, sizeof(info->shadow),
 		       GFP_NOIO | __GFP_REPEAT | __GFP_HIGH);
 	if (!copy)
 		return -ENOMEM;
-	memcpy(copy, info->shadow, sizeof(info->shadow));
 
 	/* Stage 2: Set up free list. */
 	memset(&info->shadow, 0, sizeof(info->shadow));
-- 
1.7.10.4

Re: [PATCH] block: replace kmalloc and then memcpy with kmemdup

[Konrad Rzeszutek Wilk]

On Mon, Mar 11, 2013 at 01:23:36PM +0200, Mihnea Dobrescu-Balaur wrote:

Are there performance improvements to doing it this way?

> Signed-off-by: Mihnea Dobrescu-Balaur <mihneadb@gmail.com>
> ---
>  drivers/block/xen-blkfront.c |    3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
> index c3dae2e..9620644 100644
> --- a/drivers/block/xen-blkfront.c
> +++ b/drivers/block/xen-blkfront.c
> @@ -1203,11 +1203,10 @@ static int blkif_recover(struct blkfront_info *info)
>  	int j;
>  
>  	/* Stage 1: Make a safe copy of the shadow state. */
> -	copy = kmalloc(sizeof(info->shadow),
> +	copy = kmemdup(info->shadow, sizeof(info->shadow),
>  		       GFP_NOIO | __GFP_REPEAT | __GFP_HIGH);
>  	if (!copy)
>  		return -ENOMEM;
> -	memcpy(copy, info->shadow, sizeof(info->shadow));
>  
>  	/* Stage 2: Set up free list. */
>  	memset(&info->shadow, 0, sizeof(info->shadow));
> -- 
> 1.7.10.4
> 

Re: [PATCH] block: replace kmalloc and then memcpy with kmemdup

[Daniel Baluta]

Hi Konrad,

The performance is the same using kmemdup, but we have the following advantages:
* code is cleaner
* kmemdup adds additional debugging info useful for tracking the real
place where memory was allocated [1] [2].

thanks,
Daniel.

[1] http://lxr.linux.no/#linux+v3.8.2/mm/util.c#L61
[2] http://lxr.linux.no/#linux+v3.8.2/include/linux/slab.h#L363

On Mon, Mar 11, 2013 at 4:54 PM, Konrad Rzeszutek Wilk
<konrad.wilk@oracle.com> wrote:
> On Mon, Mar 11, 2013 at 01:23:36PM +0200, Mihnea Dobrescu-Balaur wrote:
>
> Are there performance improvements to doing it this way?
>
>> Signed-off-by: Mihnea Dobrescu-Balaur <mihneadb@gmail.com>
>> ---
>>  drivers/block/xen-blkfront.c |    3 +--
>>  1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
>> index c3dae2e..9620644 100644
>> --- a/drivers/block/xen-blkfront.c
>> +++ b/drivers/block/xen-blkfront.c
>> @@ -1203,11 +1203,10 @@ static int blkif_recover(struct blkfront_info *info)
>>       int j;
>>
>>       /* Stage 1: Make a safe copy of the shadow state. */
>> -     copy = kmalloc(sizeof(info->shadow),
>> +     copy = kmemdup(info->shadow, sizeof(info->shadow),
>>                      GFP_NOIO | __GFP_REPEAT | __GFP_HIGH);
>>       if (!copy)
>>               return -ENOMEM;
>> -     memcpy(copy, info->shadow, sizeof(info->shadow));
>>
>>       /* Stage 2: Set up free list. */
>>       memset(&info->shadow, 0, sizeof(info->shadow));
>> --
>> 1.7.10.4
>>
> _______________________________________________
> Virtualization mailing list
> Virtualization@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/virtualization