On 5/18/07, Sam Ravnborg <[EMAIL PROTECTED]> wrote:
On Fri, May 18, 2007 at 05:10:02PM +0530, pradeep singh wrote:
> Hi ,
>
> following code snippet is self evident in what it does -
>
> truct net_device *alloc_netdev(int sizeof_priv, const char *name,
> void (*setup)(struct net_device *))
> {
> void *p;
> struct net_device *dev;
> int alloc_size;
>
> /* ensure 32-byte alignment of both the device and private area */
> alloc_size = (sizeof(*dev) + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST;
> alloc_size += sizeof_priv + NETDEV_ALIGN_CONST;
>
> p = kzalloc(alloc_size, GFP_KERNEL);
> if (!p) {
> printk(KERN_ERR "alloc_dev: Unable to allocate device.\n");
> return NULL;
> }
> [snip]
>
> Is the comment correct?
> Why is it 'ensure 32 byte alignment', instead of 32 bit alignment?
>
> What am i missing here?
Google turned up following useful link:
http://www.nabble.com/-RFC--changing-value-of-NETDEV_ALIGN-to-cacheline-size-t1620499.html
Thank you Sam.
This is really the answer i should have searched :-(.
A lesson to learn indeed.
Thank you once again.
--psr
snip:
The reason it's 32 is that old drivers depended on the
struct being at least 32-byte aligned because they would
embed structures DMA'd to/from the card in their private
area and just assumed that would be aligned enough for
the card's restrictions.
Sam
--
play the game
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
