Re: [PATCH] [BLUETOOTH] kmalloc + memset -> kzalloc conversion

[Paulo Marques <pmarques@grupopie.com>]

Deepak Saxena wrote:
> On Oct 12 2005, at 00:52, Marcel Holtmann was caught saying:
> 
>>Hi Andrew,
>>
>>>Confused.  This patch changes lots of block code, not bluetooth.
>>
>>I know. This is what I already mailed Deepak, but he never replied.
> 
> Sorry, got lost in the mailbox. I think I was not paying attention to
> my tab completion and included the wrong patch. Proper patch follows.
> 
[...]
> --- a/drivers/bluetooth/hci_usb.c
> +++ b/drivers/bluetooth/hci_usb.c
> @@ -134,10 +134,9 @@ static struct usb_device_id blacklist_id
>  
>  static struct _urb *_urb_alloc(int isoc, unsigned int __nocast gfp)
>  {
> -	struct _urb *_urb = kmalloc(sizeof(struct _urb) +
> +	struct _urb *_urb = kzalloc(sizeof(struct _urb) +
>  				sizeof(struct usb_iso_packet_descriptor) * isoc, gfp);
>  	if (_urb) {
> -		memset(_urb, 0, sizeof(*_urb));
>  		usb_init_urb(&_urb->urb);
>  	}
>  	return _urb;

This one doesn't keep the exact same behavior as before, as it is 
zeroing more memory than it did.

If this is not a performance critical path, then I guess it's ok (code 
size reduction, and all).

I just wanted to call some attention on this so that someone more 
knowledgeable than me in the bluetooth ways can make sure it's ok.

-- 
Paulo Marques - www.grupopie.com

The rule is perfect: in all matters of opinion our
adversaries are insane.
Mark Twain