Re: [patch 3/5] A-MSDU Rx aggregation support

[Randy Dunlap <randy.dunlap@oracle.com>]

mohamed wrote:
> On Mon, 2007-03-26 at 16:36 -0700, Randy Dunlap wrote:
>> On Mon, 26 Mar 2007 04:40:00 -0700 mohamed wrote:
>>
>>> Add A-MSDU Rx aggregation support.
>>>
>>> To support IEEE 802.11n, we need to be able to process A-MSDU frames. 
>>> The present of the HT control field indicates it is A-MSDU frame. 
>>> This patch adds support to discover and process A-MSDU frames.
>>>
>>> Signed-off-by: Mohamed Abbas <mabbas@linux.intel.com>

>>> +	while ((u8*)eth < skb->data + skb->len) {
>>> +        	unsigned int eth_len = sizeof(struct ethhdr) + 
>>> +						ntohs(eth->h_proto);
>>> +        
>>> +		frame = dev_alloc_skb(3000);
>> Why 3000?  can it be a well-defined #define??
> 3000 just a safe number I picked up. I will change to 
> dev_alloc_skb(local->hw.extra_tx_headroom + eth_len);
> Are any more space needed?

Nope, that looks fine.  Thanks.

-- 
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***