Re: [PATCH 0/4] skb paged fragment destructors
From: Eric Dumazet <hidden>
Date: 2011-12-21 14:02:22
From: Eric Dumazet <hidden>
Date: 2011-12-21 14:02:22
Le mercredi 21 décembre 2011 à 13:48 +0000, Ian Campbell a écrit :
On Wed, 2011-12-21 at 12:30 +0000, Eric Dumazet wrote:quoted
Le mercredi 21 décembre 2011 à 11:18 +0000, Ian Campbell a écrit :quoted
An order 1 allocation is in multiples of PAGE_SIZE, isn't it, even though they happen to be contiguous?Really an order-1 allocation allocates one page, a compound one.Oh, right, I see what you mean. You snipped the question about the + 2, I take it you have no idea what it is all about either?
No idea on this +2 point. I know some hardwares have limits on a fragment length (for example IGB uses IGB_MAX_DATA_PER_TXD limit), but I dont know why we add extra frags in generic skb.