Re: [DEBUG]: sk_forward_alloc assertion failures

["David S. Miller" <davem@xxxxxxxxxxxxx>]

On Sat, 15 Jan 2005 16:45:47 +1100
Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> wrote:

> > +           if (sk->sk_forward_alloc < copy &&
> > +               !sk_stream_mem_schedule(sk, copy, 0))
> 
> You want to have sk->sk_allocation here instead of 0.

The third argument to sk_stream_mem_schedule() is "int kind"
which is to be set to 1 for receive packet allocations and 0
for send packet allocations.

Why would I want to use a sk_allocation gfp mask here? :-)