netdev
[Top] [All Lists]

Re: [DEBUG]: sk_forward_alloc assertion failures

To: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Subject: Re: [DEBUG]: sk_forward_alloc assertion failures
From: "David S. Miller" <davem@xxxxxxxxxxxxx>
Date: Thu, 13 Jan 2005 19:44:51 -0800
Cc: sri@xxxxxxxxxx, anton@xxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <20050114021005.GA11962@gondor.apana.org.au>
References: <20050113171234.3fde0925.davem@davemloft.net> <Pine.LNX.4.58.0501131741430.25252@w-sridhar.beaverton.ibm.com> <20050114021005.GA11962@gondor.apana.org.au>
Sender: netdev-bounce@xxxxxxxxxxx
On Fri, 14 Jan 2005 13:10:05 +1100
Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> wrote:

> Since there is no reclaim for sk_wmem_queued it doesn't happen
> there.
> 
> So taking out the adjustments should not cause any problems of
> sk_wmem_queued, theoretically :)

If we take out the skb->truesize adjustment, as the test patch does,
then we must take out the sk_wmem_queued and sk_forward_alloc
adjustments, as the test patch also does. :-)

<Prev in Thread] Current Thread [Next in Thread>