Re: [PATCH 1/2] net: Stop decapitating clones that have a head_frag

[Eric Dumazet <eric.dumazet@gmail.com>]

On Wed, 2012-05-02 at 20:38 -0700, Alexander Duyck wrote:
> This change is meant ot prevent stealing the skb->head to use as a page in
> the event that the skb->head was cloned.  This allows the other clones to
> track each other via shinfo->dataref.
> 
> Without this we break down to two methods for tracking the reference count,
> one being dataref, the other being the page count.  As a result it becomes
> difficult to track how many references there are to skb->head.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Cc: Eric Dumazet <edumazet@google.com>
> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> ---
> 
>  net/core/skbuff.c    |    2 +-
>  net/ipv4/tcp_input.c |    9 ++-------
>  2 files changed, 3 insertions(+), 8 deletions(-)
> 
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index 52ba2b5..8703754 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -1699,7 +1699,7 @@ static bool __skb_splice_bits(struct sk_buff *skb, struct pipe_inode_info *pipe,
>  			      struct splice_pipe_desc *spd, struct sock *sk)
>  {
>  	int seg;
> -	bool head_is_linear = !skb->head_frag;
> +	bool head_is_linear = !skb->head_frag || skb_cloned(skb);
>  

Can you chose a better name than head_is_linear maybe ?

>  	/* map the linear part :
>  	 * If skb->head_frag is set, this 'linear' part is backed
> diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
> index 2f696ef..c6f78e2 100644
> --- a/net/ipv4/tcp_input.c
> +++ b/net/ipv4/tcp_input.c
> @@ -4589,7 +4589,7 @@ copyfrags:
>  		to->data_len += len;
>  		goto merge;
>  	}
> -	if (from->head_frag) {
> +	if (from->head_frag && !skb_cloned(from)) {
>  		struct page *page;
>  		unsigned int offset;
>  
> @@ -4599,12 +4599,7 @@ copyfrags:
>  		offset = from->data - (unsigned char *)page_address(page);
>  		skb_fill_page_desc(to, skb_shinfo(to)->nr_frags,
>  				   page, offset, skb_headlen(from));
> -
> -		if (skb_cloned(from))
> -			get_page(page);
> -		else
> -			*fragstolen = true;
> -
> +		*fragstolen = true;
>  		delta = len; /* we dont know real truesize... */
>  		goto copyfrags;
>  	}

Thanks