netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Simon Horman <horms@verge.net.au>
To: Chen Gang <gang.chen@asianux.com>
Cc: David Miller <davem@davemloft.net>, netdev <netdev@vger.kernel.org>
Subject: Re: [Regression net-next] include/linux/skbuff.h: "1a37e41 net: Use 16bits for *_headers fields of struct skbuff"
Date: Fri, 31 May 2013 08:48:33 +0900	[thread overview]
Message-ID: <20130530234833.GD31349@verge.net.au> (raw)
In-Reply-To: <51A7362D.4070501@asianux.com>

On Thu, May 30, 2013 at 07:21:17PM +0800, Chen Gang wrote:
> Hello Maintainers:
> 
> When not define NET_SKBUFF_DATA_USES_OFFSET, 'sk_buff_data_t' will be a
> pointer, then the last parameter of memmove() in mangle_contents() will
> be a pointer in "net/netfilter/nf_nat_helper.c".
> 
> The reason is from the patch "1a37e41 net: Use 16bits for *_headers
> fields of struct skbuff", it changed the type of 'skb->network_header'
> for saving size, it needs additional improvement.
> 
> The related warning (make EXTRA_CFLAGS=-W ARCH=m68k allmodconfig):
>   arch/m68k/include/asm/string.h:61:14: note: expected ‘__kernel_size_t’ but argument is of type ‘sk_buff_data_t’
> 
> 
> in "net/netfilter/nf_nat_helper.c"

Hi Chen,

thanks for pointing this out. I am preparing a patch to resolve this
problem.


> 
>  91 /* Frobs data inside this packet, which is linear. */
>  92 static void mangle_contents(struct sk_buff *skb,
>  93                             unsigned int dataoff,
>  94                             unsigned int match_offset,
>  95                             unsigned int match_len,
>  96                             const char *rep_buffer,
>  97                             unsigned int rep_len)
>  98 {
>  99         unsigned char *data;
> 100 
> 101         BUG_ON(skb_is_nonlinear(skb));
> 102         data = skb_network_header(skb) + dataoff;
> 103 
> 104         /* move post-replacement */
> 105         memmove(data + match_offset + rep_len,
> 106                 data + match_offset + match_len,
> 107                 skb->tail - (skb->network_header + dataoff +
> 108                              match_offset + match_len));
> 109 
> 
> 
> in "include/linux/skbuff.h":
> 
>  409 struct sk_buff {
>  410         /* These two members must be first. */
>  411         struct sk_buff          *next;
>  412         struct sk_buff          *prev;
>  413 
>  414         ktime_t                 tstamp;
>  415 
>  416         struct sock             *sk;
>  417         struct net_device       *dev;
>  418 
>  ...
>  515         __be16                  inner_protocol;
>  516         __u16                   inner_transport_header;
>  517         __u16                   inner_network_header;
>  518         __u16                   inner_mac_header;
>  519         __u16                   transport_header;
>  520         __u16                   network_header;
>  521         __u16                   mac_header;
>  522         /* These elements must be at the end, see alloc_skb() for details.  */
>  523         sk_buff_data_t          tail;
>  524         sk_buff_data_t          end;
>  525         unsigned char           *head,
>  526                                 *data;
>  527         unsigned int            truesize;
>  528         atomic_t                users;
>  529 };
>  530 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

      reply	other threads:[~2013-05-30 23:47 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-30 11:21 [Regression net-next] include/linux/skbuff.h: "1a37e41 net: Use 16bits for *_headers fields of struct skbuff" Chen Gang
2013-05-30 23:48 ` Simon Horman [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130530234833.GD31349@verge.net.au \
    --to=horms@verge.net.au \
    --cc=davem@davemloft.net \
    --cc=gang.chen@asianux.com \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).