netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
To: David Miller <davem@davemloft.net>
Cc: herbert@gondor.apana.org.au, netdev@vger.kernel.org
Subject: Re: skb_shared_info()
Date: Wed, 9 Aug 2006 09:35:24 +0400	[thread overview]
Message-ID: <20060809053524.GC17446@2ka.mipt.ru> (raw)
In-Reply-To: <20060808.175839.85396148.davem@davemloft.net>

On Tue, Aug 08, 2006 at 05:58:39PM -0700, David Miller (davem@davemloft.net) wrote:
> From: Herbert Xu <herbert@gondor.apana.org.au>
> Date: Wed, 9 Aug 2006 10:36:16 +1000
> 
> > I'm not sure whether the problem is where we store skb_shared_info,
> > or the fact that we can't put kmalloc'ed memory into
> > skb_shinfo->frags.
> 
> That's a good point.

It is second problem actually.
If we want to change memory usage in skbs, we have problems with
frag_list and place where shared info lives, since it's addon only for
1500 MTU does not change alignment.

> I guess we could do something like:
> 
> struct skb_frag_struct {
> 	union {
> 		struct page *page;
> 		void *data;
> 	} u;
> 	__u16 page_offset;
> 	__u16 size;
> };
> 
> and something clever like a special page_offset encoding
> means "use data, not page".

We can separate kmalloced data from page by using internal page structures
(lru pointers and PG_slab bit).

-- 
	Evgeniy Polyakov

  reply	other threads:[~2006-08-09  5:35 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-08-08 23:39 skb_shared_info() David Miller
2006-08-09  0:36 ` skb_shared_info() Herbert Xu
2006-08-09  0:58   ` skb_shared_info() David Miller
2006-08-09  5:35     ` Evgeniy Polyakov [this message]
2006-08-09  5:50       ` skb_shared_info() David Miller
2006-08-10 19:49 ` skb_shared_info() Evgeniy Polyakov
2006-08-11 14:00 ` skb_shared_info() Alexey Kuznetsov
2006-08-12  0:27   ` skb_shared_info() David Miller
2006-08-12  0:32     ` skb_shared_info() Herbert Xu
2006-08-13 13:04     ` skb_shared_info() Alexey Kuznetsov
2006-08-14  8:01       ` skb_shared_info() Evgeniy Polyakov
2006-08-14 13:19         ` skb_shared_info() Alexey Kuznetsov
2006-08-14 13:38           ` skb_shared_info() Evgeniy Polyakov
2006-08-15 11:30             ` skb_shared_info() Alexey Kuznetsov
2006-08-14  5:04 ` skb_shared_info() Andi Kleen
2006-08-14  7:29   ` skb_shared_info() Herbert Xu
2006-08-14  7:45     ` skb_shared_info() Andi Kleen
2006-08-14  7:50       ` skb_shared_info() Herbert Xu
2006-08-14  8:18         ` skb_shared_info() Andi Kleen

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=20060809053524.GC17446@2ka.mipt.ru \
    --to=johnpol@2ka.mipt.ru \
    --cc=davem@davemloft.net \
    --cc=herbert@gondor.apana.org.au \
    --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).