From: Christoph Hellwig <hch@lst.de>
To: David Miller <davem@davemloft.net>
Cc: shemminger@osdl.org, hch@lst.de, netdev@vger.kernel.org
Subject: Re: [PATCH 2/2] correct dev_alloc_skb kerneldoc
Date: Sat, 8 Jul 2006 12:48:12 +0200 [thread overview]
Message-ID: <20060708104811.GB31751@lst.de> (raw)
In-Reply-To: <20060707.165527.77037268.davem@davemloft.net>
On Fri, Jul 07, 2006 at 04:55:27PM -0700, David Miller wrote:
> From: Stephen Hemminger <shemminger@osdl.org>
> Date: Fri, 7 Jul 2006 15:52:55 -0700
>
> > What is the point of dev_alloc_skb anyway? all it does is add header space.
>
> In stone-age times it actually had specific semantics, but yes today
> it is just a synonym.
>
> It's going to be hard to get rid of it, every single network driver
> out there references it.
On powerpc64 it actually makes a difference because it now allocates the
right headerspace as pad to ge the higher layer protocol headers aligned
for best dma performance.
Now what lead me up to this patch is something else. On the Cell
hardware we badly need to allocate rx buffers on the local node, because
the access to memory on a different node is horrible slow (thank rambus!)
My plan is to give dev_alloc_skb a struct netdevice * argument and
introduce a alloc_netdev_node so the driver can tell what node the
device is on. This gives a significant speedup for cell. I already
have this implemented in fact but only converted a handfull of drivers.
Does this approach sound okay?
next prev parent reply other threads:[~2006-07-08 10:48 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-07-07 9:09 [PATCH 2/2] correct dev_alloc_skb kerneldoc Christoph Hellwig
2006-07-07 22:52 ` Stephen Hemminger
2006-07-07 23:55 ` David Miller
2006-07-08 3:11 ` Herbert Xu
2006-07-08 10:48 ` Christoph Hellwig [this message]
2006-07-08 20:37 ` David Miller
2006-07-13 20:36 ` Christoph Hellwig
2006-07-13 20:47 ` Randy.Dunlap
2006-07-13 20:47 ` David Miller
2006-07-14 15:24 ` chas williams - CONTRACTOR
2006-07-13 21:24 ` Stephen Hemminger
2006-07-13 21:29 ` Christoph Hellwig
2006-07-13 22:01 ` David Miller
2006-07-14 7:27 ` Samuel Ortiz
2006-07-14 6:49 ` Christoph Hellwig
2006-07-24 22:31 ` David Miller
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=20060708104811.GB31751@lst.de \
--to=hch@lst.de \
--cc=davem@davemloft.net \
--cc=netdev@vger.kernel.org \
--cc=shemminger@osdl.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).