From: Eric Dumazet <dada1@cosmosbay.com>
To: Robert Olsson <robert@robur.slu.se>
Cc: Terry <hanfang@baidu.com>,
Robert Olsson <Robert.Olsson@data.slu.se>,
Stephen Hemminger <shemminger@vyatta.com>,
Jeff Garzik <jgarzik@pobox.com>,
netdev@vger.kernel.org
Subject: Re: [PATCH] sky2: skb recycling
Date: Tue, 21 Oct 2008 14:42:37 +0200 [thread overview]
Message-ID: <48FDCE3D.9050801@cosmosbay.com> (raw)
In-Reply-To: <18685.51303.285228.177350@robur.slu.se>
Robert Olsson a écrit :
> Terry writes:
> > Hi
> > yeah. In the forwarding scenario , skb recycling should boost the
> > performance by avoiding slowpath slub alloc/free .But if using
> > multiqueue hardware and assigning proper cpu affinities could make it
> > always in the fastpath of slub alloc/free, which is faster?
>
> For perfect setup with monotone work like forwarding you can make the skb
> recycling a little faster.
>
> If you have many active NIC's and the recycling list is per device the situation
> it's probably different...
>
I suspect all this skb recycling stuff in forwarding workloads is defeated anyway...
By the copybreak feature... (RX_COPY_THRESHOLD = 256 for example on tg3)
Checking drivers/net/tg3.c, we can even see that in case of copying to a smaller skb,
we allocate it with a call to netdev_alloc_skb(), while it would be better
to use a plain alloc_skb() ...
next prev parent reply other threads:[~2008-10-21 12:42 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-21 2:09 [PATCH] sky2: skb recycling Stephen Hemminger
2008-10-21 5:18 ` David Miller
2008-10-21 19:59 ` Rick Jones
2008-10-21 20:15 ` Eric Dumazet
2008-10-21 20:38 ` Stephen Hemminger
2008-10-22 23:58 ` Rick Jones
2008-10-23 0:05 ` Stephen Hemminger
2008-10-23 0:17 ` Rick Jones
2008-10-21 5:54 ` Eric Dumazet
2008-10-21 8:25 ` Robert Olsson
2008-10-21 8:49 ` Terry
2008-10-21 12:17 ` Robert Olsson
2008-10-21 12:34 ` Terry
2008-10-21 12:42 ` Eric Dumazet [this message]
2008-10-21 13:04 ` Robert Olsson
2008-10-21 9:53 ` Eric Dumazet
2008-10-21 11:58 ` Robert Olsson
2008-10-23 5:28 ` David Miller
2008-10-21 15:09 ` Stephen Hemminger
2008-10-22 11:03 ` Jeff Garzik
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=48FDCE3D.9050801@cosmosbay.com \
--to=dada1@cosmosbay.com \
--cc=Robert.Olsson@data.slu.se \
--cc=hanfang@baidu.com \
--cc=jgarzik@pobox.com \
--cc=netdev@vger.kernel.org \
--cc=robert@robur.slu.se \
--cc=shemminger@vyatta.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.