From: Stephen Hemminger <shemminger@osdl.org>
To: "Michael S. Tsirkin" <mst@mellanox.co.il>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
netdev@vger.kernel.org, openib-general@openib.org,
Roland Dreier <rolandd@cisco.com>
Subject: Re: Dropping NETIF_F_SG since no checksum feature.
Date: Tue, 10 Oct 2006 10:43:15 -0700 [thread overview]
Message-ID: <20061010104315.61540986@freekitty> (raw)
In-Reply-To: <20061010144330.GA28175@mellanox.co.il>
On Tue, 10 Oct 2006 16:43:30 +0200
"Michael S. Tsirkin" <mst@mellanox.co.il> wrote:
> Quoting r. Stephen Hemminger <shemminger@osdl.org>:
> > Subject: Re: Dropping NETIF_F_SG since no checksum feature.
> >
> > On Mon, 9 Oct 2006 19:47:05 +0200
> > "Michael S. Tsirkin" <mst@mellanox.co.il> wrote:
> >
> > > Hi!
> > > I'm trying to build a network device driver supporting a very large MTU (around 64K)
> > > on top of an infiniband connection, and I've hit a couple of issues I'd
> > > appreciate some feedback on:
> > >
> > > 1. On the send side,
> > > I've set NETIF_F_SG, but hardware does not support checksum offloading,
> > > and I see "dropping NETIF_F_SG since no checksum feature" warning,
> > > and I seem to be getting large packets all in one chunk.
> > > The reason I've set NETIF_F_SG, is because I'm concerned that under real life
> > > stress Linux won't be able to allocate 64K of continuous memory.
> > >
> > > Is this concern of mine valid? I saw in-tree drivers allocating at least 8K.
> > > What's the best way to enable S/G on send side?
> > > Is checksum offloading really required for S/G?
> >
> > Yes, in the current implementation, Linux needs checksum offload. But there
> > is no reason, your driver can't compute the checksum in software.
>
> Are there drivers that do this already? Couldn't find any such beast ...
dev_queue_xmit() does it, all you need to do in your driver is:
/* If packet is not checksummed and device does not support
* checksumming for this protocol, complete checksumming here.
*/
if (skb->ip_summed == CHECKSUM_PARTIAL) {
if (skb_checksum_help(skb))
goto error_recovery
}
> I'm worried whether an extra pass over data won't eat up all of
> the performance gains I get from the large MTU ...
Yup, the cost is in touching the data, not in the copy.
> > > What are the helpers legal for fragmented skb?
>
> BTW, I found skb_put_frags in sky2 which seems generic enough - I even wander
> why isn't this in net/core.
>
Only because I just wrote it for my needs. If you need it, then it
can be moved to skbuff.c
--
Stephen Hemminger <shemminger@osdl.org>
next prev parent reply other threads:[~2006-10-10 17:46 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-10-09 17:47 Dropping NETIF_F_SG since no checksum feature Michael S. Tsirkin
2006-10-09 16:50 ` Stephen Hemminger
2006-10-10 14:43 ` Michael S. Tsirkin
2006-10-10 17:43 ` Stephen Hemminger [this message]
2006-10-11 0:13 ` Michael S. Tsirkin
2006-10-11 0:15 ` Roland Dreier
2006-10-11 0:26 ` Michael S. Tsirkin
2006-10-11 3:33 ` Roland Dreier
2006-10-11 3:36 ` David Miller
2006-10-11 3:42 ` Roland Dreier
2006-10-11 3:45 ` David Miller
2006-10-11 3:49 ` Roland Dreier
2006-10-11 3:50 ` David Miller
2006-10-11 2:15 ` David Miller
2006-10-11 9:05 ` Michael S. Tsirkin
2006-10-11 9:09 ` Steven Whitehouse
2006-10-11 15:01 ` Michael S. Tsirkin
2006-10-11 20:11 ` Steven Whitehouse
2006-10-11 20:52 ` Michael S. Tsirkin
2006-10-11 20:57 ` Stephen Hemminger
2006-10-11 21:23 ` Michael S. Tsirkin
2006-10-11 21:29 ` Stephen Hemminger
2006-10-11 21:42 ` Michael S. Tsirkin
2006-10-11 21:41 ` David Miller
2006-10-12 19:12 ` Michael S. Tsirkin
2006-10-13 4:22 ` David Miller
2006-10-13 6:17 ` Michael S. Tsirkin
2006-10-11 20:52 ` David Miller
2006-10-11 21:11 ` Michael S. Tsirkin
2006-10-11 9:20 ` David Miller
2006-10-11 9:46 ` Michael S. Tsirkin
2006-10-11 18:21 ` [openib-general] " Michael Krause
2006-10-11 13:11 ` [RFC] Question about potential problem in net/ipv4/route.c Eric Dumazet
2006-10-12 5:05 ` David Miller
2006-10-12 5:31 ` Patrick McHardy
2006-10-12 5:54 ` David Miller
2006-10-12 5:48 ` Eric Dumazet
2006-10-12 6:02 ` David Miller
2006-10-12 6:10 ` Patrick McHardy
2006-10-12 6:25 ` David Miller
2006-10-12 6:35 ` Eric Dumazet
2006-10-12 7:48 ` David Miller
2006-10-16 9:00 ` [PATCH] NET : Suspicious locking in reqsk_queue_hash_req() Eric Dumazet
2006-10-16 9:07 ` Eric Dumazet
2006-10-16 16:16 ` Arnaldo Carvalho de Melo
2006-10-16 16:56 ` Eric Dumazet
2006-10-16 17:39 ` Eric Dumazet
2006-10-16 20:41 ` 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=20061010104315.61540986@freekitty \
--to=shemminger@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@mellanox.co.il \
--cc=netdev@vger.kernel.org \
--cc=openib-general@openib.org \
--cc=rolandd@cisco.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 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).