From: Eric Dumazet <eric.dumazet@gmail.com>
To: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: "Kirsher, Jeffrey T" <jeffrey.t.kirsher@intel.com>,
"davem@davemloft.net" <davem@davemloft.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"gospo@redhat.com" <gospo@redhat.com>,
"bphilips@novell.com" <bphilips@novell.com>,
"Skidmore, Donald C" <donald.c.skidmore@intel.com>
Subject: Re: [net-next-2.6 PATCH 2/5] ixgbe: drop support for UDP in RSS hash generation
Date: Tue, 20 Jul 2010 18:39:37 +0200 [thread overview]
Message-ID: <1279643977.2498.89.camel@edumazet-laptop> (raw)
In-Reply-To: <4C45CAC6.1050006@intel.com>
Le mardi 20 juillet 2010 à 09:11 -0700, Alexander Duyck a écrit :
> The packets will still be hashed on source and destination IPv4/IPv6
> addresses. The change just drops reading the UDP source/destination
> ports since in the case of fragmented packets they are not available and
> as such were being parsed as IPv4/IPv6 packets. By making this change
> the queue selection is consistent between all packets in the UDP stream.
>
Excellent, this is perfect IMHO.
> The only regression I would expect to see would be in testing between
> two fixed systems since the IP addresses of the two systems would be
> fixed and so running multiple flows between the two would yield the same
> RSS hash for multiple UDP streams. As long as multiple ip addresses
> are used you should see multiple RSS hashes generated and as such the
> load should still be distributed.
>
Ack. Fortunately, one can still use RPS to spread load onto multiple
cpus in this case.
This until ixgpe fills skb->rxhash with a non null value.
If it happens one day, we shall remind _not_ filling it for UDP packets.
BTW, this reminds me a netdev discussion we had for bnx2x
http://www.kerneltrap.com/mailarchive/linux-netdev/2010/4/23/6275415/thread
And now, I understand why Toepliz hash doesnt use src port/dst port,
since this is not available on fragments, obviously...
Thanks !
next prev parent reply other threads:[~2010-07-20 16:39 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-19 23:59 [net-next-2.6 PATCH 1/5] ixgbe: dcb, set DPF bit when PFC is enabled Jeff Kirsher
2010-07-19 23:59 ` [net-next-2.6 PATCH 2/5] ixgbe: drop support for UDP in RSS hash generation Jeff Kirsher
2010-07-20 3:24 ` David Miller
2010-07-20 6:07 ` Bill Fink
2010-07-20 6:30 ` Eric Dumazet
2010-07-20 6:39 ` David Miller
2010-07-20 6:39 ` Eric Dumazet
2010-07-20 6:44 ` David Miller
2010-07-20 16:11 ` Alexander Duyck
2010-07-20 16:39 ` Eric Dumazet [this message]
2010-07-19 23:59 ` [net-next-2.6 PATCH 3/5] ixgbe: properly toggling netdev feature flags when disabling FCoE Jeff Kirsher
2010-07-20 3:24 ` David Miller
2010-07-20 0:00 ` [net-next-2.6 PATCH 4/5] ixgbe: use GFP_ATOMIC when allocating FCoE DDP context from the dma pool Jeff Kirsher
2010-07-20 3:24 ` David Miller
2010-07-20 0:00 ` [net-next-2.6 PATCH 5/5] ixgbe: fix version string for ixgbe Jeff Kirsher
2010-07-20 3:24 ` David Miller
2010-07-20 3:24 ` [net-next-2.6 PATCH 1/5] ixgbe: dcb, set DPF bit when PFC is enabled 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=1279643977.2498.89.camel@edumazet-laptop \
--to=eric.dumazet@gmail.com \
--cc=alexander.h.duyck@intel.com \
--cc=bphilips@novell.com \
--cc=davem@davemloft.net \
--cc=donald.c.skidmore@intel.com \
--cc=gospo@redhat.com \
--cc=jeffrey.t.kirsher@intel.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