From: Eric Dumazet <eric.dumazet@gmail.com>
To: Tom Herbert <therbert@google.com>
Cc: David Laight <David.Laight@aculab.com>,
David Miller <davem@davemloft.net>,
Linux Netdev List <netdev@vger.kernel.org>,
"Brandeburg, Jesse" <jesse.brandeburg@intel.com>
Subject: Re: [PATCH 1/2] net: Toeplitz library functions
Date: Tue, 24 Sep 2013 08:29:24 -0700 [thread overview]
Message-ID: <1380036564.3165.76.camel@edumazet-glaptop> (raw)
In-Reply-To: <CA+mtBx_y0hK69jcdY5e0MUzAa8hEPwExBDuK9Dn4Ceo5Hkn_iw@mail.gmail.com>
On Tue, 2013-09-24 at 08:22 -0700, Tom Herbert wrote:
> Assuming skb_rx_hash does symmetric calculation is currently
> incorrect. For instance, looks like tun.c is trying to implement a
> sort of 'flow director' logic to pair TX queues and RX queues using
> skb_get_rxhash an expecting that the value is calculated
> symmetrically. If HW is providing RX hash, this is broken and we'll
> never match the flows. We could either recompute the hash in SW or
> try to match HW hash.
Its not incorrect, its an implementation choice.
Its software in linux, we do not have to care of how its done in
hardware.
This is done to reduce conntracking cost, in case RPS is used on a
router : Same cpu will process frames in both ways.
But conntracking does not 'rely' on rxhash being symmetric, thats an
optimization to have better data locality.
next prev parent reply other threads:[~2013-09-24 15:29 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-23 22:41 [PATCH 1/2] net: Toeplitz library functions Tom Herbert
2013-09-24 0:03 ` Eric Dumazet
2013-09-24 1:39 ` David Miller
2013-09-24 2:30 ` Hannes Frederic Sowa
2013-09-24 3:35 ` Hannes Frederic Sowa
2013-09-24 5:38 ` Eric Dumazet
2013-09-24 5:45 ` Hannes Frederic Sowa
2013-09-24 13:19 ` [PATCH] net: net_secret should not depend on TCP Eric Dumazet
2013-09-24 15:13 ` Hannes Frederic Sowa
2013-09-24 15:22 ` Eric Dumazet
2013-09-24 15:28 ` Hannes Frederic Sowa
2013-09-24 15:46 ` Eric Dumazet
2013-09-24 23:51 ` Hannes Frederic Sowa
2013-09-28 22:20 ` David Miller
2013-09-25 9:00 ` [PATCH RFC] random: introduce get_random_bytes_busy_wait_initialized Hannes Frederic Sowa
2013-09-25 12:06 ` Eric Dumazet
2013-09-25 13:35 ` Hannes Frederic Sowa
2013-10-02 15:10 ` Theodore Ts'o
2013-10-02 17:18 ` Hannes Frederic Sowa
2013-10-02 19:40 ` Theodore Ts'o
2013-09-24 16:01 ` [PATCH 1/2] net: Toeplitz library functions Hannes Frederic Sowa
2013-09-24 16:14 ` Eric Dumazet
2013-09-24 16:35 ` Tom Herbert
2013-09-24 16:46 ` Eric Dumazet
2013-09-24 17:02 ` Ben Hutchings
2013-09-24 17:03 ` Tom Herbert
2013-09-24 17:34 ` Eric Dumazet
2013-09-24 17:37 ` Rick Jones
2013-09-24 17:44 ` Eric Dumazet
2013-09-24 18:02 ` Tom Herbert
2013-09-24 18:48 ` David Miller
2013-09-24 19:42 ` Hannes Frederic Sowa
2013-09-24 8:32 ` David Laight
2013-09-24 12:24 ` Eric Dumazet
2013-09-24 15:22 ` Tom Herbert
2013-09-24 15:29 ` Eric Dumazet [this message]
2013-09-24 15:39 ` David Miller
2013-09-24 15:54 ` Tom Herbert
2013-09-24 16:00 ` Hannes Frederic Sowa
2013-09-24 16:10 ` Eric Dumazet
2013-09-24 18:03 ` David Miller
2013-09-24 18:06 ` Tom Herbert
2013-09-24 18:10 ` Ben Hutchings
2013-09-24 18:24 ` Tom Herbert
2013-09-24 19:14 ` Eric Dumazet
2013-09-24 18:49 ` David Miller
2013-09-24 18:48 ` Jesse Brandeburg
2013-09-24 19:04 ` Tom Herbert
2013-09-24 16:38 ` Ben Hutchings
2013-09-24 16:32 ` Ben Hutchings
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=1380036564.3165.76.camel@edumazet-glaptop \
--to=eric.dumazet@gmail.com \
--cc=David.Laight@aculab.com \
--cc=davem@davemloft.net \
--cc=jesse.brandeburg@intel.com \
--cc=netdev@vger.kernel.org \
--cc=therbert@google.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