All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Miller <davem@davemloft.net>
To: andi@firstfloor.org
Cc: therbert@google.com, netdev@vger.kernel.org
Subject: Re: [PATCH] Software receive packet steering
Date: Mon, 20 Apr 2009 03:46:28 -0700 (PDT)	[thread overview]
Message-ID: <20090420.034628.224469114.davem@davemloft.net> (raw)
In-Reply-To: <87eivnpqde.fsf@basil.nowhere.org>

From: Andi Kleen <andi@firstfloor.org>
Date: Mon, 20 Apr 2009 12:32:29 +0200

> Tom Herbert <therbert@google.com> writes:
> 
>> +got_hash:
>> +	hash %= cpus_weight_nr(mask);
> 
> That looks rather heavyweight even on modern CPUs. I bet it's 40-50+ cycles
> alone forth the hweight and the division. Surely that can be done better?

The standard way to do this is to compute a 32-bit jenkins
hash, and do a 64-bit multiply of this value with a suitable
reciprocol.

This is what skb_tx_hash() is doing in net/core/dev.c

  reply	other threads:[~2009-04-20 10:46 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-08 22:48 [PATCH] Software receive packet steering Tom Herbert
2009-04-08 23:08 ` Stephen Hemminger
2009-04-08 23:09 ` Stephen Hemminger
2009-04-08 23:15   ` David Miller
2009-04-09 16:43     ` Tom Herbert
2009-04-09 18:23       ` Ben Hutchings
2009-04-09 21:17       ` David Miller
2009-04-09  0:36 ` David Miller
2009-04-09  4:40   ` Tom Herbert
2009-04-09  5:24     ` David Miller
2009-04-20 10:32 ` Andi Kleen
2009-04-20 10:46   ` David Miller [this message]
2009-04-21  3:26   ` Tom Herbert
2009-04-21  9:48     ` Eric Dumazet
2009-04-21 15:46       ` Stephen Hemminger
2009-04-21 18:52         ` Tom Herbert
2009-04-22  9:21           ` David Miller
2009-04-22 15:46             ` Tom Herbert
2009-04-22 18:49             ` Rick Jones
2009-04-22 20:44             ` Jesper Dangaard Brouer
2009-04-23  6:58               ` Jens Axboe
2009-04-23  7:25                 ` David Miller
2009-04-23  7:29                   ` Jens Axboe
2009-04-23  9:12               ` Jens Laas
2009-04-22 14:33         ` Martin Josefsson
2009-04-23  7:34           ` 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=20090420.034628.224469114.davem@davemloft.net \
    --to=davem@davemloft.net \
    --cc=andi@firstfloor.org \
    --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 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.