linux-crypto.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Zdenek Kaspar <zkaspar82@gmail.com>
To: linux-crypto@vger.kernel.org
Subject: Re: Best mode of operation for AES-128 on x86_64?
Date: Thu, 13 May 2010 03:25:58 +0200	[thread overview]
Message-ID: <4BEB5526.9000701@gmail.com> (raw)
In-Reply-To: <AANLkTikZrBQ8Va2oZAK-D1EVxt9wcniS5kf2W-GqcMWJ@mail.gmail.com>

Dne 13.5.2010 2:07, Pedro Ribeiro napsal(a):
> Hello,
> 
> I'm currently using a LUKS volume which contains an encrypted LVM with
> itself contains my root, home and swap partitions.
> 
> My processor is a core 2 duo 2.26ghz and I'm running x86_64. I'm using
> cbc-essiv:sha256 and AES with 128 bit key.
> 
> I was wondering, is CBC mode the fastest?
> I have a fast processor and can't really complain about the read/write
> speed, but I was wondering if I could gain any speed from counter mode
> or other mode.
> 
> I would also like to ask what are the real world speed gains when
> using AES-NI? Can anyone please point me to some benchmarks, if there
> are any?
> 
> Thank for the help,
> Pedro

Hi Pedro, you can get better speeds with lower security. But before you
change your working scheme (which is good imo, I prefer XTS..) do a
quick test and compare if it's worth the pain :) ie:

$ hdparm -t /dev/sda1
$ hdparm -t /dev/mapper/sda1_crypt

Here are some numbers you should check -
http://blog.wpkg.org/2009/04/23/cipher-benchmark-for-dm-crypt-luks

My guess for single drive you're fine with your setup..

>From measuring specific server-loads and the impact on CPU frequency
scaling with cpufreq-ondemand as governor I would recommend lowering
default threshold value (95) for this load on desktop/server. ie:

$ echo 50 > /sys/devices/system/cpu/cpufreq/ondemand/up_threshold

HTH, Z.

      reply	other threads:[~2010-05-13  1:30 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-05-13  0:07 Best mode of operation for AES-128 on x86_64? Pedro Ribeiro
2010-05-13  1:25 ` Zdenek Kaspar [this message]

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=4BEB5526.9000701@gmail.com \
    --to=zkaspar82@gmail.com \
    --cc=linux-crypto@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;
as well as URLs for NNTP newsgroup(s).