1994-09-16 - Re: thoughts on RC4

Header Data

From: “Perry E. Metzger” <perry@imsi.com>
To: jim@rand.org
Message Hash: 112b7b7f9bb964c84b9c6b7fdbadeb116bccfa2f5c17ac4f630e38d58c692e2a
Message ID: <9409161931.AA06647@snark.imsi.com>
Reply To: <9409161923.AA06121@mycroft.rand.org>
UTC Datetime: 1994-09-16 19:31:39 UTC
Raw Date: Fri, 16 Sep 94 12:31:39 PDT

Raw message

From: "Perry E. Metzger" <perry@imsi.com>
Date: Fri, 16 Sep 94 12:31:39 PDT
To: jim@rand.org
Subject: Re: thoughts on RC4
In-Reply-To: <9409161923.AA06121@mycroft.rand.org>
Message-ID: <9409161931.AA06647@snark.imsi.com>
MIME-Version: 1.0
Content-Type: text/plain



Jim Gillogly says:
> Is mbps megabits per second?

Yes. John Ioannidis has gotten the code up to 24mbit/sec on
SparcStation IIs.

> The original code plods along on my 50 Mhz '486 laptop (Borland C++ Pro)
> at a paltry 1.43mbits/s.  Turning the inner loop into obfuscated C picks up
> a little to 3.84mbits/s, and doing it with 8086-compatible assembler
> yields only 8.40mbits/s.

A 50 Mhz '486 shouldn't be that far off a SparcStation if you are
operating in the right mode...

You don't have to get very obfuscated, but moving the swap in line,
doing a bit of unrolling and playing some games with word operations
can get you pretty far...

Perry





Thread