I've been experimenting with the LX800 and via C7 lately, I got some odd results for the C7 (openssl was faster with cryptodev after padlock loaded, but not nearly as fast as the padlock engine direct from ssl) I was surprised at the overhead in the cryptodev system, though I might have caused it.
The C7 testing was done with pfSense: http://www.docunext.com/resources/mediawiki/index.php/PfSense_test_results_of_the_padlock_kernel_driver_on_a_VIA_C7 The LX800 was done with debian patched with OCF: http://www.docunext.com/resources/mediawiki/index.php/My_Notes_on_Patching_2.6.22_with_OCF Both look awesome, as both have hwrng as well. The C7 is more expensive from what I've seen but is available on bigger boards that can take more memory. Depends on what you need I guess. On 9/14/07, Holger Bauer <[EMAIL PROTECTED]> wrote: > Seth benched a soekris 55xx and got around wirespeed. alix should be the > same as the cpu and nics are the same. There are no benches for > ipsecencapsulation yet though. > > Holger > > > -----Original Message----- > > From: Eugen Leitl [mailto:[EMAIL PROTECTED] > > Sent: Friday, September 14, 2007 1:31 PM > > To: [email protected] > > Subject: Re: [pfSense-discussion] WRAP Support & Images > > > > On Fri, Sep 14, 2007 at 12:14:30PM +0200, Holger Bauer wrote: > > > fyi: http://blog.pfsense.org/?p=139 > > > > How much throughput do the new Geode LX boards have? > > RNG and AES integrated sound really good, in the same > > physical and power footprint. > > > > -- > > Eugen* Leitl <a href="http://leitl.org";>leitl</a> > > http://leitl.org > > ______________________________________________________________ > > ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org > > 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE > > > > > -- My Blogs: http://www.docunext.com/ http://www.albertlash.com/
