[PATCH 0/17] Add zinc using existing algorithm implementations

Riku Voipio riku.voipio at iki.fi
Tue Mar 26 02:46:03 PDT 2019


On Mon, Mar 25, 2019 at 09:10:08AM +0000, Pascal Van Leeuwen wrote:
> As someone who has been working on crypto acceleration hardware for the better
> part of the past 20 years, I feel compelled to respond to this, in defence of
> the crypto API (which we're really happy with ...).

Thanks for joining in.

> We have done plenty of measurements, on both power and performance, to prove
> you wrong there. Typically our HW needs *at least* a full order of a magnitude
> less power to do the actual work. The CPU load for handing the interrupts etc.
> tends to be around 20%. So assuming the CPU goes to sleep for the other 80%
> of the time,  the combined solution would need about 1/3rd of the power of a
> CPU only solution. It's one of our biggest selling points.

I actually have the kind of (relatively) low-power system with your crypto
accelerator IP. But it doesn't matter how great performance or energy saving
I would get from using crypto offloading. I will continue to use CPU only
solution for the foreseeable future.

Because of one tiny problem:

The firmware files for eip197(?) crypto accelerator are secrets we are not
allowed download anywhere from.

> Pascal van Leeuwen,
> Silicon IP Architect @ Inside Secure

Riku



More information about the linux-arm-kernel mailing list