Subject: Re: AMD Geode LX Security Block (was Re: NetBSD on the Soekris net5501 status report)
To: Jason Thorpe <email@example.com>
From: Thor Lancelot Simon <firstname.lastname@example.org>
Date: 06/18/2007 19:31:53
On Mon, Jun 18, 2007 at 04:14:54PM -0700, Jason Thorpe wrote:
> That said, my examination of the code revealed a busy-wait in the
> glxsb driver waiting for the AES operation to complete, which includes
> a bus_space_read_4() in the tight (without delay) loop.
> Seems like that ought to use a callout instead.
Oh, I've been down this road (painfully) with other accellerators. Unless
you're going to run the system at very high HZ you can't really use a
callout unless you want awful single-stream performance.
With this accellerator interrupt-driven mode might be the way to go.
Thor Lancelot Simon email@example.com
"All of my opinions are consistent, but I cannot present them all
at once." -Jean-Jacques Rousseau, On The Social Contract