Re: Kulago/Mauch DRO ISA quadrature feedback kit

["Matt Shaver" <mshaver-at-erols.com>]

> From: Doug Fortune <pentam-at-home.com>
...
> Here it is (only for pentiums, not for 386/486 as they do not have a
pentiumtimer.
...

This is a wonderful idea, although it's more in line with method #2,
"increase the granularity of interrupt scheduling ". The problem with this
method is revealed in this email from the rtlinux list:

----------
> From: Michael Barabanov <baraban-at-zappa.spb.su>
> To: rtl-at-rtlinux.cs.nmt.edu
> Subject: Re: [rtl] about local APICs
> Date: Tuesday, April 07, 1998 1:39 PM
> 
> Paolo Mantegazza (mantegazza-at-aero.polimi.it) wrote:
> > I have a single CPU Pentium II and I tried to verify that
> > is has a local APIC by using
> >  the asm instruction "cpuid" and looked at bit 9 of register
> > "edx" but it was not set.
> 
> Same with my MMX Pentium. AFAIK, local APICs can be disabled by a signal
> on a CPU pin, and it seems like most single-cpu motherboard
> manufacturers do disable in-cpu APICs.
> 
> Michael.
> --- [rtl] ---
> For more information on Real-Time Linux see:
> http://rtlinux.cs.nmt.edu/

If you know of any commonly available, single CPU, Pentium motherboards that
don't have the timer disabled this might be a good avenue of research. Other
possibilities would be to use a dual processor board (beyond my financial
reach), or try to re-enable the timer on a single CPU board by cutting a
trace (or something like that).

Matt