home .. forth .. misc mail list archive ..

Re: MISC-d Digest V98 #28 (fwd)


Forwarded message:
> From MISC-request@pisa.rockefeller.edu  Sun Jan  3 15:09:46 1999
> Resent-Date: Sun, 3 Jan 1999 15:54:45 -0400
> Date: Sun, 3 Jan 1999 14:54:37 -0500
> From: Penio Penev <penev@venezia.rockefeller.edu>
> To: MISC@pisa.rockefeller.edu
> Subject: Re: MISC-d Digest V98 #28
> In-Reply-To: <35AF129A.4FC9@nvmedia.com>
> Message-ID: <Pine.SGI.4.05.9901031441310.25560-100000@venezia.rockefeller.edu>
> MIME-Version: 1.0
> Content-Type: TEXT/PLAIN; charset=US-ASCII
> Resent-Message-ID: <"BmVToC.A.Ye.Bs8j2"@pisa>
> Resent-From: MISC@pisa.rockefeller.edu
> X-Mailing-List: <MISC@pisa.rockefeller.edu> archive/latest/831
> X-Loop: MISC@pisa.rockefeller.edu
> Precedence: list
> Resent-Sender: MISC-request@pisa.rockefeller.edu
> 
> On Fri, 17 Jul 1998, GARY B. LAWRENCE wrote:
> 
> |      I would like to get some feedback from this group on a subject that
> | interests me at the moment. I would like to see F21's be used in a 
> | multiprocessor version of a multitasking operating system. 
> 
MISC makes feasible what I call "DMA-tasking". This is not parallel,
but is I believe where MISC can assert itself very visibly. 
One CPU is the scheduler, and provides other general services perhaps. The scheduler
has it's own bus, and a "bus" of MISCs. A task switch is a CPU switch.
0-latency task switching. The bus-of-miscs can be on one chip, run cool,
and need maybe 10 more pins for 128 MISCs. 

Tricks might be possible where 2 miscs operate in echelon on the same
memory to give a larger effective machine state, but the latency thing
will sell it for general multi-media use.

Rick Hohensee  

Rick Hohensee          http://cqi.com/~humbubba
cLIeNUX   xart   kandinski  cycluphonics   ratioles  H3sm   Md., USA