Hi Tirath

I looked up the article on the IEEE Xplore web site. It's actually an
IEEE paper rather than ACM, but I found the citation on the ACM Portal
using the link you gave me. I'll have a read. I seems to be oriented
towards encryption, but I'm sure it can be used for other applications.
I'm sorry for the delay in responding. Your email is dated the 8th
August 2008 (08,08,08) but I did not receive it until today. 2008-08-24.

John.

Tirath Ramdas wrote:
> I heard Prof. Ruby Lee (Princeton) speak at a conference recently,
> and I think she has some things on parallel LUTs that might interest
> you, e.g: http://portal.acm.org/citation.cfm?
> id97107.1097149&coll=GUIDE&dl=GUIDE&CFID295850&CFTOKEN497411
>
> -tirath
>
>

--
http://www.johnkent.com.au
http://members.optushome.com.au/jekent

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

I heard Prof. Ruby Lee (Princeton) speak at a conference recently,
and I think she has some things on parallel LUTs that might interest
you, e.g: http://portal.acm.org/citation.cfm?
id97107.1097149&coll=GUIDE&dl=GUIDE&CFID295850&CFTOKEN497411

-tirath

On 08/08/2008, at 9:02 PM, John Kent wrote:

> Is it possible to design a scalable search processor that performs
> parallel searches in data ....
> It would certainly make use of the parallel capacity of FPGA hardware.
> What sort of features would such a processor have ?
> What sort of searches would it perform ?
> What sort of applications would it have ?
>
> John.
>
> --
> http://www.johnkent.com.au
> http://members.optushome.com.au/jekent
>

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

e2kcpu wrote:
> --- In f..., "Rick Collins" wrote:
>
> Personally, I like the three address instructions.
> Save a lot on my codes ...
>
> Cheers
>
But do you really need general purpose 3 address instructions?
Most of the time I think of the 3'rd address used, it becomes a
software stack. It might be handy to make that a stack and save
few address bits at the cost of slighty different addressing modes.

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

--- In f..., "Rick Collins" wrote:
>
> Perhaps you are thinking of the xr16? That was completely published
> and an LCC port was provided. The gr0040 was not fully documented and
> the web site even says the compiler was ported, but not provided.
>
Why do you actually prefer the gr0040 over the xr16 ?
(license issues aside)
Personally, I like the three address instructions.
Save a lot on my codes ...

Cheers

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

On Mon, 2008-08-11 at 15:33 +0000, rtstofer wrote:

> Leaving the internal BlockRAM unused (for the most part) makes me
> want
> to build a logic analyzer into the project. Something along the lines
> of http://www.sump.org/projects/analyzer/

I did something similar some time ago and published it
on [xsboard-users], a yahoo group for users of the XESS
boards, with the following announcement:

A few months ago, during the development of my 32-bit CPU,
I ran across a problem for which the solution might be of
general interest for some of you. My design had an error
which inhibited even the very first instruction fetch.
Simulations were of no help; they seemed to run without
errors. So I wished I had a logic analyzer to look into
the FPGA...

I came up with the following idea, which undoubtedly others
have employed as well: I implemented a bare-bones logic
analyzer on the FPGA itself, which samples up to 128 channels
synchronously into 4 block RAMs (which provide a depth of
512 values). The sampling process starts when a trigger signal
is received, and it is qualified with an enable signal, so that
uninteresting clock cycles can be skipped. When the sample
buffer is full, the samples are serially transmitted to a PC
where they are written to a file. I also wrote a simple viewing
program to list the sample values in hex.

This design is for sure no masterpiece in digital logic, but
it served its purpose well - I found the error almost instantly.

If you are interested, you can find my logic analyzer here:
http://homepages.fh-giessen.de/~hg53/LogiProbe
Please feel free to use it as you wish. I would be glad if
it could be of help for someone.

Best regards,
Hellwig

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

--- In f..., "bfranchuk@..." wrote:

> When I am done, I just have the bare machine ...
> lights and switches of and memory of course
>

That's not unlike the very earliest days of the Altair 8800. We had
256 BYTES of RAM and could use it any way we could toggle.

Things looked up fast when Digital Research release CP/M with an
editor and assembler, followed shortly by a macro assembler and,
eventually, PL/I. But everyone really wanted PL/M.

If I were bringing up bare iron, I would probably look for a cross
assembler, preferably with macro capability. I might look at GAS.

With a macro assembler I could implement many of the control
structures of higher level languages (WHILE-DO, IF-THE-ELSE,
REPEAT-UNTIL, etc) as well as data structures.

My favorite next step was to bring up Palo Alto Tiny BASIC. Then I
had something I could work with.

If that wasn't an option, I might try to port LCC. If I wasn't smart
enough to do that, I might implement a toy language like the limited
Pascal in "Algorithms + Data Structures = Programs" or just create a
language using YACC and BISON. There are C like examples around. A
cross-compiler it would have to be until I could port it to the iron.

I bought the LCC book but haven't had a need to read it. One of these
days...

Richard

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

rtstofer wrote:
> --- In f..., "bfranchuk@..." wrote:
>
>
>> (Looking for C compiler source that fits in 64K code and 64K data
>>
> space
>
>> cause thats all I have in the CPU I'm building)
>>
>
> There are decent C compilers for the 8080/Z80 platform. 64k is easily
> doable. But I have never check to see if these compilers implement
> the full ANSI C. Maybe not.
>
>
Most of what I have seen is cross compilers for
the micro computers.
> Heck, there's even a PL/I compiler by Digital Research.
>
>
I'll take almost any source , other than Jim Hendrix's Small C.
I could never figure that mess out. The original small C I like.
> On the IBM 1130, the Fortran compiler made 27 passes! The card images
> were held in core and converted in-situ to tokens. That kind of
> limited the size of the source program. Especially on machines with
> limited RAM.
>
> Richard
>
>
When I am done, I just have the bare machine ...
lights and switches of and memory of course

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

On Mon, 11 Aug 2008 15:33:50 -0000, rtstofer wrote:

>IBM1130 - the first computer I ever used programmed, circa 1969.

That takes me back! I used to swap the differently colored and
labeled bezels over the light bulbs (easy to do) just to be a pain to
the system operator. Used assembly and FORTRAN II on that machine --
university "math with fortran" classes got me access.

Jon

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

--- In f..., "bfranchuk@..." wrote:

> (Looking for C compiler source that fits in 64K code and 64K data
space
> cause thats all I have in the CPU I'm building)

There are decent C compilers for the 8080/Z80 platform. 64k is easily
doable. But I have never check to see if these compilers implement
the full ANSI C. Maybe not.

Heck, there's even a PL/I compiler by Digital Research.

On the IBM 1130, the Fortran compiler made 27 passes! The card images
were held in core and converted in-situ to tokens. That kind of
limited the size of the source program. Especially on machines with
limited RAM.

Richard

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...

rtstofer wrote:
> IBM1130 - the first computer I ever used programmed, circa 1969.
>
>
I used one of them things in 1980, the PC of the day (1969)
- Dual floppy,64kb and a printer. :)
> I started looking at the PDP project and coded the ALU but somehow the
> project never really caught my fancy. Running Unix V6 would be a trip
> but other projects got in the way.
>
>
Some how after V6 everything became BLOAT-NIX.
(Looking for C compiler source that fits in 64K code and 64K data space
cause thats all I have in the CPU I'm building)
> Richard
>
>

To post a message, send it to: f...
To unsubscribe, send a blank message to: f...