Re: A useless machine

From	Michael S <already5chosen@yahoo.com>
Newsgroups	comp.arch
Subject	Re: A useless machine
Date	2026-02-19 23:12 +0200
Organization	A noiseless patient Spider
Message-ID	<20260219231241.0000718c@yahoo.com> (permalink)
References	(2 earlier) <10msdau$3vgrg$2@dont-email.me> <10msqc3$4011$1@dont-email.me> <20260218182251.00004e44@yahoo.com> <10n61l5$36tk6$1@dont-email.me> <10n6pup$3efqs$1@dont-email.me>

Show all headers | View raw

On Thu, 19 Feb 2026 11:51:03 +0100
Terje Mathisen <terje.mathisen@tmsw.no> wrote:

> Robert Finch wrote:
> > On 2026-02-18 11:22 a.m., Michael S wrote:  
> >> On Sun, 15 Feb 2026 10:56:51 -0500
> >> Robert Finch <robfi680@gmail.com> wrote:
> >>  
> >>> I have coded a version that tests up to 8192 number
> >>> simultaneously. It is supposed to be able to run at over 200 MHz.
> >>> Some of the number would take a large number of iterations though.
> >>>
> >>> I found something strange, the code to count for more numbers does
> >>> not increase the size of the core very much. (But it takes longer
> >>> to build.) I am not sure if this occurs because there is a mitsake
> >>> somewhere, or if the tools are able to figure out how to share all
> >>> the logic. I am thinking the logic for adders is shared somehow.
> >>> 128-bit arithmetic is being used.
> >>>  
> >>
> >> There is a mistake in your code. No other explanation is possible.
> >>
> >> You should look for 3 sorts of mistakes
> >> 1. Core itself.
> >> The most likely mistake is failing to look for one of the ending
> >> conditions. You have to look for 3 conditions:
> >> a) - value produced at current step is smaller than initial value.
> >>   That is successful ending. Here you signal to the scheduler that
> >> you are ready to accept the next value.
> >> b) - overflow. Next value exceeds the width of register.
> >> c) - timeout. You were running for more than N steps. N=8192 looks
> >> like the reasonable value.
> >> a) and b) are suspected failures.
> >> The handling for b) and c) is similar - you raise alert signal and
> >> stop. Your test unit has to provide supervisor circuity with the
> >> mean to fetch the initial value that caused suspected failure and
> >> to clear alert condition.
> >> After alert is cleared your unit returns to the same state as after
> >> success - waiting for next initial value.  
> > 
> > I found the mistake. Core was working, but groups of cores were
> > not. Done signals were missing.
> >   
> >>
> >> 2. Scheduler.
> >> In proof-of-concept design the scheduler does not have to be very
> >> realistic. For example, it does not have to be efficiently
> >> pipelined. But it should be able to monitor ready signals of all
> >> test units and to supply different initial data to different units.
> >> Supplying the same data to all units simultaneously is a sort of
> >> mistake that explains your observed behavior.
> >>  
> > There is really simple scheduling ATM. There is just a bunch of
> > done signals that are gated together. Meaning the slowest
> > calculation of the group will slow everything else down. Once all
> > the dones are true the next set of numbers are loaded.  
> 
> The obvious choice is to have a work-to-do list, each work item 
> containing a start-end set of numbers to test.
> 
> Each worker thread would simply finish its current task, store
> the/any results, and then ask for the next available task.
> 
> A single producer - multiple consumers queue like this is very simple
> to get right with minimum admins: Just a single LOCK XADD to update
> the queue back index, grabbing the corresponding task.
> 

There is no XADD in FPGA, just gate, multipliers and memory blocks.

> I used this exact setup for my wire speed multi-threaded ntpd deamon 
> about two decades ago. (I did need a few more XADD instructions to 
> handle queue full/empty situations which are not applicable for the 
> current problem.)
> 
> Terje
> 

Did your ntpd deamon scale to 300-400 threads?
300 is the # of test units that fits in FPGAs that appears to be the
most cost-effective for this sort of job, e.g. 10CX220YU484E5G.

Thread

A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-14 09:57 +0000
  Re: A useless machine MitchAlsup <user5857@newsgrouper.org.invalid> - 2026-02-14 18:48 +0000
    Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-14 20:18 +0100
      Re: A useless machine MitchAlsup <user5857@newsgrouper.org.invalid> - 2026-02-14 21:22 +0000
        Re: A useless machine Stephen Fuld <sfuld@alumni.cmu.edu.invalid> - 2026-02-14 13:30 -0800
        Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-15 12:50 +0100
          Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-21 13:48 -0600
            Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-21 21:40 +0100
              Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-21 15:17 -0600
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-22 04:50 -0600
              Re: A useless machine Tim Rentsch <tr.17687@z991.linuxsc.com> - 2026-03-12 21:01 -0700
            Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-22 14:43 +0100
              Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-22 12:38 -0600
                Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-23 10:51 +0100
                Re: A useless machine MitchAlsup <user5857@newsgrouper.org.invalid> - 2026-02-26 01:27 +0000
                Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-26 09:21 +0100
                Re: A useless machine Tim Rentsch <tr.17687@z991.linuxsc.com> - 2026-03-12 21:11 -0700
      Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-15 14:58 +0100
      Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-21 12:27 -0600
        Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-22 15:14 +0100
      Re: A useless machine Stefan Monnier <monnier@iro.umontreal.ca> - 2026-02-22 11:41 -0500
        Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-23 10:53 +0100
          Re: A useless machine Stefan Monnier <monnier@iro.umontreal.ca> - 2026-02-23 10:52 -0500
  Re: A useless machine Tim Rentsch <tr.17687@z991.linuxsc.com> - 2026-02-14 21:19 -0800
  Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-15 03:58 -0500
    Re: A useless machine David Brown <david.brown@hesbynett.no> - 2026-02-15 13:14 +0100
      Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-15 15:46 +0000
      Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-15 10:56 -0500
        Re: A useless machine MitchAlsup <user5857@newsgrouper.org.invalid> - 2026-02-15 19:16 +0000
          Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-15 22:40 +0100
            Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-26 19:31 +0000
        Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-18 18:22 +0200
          Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-18 22:56 -0500
            Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-19 12:10 +0200
              Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-21 01:56 -0500
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-21 19:23 +0200
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-21 20:36 +0200
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-22 10:46 +0200
                Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-22 12:45 +0100
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-22 16:35 +0200
                Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-22 19:12 -0500
                Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-22 19:47 -0500
                Re: A useless machine EricP <ThatWouldBeTelling@thevillage.com> - 2026-02-21 15:39 -0500
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-22 03:16 +0200
                Re: A useless machine EricP <ThatWouldBeTelling@thevillage.com> - 2026-02-22 14:48 -0500
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-22 23:25 +0200
            Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-19 11:51 +0100
              Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-19 23:12 +0200
                Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-20 15:11 +0100
            Re: A useless machine Stefan Monnier <monnier@iro.umontreal.ca> - 2026-02-19 15:27 -0500
              Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-20 11:41 +0200
              Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-22 18:41 -0600
                Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-23 11:10 +0000
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-23 17:24 -0600
                Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-24 06:44 +0000
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-24 01:38 -0600
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-24 03:16 -0600
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-24 04:52 -0600
                Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-24 11:16 +0000
                Re: A useless machine BGB <cr88192@gmail.com> - 2026-02-24 13:19 -0600
                Re: A useless machine Tim Rentsch <tr.17687@z991.linuxsc.com> - 2026-03-12 21:14 -0700
          Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-20 08:01 +0000
    Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-15 15:35 +0200
      Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-15 18:33 +0200
        Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-15 19:19 +0000
          Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-16 12:31 +0200
            Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-17 15:41 +0200
              Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-17 18:17 +0200
    Re: A useless machine anton@mips.complang.tuwien.ac.at (Anton Ertl) - 2026-02-15 22:41 +0000
  Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-15 14:52 +0100
    Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-15 16:45 +0000
      Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-15 18:01 +0100
        Re: A useless machine Thomas Koenig <tkoenig@netcologne.de> - 2026-02-15 19:42 +0000
          Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-15 22:59 +0100
            Re: A useless machine Robert Finch <robfi680@gmail.com> - 2026-02-15 21:53 -0500
              Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-16 12:40 +0200
                Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-16 20:34 +0200
              Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-16 16:08 +0100
  Re: A useless machine Stephen Fuld <sfuld@alumni.cmu.edu.invalid> - 2026-02-16 07:50 -0800
    Re: A useless machine Michael S <already5chosen@yahoo.com> - 2026-02-16 18:31 +0200
    Re: A useless machine Terje Mathisen <terje.mathisen@tmsw.no> - 2026-02-18 21:56 +0100

csiph-web