Shop OBEX P1 Docs P2 Docs Learn Events
Tachyon NEON V5 (FAT32 and Ethernet Servers in 32kB EEPROM!) - Page 17 — Parallax Forums

Tachyon NEON V5 (FAT32 and Ethernet Servers in 32kB EEPROM!)

1141517192025

Comments

  • MJBMJB Posts: 1,235
    ErNa wrote: »
    ...

    Hi ErNa,
    I am traveling and don't have your Email addresss with me.
    Just wanted to ask again if you have a working build for the IoT5500+P8
    - which exact files from Peter's repository to load for a full
    SD+NW build.
    AFAIR there was an issue with 5.5 vs. 5.4 ??

    I want to set up a server that will log to SD data sent from ESP8266 remote sensors. And makes them available via WEB & FTP

    you have My Email ... thanks
    MJB
  • @MJB - just let me know specifically what it is you need it to do and I'm sure I can sort out any loose ends.
    While it is true I am focused on P2, it is also true that any issues to do with the W5500 will impact its use on the P2, so I would be happy to devote some time to sorting out any issues.

    @ErNa - same goes for you, so please let me know where I can help.
  • MJBMJB Posts: 1,235
    edited 2019-06-09 09:19
    @MJB - just let me know specifically what it is you need it to do and I'm sure I can sort out any loose ends.
    While it is true I am focused on P2, it is also true that any issues to do with the W5500 will impact its use on the P2, so I would be happy to devote some time to sorting out any issues.

    @ErNa - same goes for you, so please let me know where I can help.

    Hi Peter,
    great to see your progress on P2.

    I was hoping there are IoT5500+P8 users out there having a latest build config they could share.
    Yesterday I downloaded the full Tachyon dropbox share again.

    The V5 contains a
    TACHYON.SPIN (V5.2) which is much larger (mostly ROMS) than the
    TACHYON5r4.SPIN (V5.4) - but also quite a number of other changes.

    There is also a subfolder V5.4 with some slightly different files

    so I take the subfolder to build a IoT5500+P8 ?
    TACHYON5r4.SPIN
    Extend
    Easyfile
    Easynet
    FTP
    HTTP
    

    AND - there are NO ISSUES (yet ..)

    It is a while since the Spinneret HTTP+ version, but I am confident
    to use adapted HTTP/FTP/Telnet to implement s.th. much simpler than MQTT for my use

    - receiving ESP8266 remote sensor data
    - logging to SD
    - presenting via HTTP
    - download via FTP
    - Ping-Pong to other P1+Tachyon systems
    - a little QVGA-LCD-display
    - Bluetooth-Serial might be nice ...
    ...

    EDIT: and one future day ... a P2D2+Eval-Board ;-) :-)

    EDIT2: & btw. having now a STABLE & MATURE Tachyon on the P1, not changing every few days, will make it's use a bit less challenging ;-)




  • tritoniumtritonium Posts: 539
    edited 2019-06-09 10:45
    Hi
    having now a STABLE & MATURE Tachyon on the P1, not changing every few days, will make it's use a bit less challenging ;-)

    AHA- that's what I've been waiting for- no disrespect but I found forth a challenge without continual improvements (meaning changes). Of course improvements are great but can be hard to keep up with.

    So
    Where can I find the 'STABLE & MATURE Tachyon on the P1' files and instructions please?
    There seem to have been so many versions.


    Dave
  • MJBMJB Posts: 1,235
    tritonium wrote: »
    Hi
    having now a STABLE & MATURE Tachyon on the P1, not changing every few days, will make it's use a bit less challenging ;-)

    AHA- that's what I've been waiting for- no disrespect but I found forth a challenge without continual improvements (meaning changes). Of course improvements are great but can be hard to keep up with.

    So
    Where can I find the 'STABLE & MATURE Tachyon on the P1' files and instructions please?
    There seem to have been so many versions.


    Dave

    in Peter's footer is a link to his dropbox folder
    V5 is the latest for P1
    the P2 folder is the latest TaqOS and Tachyon for the P2
    in MORE is a load of examples and demos

    latest version is
    version long 540_181011

    so unchanged since 8 month now ...
  • @MJB
    so unchanged since 8 month now ...

    That's great- I guess then its pretty stable so no gotchas to watch out for.

    Thanks

    Must give it a try in preparation for P2


    Dave

  • I decided that I needed to tidy up those files, especially V5, so I've started a README.TXT and created folders for organize it better. The current kernel itself along with archives etc has its own KERNEL folder, as do the MODULES such as EXTEND and EASYFILE etc. I hope to gradually organize it in such a way that files can be found easily and without too much clutter.

    I'm not too concerned with the older versions though since V5 upgraded V4 which was the first wordcode version that replaced the last bytecode version V3. Maybe I will write a little ABOUT.TXT in the main TACHYON folder to explain all this.

    If anyone asks about documentation then I will simply point out that while it seems and will always be "incomplete", that nonetheless I have written enough that others can add to it or at least fill in what they know and then ask questions. Please feel free to add sections to the existing documentation or create new documents which can be linked in.
  • MJBMJB Posts: 1,235
    edited 2019-06-10 22:20
    SD CARD issue

    directory is not shown correctly

    this 4GB SDHC card worked before with Tachyon - not sure which exact version
    and contains a number of files.

    the more verbose version of .CARD gives a reboot in the SPEED line
    --------------------------------------------------------------------------------
    ...  MOUNT -->
    CARD: NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,974us
    FAT: #0000.0000 TXX  TXT  0 bytes (0kB clusters)
     ok
    ...  .CARD --> NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,977us ok
    ...
    ...  pri !SD 0EXIT
    ...        CR ." MFG:    "  cid C@ .BYTE
    ...        CR ." OEM:    "  cid 1+ 2 CTYPE
    ...        CR ." PROD:   "  cid 3 + 5 CTYPE
    ...        CR ." REV:    "  cid 8 + C@ .BYTE
    ...        CR ." S/N:    "  cid 9 + U@ .LONG
    ...        CR ." DATE:   "  cid #13 + C@
    ...        8 << cid #14 + C@ OR
    ...        DUP 4 >> #2000 + #10 PRINTNUM ??? in .CARD at PRINTNUM
    
    ...        ." /"  >N #10 PRINTNUM ??? in .CARD at PRINTNUM
    
    ...        CR ." CRC:    "  cid #15 + C@ 2/ .BYTE
    ...        CR ." OCR:    "  ocr @ .LONG
    ...        CR ." SPEED:  "  CNT@ 256 FOR I SECTOR NEXT CNT@ SWAP - ABS CLKFREQ 1000000 / / 8 >> .
    ...        cid 16 DUMP csd 16 DUMP
    ...      ;
    ...  .CARD -->
    MFG:    73
    OEM:    BG
    PROD:   NCard
    REV:    10
    S/N:    5957.3321
    DATE:   /
    CRC:    6C
    OCR:    C0FF.8069
    SPEED:  |
    
      Propeller .:.:--TACHYON--:.:. Forth V5r4 NEON 540181011.1000
      *** MODULES ***  Propeller .:.:--TACHYON--:.:. Forth V5r4 NEON 540181011.1000
    3F8E: EASYFILE         SDHC card + FAT32 Virtual Memory File System V1.2 171024-0000
    2ECA: TOOLS            DEV TOOLS
    1A00: EXTEND           Primary extensions to TACHYON V5 kernel  - 18104-0900
    
    
    FREQ = 80.00MHZ
    *** INITS ***
    MOUNT 45CC
    NO ROMS
    *** I2C ***
    $42 I/O
    $A0 EE/RTC
    I/O =  31 :UHUU 27 :DDHL 23 :LUDD 19 :~~~~ 15 :~U~~ 11 :~~UU 7 :DD~~ 3 :~UUU
    INTERCOM:
    
    CODE:$5194 = 20372 bytes
    NAME:$5400 = 8192 bytes
    DATA:$797D = 1133 bytes
    FREE:      = 620 bytes
     Data Stack (0)
    Mon, 01 Jan 2001 00:00:00 UTC
    CARD: NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,974us
    FAT: #0000.0000 TXX  TXT  0 bytes (0kB clusters)
    
    --------------------------------------------------------------------------------
    ...
    ...  cid 16 DUMP csd 16 DUMP -->
    0000.7735:  73 42 47  4E 43 61 72  64 10 21 33  57 59 00 C3  D9    sBGNCard.!3WY...
    0000.7745:  40 0E 00  32 5B 59 00  00 1D E9 7F  80 0A 40 00  69    @..2[Y.......@.i ok
    ...
    ...  CR ." SPEED:  "  CNT@ 256 FOR I SECTOR NEXT CNT@ SWAP - ABS CLKFREQ 1000000 / / 8 >> . -->
    SPEED:    0M
    SPEED:  |
    
      Propeller .:.:--TACHYON--:.:. Forth V5r4 NEON 540181011.1000
      *** MODULES ***  Propeller .:.:--TACHYON--:.:. Forth V5r4 NEON 540181011.1000
    3F8E: EASYFILE         SDHC card + FAT32 Virtual Memory File System V1.2 171024-0000
    2ECA: TOOLS            DEV TOOLS
    1A00: EXTEND           Primary extensions to TACHYON V5 kernel  - 18104-0900
    
    (..sone stuff removed ..)
    
    ...  ls
    CARD: NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,974us
    FAT: #0000.0000 TXX  TXT  0 bytes (0kB clusters)
    
    ...  DIR -->
     ok
    ...  MOUNT -->
    CARD: NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,974us
    FAT: #0000.0000 TXX  TXT  0 bytes (0kB clusters)
     ok
    ...  .CARD --> NCard BG73.10 #5957.3321 2012/3 !C0FF.8069 14,978us ok
    ...
    

    EDIT:
    the Directory looks OK
    ...  $800000 $100 SD DUMP -->
    0080.0000:   4D 4A 42 53  44 20 20 20  20 20 20 08  00 00 00 00    MJBSD      .....
    0080.0010:   00 00 00 00  00 00 77 A3  64 4C 00 00  00 00 00 00    ......w.dL......
    0080.0020:   41 4A 41 58  30 30 30 31  48 54 4D 20  00 08 94 A4    AJAX0001HTM ....
    0080.0030:   64 4C 64 4C  00 00 F0 92  F8 44 03 00  CE 02 00 00    dLdL.....D......
    0080.0040:   41 4A 41 58  30 30 30 31  48 54 58 20  00 0E 94 A4    AJAX0001HTX ....
    0080.0050:   64 4C 64 4C  00 00 F4 92  F8 44 04 00  5D 00 00 00    dLdL.....D..]...
    0080.0060:   41 4A 41 58  30 30 30 32  48 54 58 20  00 11 94 A4    AJAX0002HTX ....
    0080.0070:   64 4C 64 4C  00 00 23 A1  F4 44 05 00  5E 00 00 00    dLdL..#..D..^...
    0080.0080:   46 41 56 49  43 4F 4E 20  49 43 4F 20  00 14 94 A4    FAVICON ICO ....
    0080.0090:   64 4C 64 4C  00 00 F8 54  D7 44 06 00  85 1F 00 00    dLdL...T.D......
    0080.00A0:   46 49 52 4D  57 41 52 45  52 4F 4D 20  00 18 94 A4    FIRMWAREROM ....
    0080.00B0:   64 4C 64 4C  00 00 A4 54  D7 44 07 00  00 80 00 00    dLdL...T.D......
    0080.00C0:   48 4F 4D 45  20 20 20 20  48 54 4D 20  00 21 94 A4    HOME    HTM .!..
    0080.00D0:   64 4C 64 4C  00 00 F8 54  D7 44 08 00  9D 2C 00 00    dLdL...T.D...,..
    0080.00E0:   48 4F 4D 45  32 20 20 20  48 54 4D 20  00 24 94 A4    HOME2   HTM .$..
    0080.00F0:   64 4C 64 4C  00 00 60 75  DB 44 09 00  E4 01 00 00    dLdL..`u.D...... ok
    
    

    this as well
    ...  $800000 9 >> SECTOR lsdirs -->
    AJAX0001.HTM    .....a 8000.984A  718  m2014/07 /24   18:23   c2018/03 /04   20:36 920.  a2018/03 /04
    AJAX0001.HTX    .....a 0000.984A   93  m2014/07 /24   18:23   c2018/03 /04   20:36 920.  a2018/03 /04
    AJAX0002.HTX    .....a 8000.984A   94  m2014/07 /20   20:09   c2018/03 /04   20:36 920.  a2018/03 /04
    FAVICON .ICO    .....a 0000.984A 8069  m2014/06 /23   10:39   c2018/03 /04   20:36 920.  a2018/03 /04
    FIRMWARE.ROM    .....a 8000.984A  32k  m2014/06 /23   10:37   c2018/03 /04   20:36 920.  a2018/03 /04
    HOME    .HTM    .....a 0000.984A  11k  m2014/06 /23   10:39   c2018/03 /04   20:36 920.  a2018/03 /04
    HOME2   .HTM    .....a 8000.984A  484  m2014/06 /27   14:43   c2018/03 /04   20:36 920.  a2018/03 /04
    HTTP001 .HTM    .....a 0000.984A  388  m2014/06 /23   10:39   c2018/03 /04   20:36 920.  a2018/03 /04
    HTTP404 .HTM    .....a 8000.984A  564  m2014/06 /23   10:39   c2018/03 /04   20:36 920.  a2018/03 /04
    JGHOME  .HTM    .....a 0000.984A 2584  m2014/08 /01   16:07   c2018/03 /04   20:36 920.  a2018/03 /04
    JGHOME  .HTX    .....a 8000.984A  108  m2014/08 /01   16:40   c2018/03 /04   20:36 920.  a2018/03 /04
    JGHOME~1.TES    .....a 0000.984A 1219  m2014/08 /02   21:41   c2018/03 /04   20:36 920.  a2018/03 /04
    JGINDEX .HTM    .....a 8000.984A 9810  m2014/07 /29   22:19   c2018/03 /04   20:36 920.  a2018/03 /04
    LOG0001 .TXT    .....a 0000.984A 130k  m2014/06 /26   18:19   c2018/03 /04   20:36 920.  a2018/03 /04  ok
    

    trying to figure out where it goes wrong ...
  • Yesterday was a day of total frustration! I took a brand new QuickStart board, loaded it with the latest version of V5Neon,(using Prop Tool) added the latest Extend (using TeraTerm) and put the Forth tutorial on my iPad. First exercise, (. 3 4 + = ) I get result 7, next I try (. 5 2 + = ) and I get 32,767 ! Can’t even get to second exercise. Next try storing string “Hello world “ and printing that out. No luck. Reboot computer to see if that is the problem. Move QS board to another USB HUB. I have to use USB hubs as laptop only has 3 ports. Now all I get when I press reset button on QS board is gibberish, verify baud rate of TeraTerm, it is 115200, worked before. Try another QS board, same results. I think that I have some how managed to crash the PLL in 3 QS boards. Am I getting too old and senile to be doing this stuff? Do I just need to List all of the electronics stuff on EBAY and take up solitaire as entertainment? My wife would be happy if all of my “junk” was gone.

    End of rant. Any one have any suggestions as t why I can’t get Tachyon V5 to do two consecutive adds or why I seem to keep blowing up QS boards?

    Jim
  • ErNaErNa Posts: 1,738
    Quick Start board in the beginning only had 32 kBytes of flash, if I remember correctly
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2019-06-30 00:42
    @RS_Jim

    It's always good to copy and paste what's on your terminal as code into your post or take a screenshot so we can see what happened. However it sounds like even the first exercise was back to front because it we break this up onto separate lines with my comments in brackets, this is how to add and print:
    3 ( push 3 onto the top of the stack )
    4 ( push 4 onto the top of the stack over 3 )
    + ( pop the top two numbers 4 then 3 and add them then push the result to the top of the stack )
    PRINT ( pop the top of the stack and print the number - note you can use the period symbol instead but PRINT is more discernible )

    Just think of Forth as assembler language that you can combine together on the same line. Each instruction such as 3 (akin to PUSH #3) still executes left to right. Your code would have tried to print immediately before it had anything valid so it would print whatever was on top of the stack (or was on top if the stack is empty). The = sign is a comparison operation, just as it is with many languages and has no place (at the end of a statement) in any calculations except on a calculator. Don't think "reverse", just think left to right, pick up 3 apples in one hand, then pick up another 4 apples with the other hand, then add what you have in the two hands together into the one hand, then say or print what that is while you now discard (or eat) those apples.

    The Tachyon source automatically adjusts for 5 or 10MHz crystals but there is another section at the start that allows for 6MHz crystals so just make sure you have the 5/10 enabled with the old '{ comment trick which is employed since the Spin compiler does not have any conditional compilation directives.

    I doubt very much that QS are blowing up but I have blown boards before when the laptop was plugged into power with the USB lead connected, and the board was powered from another source. If the power sources are "badly grounded" you could get ground loop currents that can even wreck your USB port on the laptop. I find it best to run the laptop off batteries in these situations but many of my boards are powered off the USB port and so it doesn't really matter. Desktop computers normally have a good ground but those wall warts and laptop power packs without an earth can give you minor shocks but those minor shocks are due to the neutral being capacitively coupled to the DC common and they can kill boards that are independently powered.

    TAKE NOTE anyone using powered laptops connected via the USB cable to separately powered boards.

    Here, the source code has the 5/10MHz section enabled because it comments out the { comment with a ' comment while there is a matching ' before the closing brace.
    CON { *** CLOCK MODES *** }
    '{ 5/10MHZ CRYSTAL
    _clkmode        = xtal1 + pll8x
    _xinfreq        = 10_000_000            ' <--- AUTOMATIC 5 or 10MHz operation change at boot
    sysfreq         = 80_000_000
    '}
    { 6MHZ CRYSTAL
    _clkmode        = xtal1 + pll16x
    _xinfreq        = 6_000_000
    sysfreq         = 96_000_000
    '}
    baud            = 115200                ' <-- user change - tested from 300 baud to 3M baud
    

    BTW, if your board only has 32k of EEPROM then Tachyon will still run fine, it is just that the upper 32k of 64k EEPROMs are used to store ROM images, that is, the <2k binary that gets loaded into cogs as "PASM objects" such as serial ports and VGA etc.
  • Peter,
    Thanks for the comprehensive reply. I am getting senile, I never thought about ground loops destroying my boards! I should have thought of that as I spent 40+ years in the video industry and we were always on the look out for ground loops. I will see if my QS boards will function on an unpowered (externally) USB port Yes, I was putting the print command first and then doing the stack ops for the arithmetic.
    Did you see my PM?
    Thanks for helping keep my head on somewhat straight.

    Jim
  • @RS_Jim - hopefully those boards aren't fried then.

    You posted into the big ACTIVITY message box rather than clicking the little MESSAGE button in the top right.
    Why oh why do that lay it out that way! :)


  • Thanks Peter,
    Will check the boards later today.
    Jim
  • Peter,
    Finally did some checking on my setup for ground loops and found the culprit. My laptop is connected via a three wire grounded power cord. So I started looking into my powered USB hub and finally found the issue. The outer conductor of the power cord has a potential of 42 VAC to ground no matter how the unpolarized plug was inserted into the outlet. So, now if I want to use a powered USB hub, I have to find one that has a grounded power supply or use one that I have that I can run off a 10,000 MAh battery. This one is connected to power via a micro USB to standard USB cable. The Phone Battery has a standard USB port on it so I can plug right into it. I think I better trash this hub or find a 2.5 amp power supply that has a zero potential to ground.
    Jim
  • RS_Jim wrote: »
    Peter,
    Finally did some checking on my setup for ground loops and found the culprit. My laptop is connected via a three wire grounded power cord. So I started looking into my powered USB hub and finally found the issue. The outer conductor of the power cord has a potential of 42 VAC to ground no matter how the unpolarized plug was inserted into the outlet. So, now if I want to use a powered USB hub, I have to find one that has a grounded power supply or use one that I have that I can run off a 10,000 MAh battery. This one is connected to power via a micro USB to standard USB cable. The Phone Battery has a standard USB port on it so I can plug right into it. I think I better trash this hub or find a 2.5 amp power supply that has a zero potential to ground.
    Jim

    Yes, the usual problem. Most people do not know that any transformer will have a capacitive coupling from the primary to he secondary side. This will normally put an AC potential of half the input voltage on the secondary winding. This can only be prevented by grounding (to earth) of the secondary side. I had a customer who blew up his PC when he had supplied a USB device from an ungrounded PSU and then connected the USB port to the PC. That was in Germany with 240V AC. The USB connector from the device was a 120V AC relative to the grounded PC. The chip did not like this.
    I take great care to always have everything properly grounded. No to pin power connectors allowed here.
  • Reinhardt,
    This power supply is a switcher, not transformer based. If I had a 12 Volt 4 amp transformer, I would probably build my own 5 volt supply for the HUB. At this point I have removed the troublesome HUB completely and reordered my USB connections.
    Jim
  • Peter,
    I am back to attempting a port of the Altimeter code to Tachyon. Is there a reasonable way to calculate the P0/P1 raised to the 0.190264 power or should I adapt PhiPi’s cubic lookup to make my barometer function in Tachyon? Also to read a 24bit value in I2C is the I2C code in Tachyon reading 8 bits at a time and I need to left shift to do all 24 bits?
    Jim
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2019-07-06 22:42
    Hi Jim, I managed to dig up that Altimeter code (after I eventually found it) so I will look at how it can be implemented based upon PhiPi's excellent documentation. There's already some 64-bit mixed mode maths in Tachyon but it would be easy enough to fill that in for this requirement. In fact there is the built-in F32 floating-point ROM that could be loaded into a cog if need be. However if integer methods are available using a table then that is where it is best to start.
    In Tachyon this table is entered like this:
    TABLE alt_table
    4433077 , 2323037 , 2025546 , 1832444 , 1686113 , 1566963 , 1465785 , 1377458 ,
    1298823 , 1227784 , 1162875 , 1103025 , 1047431 ,  995470 ,  946651 ,  900579 ,
    856930 ,  815437 ,  775876 ,  738056 ,  701815 ,  667012 ,  633527 ,  601252 ,
    570095 ,  539971 ,  510808 ,  482539 ,  455106 ,  428455 ,  402538 ,  377311 ,
    352735 ,  328773 ,  305392 ,  282561 ,  260253 ,  238441 ,  217101 ,  196211 ,
    175750 ,  155700 ,  136041 ,  116758 ,   97835 ,   79257 ,   61010 ,   43082 ,
     25459 ,    8132 ,   -8910 ,  -25680 ,  -42185 ,  -58436 ,  -74439 ,  -90205 ,
    -105740 , -121053 , -136149 , -151036 , -165720 , -180207 , -194503 , -208614 ,
    -222544 ,
    

    The I2C code is very easy and yes, if you want 24-bit values you simply build them and given that the device outputs the bytes descending order (23..16,15..8,7..0) like this:
    ackI2C@ 8<< ackI2C@ OR 8<< nakI2C@ OR
    
    You don't need to visualize this in your head, just interact with Tachyon, in this I have TAQOZ in front of me:
    TAQOZ# $12 8<< $34 OR 8<< $56 OR .L --- $0012_3456 ok
    

    Be careful about porting code to Tachyon, while it is tempting to just convert each routine, you may end up with a hard to read and debug mess of code. The idea is to extend the language as necessary so that the program should be readable and self-commenting. Figuring out where and how to factor snippets and choosing an apt name is a rewarding art.

    I will provide some updates to the kernel to extend the 64-bit maths to include simple add and subtract as well, just to help you along. But don't be too worried if some of your routines look a bit "ugly" or inefficient, as long as it works. I call this Q&D (quick & dirty) because it lets me check out functionality first, and then I can improve it later, but not too much later, do that incrementally.

  • As a lifelong COBOL programmer, used to the most verbose language known to mankind, I kind of shiver when reading 'self-commenting' and thinking about FORTH.

    I do get the concept, I even like it, but try after try I fail miserable in understanding any of your code. Has maybe to do with old dogs and new tricks, but I consider FORTH as a write-only language...

    Mike
  • Hey Mike, hehe, i say "10 ms" you say delay.ms(10), but hey what, that's what you've been programmed to do! :)
  • Exactly my problem. I love the general idea, but can't get my head to wrap around it.

    Maybe I need to spend a couple of month in Australia to fix that problem. Something I have on my bucket list anyways...

    Mike
  • msrobotsmsrobots Posts: 3,701
    edited 2019-07-07 03:49
    No, I need to be serious here, but I think to be able to call a TAQOZ subsystem from any program is very desirable and still not solved, yet.

    Thankfully @"Peter Jakacki" finally got the idea that this might be useful, co-existing with Phyton, hidden in some HUB area using one COG.

    I plead for a more general usable version, so that I could include a TAQOZ sub system into my fastspin program. The goal is to get a TAQOZ console connected to a virtual serial driver (without using pins).

    Then one can execute TAQOZ snippets from other languages, and share the useful one-liners (or more).

    I can do most of it with the build-in EVAL TAQOZ, alas I need 4 smartpins for communication. If I would understand how to redirect the console to some hub-addresses as mailbox, I could send that and free the used 4 pins again.

    Since I am patching the TAQOZ rom a bit this code may not run on the next silicon. But the goal is still there to use snippets on a TAQOZ sub system...

    Now make nails with heads.

    Starting with a TAQOZ loaded from ROM, what definitions would be needed to switch console input and output from serial to a simple mailbox-system, say two longs, one in one out, -1 when ready/unused else value, set to -1 after transmit/read. Something simple stupid, may should be able to block, speed is not the problem. just bi-directional solid transfer of console I/O byte wise between TAQOZ sub system and rest of the P2 using two longs in HUB.

    If I had that FORTH code I could make a TAQOZ subsystem usable from fastspin/PASM running on the rest of the chip.

    And basically TAQOZ does console redirection, I know that much, but I do not now how to do that and hope that somebody in the know of magic will throw some code at me to do that. The rest I might figure out by myself.

    Help!

    Mike
  • msrobotsmsrobots Posts: 3,701
    edited 2019-07-07 08:32
    shoot. I do not meant TAQOZ sub system. NO, NO, it has to be named TAQOZ-MCP. Please no offense, sub-system; shoot.

    Mike, Mike, Mike...

    In my current implementation it does even run on COG 0.

    So my fastspin program does its bootstrap, leaves space for TAQOZ, loads and runs its Payload in COG1 not COG0 and silently slides COG0 into TAQOZ by just jumping into the ROM.

    To do so I need to modify the fastspin generated PASM file a little bit and just compile it again.

    I just need to get rid of pins 1-0-63-62 used now to talk to TACOZ.

    Mike

    edited pin numbers, @clusso99 got me...
  • Cluso99Cluso99 Posts: 18,066
    "I just need to get rid of pins 1-0-61-60 used now to talk to TACOZ."

    Don't you mean 1-0-63-62 ???

    BTW it doesn't help to live downunder in coming to grips with TAQOZ. Sometimes I can understand the one-liners, and other times I'm just amazed.
  • Cluso99Cluso99 Posts: 18,066
    I think this is the part of ROM you need to patch to intercept the serial I/O - the SEROUT part anyway
                       '************************************* SERIAL I/O **************************************
    ff34c 17a f6030621 SEROUT		mov	txpin,tos
    ff350 17b fd9ffbb4 		jmp	#DROP
                       
                       
    ff354 17c f6001a21 CONEMIT		mov	R1,tos
    ff358 17d f607063e TXR1		mov	txpin,#tx_pin
    ff35c 17e fc201b83                 wypin   r1,txpin             '..send byte
    ff360 17f fd64001f 		waitx	#0
    ff364 180 fd730640 .wait           testp   txpin         wc      '..wait for buffer empty
    ff368 181 3d9ffff8  if_nc          jmp     #.wait
    ff36c 182 fd9ffb98                 jmp     #DROP
    ff370 183 0000003e txpin		long	tx_pin
    

    and this looks to be the receive section tho a quick look doesnt jump out and say how it works
                       '***************************************** HUBEXEC CODE ***************************
                       '
    fe9b4              		orgh
    fe9b4                              alignw
    fe9b4              _hubexec
                       
                       ' !SP - init the data stack pointer
    fe9b4     f6042200 INITSP          mov     datptr,#datstk
    fe9b8     f6042a00                 mov     depth,#0
    fe9bc     ff6f56df         _ret_   mov     tos,##$DEADBEEF         	' marker
    fe9c0     060442ef 
                       
                       {
                       		dirl	#rx_pin			'disable receiver
                       		wrpin	#%00_11111_0,#rx_pin	'configure rx_pin for asynchronous receive, always input
                       
                       		wrpin	#%01_11110_0,#tx_pin	'configure tx_pin for asynchronous transmit, always output
                       		dirh	#tx_pin			'enable transmitter
                       }
                       
    fe9c4              InitTaqoz
    fe9c4     fd601601                 cogid   fx
    fe9c8     fed0f3f6                 loc     PTRA,#@IDLE&$FFFF               ' default startup into Instruction Pointer
    fe9cc     fb9c160c                 tjnz    fx,#INITSTKS
                       ' COG 0 CONSOLE '
    fe9d0     fd647c41                 dirh    #tx_pin                 	'set tx output high
    fe9d4     fc2c1a3e 		wypin	#$0D,#tx_pin
    fe9d8     fed0f4ae                 loc     PTRA,#@TERMINAL&$FFFF
    fe9dc     fd647e41 		dirh    #rx_pin
                       		'setse1	#%001_111111
    fe9e0     fa9c3c3f 		rdpin	fz,#rx_pin wc			' clear rx?
    fe9e4     fd640025 		setint1	#0
    fe9e8     f6044000 		mov 	rxwrC,#0
    fe9ec     fc64400c 		wrlong 	rxwrC,#rxrd
                       
    fe9f0     fd677e20 		setse1  #%110<<6+rx_pin        		'set se1 to trigger on  rx char event?????
    fe9f4     ff0007f5                 mov     ijmp1,##@taqoz_rxisr     	'set int1 jump vector to receive buffer
    fe9f8     f607e810 
    fe9fc     fd640825 		setint1	#4
                       
    fea00     fdbfffb0 INITSTKS        call    #INITSP
    fea04     f6042630 INITLP          mov     lpptr,#lpstk
    fea08     f6042420                 mov     braptr,#brastk
    fea0c     06042840 INITRP  _ret_   mov     retptr,#retstk
                       
    fea10     f0643a08 taqoz_rxisr	shl 	rxlong,#8
    fea14     fa8c3c3f                 rdpin   fz,#rx_pin       		' recv byte (bits31:24) from rx pin
    fea18     f0443c18 		shr	fz,#24
    fea1c     f5483a1e 		or	rxlong,fz wz
    fea20     ff0d8d8d 		cmp	rxlong,##$1B1B1B1B wz		' esc esc esc esc break'
    fea24     f20c3b1b 
    fea28     af0007f6 	if_z	coginit #0,##@RESET
    fea2c     acec0164 
    fea30     f6003e20 		mov	rxwrP,rxwrC
    fea34     f1043f80 		add	rxwrP,#rxbuffers
    fea38     fc403c1f 		wrbyte	fz,rxwrP
    fea3c     ff000007 		incmod	rxwrC,##rxsize-1
    fea40     f704407f 
    fea44     fc54400e 		wrword	rxwrC,#rxwr
    fea48     fb3bfff5 		reti1
                       '_rxrd		long	rxrd
                       '_rxwr		long	rxwr
                       
    fea4c     fae4180c READRX		rdword	r0,#rxrd
    fea50     f2081820 		cmp	r0,rxwrC wz
    fea54     ad900018 	if_z	jmp	#@rr1
    fea58     f6001a0c 		mov	r1,r0
    fea5c     f1041b80 		add	r1,#rxbuffers
    fea60     fac0140d 		rdbyte	acc,r1
    fea64     ff000007 		incmod	r0,##rxsize-1
    fea68     f704187f 
    fea6c     fc54180c 		wrword	r0,#rxrd
    fea70     fd800142 rr1		jmp	#PUSHACC
                       
    fea74     fd640025 DEBUGGER	setint1	#0
    fea78     fd8fca88 		jmp 	#\_Start_Monitor	' \ forces absolute address '
    
  • Cluso99Cluso99 Posts: 18,066
    Ok a bit more..
    fe9c4              InitTaqoz
    fe9c4     fd601601                 cogid   fx
    fe9c8     fed0f3f6                 loc     PTRA,#@IDLE&$FFFF               ' default startup into Instruction Pointer
    fe9cc     fb9c160c                 tjnz    fx,#INITSTKS
                       ' COG 0 CONSOLE '
    fe9d0     fd647c41                 dirh    #tx_pin                 	'set tx output high
    fe9d4     fc2c1a3e 		wypin	#$0D,#tx_pin
    fe9d8     fed0f4ae                 loc     PTRA,#@TERMINAL&$FFFF
    fe9dc     fd647e41 		dirh    #rx_pin
                       		'setse1	#%001_111111
    fe9e0     fa9c3c3f 		rdpin	fz,#rx_pin wc			' clear rx?
    fe9e4     fd640025 		setint1	#0
    fe9e8     f6044000 		mov 	rxwrC,#0
    fe9ec     fc64400c 		wrlong 	rxwrC,#rxrd
                       
    fe9f0     fd677e20 		setse1  #%110<<6+rx_pin        		'set se1 to trigger on  rx char event?????
    fe9f4     ff0007f5                 mov     ijmp1,##@taqoz_rxisr     	'set int1 jump vector to receive buffer
    fe9f8     f607e810 
    fe9fc     fd640825 		setint1	#4
    
    this sets up the rx interrupt vector to taqoz_rxisr in ijmp1
    and here is the rxisr
    fea10     f0643a08 taqoz_rxisr	shl 	rxlong,#8
    fea14     fa8c3c3f                 rdpin   fz,#rx_pin       		' recv byte (bits31:24) from rx pin
    fea18     f0443c18 		shr	fz,#24
    fea1c     f5483a1e 		or	rxlong,fz wz
    fea20     ff0d8d8d 		cmp	rxlong,##$1B1B1B1B wz		' esc esc esc esc break'
    
  • yes @Cluso99 I already do patch TAQOZ in the rom to intercept the pins with other smartpins, before I start it, and can intercept the TAQOZ console. But I would like to have the pins free for the main program written in whatever and access the TAQOZ console via HUB mailbox, not serial pins.

    So after intercepting TAQOZ (what I can) I would like to send some TAQOZ code to TAQOZ to redirect its I/O to a mailbox and release 63.62 (yes, I miss typed) so I can release 1,0 I am using to cheat TACOS in believing it talks to a serial connected port.

    And to do that I need some smart mind to write me a console access to TACOZ over a HUB mail box, so I need some TAQOZ code to switch the TAQOZ console to talk over some HUB mail box, not pins serial.

    Is my English really so bad?, I try to explain this now since about 8 month?

    feeling helpless to explain further

    see here : https://forums.parallax.com/discussion/168621/coexistence-with-taqoz
    and here https://forums.parallax.com/discussion/169620/hijacking-the-p2-eval-rom

    Mike


  • what I need is some TAQOZ code to not redirect the TAQOZ console to VGA/PS2 or serial or LCDs or the led accessory board to display strings, just simply to a mailbox in HUB usable from PASM.

    For someone fluent in TAQOZ maybe a 2 or three liner not a one liner. For me impossible.

    Mike
  • ErNaErNa Posts: 1,738
    Hi Mike, I also one moment see clearly what I have to do, the next moment I feel to be in a dessert. While writing sophisticated code one moment, I'm stuck the next. Tachyon tasts sweet, like honeydew, but I'm looking anxiously for the spider. ... So we have to find a way to communicate the principles a simple way, what I didn't reach to now. Be the forth with you.
Sign In or Register to comment.