Shop OBEX P1 Docs P2 Docs Learn Events
Can I read/pull my own code from my prop? — Parallax Forums

Can I read/pull my own code from my prop?

turbosupraturbosupra Posts: 1,088
edited 2014-09-26 05:31 in Propeller 1
I left my code on my thumb drive at home and would like to work on my prop before I get back home, is there a utility for this?

Comments

  • PublisonPublison Posts: 12,366
    edited 2014-09-25 07:17
    turbosupra wrote: »
    I left my code on my thumb drive at home and would like to work on my prop before I get back home, is there a utility for this?

    Not really, unless you can read bytecode.
  • turbosupraturbosupra Posts: 1,088
    edited 2014-09-25 07:23
    No I can't, and I would want to be able to load it into the prop tool and then continue to update it ... sounds like this is not possible
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-25 07:27
    I have a utility that will generate Spasm source from a binary file. It also attempts to generate Spin source from the Spasm source, but it generates generic variable and method names. It also has to interpret jumps and guess at the high level command that generated them. I think it generates if and elseif for forward jumps and repeat while for backward jumps. After all the editing you'll need to do on the generated Spin code it's probably better to just wait till to get home to work on it.

    If you want me to try to generate a source file just extract a binary file from the EEPROM and post it to this thread or PM me.
  • turbosupraturbosupra Posts: 1,088
    edited 2014-09-25 07:34
    That'd be great Dave. How do I get the binary off of my prop?
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-25 07:43
    You need to load a program into RAM that reads the EEPROM and dumps the hex code to the serial port. You then copy the hex code from PST or whatever terminal emulator you use and store it in a text file. I can convert the hex file to a binary file. I'll see if I can write a hex dump program in a few minutes, or maybe somebody already has one that they can post.
  • PublisonPublison Posts: 12,366
    edited 2014-09-25 07:48
    Dave Hein wrote: »
    I have a utility that will generate Spasm source from a binary file. It also attempts to generate Spin source from the Spasm source, but it generates generic variable and method names. It also has to interpret jumps and guess at the high level command that generated them. I think it generates if and elseif for forward jumps and repeat while for backward jumps. After all the editing you'll need to do on the generated Spin code it's probably better to just wait till to get home to work on it.

    If you want me to try to generate a source file just extract a binary file from the EEPROM and post it to this thread or PM me.

    I stand joyfully corrected. :)
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-25 08:03
    Here's a quickie EEPROM dump program. It compiles, but I haven't tested it. Make should you load it into RAM and not into EEPROM or you will wipe out the program that is in EEPROM. In the Prop Tool F10 will compile and load into RAM. F11 loads into EEPROM. Do not push F11.
    spin
    spin
    eedump.spin
    523B
    spin
    spin
    spin_i2c_driver.spi.spin
    3K
  • turbosupraturbosupra Posts: 1,088
    edited 2014-09-25 08:20
    Ok, here it is, the only thing I tweaked on your code is to add a 3 second wait before main starts. Thank you.
    Dave Hein wrote: »
    Here's a quickie EEPROM dump program. It compiles, but I haven't tested it. Make should you load it into RAM and not into EEPROM or you will wipe out the program that is in EEPROM. In the Prop Tool F10 will compile and load into RAM. F11 loads into EEPROM. Do not push F11.
    txt
    txt
    binaryDump1.txt
    65K
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-25 08:54
    Well I tried it, but my despin program wasn't quite as far along as I remembered it to be. It generated some Spasm code, and attempted to group it together into Spin statements, but it crashed at some point into the file. Here's the binary file and the partially generate Spin/Spasm file. Sorry I couldn't be of more help.
    binary
    binary
    test2.binary
    19K
    spin
    spin
    test2.spin
    8K
  • Mark_TMark_T Posts: 1,981
    edited 2014-09-25 09:50
    turbosupra wrote: »
    I left my code on my thumb drive at home and would like to work on my prop before I get back home, is there a utility for this?

    There is a way to prevent this happening again, keep the drive on your keychain... Doesn't stop
    you locking yourself out because the drive's still in the machine though!

    Backup to cloud and USBdrive?
  • Duane DegnDuane Degn Posts: 10,588
    edited 2014-09-25 10:02
    Mark_T wrote: »
    Backup to cloud and USBdrive?

    I keep my Prop files in my Dropbox. No more trying to sync files with a USB drive.

    I also like to know which program is presently loaded in a Propeller. To do this I add a short DAT section at the top of the program where I keep the name of the program.
    DAT programName  byte "Hexapod140924c", 0  
CON

    Near the beginning of the program I send this string to the terminal so I know which version of the code is presently loaded.

    This has been a huge convenience.
  • turbosupraturbosupra Posts: 1,088
    edited 2014-09-25 10:18
    If you saw my keychain, you would not be encouraging more items to be placed on it.

    Just me being diligent to double check I have everything would be the best solution, the cloud would also be a good solution.

    I like Duane's naming system, I have yet to setup a rx routine in this code, but that is in the plans.

    Mark_T wrote: »
    There is a way to prevent this happening again, keep the drive on your keychain... Doesn't stop
    you locking yourself out because the drive's still in the machine though!

    Backup to cloud and USBdrive?
  • turbosupraturbosupra Posts: 1,088
    edited 2014-09-25 10:55
    Thank you for trying Dave
    Dave Hein wrote: »
    Well I tried it, but my despin program wasn't quite as far along as I remembered it to be. It generated some Spasm code, and attempted to group it together into Spin statements, but it crashed at some point into the file. Here's the binary file and the partially generate Spin/Spasm file. Sorry I couldn't be of more help.
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-25 11:51
    I ran the binary file through an earlier version of despasm that did not try to group Spasm code together into Spin statements and it was able to get through the entire binary file. I attached test2.spasm within test2.zip below. Of course this does not help you much, but in theory a Spin source could be regenerated from this.
    zip
    zip
    test2.zip
    36K
  • RaymanRayman Posts: 15,391
    edited 2014-09-26 02:59
    That's pretty neat Dave.
    I took a look, just out of curiosity...

    I think if it were my Spin code lost and I knew what some of your symbols meant, I could remember what my Spin code was...
  • Dave HeinDave Hein Posts: 6,347
    edited 2014-09-26 05:31
    There's no official assembly language for Spin bytecodes, so I invented one and called it spasm. I've attached the documentation for the Spasm assembler that describes the opcodes.

    The grouping for the Spin statements can be determined by following through the stack usage for the opcodes, and locating when the stack goes down to zero. Exceptions to this rule are instructions like repeat that load parameters on the stack that are used during the entire loop.
    txt
    txt
    spasm.txt
    15K
Sign In or Register to comment.