Shop OBEX P1 Docs P2 Docs Learn Events
Attention C3 owners: Catalina test request — Parallax Forums

Attention C3 owners: Catalina test request

RossHRossH Posts: 5,519
edited 2011-04-26 15:57 in Propeller 1
All,

I have had some reports of some problems with Catalina on the C3 that I cannot reproduce on the hardware I have, and I would appreciate it if any C3 owners can run a simple test and post their results here.

The attached zip file contains two batch files (sorry, Windows only) to run two tests. The first test reprograms the SPI Flash with a Catalina program and executes it, and the second does the same but also reprograms the EEPROM.

Running the test requires a PC connected to the C3 via USB, and a TV and keyboard connected to the C3. Each of the tests takes a couple of minutes to execute, and requires some interaction. Please run whichever test you feel comfortable with and post the results.

There is a README.TXT in the attached file with step-by-step instructions.

Thanks in advance!

Ross.
zip
zip
C3_tests.zip
65K

Comments

  • TharkunTharkun Posts: 67
    edited 2011-04-20 12:57
    Hi Ross,

    i did some tests for you with my C3, here is the result from my DOS-Window:

    C:\A>test_1
    Using Propeller (version 1) on port COM5 for first upload
    Using port COM5 for subsequent uploads
    Warning: no sync received for addr 00008000
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Error: too many retries for addr 00008200
    C:\A>test_2
    Using Propeller (version 1) on port COM5 for first upload
    Using port COM5 for subsequent uploads
    Warning: no sync received for addr 00008000
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Error: too many retries for addr 00008200
    Using Propeller (version 1) on port COM5 for upload

    C:\A>


    On my C3-Monitor i got a bluescreen with blinking cursor, that's all !?!?!?



    Tharkun
  • RossHRossH Posts: 5,519
    edited 2011-04-20 15:40
    Tharkun wrote: »
    Hi Ross,

    i did some tests for you with my C3, here is the result from my DOS-Window:

    C:\A>test_1
    Using Propeller (version 1) on port COM5 for first upload
    Using port COM5 for subsequent uploads
    Warning: no sync received for addr 00008000
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Error: too many retries for addr 00008200
    C:\A>test_2
    Using Propeller (version 1) on port COM5 for first upload
    Using port COM5 for subsequent uploads
    Warning: no sync received for addr 00008000
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Warning: no sync received for addr 00008200
    Error: too many retries for addr 00008200
    Using Propeller (version 1) on port COM5 for upload

    C:\A>


    On my C3-Monitor i got a bluescreen with blinking cursor, that's all !?!?!?



    Tharkun

    Thanks Tharkun,

    That's even worse than I thought - the Flash program loader is not even starting correctly! Can you tell me what version of Windows you are using?

    Ross.
  • Bill HenningBill Henning Posts: 6,445
    edited 2011-04-20 16:11
    Ross,

    Try printing the JDEC ID and status of the flash chip. Perhaps the specific chip is not exactly the same as the one you have a driver for?

    Also try looking at the status after every write.

    Regards,

    Bill
  • RossHRossH Posts: 5,519
    edited 2011-04-20 17:01
    Bill Henning wrote: »
    Ross,

    Try printing the JDEC ID and status of the flash chip. Perhaps the specific chip is not exactly the same as the one you have a driver for?

    Also try looking at the status after every write.

    Regards,

    Bill

    @Bill,

    Thanks for the suggestion - I'll look into that. In this case, it appears that the very first chip erase request is never returning. It may be that there is much more variability amongst individual C3's than I thought!

    Does anyone know if all C3's use the same Flash chip?

    Ross.
  • David BetzDavid Betz Posts: 14,516
    edited 2011-04-20 17:19
    RossH wrote: »
    @Bill,

    Thanks for the suggestion - I'll look into that. In this case, it appears that the very first chip erase request is never returning. It may be that there is much more variability amongst individual C3's than I thought!

    Does anyone know if all C3's use the same Flash chip?

    Ross.

    Mine is an Atmel 26DF081A. Also, I have a prototype C3 not a production one. I don't know if that matters.
  • RossHRossH Posts: 5,519
    edited 2011-04-20 19:18
    David Betz wrote: »
    Mine is an Atmel 26DF081A. Also, I have a prototype C3 not a production one. I don't know if that matters.

    Thanks David,

    Anyone have other FLASH chips in their C3?

    Ross.
  • TharkunTharkun Posts: 67
    edited 2011-04-21 00:34
    RossH wrote: »
    Can you tell me what version of Windows you are using?

    I am using Windows 7.

    Flash: Atmel 26DF081A
  • RossHRossH Posts: 5,519
    edited 2011-04-21 06:01
    Tharkun wrote: »
    I am using Windows 7.

    Flash: Atmel 26DF081A

    Thanks Tharkun,

    Same Flash chip as mine and David's. I suppose all the C3's must be the same. But out of a sample of three C3's, each one behaves completely differently when programming the Flash - and unfortunately (!) mine works flawlessly every time, so it's going to be difficult to track down.

    I'll revisit the code that does the Flash programming and will post a new test soon,

    Ross.
  • KMyersKMyers Posts: 433
    edited 2011-04-21 14:10
    Hi late to the party as usual due to issues here at home. Run both bat files multiple times with NO errors. I have the same memory and am running Vista 32 bit. My C3 is one of the first full kits shipped.

    I did have an issue when I ran the test suite that comes with it, the flash would fail ramdonly when loaded but never after a reset. Hope this helps and let me know if I can do anything else to help!
  • David BetzDavid Betz Posts: 14,516
    edited 2011-04-21 14:15
    I guess I should say that I was running Windows 7. Both of the people who had failures were running Windows 7. What are you running Ross?
  • RossHRossH Posts: 5,519
    edited 2011-04-21 17:19
    KMyers wrote: »
    Hi late to the party as usual due to issues here at home. Run both bat files multiple times with NO errors. I have the same memory and am running Vista 32 bit. My C3 is one of the first full kits shipped.

    I did have an issue when I ran the test suite that comes with it, the flash would fail ramdonly when loaded but never after a reset. Hope this helps and let me know if I can do anything else to help!

    Thanks for letting me know! I was beginning to doubt my own sanity :lol:

    Ross.
  • RossHRossH Posts: 5,519
    edited 2011-04-21 17:19
    David Betz wrote: »
    I guess I should say that I was running Windows 7. Both of the people who had failures were running Windows 7. What are you running Ross?

    Hmmm. I run Windows XP. I guess I'll have to find a Windows 7 box to test on.

    For those people that had failures - what clock speed do you run your C3 at? If you are overclocking it (by using a different crystal) can you also try to run the tests at 80Mhz?

    Thanks,

    Ross.
  • David BetzDavid Betz Posts: 14,516
    edited 2011-04-21 17:49
    RossH wrote: »
    Hmmm. I run Windows XP. I guess I'll have to find a Windows 7 box to test on.

    For those people that had failures - what clock speed do you run your C3 at? If you are overclocking it (by using a different crystal) can you also try to run the tests at 80Mhz?

    Thanks,

    Ross.

    I'm running with the standard crystal at 80mhz.
  • jhhjhh Posts: 28
    edited 2011-04-21 18:47
    I'm on Linux, if I installed Catalina, would I get the "payload" binary that would let me convert those batch files to shell, and run the tests? (Never done any Catalina but it sounds pretty interesting.)
  • RossHRossH Posts: 5,519
    edited 2011-04-21 19:15
    jhh wrote: »
    I'm on Linux, if I installed Catalina, would I get the "payload" binary that would let me convert those batch files to shell, and run the tests? (Never done any Catalina but it sounds pretty interesting.)

    Hi jhh,

    Yes - the batch files are very trivial. If you have a C3, I'd be interested in the results.

    Ross.
  • RossHRossH Posts: 5,519
    edited 2011-04-21 19:15
    David Betz wrote: »
    I'm running with the standard crystal at 80mhz.

    Drat! Another theory down the drain!
  • David BetzDavid Betz Posts: 14,516
    edited 2011-04-21 19:20
    Maybe I should dig out my ZOG loader and try that to see if it also has trouble writing the flash.
  • RossHRossH Posts: 5,519
    edited 2011-04-21 19:34
    David Betz wrote: »
    Maybe I should dig out my ZOG loader and try that to see if it also has trouble writing the flash.

    Hi David,

    I'd be really interested in finding that out - because my FLASH programming code is (or so I thought!) just your original code taken from your caching SPI driver :lol:

    Ross.
  • David BetzDavid Betz Posts: 14,516
    edited 2011-04-21 19:38
    RossH wrote: »
    Hi David,

    I'd be really interested in finding that out - because my FLASH programming code is (or so I thought!) just your original code taken from your caching SPI driver :lol:

    Ross.

    Oh boy, I guess this is all my fault! :-(
    I'll try checking out zogload tomorrow. I had wanted to convert that to write flash images in Catalina format anyway.
  • RossHRossH Posts: 5,519
    edited 2011-04-23 23:36
    All,

    I've just been reading the datasheet for the SPI Flash chip on the C3 (AT26DF081A) and I found it can take up to 14 seconds to erase the flash! I've never seen it take anything like this amount of time on my C3, but presumably it can vary from chip to chip (and perhaps it also depends on the current flash contents) - so the problem some people have seen could be as simple as payload not waiting long enough for the Flash to erase.

    Could anyone who tried and failed to run the tests retry them with a modifed test_1.bat file? All you need to do is increase the timeout for each retry to 4000ms (i.e. 4 seconds), so that payload should still work even if the erase takes up to 20 seconds (it retries 5 times).

    Here is what the modified test_1.bat should look like:
    @echo off
payload XMM xbasic.binary -t 4000
    Thanks,

    Ross.
  • kuronekokuroneko Posts: 3,623
    edited 2011-04-23 23:47
    @Ross: Are you just waiting (and hope) or polling the status register (as suggested in the datasheet)?
  • RossHRossH Posts: 5,519
    edited 2011-04-23 23:54
    Hi Kuroneko,

    The erase program code polls the status register, and is probably working ok - but my payload loader simply times out after 5 attempts to send the next protocol packet - so if the erase is not complete within 5 seconds it will declare the load a failure.

    Ross.
  • RossHRossH Posts: 5,519
    edited 2011-04-24 21:11
    Just bumping this, as I'd really appreciate it if someone with both a C3 and Windows 7 could try the tests with the modification I described here (i.e. use the option -t 4000 on the payload command in test_1.bat).

    This could help me figure out whether I have a problem with Windows 7, or with the Flash programming code!

    Thanks,

    Ross.
  • CogSaverCogSaver Posts: 17
    edited 2011-04-26 11:37
    Tested successfully on my C3 from Windows 7 today.

    Initially I got the same error as Tharkun, but the modified test_1.bat seems to solve the problem.
  • RossHRossH Posts: 5,519
    edited 2011-04-26 15:57
    CogSaver wrote: »
    Tested successfully on my C3 from Windows 7 today.

    Initially I got the same error as Tharkun, but the modified test_1.bat seems to solve the problem.

    Great! Thanks CogSaver.

    I'll see if there's something I can do to make this automatic. I may need to add a status to my serial protocol that allows the C3 to say something like "I saw your packet, but cannot proces it yet - send it again in later".

    Ross.
Sign In or Register to comment.