[can-do-users] CAN-Do info

[Stephen Moraco]

Enric,
    After looking at the code, I don't see how it can be looping, reading
status.
    I'm still trying to find out why this could happen...

    Anyway, I'm writing because I saw that we will have better debug output
in
    this case, if you use level 6 messaging, not level 5 as we had been
doing.

    Please run your tests with debug messaging level 6 selected.

Regards,
Stephen
--
Stephen M Moraco


On Wed, Aug 26, 2009 at 11:01 AM, Stephen Moraco <kz0q at amsat.org> wrote:

> Enric,
>     You are correct in that it appears to be working differently.  However,
>     this can only be due to how the CAN232 hardware is responding.   I
> suggest
>     that something was not configured/powered up correctly.
>
>     Please set this up again and run with logging again.  If you do not get
> C-V-S6 sequence
>     then it is not correct.
>
>     While you are doing that, I'll look (tomorrow night) into flasher
> source code to see if there is
>     any way it can get into this C-F-F-F-F-F  loop.  I have never seen this
> before.  It is like
>     it somehow got past the startup code (but it should not be able to.)
>
> Regards,
> Stephen
> --
> Stephen M Moraco
>
>
> 2009/7/23 Enric Fernandez Murcia <efermu at gmail.com>
>
> Hi Stephen,
>>
>> Today flasher seems to work different from yesterday (it's really
>> sorprendent). As far as I can see, dvcCheck232 steps now are:
>>
>> C
>>
>> F
>>
>> Two days before, were:
>>
>> C
>>
>> V
>>
>> S6
>>
>> (when stack smashing is detected)
>>
>> But following the CAN232 datasheet from builders, routines I may follow
>> are:
>>
>> C
>>
>> V
>>
>> S6 (500kbit)
>>
>> X1 (Auto poll/send ON)
>>
>> U2 (57600 baud)
>>
>> Z1 (time stamp ON)
>>
>> O
>>
>> F (datasheet tolds this command only works if port is open)
>>
>>
>> What is your opinion about it?
>>
>> Two logs are attached, export command doesn't works correctly.
>>
>>
>> Regards,
>> Enric.
>>
>>
>>
>> 2009/7/21 Stephen Moraco <kz0q at amsat.org>
>>
>>  Enric,
>>>     You don't need to install the .deb file when you are building from
>>> newer source.
>>>     You can ignore installing the .deb.
>>>
>>>      You don't need to do a 'make install' as it will run just fine from
>>> your local directory
>>>      after the 'make'.
>>>      So let's not do a 'make install'
>>>
>>>      You copied more into the log and part of it indicates that you are
>>> talking to the
>>>      CAN232 until a stack "smash" occurred (whatever that is.... it's a
>>> new term for me.)
>>>
>>>      So, can you run it again dvcChk232, that is. and see if this happens
>>> all the time?
>>>      Please send me the logs from your attempts.
>>>
>>>
>>>       Excerpt from your console log:
>>> ------------------------------
>>>
>>> DBG-3: CAN232IO::openDevice() Retrieving F/W, H/W versions
>>> DBG-5: --------Sent to Module--------
>>>     0: 56 0D                                           | V
>>> DBG-5: --------Rcvd fm Module-B------
>>>     0: 56 31 32 32 30 0D                               | V1220
>>>  CAN Controller: Lawicel CAN232- H/W-v12, S/W-v20
>>> DBG-5: --------Sent to Module--------
>>>     0: 53 36 0D                                        | S6
>>> DBG-5: --------Rcvd fm Module-B------
>>>     0: 0D                                              |
>>> DBG-5: --------Sent to Module--------
>>>     0: 4F 0D                                           | O
>>> *** stack smashing detected ***: ../flasher terminated
>>> ======= Backtrace: =========
>>>
>>> ------------------------------
>>>
>>>
>>>     When we can get you past this you will be able to flash your devices!
>>>
>>> Regards,
>>> Stephen, KZ0Q
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gag.com/pipermail/can-do-users/attachments/20090827/a106ca23/attachment.htm>