Reply by Jaromir Subcik February 20, 20052005-02-20
Hi Joerg,

> What really bugs me are these generic error
messages that mean
> nothing: "Failed to initialize device". Yeah, right. Like why?
> Considering that this USB programmer contains a fat MSP plus a TUSB it
> can't be rocket science to provide some feedback. Such as "cannot
> match target voltage" or "xyz pin stuck" or "reset
failed" and so on.
> Something meaningful. Even the engine controller in my car can do that
> and this is not a luxury vehicle.

You are right, not only this message is meanigless. But as you mentioned
"reset failed" it brings me an idea - check reset circuit if it is
O.K.
You may try to reset MSP just before starting (switching to) C-SPY debugger.
Sometimes it helps.

Jarda


Beginning Microcontrollers with the MSP430

Reply by lee_...@... February 20, 20052005-02-20
--- In msp430@msp4..., "microbit" <microbit@c...> wrote:
> ...
> Reminds me of the ever-puzzling legacy BIOS error when your 
keyboard's not connected :
> (or faulty ?)
> 
> Keyboard error : press F1 to continue....
> 

As the PC keyboard typically labels the "carriage return" key 
as "Enter", for 20 years I have resisted all attempts to have on-
screen or printed instructions of "Press Enter to exit". A recent 
embedded app forced me to break my own rule per customer spec of the 
screen prompts.

Lee




Reply by Matthias Weingart February 20, 20052005-02-20
On Sun, Feb 20, 2005 at 11:10:26PM +1100, microbit wrote:
> 
> Hi Joerg,
> 
> > It reminds me of an episode in Jack Ganssle's Embedded Muse: He
passed
> > an ATM in Germany that had an error message "This application
closed
> > due to an illegal procedure...". Wow, what a level of diagnostic
> > precision. Jack took a picture of it:
> > 
> > http://www.ganssle.com/tem/tem106.pdf

Well this is message is a Microsoft standard message I can reproduce nearly
every week on a win 9x machine. But the ATM's (information terminals) at
the
Munich electronica were much more funny. I got a similar error. After
pressing "print" the application does not respond anymore and it took
5
minutes until it shows this message ("illegal function call") (In the
meantime the asian tourist before me was giving up to wait for the print
out). I was able to moving this error message and -funny- behind it it shows the
correct "paper out" message. Moving both windows at the corners of the
screen I could using the terminal again.

The reasons for unstable applications are simple:
Boss to coder: "implement features, proper handling of error messages do
not sell"

        Matthias

Reply by microbit February 20, 20052005-02-20
Hi Joerg,

> It reminds me of an episode in Jack Ganssle's
Embedded Muse: He passed
> an ATM in Germany that had an error message "This application closed
> due to an illegal procedure...". Wow, what a level of diagnostic
> precision. Jack took a picture of it:
> 
> http://www.ganssle.com/tem/tem106.pdf

read the article, a-MUSE-ing :-)

Reminds me of the ever-puzzling legacy BIOS error when your keyboard's not
connected :
(or faulty ?)

Keyboard error : press F1 to continue....

WTF ??? Who came up with that... :-)


-- Kris
"Follow the White Rabbit"






Reply by Joerg Schulze-Clewing February 19, 20052005-02-19
Hello Dieter,

> last week i started using a USB FET and it did not
work until i 
> shortened the 20 cm JTAG cable down to 15 cm. The have a note somewhere 
> that this connection should not exceed 20 cm length....

Yes, it has to be short. But I am using the pod from TI and it is
under 20cm.

What really bugs me are these generic error messages that mean
nothing: "Failed to initialize device". Yeah, right. Like why?
Considering that this USB programmer contains a fat MSP plus a TUSB it
can't be rocket science to provide some feedback. Such as "cannot
match target voltage" or "xyz pin stuck" or "reset
failed" and so on.
Something meaningful. Even the engine controller in my car can do that
and this is not a luxury vehicle.

Instead, users like us have to print out the USB pod's schematic (just
did that), figure out how it works, schlepp a fast storage scope to
the MSP bench and try to diagnose which signals on the JTAG connector
look normal and which signals don't.

It reminds me of an episode in Jack Ganssle's Embedded Muse: He passed
an ATM in Germany that had an error message "This application closed
due to an illegal procedure...". Wow, what a level of diagnostic
precision. Jack took a picture of it:

http://www.ganssle.com/tem/tem106.pdf

> There was another issue installing the USB FET
drivers. Under XP SP2 
> they could no be found automatically when i connected the USB FET the 
> first time, but i had to manually direct the installer to the correct 
> subdirectory of /programs/IAR/...

Well, yes, but I would expect a workbench type software to present an
error message when it can't find a driver. That ain't so difficult.

Regards,

Joerg.




Reply by Dieter Teuchert February 19, 20052005-02-19
Hello Joerg,

last week i started using a USB FET and it did not work until i 
shortened the 20 cm JTAG cable down to 15 cm. The have a note somewhere 
that this connection should not exceed 20 cm length. In my case i have 
to use an extra adapter to connect my MSP430 modules. After that the USB 
FET did not fail again.

The same happened 2 years ago when i started a Blackfin project. It 
appears like with USB those JTAG debugging ports can be driven pretty 
close to their limits.

There was another issue installing the USB FET drivers. Under XP SP2 
they could no be found automatically when i connected the USB FET the 
first time, but i had to manually direct the installer to the correct 
subdirectory of /programs/IAR/...

Regards,
Dieter Teuchert


Joerg Schulze-Clewing wrote:

>Hello Micah,
>
>Thanks. Finally one of the assembler examples, also blinking LED,
>built without errors. The C examples don't.
>
>So, I joyfully sent it to debug for download. Got another bloody nose:
>"Fatal error: Failed to initialize device". Oh drat. Tried with
and
>without external power, reset the UIF several times (USB first, then
>target). Tried two different F1232 headers. Now I wrote to TI, see
>what they say.
>
>So far this is many times more frustrating than it was in 1994 with
>the old 89C51. Maybe I should go back...? Nah, not yet.
>
>Regards,
>
>Joerg.
>
>
>
>
>
>.
>
> 
>Yahoo! Groups Links
>
>
>
> 
>
>
>
>
>  
>

-- 
Dipl.-Phys. Dieter Teuchert
Software und Systeme
Postanschrift:



Telefon:
Telefax:
EMail Firma:
EMail perslich:
Internet:
	Rommelstr. 6
D-76571 Gaggenau
Germany

+49 7225 989253
+49 7225 989254
info@info...
dieter@diet...
_http://www.cadt.de_






Reply by Micah Stevens February 18, 20052005-02-18
As much as I have some issues with IAR, I think you're pretty unique in
having 
this much trouble. I'd uninstall it and install it again if I was you. If I

recall I had to do that once a while back to get things to work at all. 

Or try out crossworks, I hear it's pretty nice, if only it worked. :) 

-Micah 


On Friday 18 February 2005 02:48 pm, Joerg
Schulze-Clewing wrote:
>  Hello Micah,
>
>  Thanks. Finally one of the assembler examples, also blinking LED,
>  built without errors. The C examples don't.
>
>  So, I joyfully sent it to debug for download. Got another bloody nose:
>  "Fatal error: Failed to initialize device". Oh drat. Tried with
and
>  without external power, reset the UIF several times (USB first, then
>  target). Tried two different F1232 headers. Now I wrote to TI, see
>  what they say.
>
>  So far this is many times more frustrating than it was in 1994 with
>  the old 89C51. Maybe I should go back...? Nah, not yet.
>
>  Regards,
>
>  Joerg.
>
>
>
>
>
>  .
>
>
>
>
>
>  .

Reply by Joerg Schulze-Clewing February 18, 20052005-02-18
Hello Micah,

Thanks. Finally one of the assembler examples, also blinking LED,
built without errors. The C examples don't.

So, I joyfully sent it to debug for download. Got another bloody nose:
"Fatal error: Failed to initialize device". Oh drat. Tried with and
without external power, reset the UIF several times (USB first, then
target). Tried two different F1232 headers. Now I wrote to TI, see
what they say.

So far this is many times more frustrating than it was in 1994 with
the old 89C51. Maybe I should go back...? Nah, not yet.

Regards,

Joerg.




Reply by Micah Stevens February 18, 20052005-02-18
Oh yeah, they strip attachments. I forgot. Sorry.

Just the simple blinking LED one.. The first one in the list. 

-Micah 

On Friday 18 February 2005 01:52 pm, Joerg
Schulze-Clewing wrote:
>  Hello Micah,
>
>  Which example code did you try? One that came with the IAR SW?
>
>  I couldn't get an attachment, neither via Yahoo nor the gmane news
server.
>
>  Regards,
>
>  Joerg.
>
>
>
>
>
>  .
>
>
>
>
>
>  .

Reply by Joerg Schulze-Clewing February 18, 20052005-02-18
Hello Micah,

Which example code did you try? One that came with the IAR SW?

I couldn't get an attachment, neither via Yahoo nor the gmane news server.

Regards,

Joerg.