Forum : ARM
| Post Information | Post |
|---|---|
|
June 25, 2009 - 7:46pm
|
We have a board with the ST912faw244 on it. Every time, the Seggar JTAG shipped by IAR will program the parts whether it is a brand new board or reprogramming a board that already has code. I am trying to set up a production programming station with the Raisonance Rlink which just arrived in the mail today. I installed it on a lap top. If I download code to a board of ours that has already been programmed with different code, it works fine every time. But, if we take a brand new board that has never had any code in it - I try to program it, it looks like it goes quickly through the erase step, but when it starts the program step, it never starts drawing the progress bar, stops and then pops up an alert window that says "OPI Driver Error" in the title bar and "error reported by device" right above the ok button. This happens every time, even after powering up everything, different memory images, and on different brand new boards. The same board can be programmed everytime with the other company's JTAG device and the Raisonance RLink can always program a board that already has code in it. any ideas? Eric Krieg |
Hi
Can you confirm you use the latest version of our software and if not canyou try them.
http://www.mcu-raisonance.com/mcu_downloads.html
What is the diffrence between new board and the older one? Are they supposed to be the same boards?
Often a difference between our programmer and other JTAG programmer can be the RST signals handling.
Can you check this point from http://raisonance-forum.xsalto.com/viewtopic.php?id=2497
Can you also try to use STR9_pgm.exe to program your boards (it is added in the path).
Just run STR9_pgm.exe without argument to get the help.
I would like to know ifd you can erase and program with this executable and if not what is the error message.
Regards,
Matloub
yes, I have the latest software. I have had IAR's latest program also have the same problem. I have the RST and TRST signals correct. I have a Raisonance Jan 2009 verison of str9_pgm.exe in my raisonance/ride/bin directory. But I'm not sure the whole structure of the command line and chain files and everything else needed to set up str9_pgm.exe to work off a command line for my environment. When a brand new board that has never been programmed comes in, all the raisonance JTAG programming programs I have tried and the IAR JTAG programmer fail to program it that first time. But if I take that same board over to where I have a SEGGER USB JTAG programmer hooked up through an IAR IDE, it will program it. After the Seggar programmer has programmed the board the first time, then the same board can be easily programmed with any raisonance or IAR USB JTAG tool.
An ST support person was in here in person trying to get it to work. ST has asked us to try the same thing with a newer version of their ARM chip on our board, so we will try that and let you know what we find.
thanks for your efforts,
Eric
Hi,
There is nothing in the programming spec that could explain what you describe.
The only possible explanation that I can imagine is that the parts require some kind of factory pre-programming (calibration, ...) and you happen to have got parts that were not configured.
Did you get the chips from a standard distribution process?
Are you sure they are not prototypes?
The suggestion from the ST support person is the same I would make. Please let us know the results of these tests.
Best Regards,
Vincent
We got the parts unprogrammed from ST - I have tried it with both newer and older parts. One ST person told me that the problem may be that the Reset line on the JTAG is not connected on our target circuit board. We were originally told to make it that way, and it does work that way with the Segger programmer. When I next get a new board unprogrammed to test, I will experiment with connecting that line in or pulling different lines high or low with a 10 K resistor and post my results here in case it can help anyone else.
A strange thing I have found with some samples of 805 date code ST912 parts - 4 our of 5 of them would crash after I initialized the ENET part of the chip - the same code ran fine on older series chips.
thanks for considering this,
Eric Krieg
Hi,
According to the STR9 programming spec from ST that I have, we cannot guarantee that programming will work if the RST signal is not connected from the programmer (RLink) to the STR9. If the signal is not connected, I expect that the success of the programming will depend on the previous content of the flash, which would explain what you observe. It could also depend on the algorithm that the programmer chooses for programming, as there are several possible choices for us there. That would explain why some programmers can program it without RST and not others. But you are out of spec when doing this and even if you are using a programmer that can program without RST, there might be other problems arising later, so it is not recommended.
So you must connect the RST signal. Be careful not to connect RST and TRST together. They are two different signals, both required and independent.
Best Regards,
Vincent
Vincent,
Yes, now that we have changed our JTAG connection on our board to include controlling the reset line, your programmer is now able to program brand new ST 912 parts. Strange that from the beginning, I thought the ST people told us to just use the wiring that we had for the JTAG. I have no idea how many other people may have this problem too.
Thanks for your help,
Eric