Hello,
After creating and programming a "bare-metal" project (not using ISF) into FXLC95000 no BDM connection can be established.
Tools: CodeWarrior 10.6, P&E USB Multilink Universal
Could you please advise on a possible reasons?
Thanks,
ilya
Solved! Go to Solution.
Hi Ilya,
I answered already several days ago but apparently it was "rejected by a moderator", and I just saw the notification now.
Anyway, here it is again:
Looking at the S19 record file, it seems that initial Stack Pointer (0x008040EC) is beyond the 16kBytes RAM area (0x0080000-0x00803FFF) so you may check the linker file. I don’t know how this exception is handled, nevertheless the BDM adaptor should be able to take control of the target during the POR reset.
To make sure that the Target “waits” the BDM signal after a POR, I suggest you do the following when the CW pop-up window instructs you to power cycle the DUT:
Let me know if this workaround solves your connection issue.
Regards, Jacques.
Hello Ilya,
In case you are using your own board, please make sure you have the following configuration in order to program the device:
I hope this information will be useful for you.
Have a great day.
David
David,
Many thanks for your response.
But it is a KITFXLC95000EVM board
Hi Ilia,
Can you give us more information such as:
- are you using the KITFXLC95000EVM development board or a custom one,
- do you get any warning/error message in the CW console or as a pop-up window,
- what are the Debug Configuration exact settings.
Any screenshot would be valuable to help us diagnose your issue.
Ultimately you may also provide your CW project.
Regards, Jacques.
Hi Jacques,
Do you have any ideas?
Hi Ilya,
I answered already several days ago but apparently it was "rejected by a moderator", and I just saw the notification now.
Anyway, here it is again:
Looking at the S19 record file, it seems that initial Stack Pointer (0x008040EC) is beyond the 16kBytes RAM area (0x0080000-0x00803FFF) so you may check the linker file. I don’t know how this exception is handled, nevertheless the BDM adaptor should be able to take control of the target during the POR reset.
To make sure that the Target “waits” the BDM signal after a POR, I suggest you do the following when the CW pop-up window instructs you to power cycle the DUT:
Let me know if this workaround solves your connection issue.
Regards, Jacques.
Jacques, it worked!
I believe I tried to do the same before ( to prevent a target to run from POR ) but it didn't work... So at first there was no such optimism after reading your comment, but I was pleasantly surprised when I did all the steps and it worked this time :smileyhappy: Thank you!
Excellent !
As you are dealing with a "bare-metal" project, I was wondering wether you intend to program and process internal accelerometer measurement.
For your information, we are preparing a "bare-metal" project exemple to show customer how to do so.
It should be available within a few weeks, let me know if you are interested.
Regards, Jacques.
Jacques,
Thank you, that would be great.
Yes, I intend to process accelerometer data and to calculate FFT using MAC.
Thanks, ilya
Hi Jacques,
- it is KITFXLC95000EVM board, not a custom one.
- I constantly get a popup dialog asking to power cycle a board. But this doesn't help I assume a programming adapter can not "catch" a target during POR or something. This started to happen when I programmed a project.
After many unsuccessful connection attempts a chip was re-soldered, it was replaced by fresh one. This worked, I was able to "Attach" with debugger (w/o Flash programming). But when I tried to program my project again, the same issue appeared - no BDM communication.
- Debug configuration:
I tried to attach my s-record but I guess it was a reason why my previous post was rejected.
Thank you for your reply,
ilya