Dear All,
USBDM has been updated to V4.10.2
Please post any queries on this version in a separate thread - I really can't cope with this new-fangled setup!
Documentation available at: SourceForge
Applications available at: SourceForge
Source code is available at: GitHub eventually.
bye
Note
History
V4.10.2 (October 2012) -
This is intended as a interim release to support Codewarrior 10.3 Beta.
It has not been extensively tested.
Use only if you wish to experiment with CW10.3 or one of the following changes is of interest.
V4.10.0 (September 2012) -
OSJTAG (build30_rev21) | Tower K-40 | 1m 20s | 53s |
USBDM-JTAG (4.10.0) | Tower K-40 | 47s | 23s |
USBDM-SWD (4.10.0) | USBDM-SWD | 21s | 17s |
Dear pgo,
I downloaded the file "usbdm-firmware-master/USBDM_CF_JMxxCLD",but there is not .s19 in it. pls check it and reply.
B.R.,
JF
Hi JF,
If you are referring to the GIT repository then you should find a .sx file. A lot of the Freescale tools produce S19 files with this extension.
In any case, if you want the release files rather than source code then you should use the installer version on Sourceforge.
bye
Dear PGO,
I have tried to debug/program MC9S12XDG128 with TBDML using the latest version of Code Warrior for HC's. When USBDM device have tried to erase flash memory, there was the following error: " Error while loading diagnostic algorithm to target system. The chip may be secured, or the derivative selected may be wrong.". Since derivative was selected correctly, I tried to unsecure the chip by clicking on "Unsecure" in Code Warrior menu. The script seemed to run well, I have selected 8 MHz bus clock, since I am using 8MHz crystal, but after that operation I was still receiving the same error. I am sure, that all connections are fine, and I have signals MODA, MODB pulled down to ground, so considering the MODC/BKGD pin is pulled up to VCC, chip should be in special single chip mode, and USBDM should work fine.
Here is the log from command line. Could you suggest me what to do, please.
Hi Pawel,
Could you try programming the device with your file using the stand-alone USBDM programmer. Make sure you select the unsecure option.
Try again with Codewarrior and let me know the result.
Thanks
Thank you very much for the advice. First I have updated the USBDM firmware to be able to use the newest HCS12 Flash Programmer, then I program flash memory with some simple program compiled with CodeWarrior. Now it works fine both with Flash Programmer and Code Warrior.
Kind regards :smileyhappy:
Dear pgo
Thank you for your great job. I'm wondering whether I can use the usbdm with IAR. Can I just choose the PE MICRO‘s driver in IAR to let it work with USBDM? I'm planning to make my own USBDM now, so I want to make sure I can make it work with IAR, not only Codewarrior. If it does not, guess someday I can try to make some contributions like you too.:smileyblush:
cheers
Sorry but USBDM does not work with IAR.
AFAIK there is no information available on the interface used. If you know of any material I will be happy to consider adding some support,
bye
Dear Pgo,
Does the USBDM not support the MPC55xx/MPC56xx devices?
Sorry Jiang, USBDM does not support MPC55 devices.
I had a look at these and concluded it would be a great deal of work to implement the required functions due to the complexity and variety of the debug interface.
bye
Dear pgo,
Thanks for your reply.
Dear pgo,
I found an issue with USBDM in CW10.3b debugger for ColdFire+ (if I remember there's the same issue with CW10.2), D2 to D7 and A0 to A3 registers are always read zero, so this is not very convenient in some cases.
regards,
Emmanuel
Hi Emmanuel,
I had thought this only occurred in 10.3 Beta.
In any case it is a bug in Codewarrior rather than USBDM I believe. It also occurs for OSBDM. I have reported it to Freescale previously so it should hopefully be fixed soon.
If you turn on the log you can see that USBDM GDI layer is being asked to read non-existent registers.
bye
Hi pgo,
This issue seems to be still there with CW10.3
pgo a écrit:
Hi Emmanuel,
I had thought this only occurred in 10.3 Beta.
In any case it is a bug in Codewarrior rather than USBDM I believe. It also occurs for OSBDM. I have reported it to Freescale previously so it should hopefully be fixed soon.
If you turn on the log you can see that USBDM GDI layer is being asked to read non-existent registers.
bye
Hi Emmanuel,
Yes it is still present in 10.3.
Nothing much I can do about it. You can try posting a bug report as I did during the 10.3Beta and see if you get any further.
Note however that USBDM is not supported by Freescale so it may well be ignored even though the error is in Codewarrior.
Sorry!
bye
true, USBDM is not supported by Freescale, but OSBDM is supported. I tried with a TWR-MCF51JF board and there's the same problem.
Hi Emmanuel,
If you have an independent check, as you have indicated, then I suggest you submit a service request.
That is what I did previously quoting the OSBDM-JM60 failure.
bye
Hi pgo,
it's already done :smileyhappy:
Dear pgo:
Thanks for your excellent work on the USBDM.
I work a whole day and night to test the new USBDM 4.10.2 version,
because I want to teaching my students learning K10 and KL25 in next
year, I just prepare the hardware and usbdm for them.
The hardware is a K10 board and USBDM CF/DSC/kinetis version.
Firstly, I use the version of 4.9.4b to work with stand-alone arm programmer
within the USBDM software, and it work very well.
Secondly, I uninstall the old 4.9.4b version and install the newest 4.10.2 version,
and the stand-alone arm programmer still work very well.
Thirdyly, I test a project with more than 120KB data with 4.9.4b and codewarrior 10.2,
I enable the verify option in debug setting, and everything is OK.
Finally, I test the same project with 4.10.2 and codewarrior 10.3beta, and when I enable
the verify option, codewarrior will report error at address of 1800. But when I power off and then
power on the hardware, the code same to work well.
So, is there something error with download process with codewarriror 10.3beta and USBDM 4.10.2?
Many thanks,
Tao
Hi Tao,
I am not aware of any problems with that combination.
If the project is not confidential, would you be able to post the exported project and I will investigate the problem?
You could also try the following tests if convenient:
Also
It is possible that the verify in Codewarrior is reporting differences in the security values programmed to the target. Does Codewarrior report the actual address failing?
bye
PS. If you are only interested in the Kinetis devices I would recommend the simplified JS16 based SWD version of the programmer. It is considerably faster that the JTAG based programmer and works with a wider range of Kinetis devices..
Hi, pgo:
I finished the test steps that you suggested.
1, I program the K10 chip using Codewarrior 10.3 beta, it report error, the figure is in the attach files.
2, I only verify with USBDM 4.10.2's arm standalone programmer, it report everything is OK, and the board working well
after I re-power it.
3, The stand alone programmer can program and verify the K10 board no any problem.
and I will test the same board with SWD version of USBDM hardware next day.
The attached file is my project and error figure.
Many thanks,
Tao