Is there a jumper setting for this?
You don't have to disable OpenSDA at all.
Simply connect the JTAG connector.
Thanks Daniel. I did, I connected a Lauterbach ICD and as soon as i do that the 3.3V and 5V is pulled down. Perhaps something wrong with the ICD or connector. I did notice on page 5 of the schematics (in really small font) that a trace has to be cut to isolate the MCU from the opensda. I can also try connecting another ICD and see what happens.
I don't have any problem programming the board with Lauterbach.
Lauterbach connected to J10.
Powered from SBC, J8.1-2.
Do you have only one board?
I got the same issue,is it possible to share me your startup Scripts of lauterbach?
My lauterbach:LA-3500 DEBUG-USB3
My demo board:S32K148EVB-Q176 2017 NXP B.V.
The latest version of scripts can be found on the Lauterbach homepage.
I had no problems to connect to the board using J9. If any cable is connected to the OpenSDA USB port, then I would suggest removing it for testing.
Additionally please check, that R113 and R121 near J10 are not soldered. On debugger side these signals are connected to GND. If the resistors are soldered, then the voltage will drop down, when plugging in the debug cable.
In case R113 and R121 are not soldered, but the board voltage still drops down, then what kind of debug cables and adapters are used?
I had already got the S32K148EVB startup script on the Lauterbach homepage.I did't not connect any USB cable to the openSDA port when I debug with Lauterbach.
My Lauterbach adaption is 14-pin OnCE connector,So I only connected the necessary signals:
(3)GND with only one wire;
So,does the resistor R113 and R121 still need to be removed?
When I start the Trace32,software indicates the MCU in reset mode and so the hardware LED D11 does the same things.And then I load the startup script,Trace32 indicates "Syntax error",I read the script files and found the script was not work at all.
So,I have no idea which steps goes wrong.Do you have any suggestions? Thanks a lot.
Hello Wang Yang,
Let's first focus on the connection. The reset LED should not turn on, when the debugger is connected or if the debugger is stared, but in "Down" state.
Regarding R113 and R121: I found our board here and both resistors seems to be not soldered by default. They are very near to the debug header, so that you can easily check.
I do recommend to connect as many ground signals as possible. The debugger uses strong divers to support higher frequencies. This can cause electrical issues with the debug signals even on lower JTAG frequencies.
You said you have an 14-pin OnCE connector. Could you please provide for information about the type of debug cable and your adaption to the target board? Could you please post an picture of your full debugger and target setup?
Inside the debugger: What do the windows:
The result may also explain the syntax error in the scripts, which we need to check after it.
Thanks for your reply.
(1) As you say,the R113 and R121 are not solder on my S32K148EVB;
(2) I will add as many ground signals as I can to eliminate the potential electrical issues;
(3) My 14-pin OnCE adaption is a standard 2.54mm pin pitch connector that is not match with the S32K148EVB J10(1.27mm pin pitch),so I made a crude cable and the pin definition has been checked.Take the pictures as mentioned below for reference:
4-pin OnCE adaption definition:
Power up the debugger and EVB,the software indicates:
the hardware status:
Load the startup script:
The software indicates:
So,any further suggesitons?Thank you.
Could you please tell me, what VERSION.HARDWARE show? It is important to know, if you have an AUTO or AUTOPRO cable. Only the AUTOPRO is supported for general Cortex-M controller, because the AUTO cable has only limited support for debug protocol capabilities. A Cortex-M license is not avaialbe for the AUTO cable.
Inside the pictures there is a chain of adapters use and it is not clear, which pin of the debug cable is exactly connected to which pin on the board. But the first 10 pins of the AUTO-26 header are compatible to the first 10 pins of the target header, so they can be 1:1 connected. When removing the solder bridges for trace R114, R122, R136, R141 and R148 on the board, then you can directly plug the AUTO-26 cable into the target (but you need to remove the polarization notch on the board). This would remove the adapter chain in between.
The recommend tool for this controller would be the µTrace, because it fully supports all 20 pins of the target header including the trace signals and allows to use all debug and trace features of the board and controller.
Thanks to your specific answers.
I took a full picture of my LAUTERBACH,to see is it helpful to identify the VERSION.HARDWARE?
According to your answers,may be the debug cable is not match?I see it's a DEBUG-MPC5XXX-AUTO cable.MPC5XXX is power architecture which is different from the cortex-m.
Here is my cable definition list, the definition with same color has been connected together:
So,any further suggestions?Thank you.
VERSION.HARDWARE was intended to be executed inside the command line of TRACE32. It will open a window with all necessary information. But the pictures already show all necessary information. The cable is not supported by Cortex-M licenses. In your case it allows only JTAG, so that you would need to change the DEBUGPORTTYPE in the script from SWD to JTAG.
I have checked a debug cable like you have together with a software back to 2018/09 and could connect to the target with such a cable type. I have connected the 26pin cable directly to the target by just breaking one of the polarity notch, and it works even without removing R114, R122, R136, R141 and R148 on the board. I would suggest trying the same on your side.
You can also check, if reset is toggling on the right pin on the target connector by executing the commands:
JTAG.PIN NRESET 0 // Assert nReset pin
JTAG.PIN NRESET 1 // De-assert nReset pin
Retrieving data ...