Hi,
My project is based on RT685, utilizing RT685 without XPresso, but incorporating
complex makeup and external tools to build and program the Flash.
Upon powering on, the application is uploaded from the Flash to the RAM.
I would like to integrate the following options into VS Code:
Launch JLINK GDB server.
Attach to the current running firmware and initiate debugging.
Upload arm.elf to the RAM.
Could you please provide guidance on the best method to implement this?
I've discovered that cortex-debug can be utilized to launch the GDB server, as shown in the example below.
However, upon this launch, the "executable" is also uploaded.
How can I exclusively initiate the GDB server without uploading the "executable"?
Best Regarrds,
/H
{
"version": "0.2.0",
"configurations": [
{
"type": "cortex-debug",
"request": "launch",
"name": "RT685--Debug J-Link",
"cwd": "${workspaceRoot}",
"executable": "${workspaceRoot}/debug/dmc.elf",
"serverpath": "C:/Program Files/SEGGER/JLink/JLinkGDBServerCL.exe",
"servertype": "jlink",
"device": "MIMXRT685S_M33",
"interface": "swd",
"serialNumber": "", //If you have more than one J-Link probe, add the serial number here.
"runToEntryPoint":"main",
"showDevDebugOutput":"raw",
}
]
}
}
Hi,
If you want to benefit from our NXP project and debug support, first of all, please switch to NXP extension which relies on "cppdbg" support, not "cortex-debug". See our extension documentation here: https://github.com/nxp-mcuxpresso/vscode-for-mcux/wiki
Then, simply right click on a project (in NXP Projects view) and select "Attach".
Or, if you prefer to work directly with launch.json, please be sure you have
Hi,
Is it correct that in order to use the NXP extension for VS Code
It's mandatory to port my project to NXP Xpresso,
or it can be used without any Xpresso project ?
Best Regards,
Shlomo
Yes, it should be an MCUXpresso project type. But this should be achievable (if you have a pure cmake project) by importing it in our extension. Please check on: https://github.com/nxp-mcuxpresso/vscode-for-mcux/wiki/Working-with-Pure-CMake-Projects#importing-a-...
Also note that even if in doc is specified a zip cmake project as example, it will also work for a project folder.
Regards,
Cristian
Hii,
This will prevent the automatic upload of the executable upon GDB server launch. Make sure to keep the "serverpath," "servertype," and other relevant settings intact. This way, you can separately upload the "arm.elf" to RAM without it being uploaded automatically during GDB server launch.