STM32CubeIDE for Visual Studio Code (MCUs)

I am using STM32CubeMX 6.16.1 to generate project with external Flash loader for STM32N6. When I generate for STM32CubeIDE and build the extmemloader then STM32CubeProgrammer can parse it (the .elf renamed to .stldr). But when I generate CMake project and build with vscode (with STM32 extensions) then STM32CubeProgrammer fails to parse. Compiler/linker removes the StorageInfo structure from output completely. I fooled it to keep it. Now it doesn't fail immediately, but still can't parse any info either.Here's the CubeProgrammer EL list view. First is vscode built loader, second is Eclipse built:I compared the ELF files, Map files, compiler and linker flags. There are plenty of differences and I haven't figured out what exactly causes it. I changed the flags but nothing helped. I noticed vscode uses GCC 14 and STM32CubeIDE uses GCC 13 but tricking vscode to use GCC 13 didn't solve it either. 

Hi Starting my first project testing VSCode integration of ST tools instead of using Eclipse... Just reporting an issue with build analyzer, it is indicating that I have 188% of FLASH used...The compiler output gives the correct information:It seems an issue parsing the map or elf file to operate the tool. 

Hi,I can't debug my application through GDB and ST-Link when the computer has no internet connection. I get this error:All bundles, both System and project, are up-to-date. When I connect to the internet, I can debug fine, but when I then disconnect and immediately try again (without exiting VS Code), I can't debug anymore.IMO debugging should be possible without an internet connection, e.g. when having to fix an error in the field.Regards,Niklas

Hello,I have a question about vscode with STM32 extension. I have setup a project with vscode and stm32 extension, and after compiler, it can do debug, and I can see the Global Variables on the VARIABLES window. as belowbut when I use vscode to do another design not with stm32 extension. I do the debug, but I cannot see the Global Variables on the VARIABLES window. as below, only Locals and Registers.Why no Global Variables? what is the difference among the 2 vscode?Thanks. 

After installing under STM32Cube key actions, and all other headings, "There is no data provider registered that can provide view data." is displayed. I have tried adding the path to Cube MX. In the STM32Cube Core plugin, runtime status shows not yet activated. Manually activating does not work. Running on Windows 11 arm64.If I roll back to a version 2, everything seems to function, however I need version 3 working. Any suggestions?

How can I attach to a running STM32 target while using its specific JTAG address?I want to debug on an already-running STM32 without altering its memory state (no reset, no download of new firmware). I currently do this on STM32CubeIDE by setting my debug configuration to "No reset" in the "Debugger" tab and setting "Download" to "false" in the "Startup" tab.The target STM32 is in a JTAG chain with other devices.On the VSCode extension: If I select "attach" in "request" the "serverJtagCfg" option becomes invalid. Not entirely sure where to look for this. Any help is appreciated.Thanks

Starting a debug session in VS Code using the RTOS proxy plugin for FreeRTOS and a J-Link Adapter consistently fails.Debugging the exact same application without RTOS awareness (serverRtos removed in launch.json) works. So debugging using JLink and the board works. This indicates that there is an issue related to the RtosProxy or RTOS awareness in general.Is there any further configuration needed that than the launch.json below?Environment:STM32CubeIDE for Visual Studio Code v3.7.0Packages in Bundles Manager are all on the newest versionlaunch.json configuration{ "version": "0.2.0", "configurations": [ { "type": "jlinkgdbtarget", "request": "launch", "name": "STM32Cube: STM32 Launch JLink GDB Server", "origin": "snippet", "cwd": "${workspaceFolder}", "preBuild": "${command:st-stm32-ide-debug-launch.build}", "runEntry": "main", "imagesAndSymbols": [ { "imageFileName": "${command:st-stm32-ide-debug-launch.get-projects-binary-from-context1}" } ], "serverRtos": { "enabled"

Hello! I am using an STM32U545 and VS Code for programming. I am experiencing the following issue: I am working with STOP mode, and once the MCU enters this state, I can no longer reprogram it. When using STM32CubeIDE, I did not encounter this problem (I believe this is related to the connect-under-reset feature, which VS Code seems to lack when programming).Is there anything I can do to avoid this issue, other than using STM32CubeProgrammer to erase the memory or adding a delay at the start of the program to give myself time to program the device after a manual reset?

Hello,I want to ask if it is possible to install STM32Cube extensions into Cursor.I looked in the normal extension list but I could not find them.Is this supported now or will it be supported in the future?Thank you.

Hello,converting a runnig project from IDE to VisualStudioCode, I have a strange problem:all global variables are not initialized...example:uint32_t TestTemp1=0x1234;When running after the main(), if i check the TestTemp1 value I don't found the expected value (0x1234), but a random value.... After checking I Found that the problem was in linker file:This NOT INITIALIZE !/* Specify the memory areas */ MEMORY { FLASH (rx) : ORIGIN = 0x08000000, LENGTH = 2048K DTCMRAM (xrw) : ORIGIN = 0x20000000, LENGTH = 128K RAM_D1 (xrw) : ORIGIN = 0x24000000, LENGTH = 512K RAM_D2 (xrw) : ORIGIN = 0x30000000, LENGTH = 288K RAM_D3 (xrw) : ORIGIN = 0x38000000, LENGTH = 64K ITCMRAM (xrw) : ORIGIN = 0x00000000, LENGTH = 64K } /* Initialized data sections goes into RAM, load LMA copy after code */ .data : { . = ALIGN(4); _sdata = .; /* create a global symbol at data start */ *(.data) /* .data sections */ *(.data*) /* .data* sections */ *(.RamFunc) /* .RamFunc sections */ *(.RamFunc*) /

Hey Devs, I migrated my V2 extension to V3 (3.6.4) on Dec 11 2025 (just before the v3.7.0 release), updated my debug workflow / launch.json etc, my STM32H745I discovery board was running my TouchGFX application (basically stock TouchGFX Designer generated app w/ CMake) fine with Dual Core debugging (new shared debugger / simplified Launch/Tasks.json is a massive improvement on my old/convoluted tasks/launcher workflow, my sincere thanks).Just tried to re-run the same code stack with my code with the new 3.7.0 extensions, hit a launch error relating to my launch.json:Extension seems to have lost context of my ide.store.json files in the project subfolders:{ // 'CM4/.settings/ide.store.json' - This line not in actual file "device": "STM32H745XIH3", "core": "Cortex-M4", "order": 0, "toolchain": "GCC" }(Similar for the CM7 version targeting the M7 core)Re-running the STM32Cube Project setup configuration didn't seem to fix it, so I just added 'deviceName' and 'deviceCore' manuall

I normally use the CMake Tools extension (by Microsoft) for compiling cmake projects in visual studio code.  After installing the STM32CubeIDE extension, the CMake Tools extension stopped working.  Presumably to do with the stm extension changing to using cube-cmake.  After uninstalling all STM32CubeIDE extensions, reinstalling CMake Tools, restaring VS code, the CMake Tools extension is still still broken.  Trying to configure a project through the IDE gives the following output:[cmakeExecutable] CMake executable not found in cache. Checking again.[proc] Executing command: cube-cmake --version[proc] The command: cube-cmake --version failed with error: Error: spawn cube-cmake ENOENT[proc] Executing command: cube-cmake -E capabilities[proc] The command: cube-cmake -E capabilities failed with error: Error: spawn cube-cmake ENOENT[main] Building folder: core2-firmware [cmakeExecutable] CMake executable not found in cache. Checking again.[proc] Executing command: cube-c

Hello,I am trying to get the VS-Code extension to work on my Intel-Mac. I already have it running on my Windows machine no problem.The problem is, that the extensions won't activate. I have created a new project using CubeMX and choose CMake as the toolchain. Then I installed the STM32 CLI tools, as well as cmake and the ARM compiler.Then I am opening the project in VS-Code and CMake asks for a target. I choose Debug.CMake extension output:[proc] Executing command: /usr/local/bin/cmake --version [proc] Executing command: /usr/local/bin/cmake -E capabilities [kit] Successfully loaded 2 kits from /Users/jan/.local/share/CMakeTools/cmake-tools-kits.json [presetController] Reading and validating the presets file "/Users/jan/Dev/fischer-technik-motor-controller/CMakePresets.json" [presetController] Successfully validated /Users/jan/Dev/fischer-technik-motor-controller/CMakePresets.json against presets schema [proc] Executing command: uname -s [presetController] Expanding presets file /Users

Using:VSCode version: 1.107.1 on Windows 11.STM32CubeIDE for Visual Studio Code version 3.7.0.I created STM32N6 project that builds two targets: FSBL and Application. I follow the instructions here: https://community.st.com/t5/stm32-mcus/how-to-debug-the-stm32n6-using-vs-code/ta-p/833499.clangd contents:CompileFlags: Add: - '-ferror-limit=0' - '-Wno-implicit-int' CompilationDatabase: Appli/build Diagnostics: Suppress: - unused-includes - unknown_typename - unknown_typename_suggest - typename_requires_specqual If: PathMatch: Appli/.* --- CompileFlags: Add: - '-ferror-limit=0' - '-Wno-implicit-int' CompilationDatabase: FSBL/build Diagnostics: Suppress: - unused-includes - unknown_typename - unknown_typename_suggest - typename_requires_specqual If: PathMatch: FSBL/.*.vscode/c_cpp_properties.json contents:{ "configurations": [ { "name": "STM32_Appli", "compileCommands": "${workspaceFolder}/Appli/build/compile_commands.json" }, { "name": "STM32_FSBL",

Using:VSCode version: 1.107.1 on Windows 11.STM32CubeIDE for Visual Studio Code version 3.7.0.I created STM32N6 project that builds two targets: FSBL and Application. I follow the instructions here: https://community.st.com/t5/stm32-mcus/how-to-debug-the-stm32n6-using-vs-code/ta-p/833499This is a single folder project, not multi-root. The problem is that when I navigate to source code which is not under FSBL or Appli then clangd does not know which complication database to use.This disturbs during coding and during debugging when I step through grayed out code with false errors highlighted.CMake (or some STM32 utility) generates .clang file into the root directory. It's stripped down version looks like this:CompileFlags: CompilationDatabase: FSBL/build If: PathMatch: FSBL/.* --- CompileFlags: CompilationDatabase: Appli/build If: PathMatch: Appli/.*But there is no path match for Drivers and Middlewares. If I open HAL GPIO driver, I can see in the STM32cube clangd lo

Hello, I am trying to set up a project using STM32Cube Extension in VS Code. When I attempt to select the Nucleo‑F207ZG board in the Board/Device selector, the board does not appear in the list.The stm32cube_core output shows the CMSIS_PACK_ROOT environment variable, and under that directory the file .Download\cache\scanner_391867825626163.cache contains entries for nucleo-f207zg as well as STMicroelectronics.nucleo-f207zg_hw-board.1.0.0.pdsc. Despite these files being present, the board is still not discovered by the IDE. Could you please advise what additional checks or steps I should perform to resolve this issue?

Hi for all,I use Cube MX to generate code with CMake and then use Vscode with needed extension to compile my code.My question is : i want to use library that contain multiple folders to drive stepper motor, how to add this library to my project.best regard for all thanks for any reply 

Hi,I can't seem to find the launch configuration setting to specify a device in the JTAG chain by its position and IRpre length. I'm using a J-link programmer and I can usually do this from the command line by passing the -jtagconf x,y argument to the J-link executable. However, the debug console in VScode shows that the debug launch config called the Segger J-link GDB server without this argument. I can't seem to find how to make the debug config to append this to the J-link server call Is this supported by the STM32Cube extension?Thanks

Hello, I am using a custom environment for development and debugging for NUCLEIO-STM32F767ZI.It is based on a custom devcontainer, VSCode and gdb-multiarch. On the server side I have OpenOCD, running on a ubuntu 24.04 PC. This is connected via USB to the board. The Board itself is not changed in any way, maybe I changed jumpers at some point, I would need to check this if this might effect the behavior. The first connection after Powerup works always as expected, the Firmware is flashed and works as expected.If I rebuild the project or flash the firmware again, the MCU is in some kind of undefined state.Calls to different HAL functions fail, return Errors. As some return values are not verified within the generated calls, I don´t see the errors immediately, but very late during debugging. Missing interrupts as an example. The code is a simply generated STM32CubeMX project, without any changes. I am using only GPIOs and Systick interrupt so far. I need the right Reset com

When creating a project in STM32CubeMX, ST ARM Clang was selected. Compilation and debugging in VSCode all work, but some variables are optimized out, which seems like the compiler optimization is set too high. However, it's unclear where to set the compile flag -O0.CMakeLists.txt like this, does not work# Define the build typeif(NOT CMAKE_BUILD_TYPE)set(CMAKE_BUILD_TYPE "Debug")set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -g -O0")endif() 

Just like the normal process, I generated the H7S3L8 project using CubeMX.Then I opened the project folder with VS Code, and STM32 VS Code extension detected that it is an STM32 project.But afterward it threw a problem.It required me to manually configure the specific model of the STM32. It seems that the CubeIDE for VS Code extension did not correctly recognize the model.So I just chose the NUCLEO-H7S3L8.But when I click the 'Save and close' button, it just shows the clicked animation effect, and then nothing happens. The extension still can't properly load the project and perform automated configuration.Why is this happening?I am quite sure that other projects (such as STM32F401RCT6) can be properly recognized and loaded; moreover, if I manually go to 'Setup STM32Cube project(s)', the 'Save and close' button works normally for those projects.It seems that only the NUCLEO-H7S3L8 or STM32H7XX series has this problem, or am I the only one experiencing it?I speculate that this may b

Question 1The first image shows the toolchain configuration of my previous project, and the second image shows the configuration displayed in my newly created project. Why do I need to reinstall the toolchain when creating a new project on the same computer?Question 2I remember that when I first used the extension, these tools were all specified through scripts. However, after using it for a period of time, I found that CMake, programmer, and other tools could not find the specified paths, so I had to add them all to the environment variables. 

I have recently started a project in VSC using an STM32 device.  I first used STM32CubeMX to generate code, so that created most of the project.  I am using CMAKE.Initially there was no issue, and everything worked fine after disabling intellisense.  However, I have now become plagued by the stubborn red squiggles which refuse to disappear whatever I do.  I would prefer to continue using VSC, but will have to ditch it and move to another IDE if this problem does not go away, as I cannot work like this.  Further down the file there are many other red squiggles.  The project compiles fine, no issues there.  An example of the problem:No matter what I do, the first #include has red squiggles under it.  If I change the order of the #includes, then it is always the first one which has the squiggles.  settings.json file:c_cpp_properties.json file:Any help would be gratefully appreciated, please.

Hi everyone,I am migrating from STM32CubeIDE to VS Code using the official STM32Cube for VS Code extension (v3.4.11 pre-release). While I can build and debug my project, I am struggling to get the Live Watch feature to work.My Environment:IDE: VS Code with the latest STM32 official extension.Hardware: Nucleo board with onboard ST-LINK V3.MCU: STM32H7 series.Project Structure: Generated via STM32CubeMX as a CMake project.Project Name: projectBlink.The Issue: In the "STM32CUBE LIVE WATCH" tab, I constantly see the message: "There is no data provider registered that can provide view data.".What I've tried:Code: I declared my variable as volatile uint8_t count = 0; to prevent compiler optimization.Debug Session: I am in an active debug session using the STM32 Cube GDB Server.Adding Variables: When I click the "+" icon and enter the variable name, nothing happens, and the "no data provider" error persists.Version Check: I noticed some community posts suggesting that the v3.x pre-release mig

Hi everyone, I am developing an application for the NUCLEO-H7S3L8 kit using VS Code with the STM32 VS Code Extension. I have successfully built both the 'Boot' and 'App' projects.Currently, I am struggling with flashing the 'App' project firmware into the external Flash memory. In STM32CubeIDE, selecting an External Loader (.stldr) is very intuitive within the Debug Configuration, but I haven't found the equivalent setting in VS Code for the STM32H7RS series.Could you please guide me on how to configure VS Code to recognize the External Loader and the proper workflow for flashing code to external memory for this specific MCU?PS: On STM32cubeIDE, I'm referring to this example: https://github.com/ST-TOMAS-Examples-ExtMem/stm32h7rs_ospi/tree/main. Now, using the same example, how would I implement it if I used VS Code as my development environment?