STM32 MCUs Software development tools

Post edited by ST moderator to be inline with the community rules especially with the code sharing. In next time please use </> button to paste your code. Please read this post: How to insert source code# STM32CubeIDE ETH Configuration Bug Report ## Issue SummarySTM32CubeIDE fails to generate proper GPIO initialization code for Ethernet peripheral, resulting in non-functional MDIO/MDC communication with PHY chips. ## Environment- **STM32CubeIDE Version**: 1.19.0- **MCU**: STM32F407ZGT6 (LQFP144 package)- **PHY**: YT8512C- **Interface**: RMII mode- **Date**: September 28, 2025 ## Problem Description ### Configuration in .ioc File (Correct)The STM32CubeIDE project configuration file (`.ioc`) shows correct pin assignments:```PC1.Mode=RMIIPC1.Signal=ETH_MDCPA2.Mode=RMII  PA2.Signal=ETH_MDIOETH.MediaInterface=HAL_ETH_RMII_MODE``` ### Generated Code Issue (Incorrect)However, STM32CubeIDE **fails to generate the required `HAL_ETH_MspInit()` function** in `s

when you install stm32cubeide first time without login the myST, and coincidentally you opened a former project，you will find that code generator won't work without any error message. I've been working on the previous code all day long lol.

When I try to login in the STMCubeIDE(myST -> Login), but I can't type my account and password into the text box of the pop up login window. It seem's like not bscause of network problem as the pop up login window correctly loaded, and I the "login" button and "Create Account" button works properly.and the command line output is several lines of:    1 : Invalid condition id : UX_CORESTACK_Condition cause : nullHow can I fix it?(platform: Arch linux, Hyprland)

Hi teams,Board: Nucleo-U083RCCPU: STM32U083RToday, I faced a problem with my new board while performing RDP regression using the STM32 Cube programmer.The condition of the board prior to the issue,* Option bytes value: At default level.* Flash data: Default data [LED sample].Step that I followed,1.Initially, I have configured the OEM1 key values with a password, resulting in the transition of the OEM lock bit status from "not active" to "active". 2. I then configured only the RDP level in the Option bytes tab, changing it from Level 0[0xAA] to Level 1[0xBB], while the other OB values remained in their default state.3. I disconnected the programmer and subsequently reconnected it, allowing the changes to be reflected shortly thereafter.4. To perform the regression in the STM32Cube programmer, I adhered to the procedures outlined in UM2237 User Manual [section: 2.9.2 RDP regression with password].4.1. STM programmer connection settings4.2.  I have written the OEM

I find one example that helps me at the "example selector" . I can open, edit, and run it. if I want to copy it, I have this error. Does anyone know the solution? It creates the project, but the subfolders are not fully copied, like 80% are not there. 

Dear ST Support Team,I am currently working with STM32CubeIDE and using the NUCLEO-H755ZITx development board. On one of the boards, I am consistently encountering the following error during debugging:'filename_Debug' has encountered a problem. Error in final launch sequence. Failed to start GDB server.I have attached a screenshot of the error for your reference.This issue is blocking my development progress, and I would appreciate it if you could help me understand the root cause and suggest a resolution at the earliest.Thank you for your support.Best regards,Prathiksha H L

For the STM32H743IIKx micro, CubeMX 6.15.0 unnecessarily creates a configuration for the R1 pin when creating an LTDC peripheral when using the RGB666 Display Type.This should not be configured. The required R* pins for this Display Type are R2-R7, as per G2-G7 and B2-B7.This did not occur in previous CubeMX versions, including 6.14.1.This bug means that it is not possible to upgrade to this version of CubeMX with an existing project that already uses the pin(s) required for LTDC_R1. 

I just downloaded the prerelease (3.something) of the official VS Code extension and created a new project.  By and large, I like it.  It's especially nice that there is an "empty project" option for those that don't want to use HAL or LL functions.  But I was surprised that this does *not* provide the standard CMSIS header files (stm32l4xx.h and so on, for my project on a Nucleo L476).  In fact, searching my entire hard disk, the CMSIS headers were never installed with the extension.  Was there another software bundle I should have downloaded?  I don't see anything in the list.Besides the basic headers, it would be great if there were also a check-box to say if you want to use the DSP library (or other CMSIS libraries), then automatically configure the cmake project accordingly.  But for sure the basic header files should be included.  I mean, I guess you can argue that "empty" means "empty" ;) but in that case, why provide the startup assembly

A new question moved from this post. I have the opposite issue with STM32CubeIDe v1.18 / MX v6.14.0, I want my semaphore initial state to be depleted, but that is not applied to my code:     /* creation of DebugUartStringReadySem */ DebugUartStringReadySemHandle = osSemaphoreNew(1, 1, &DebugUartStringReadySem_attributes);     My .ioc file seems ok:     rtos.0.binarySems.0=allocationType,Dynamic;controlBlockName,NULL;initialState,Depleted;name,DebugUartStringReadySem      

Hi,Many thanks for the experimental incremental flash feature! For me it speeds up the dev cycle considerably.  (Note https://wiki.st.com/stm32mcu/wiki/STM32CubeProgrammer_errata_2.21.x seems to be broken.)Dan 

In my project for an STM32H7S3, I'm using the Ethernet module, and it works fine.However, I don't need the ETH_CLK pin. In RCC settings, there's an option which suggests I can turn it off; however, clicking the tickbox has no effect. (Note also that the text for this option is "Digital clock input" - even though the pin is an output.)Instead, I can set the pin itself to its Reset state, and that works. However, next time I open the IOC file, that setting is missing, and I have to change the pin state again. In addition, if I save the IOC file without making any changes at all, I can see in the diff that the settings for that pin have been changed:Previously, the generated value of ETH_CLK was outside its valid range (HSE was too low, and any setting from PLL3S was too high), and even though I wasn't using it, Cube wouldn't let me generate code without disabling this pin, because ETH_CLK is an input to the ETH Clock Mux, and presumably all inputs to this mux must be valid

The VSCode extension RTOS Views (and STM32Cube RTOS Views) does not support ThreadX, which is one of the two RTOS officially offered by ST. Are there any plans to add this support, or make the new toolchain compatible with the Microsoft Embedded Tools extension? In this state the new toolchain is quite useless for ThreadX applications.Best regards

Hello,I am trying to test the sbsfu project which is provided on this link. and I am using Unit Nucleo-F413ZH.I am following all the instructions step by step but on STM32Cubeprogrammer after opening file "SBSFU.elf", downloading it and then reading it by memory, I see these messages in the Log section:after it i try to compare the memory with the SBSFU.elf, and then I get these warnings: When I check the bytes, I don't see any difference but this warning appears, and I cannot see the welcome message on Tera Term.I tried all the process with other identical unit and everything goes right, but on this unit I encounter this issue.Also It's good to mention that I can run and execute any test projects via STM32CubeIDE. Thanks in advance. 

STM32Cube Debug Generic GDB Server (version 0.9.0)Using the code below, I can debug normally.After updating from version 0.9.0 to later versions,There was an error during debugging。I tried writing to the absolute path, and it worked fine.But the problem is, before updating the plugin, I didn't need to do this.Since I have many projects, I don't want to modify them one by one.So after the plugin update, is it that querying system environment variables is no longer supported or is there some other issue?If the plugin no longer supports it, then I will have to modify each project one by one or stick with an earlier version of the plugin.

I don't know if this is caused by the vscode extension or stm32cubeclt, but after the installation(everything related to the extension works), I noticed that the system path variable had been modified to add stm32cubeclt bin directories(I found this script: /etc/profile.d/cubeclt-bin-path_1.15.1.sh).Is this normal? I want to use my system default cmake, is there a workaround instead of using /usr/bin/cmake every time?I'm using debian 12.

On the download page for the STM32CubeCLT software, there is only a single download option available for macOS (STM32CubeCLT-Mac), which downloads the x86_64 version of the software.Are there plans to add a download option for AArch64 macOS?

I have an STM32H753ZI core board and I'm looking for a driver for the ILI9341 TFT. Any ideas?Thank you very much.

Hello Experts,I am looking for a solution in which we can create STM32 project with .ioc file and generate the code but cmake as build tool instead of make.I know we have couple of options while creating new project as follows:We can either create STM32 project which allows to have .ioc file and can generate code but that is make file-based project.If create CMake project, then it doesn't have .ioc and code generation feature.In past, I created STM32 project and on top of that added CMake list files but want to know is there any better option.Thanks in advance.

Defining the underlying type of an enumeration has been a difficult task until C23 added the feature.  Previously different compilers used different methods and results could vary unexpectedly. As seen in the image below, STM32CubeIDE v 1.19.0 can recognize this C23 feature and flag mistakes, such as the out-of-range value 0x101 for the 8-bit enum.As seen in the gcc output below, the mistake is caught by gcc:This is all as it should be.  The problem is when I remove the fake out-of-range command, gcc works as expected but STM32CubeIDE flags a syntax error:The build finished with no errors or warnings so gcc 13 understands at least this part of the C23 standard.Is there a configuration within STM32CubeIDE to prevent displaying a Syntax error for code the compiler understands? PS  I checked the generated assembly and it appears that the correct 8-bit code (ldrb) is being generated:   

Just upgraded to the latest version of STM32Cube for VS Code extension (3.6.4) and now the extension stopped to work. I've tried reinstalling the extension pack several times. Every time after the bundle was installed an error show up saying "An error occurred while synchronizing metadata. Please try run "STM32Cube: Install prerequisite bundle" and "STM32Cube: Synchronize pack list". Or work offline."Running "STM32Cube: Install prerequisite bundle" from the VS Code command palette will result in a message saying "No STM32Cube bundle to install", subsequently running "STM32Cube: Synchronize pack list" will return the original metadata synchronization error.I checked my HOME/Library/"Application Support"/stmcube/ directory, the bundle/ folder seems to contain what I expected, things like cmake, gnu-gdb-for-stm32 or the stink-gdbserver, etc. However, the packs/ folder is empty, not sure if that's the problem. I checked the output log of STM32Cube Core and found the fol

Suddenly the "Import CMake project" has disappeared from my VScode.It has been there until recently, don't know why it disappeared, maybe auto update.Version is 3.6.4I also get this error:Running the "Install prerequisite bundle" does nothing, just returns ok.Running the "Synchronize pack list", returns the error above again.  

Hello,I'm having trouble getting the latest version of the STM32 VS Code Extension to work on my remote development setup. The extension works perfectly on my local Windows machine, but fails when I connect to my Linux VM via the Remote-SSH extension.Environment:Host OS: Windows 11Remote OS: Ubuntu 22.04 (running in a VMware VM)VS Code Version: 1.105.1STM32 VS Code Extension Version: 3.6.4Connection Method: Remote-SSHProblems Encountered:Cannot Create New Project: The "Create  empty project" command fails to open the project creation wizard. Nothing happens.  Bundles Manager Server Unreachable: When trying to access the bundles manager for new MCUs or firmware packs, I receive a connection error. The log shows it cannot connect to the server.  What I've Tried:The extension functions correctly on my local Windows host. I can create projects and access the bundles manager without any issues.I have checked the network connectivit

Hello,I'm encountering severe compilation errors in an STM32H7 dual-core project (eth_project_CM7) using FreeRTOS (v10.4.6) and LWIP configured via STM32CubeMX (version 6.9.2). All errors point to FreeRTOS header files (task.h, list.h).The core issue appears to be related to header file inclusion order and the compiler's inability to find FreeRTOS basic type definitions (BaseType_t, TickType_t, etc.), despite thorough configuration checks.1. Project DetailsMCU: STM32H757XI (Cortex-M7 core project: eth_project_CM7)IDE: STM32CubeIDE (latest version)Middleware: FreeRTOS (M7 enabled), LWIP enabled.Language Standard: GNU11 (-std=gnu11) is set in the C Compiler.2. Example Errors (Cascading Failures)The errors are typical of missing type definitions, causing a cascade of syntax errors throughout the FreeRTOS headers:#error "include FreeRTOS.h must appear in source files before include task.h"unknown type name 'BaseType_t'expected declaration specifiers before 'TaskHandle_t'old-style parameter

On my computer the `stldr` files are stored like `/opt/stm32cubeprog/bin/FlashLoader/0x469.stldr` and I'm sure they exists and have correct permission. Why would it try to open the file '0x' ? Is it just a bug of string concat? Currently I cannot flash the chips because of this bug.

Trying to get the STM32CubeIDE for VSCode working on MacOS. I am on the latest version I think it is 3.6.4.I installed the extension according to instructions and I made a project for my NUCLEO-446RE with CubeMx.When I open the folder in VSCode it throws a Bad CMake Executable Error when I open the Setup STM32Cube project(s) section. In the search box in this section it also seems the NUCLEO-446RE is missing as well.I am not sure what is going on I tried uninstalling everything and reinstalling. I do have it in its own profile the only extensions in there are copilot and the STM32 Bundle in its entirety.Another note: I do not have a local cmake installed so there should be no conflict there.Any help would be appreciated.