STM32CubeMX (MCUs)

CubeMX does not generate sourcecode completely for the chosen STM32H723.Even for a minimum System, several CMSIS includes are not present.Please see my IOC file.CubeMX Version 6.16.1CubeIDE Version 2.0Firmware STM32Cube_FW_H7_V1.12.0 gepatched to 1.12.1A similar file which includes LwIP also has many LwIP files not included.Please help ....

1. Run StmCubeMX and then chose upgrade in Ubuntu2. After download it error with something alike it cannot unzip the file.  Reason:1. it seems CubeMx uses the .installationinformation hidden file in its folder to find the installation folder.INSTALL_PATHt\00$.....<your patch>/STM32CubeMXt\These are settings  in a binary format, so there are some weird character. Anyhow it cannot be changed by user.Solutions:1. Move CubeMX back to where is was installed (not desired) after upgrade, move it back where  it is desired.2. Remove and Reinst

Hi,this is a small but annoying little bug: Sometimes after running code generation, apparently the STM32CubeMX application keeps some files open in the "cmake" subdirectory. This then prevents me from modifying or deleting (to force re-generation on the next code generation run) the files. It also interferes with my Post Generation user script which modifies those files.This can be seen e.g. via ProcessExplorer:It would be helpful if STM32CubeMX would close all file handles after code generation, ideally before calling the post generation user scripts.Thanks! 

I'm using "STM32CubeIDE for Visual Studio Code" on Ubuntu 22.04.  I have a situation where I'm making a driver for a chip where it very much benefits me to make the driver into a class and instantiate several instances of it.Generally speaking, what are the best practices here, if I still want to use the CubeMX code generation?At the moment, here's what I'm doing:Add `CXX` to the languages in the `./CMakeLists.txt` file.  Also add the driver.cpp file to the target_sources() directive in `./CmakeLists.txt`. Change the file extension of main.c to main.cppChange the ./cmake/stm32cubemx/CMakeLists.txt to reference the main.cpp fileI was planning on automating this with a pre and post script which would do a `mv main.cpp main.c` and `mv main.c main.cpp`. And then also changing the generated CMakeLists.txt to reference the main.cpp.However, it seems that a simple bash script isn't being run. &

Hello STM Community,I'm encountering persistent UART Overrun (ORE) errors on the STM32H573RIT6 when using code generated from a fresh STM32CubeMX project (.ioc file).Setup and Reproduction:Created new STM32CubeMX project for STM32H573RIT6Generated code and tested UART reception with:HAL_UART_Receive_IT()HAL_UARTEx_ReceiveToIdle_IT()Result: Continuous ORE errors in both interrupt and IDLE modesKey Observations:Issue persists even at:Reduced system clock frequenciesLower baud ratesWith non-blocking ISR implementationDMA Mode Works

Hi,This is for STM32U5 and STM32H7 series MCUs.I was starting to use ThreadX because it looked like it was preferred over FreeRTOS. Now it looks like it is going the other way FreeRTOS is preferred over ThreadX ... or so I read/watched.I liked ThreadX with FileX because they seemed better integrated than FreeRTOS and say FATfs. But I don't have a lot of experience with RTOS in general. I've used a bit of FreeRTOS before but it looked like ThreadX was a better option for STM32U5 series MCU (in cubeMX).The video I watched and info on going back to FreeRTOS  with separate FileX (from STM) said that it was going to happen by the end of 2025. I prefer to stay with one RTOS, whichever is going to make my life easier but I also need a file system.We're at the end of 2025 so I wonder if this switch from ThreadX to FreeRTOS with separate FileX has already been implemented in the new STM32cubeIDE v 2.0? BTW I'm currently using STM32CubeIDE v

The X-CUBE-AZRTOS-H7RS middleware is missing the setting to change the USBX System Stack Size which in turn means that the USBX_MEMORY_STACK_SIZE define in the app_usbx.h file can not be changed from its default of 1024. And because the define is not in a USER CODE section, any manual change will be overridden when re-generating code in CubeMX.Here is an example when using USBX host stack, but it's the same for the device stack.STM32H7RSxx:STM32H7A3ZIT:

Hi everyone,I'm currently porting the App of STM32U5 SBSFU example to CubeMX and "smooth" is not the word I would use to describe the process.Namely, the SBSFU example includes Bootloader, Loader and App projects with memory map being in separate centralized header files included by all these projects. These files (region_defs and flash_layout) are not fully polished (some bugginess and inconsistencies) but they do ease the memory management a lot by using defines and not hardcoding; these defines are then used in projects' linker files in order to define memory regions.In contrast, CubeMX just provides a UI representation of the memory space where users can input all the memory bounds, be it flash or RAM. The trouble is, we can only enter numbers and not constants, so generated code is full of hardcoded address bounds.This does not play well with SBSFU Example - since these address bounds are not only used in LD files, but rather for defining FLASH areas

Hello,I have been trying to configure the ethernet RMII with PTP synchro on cube MX. The idea is to link the PPS signal to trigger the timer. I can't manage to get the remap to be done to the ETH_PTP_PPS as is not displayed in the dropdown. According to the same cube mx that option should appear. On a side note when I enable the RMII_PTP_syncro I also loose all the config information of ethernet which i find rather strange. Do any of you have any suggestions on what's the proper way of configuring this? Is this even possible or is just a bug of the prompt that implies is possible? Thanks in advance ;)

Hi, I am using CubeMX 6.16.1 to work on a STM32F777ZIT6 target. However, the generated code does not have the Macro definitions of the GPIO pins in main.h. The GPIO init part IN MX_GPIO_Init() is also empty, as follows:/** * @brief GPIO Initialization Function * None * @retval None */ static void MX_GPIO_Init(void) { /* USER CODE BEGIN MX_GPIO_Init_1 */ /* USER CODE END MX_GPIO_Init_1 */ /* GPIO Ports Clock Enable */ __HAL_RCC_GPIOE_CLK_ENABLE(); __HAL_RCC_GPIOF_CLK_ENABLE(); __HAL_RCC_GPIOH_CLK_ENABLE(); __HAL_RCC_GPIOC_CLK_ENABLE(); __HAL_RCC_GPIOA_CLK_ENABLE(); __HAL_RCC_GPIOB_CLK_ENABLE(); /* USER CODE BEGIN MX_GPIO_Init_2 */ /* USER CODE END MX_GPIO_Init_2 */ } I also tried CubeIDE 1.19.0, have the same issue.Did I do anything wrong? I think it is really important to get this function work, as it would be cumbersome to manually init the GPIO pins. I

I am working with a custom board populated with an STM32H7S7L8H6H and a hyperbus flash S26HS01GTGABHB030 connected on XSPI1.My intent is to execute in place on the hyperbus flash but have not been successful thus far.

I generated a project using STM32CubeMX for the STM32N6570-DK board.Steps performed:Generated code using CubeMX (default configuration)Imported project into STM32CubeIDEBuilt the project successfully (no build errors)Attempted to run/debug the applicationDuring programming / start of execution, the debugger stops with the following message:Break at address “0x18003A1A” with no debug information available, or outside of program code

Hi,I encountered a strange problem while using stm32cubeMX in a new project.Everytime i saved the code files and then regenerated the code using stm32cubeMX,the user code will gone.A example,"#include "led.h" " have been insert in the user code zone,then i saved the project,and regenerated code using stm32cubeMX.then "#include "led.h" " is missing...

I have just installed STM32 cubeMX v6.16.1 on my Windows 10 machine have the following issue:In the Pinout&Configuration step we can hover mouse over items in the left tree view and eventually we can click on links to Related Documents. However those documents open in Acrobat Reader on my machine, even tough I have configured PDF-XChange Editor as default PDF viewer (Windows Settings, Apps, by ending, .PDF -> PDF XChange Editor).When I double-click PDF files in Windows Explorer they properly open in PDF Xchange, not Acrobat Reader. Having PDFs open in Acrobat is really annoying due to the constant popups and waste of screen space with useless tabs.

I am developing a product using the NUCLEO-N657X0-Q dev board and have generated the configuration setup from CubeMX and developed the code in the CubeIDE. When I need to layout my own circuit board I will want to use the STM32N657A0 variant which has 169 pins and I was wondering if I am able to use CubeMX to change the existing configuration done for the Nucleo board to the STM32N657A0 169 pin device without destroying the code I have already developed? I have written the code using resources and pin outs available on the STM32N657A0 device but the actual pins are different on the two devices so need to know if I can make the transition using CubeMX or whether it will have to be done manually?

I'm trying to generate code using CubeMX for the Nucleo-U031R8 demo board to output a clock signal to the MCO2 pin. But the generated MX_GPIO_Init function makes the following assignment:GPIO_InitStruct.Alternate = GPIO_AF0_TRACE;Naturally, the compiler throws an error: "GPIO_AF0_TRACE undeclared".If I manually change the code to the following:GPIO_InitStruct.Alternate = GPIO_AF0_MCO2;...the code compiles without errors.Is it possible to fix this bug in the next CubeMX release?

Hi! I am working on creating a new build system for my for a software platform using CMake to help with management of different targets and toolchains. I have an .ioc file that is currently in use with the existing build system based on CubeIDE that contains all middleware (FreeRTOS, LwIP, TouchGFX) additions and configurations, so I am not changing anything. I am trying to make the build as modular as possible so I am trying to link the stm32cubemx and touchgfx CMake targets directly. I am using build presets to manage configuration. Creating the build system works fine, but when I try to run the build I get compilation errors because the TouchGFX related files can't find the user constants defined in CubeMX for TouchGFX, as shown below: 

Hi, I'm developing a project for STM32N657X0HxQ MCU and I need to run a NN model on it. I installed the 10.2.0 package for X-CUBE-AI and I've selected Core for Artificial Intelligence X-CUBE-AI in the Software Packs component selector. However, whatever Device Application I select, I always get when I'm on the Configuration Tab in Pinout & Configuration "Warning: This peripheral has no parameters to be configured" and I cannot upload my NN model, even when Artificial Intelligence X-CUCE-AI checkbox has been selected.I'm using the 6.16.1 Version of STM32CubeMX.Thank you for the cooperation

When creating a new CMake-based project in STM32CubeMX, the "Toolchain Folder Location" field is auto-populated with the project folder path upon entering a project name. This behaviour is misleading because the field is intended (at least by its description) to specify the location of the actual toolchain binaries (e.g., GCC ARM) and it is unlikely that a toolchain is installed in a specific project folder.I have tried to understand what this is by consulting the User Manual UM1718 Revision 49, and there were two references that matched the text "toolchain folder", both of which provided no helpful guidance on what this is actually used for.

After generating C code for a CMake project, I'd like to use some existing code that I've developed elsewhere.Is there good information or guidance on how to do this properly?For example, should compile my code as a static library and bring it into my project? If so, what is the recommended way to do this (how do I ensure compatibility, etc)? Or should I just copy-paste the source code files and headers into the project?

Split from IDE Version 2.0.0 - why remove MX ? as this has nothing to do with that topicPinout view is bad also.In center of window useless hole with ST logo. Wastefull use of GUI area. 80% of area is gray hole in center or white hole at the borders.Much better to use table with pins info. In the table can be shown some usefull information in addition to pin name (type of pin, for example). Current design can be used as additional presentation.CubeMX design is created by managers wich have no undestanding of developer needs. 

I used version 6.14 beforeand could choose OrCAD's package There is no option for "Cadence" in version 6.16.1

I'm trying to use dual layers in LTDC of STM32N657X0HxQ controller. In the STM32CubeMX, I tried to set the Secure bit of RISUP for LTDC_L2 in RIF setting view, but the box is in gray color and is not selectable as follows.When I generate code with this box unchecked, my program does not work. So I manually changed the Secure bit in source code, which made LCD display work fine.How can I change the Secure checkbox in STM32CubeMX?My STM32CubeMX is version 6.16.1 for Windows 11.I've attached the ioc file.

Code generated by CubeMX use Windows line-endings, which generate huge problems under other operating systems (repository, editors, etc.). You have to strip CR-LFs after every code generation.Somebody should use standard method of platform-independent text file generation (StringBuffer and System.lineSeparator()).

STM32CubeMX standalone hangs on File → New Project (infinite loading, no error message). The problem is fully reproducible.I have tried:– reinstalling CubeMX– different CubeMX versions– Java reinstall– offline mode– clearing STM32CubeMX cache/AppData STM32CubeIDE (latest) works normally but I can not generate projects with CubeMX. OS: Windows 11 pro Has anyone experienced this issue recently or knows a workaround / fix?Any advice from ST would be appreciated.