STM32CubeMX (MCUs)

Hello all, I recently install CubeIDE and I want to do a blinky example on the User led for the nucleo-U545REI see that the GPIO for the USER LED is PA5 So in cubeIDE, I open CubeMX, configure PA5 PIN TO GPIO_OUTPUT and try to generate the code by clicking on the button "Device configuration tool Code Generation", but nothing happens Can someone explain me why the generation is not working ? Thanks

When I try and enable UART5 in the  .n657_dk2.ioc file,  the development system seems to go into an infinite loop with the mouse cursor spinning endlessly.Bug1: To replicate,1- open the n657_dk2.ioc file2- goto the connectivity section3- enable UART5 in FSBL4- change mode to Asynchronous5- set Hardware Flow control to RTS/CTS.system hangs....Bug 2: if you try and save the file, it corrupts the file and you cannot edit the .ioc file anymore..ioc file attachedIMPORTANT!!! I need a fast work around so I can go to layout this week.... 

When the TouchGFX Graphics Application is selected, the supplementary TouchGFX code is generated; however, the TouchGFX library itself is not included in the project. I tried all available code generators, but none of them added the library.

Hi,When configuring a DMA channel in Circular mode with the LL code generator for STM32H523, the generated initialization code is incorrect.It creates a DMA Link Node (LL_DMA_LinkNodeTypeDef) on the stack, which is invalid since the DMA continuously accesses this structure. The link node should be placed in persistent memory instead.Additionally, the generated code does not set the source address, destination address, or block length when the link node is created, although these can be manually set later.Regards,Peter

Hello Community, I have tried to configure TIM8 in CubeMX (Version 6.15.0) with device STM32N645B0 (VFGBA198). But TIM8 is not available in the Timers Group. All other timers are displayed, only TIM8 is missing. Also trying to activate TIM8 over the pin directly does not work. Any ideas? Best regards Thomas

I am using Keil MDK Pro ver 5.38, Compiler ver 6.19, Middleware 8.0.0 (Also tried lower and upper versions). I am using STM32F746NGH6 MCU . I generated a USB FS Host only code in CubeMX. Exported to keil. Compiled the code and got zero errors and zero warningsHowever, after adding RTX5 and middleware components to keil project, the error of MX_Device.h file not found. Ther is no clear solution for this problem on unternet anywhere. Why STM32CubeMX is not generating this file? What is the way forward? 

I want to use STM32FC091 USART1 for a MODBUS communication on RS485 with DMA. The STM32FC091 USART1 has a full implementation and supports using DMA, receiver timeout interrupt, Modbus communication and driver enable features.I just successfully tested USART1 with DMA on RS485 with my NUCLEO-C092RC. With STM32CubeMX I set USART1 in asynchronous mode and it properly works. But when I try to select Modbus Communication mode for USART1 I get:

Not sure if this belongs in CubeMX or STM32 code for VS.I have a CMAKE project generated from CubeMX.When I build within VS Code IDE with the STM32 Extension I get several nocent warnings I would like to eliminate.Some of these are deep in the HAL and/or the generated code produced by CubeMX.ex:warning: declaration of 'hrtc' shadows a global declarationwarning: cast increases required alignment of target type [-Wcast-align]etc.I tried several modifications to my top level CMakeLists.txt suggested by AI to suppress these warnings but NONE work.I know this line:include("cmake/gcc-arm-none-eabi.cmake")brings in all the gcc compile options but I don't want to modify this file as it gets regenerated every time I do a CubeMX "GENERATE CODE"

Hello. I'm using STM32CubeMX with STM32f777ZI MCU to create a CMAKE project.The generated linker script file (STM32F777XX_FLASH.ld) has a really simple definition of RAM memory regions:/* Specify the memory areas */MEMORY{RAM (xrw) : ORIGIN = 0x20000000, LENGTH = 512KFLASH (rx) : ORIGIN = 0x8000000, LENGTH = 2048K}There is no separation of SRAM1, SRAM2, DTCM and etc. For example - I want to enable cache in SRAM1 and I need to have a proper section definitons. I want to use in my code something like:__attribute__((aligned(sizeof(uint32_t)), section ... and etc.Is it possible to force STM32CubeMX to generate more detailed RAM segmentation for F7 cores?

STM32H5 USB Custom HID CubeMX issue.When setting parameter "UX_DEVICE_CLASS_HID_EVENT_BUFFER_LENGTH" to 64 in CubeMX, the corresponding define in the "ux_user.h" file is commented out.Setting this parameter to any other valid value (between 0 and 255, excluding 64) in CubeMX generates an uncommented define.I presume CubeMX is doing this because it assumes the default value from USBX is 64 bytes.In fact a default value is provided by usbx in "ux_device_class_hid.h", but this value is actually 32 bytes.Thus, using out of the box CubeMX generated code when "UX_DEVICE_CLASS_HID_EVENT_BUFFER_LENGTH" set to 64 results in very nasty memory overwrites in the USBX device structures that subsequently result in hard fault.Please update CubeMX to not comment out this define when set to 64.Thank you.    

Hello Forum,I've noticed a feature/bug/behaviour in CubeMX 6.15.0 that I'm unsure about so could do with some advice.  I've been configuring an STM32U5A9ZJT 144 pin MCU for a project I'm working on.  This has been an evolutionary process.  To make the point I'll just focus on two peripherals, LPUART1 which has already been configured and I2C3 that I want to add in and configure.I start with LPUART1 configured as follows:Which has pins assigned like this on the MCU:

Hi everyone,  can anyone provide total DCMIPP camera interface STM32CubeMX    .ioc file therefore I can create a project and My applications. Please help me to implement the B-CAMS-IMX camera.I also try with BSP drivers that are available in STM32N6570_discovery_bus.c and .h , STM32N6570_discovery.c and .h file but in IMX335.c file I am unable to read device ID properly.So, please provide proper DCMIPP camera interface STM32CubeMX .ioc Thankyou.

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. 

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);  

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:

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

Hello!I am using the NUCLEO-H755ZI-Q development board. In STM32CubeMX, I selected "Start My project from ST Board" and chose the corresponding model. I generated both an STM32CubeIDE project and an MDK project. The M4 and M7 firmware compiled from the STM32CubeIDE project can be downloaded to the board and run. When the button is pressed, the light can flash. However, the M4 and M7 firmware compiled from the MDK project cannot be run when downloaded to the board. The light does not flash. What could be the reason? Logically speaking, the same CubeMZX sample was used, only the generated projects are different. Please help me. Thank you!

I'm looking to utilize an FTP server with the STM32F7NI-Discovery board. Specifically I'm looking at taking Nx_Webserver demo for that board (which works great) and adding the ftp module to it that NetxDuo provides. However, I'm seeing that the F7 chips don't have a way to add the FTP module to NetxDuo using the STMCube IDE, where that option is available for the H7 chips. Is there anything preventing me from getting an ftp server running by manually copying the addon from the NetxDuo Github repo and modifying the WebServer Demo project to utilize FTP on the F7 board? I've gone ahead and added the files to the project and it compiles just fine, but I run into memory management issues when trying to create an FTP server with it. That could very easily just be my own inexperience with the tools, though. 

I have a project for STM32F777 generated by STM32CubeMX 6.15.0.The macro definition of #define assert_param(expr) exists in 2 include files - stm32f7xx_hal_conf.hand stm32_assert.hBoth files are included on different levels of the project. Is there a reason to have two indentical definitions in the code?  

Hello all, I'm having problems with I2C1, i have a code that works perfectly while i configure the HCLK to 8MHz but it stops working when i set any higher frequency. Has anybody faced similar problem? I am using STM32CubeIDE for setting the project, this microcontroller has a separate clock from HCLK, and i can't understand why I'm having this problem.   Thanks in advance.

Hi all,I am trying to implement ADC acquisition on one channel, with 14 bit, 4x oversample, at 1 Msmps (1 megasample per second). This is possible only using dual interleaved mode. Moreover, I want to acquire continuosly with circular DMA.But it is not clear to me how to set the right values in CubeMX, and I can't use the datasheet (which I have read) to understand, it seems CubeMX and datasheet use different names for the same thing.Let start from my ADC1 settings:Are the settings right until now? What I don't understand is the role of "DMA Access Mode" and "Conversion data management mode": I can disable DMA Access Mode and still have Conve

Hello,I have a project on STM32F091 generated with STM32CubeMX.Connectivity included USART3, USART4 and USART5.It created this code that has a problem on the line with huart5 ("(&huart5);" is written twice and it give a warning during build).Do you confirm? What the procedure to correct in next CubeMX version?void USART3_8_IRQHandler(void) { /* USER CODE BEGIN USART3_8_IRQn 0 */ /* USER CODE END USART3_8_IRQn 0 */ HAL_UART_IRQHandler(&huart3); HAL_UART_IRQHandler(&huart4); HAL_UART_IRQHandler(&huart5);(&huart5); /* USER CODE BEGIN USART3_8_IRQn 1 */ /* USER CODE END USART3_8_IRQn 1 */ }regardsjeromeEdited to apply source code formatting - please see How to insert source code for future reference.

Thread split from this discussion; specifically, this post.   My mistake. Doesn't make sense to have the nominal SJW mixed with the data bit configuration. It should be inside the Bit timing parameter tree view along with the other nominal configurations. 

A recent update to CubeMX (Version 6.15.0) has caused newly generated linker files (Such as CM7/stm32h747xx_flash_CM7.ld) to generate with the following code:  /* Remove information from the standard libraries */ /DISCARD/ : { libc.a:* ( * ) libm.a:* ( * ) libgcc.a:* ( * ) } Previously, we generated this code on version 6.14.1 and got the following instead:  /* Remove information from the standard libraries */ /DISCARD/ : { libc.a ( * ) libm.a ( * ) libgcc.a ( * ) } This change causes the new version to unconditionally discard C++ standard library code which is a valid thing to do for pure C code, but not for C++ which leverages generated C as ST addressed in this forum post. 

In the IOC editor, under PWR, and Monitoring, I have set below