STM32 MCUs Software development tools

Hello,when enabling DMA for a Timer channel in PWM mode, the transfer direction is set to Memory to Peripheral.This s very questionnable.Is it a deliberate choice or a bug?It does not make sense to select Peripheral to Memory for the DMA channel as it is used to wrtie to the device.I fell in the trap several time, wasting time to look up for an issue in my code...Very boring.STM32 device is STM32F429, CubeIDE version 1.19.0 (same problem with all versions since I started using CubeIDE a couple of years before.) 

The longevity statement in stmcufinder does not list e.b. U0/3/5 and states that all periods start 2020. An update should be considered.

Hi there! Not entirely sure if I'm missing anything here: I've been running issue when usign FatFs library with SDMMC1 on a STM32H755ZI due to the MBR sector readback been partially incorrect. Specifically during check_fs() where we check location where the ASCII for "FAT32" is store in the MBR sector (first 512bytes). The card can be verified to function correctly using HxD tool on Windows:However, through SDIO on the STM32 platform, I was getting different result for the byte [82:86], as the location is completely cleared. byte[32:24] seems to be offset to a completely different location instead:Here is the code snippet where I used HAL library to grab the first (block) 512bytes: const uint32_t BlockAdd = 0; const uint32_t NumberOfBlocks = 1; HAL_StatusTypeDef status = HAL_SD_ReadBlocks( hsd, (uint8_t*)pData, BlockAdd, NumberOfBlocks, Timeout ); if (status == HAL_OK) { if (HAL_SD_GetCardState(hsd) != HAL_SD_CARD_TRANSFER) { return HAL_ERROR; } }If anyone have any

Hi,I'm having problems entering in DFU mode on Windows using a sensortile.box pro. The board doesn't even show up on device manager howerver using the same board, with the same USB cable, on the same USB port on the same PC running Ubuntu the board enters DFU mode normally. I tried reinstalling CubeProgrammer more than once and also removing and reinstallig drivers but to no avail.Has anyone had the same problem?

Hey! I'm trying to get USB up and running on a STM32U595VJT6. From the reference manual, it says it's supported.   However, when in STM32CubeMX, the USB_OTG_HS is grayed out. This is on a fresh project, so there's no pin conflicts or similar. This is how it looks:   Do you have any ideas what the problem might be?  

Post edited by ST moderator to be inline with the community rules for the code sharing. In next time please use </> button to paste your code. Please read this post: How to insert source code. Hello to all, I'm coding a simple project for STM32WL55IC7 MCU, for the NUCLEO-WL55JC2 board. I didn't use the BSP for the board, in other words i didn't select the NUCLEO board option, but I started the project using the MCU. The Project name is STM32WL55IC7_foo. The CubeIDE build the following 3 projects: 1. STM32WL55IC7_foo  2. STM32WL55IC7_foo_CM0PLUS (the code for Cortex M0+ code) 3. STM32WL55IC7_foo_CM4 (the code for Cortex M4 code) When I try to compile the project, while the CM4 core code is compiled without any error, the code for the M0PLUS core issued errors, on the following snippet during the EXTI interrupt initialization: /* EXTI interrupt init*/ HAL_NVIC_SetPriority(EXTI1_0_IRQn, 1, 0); HAL_NVIC_EnableIRQ(EXTI1_0_IRQn); HAL_NVIC_SetPriority(EXTI15_4_IRQn, 1, 0); H

When trying to build an STM32CubeMX V6.13.0 generated cmake+gcc project with clang there are two files that generate errors:sysmem.cthe linker script (STM32L431RCTX_FLASH.ld and stm32l476rgtx_flash.ld)sysmem.c produces the following error:Core/Src/sysmem.c:30:35: error: use of undeclared identifier 'NULL'[build] 30 | static uint8_t *__sbrk_heap_end = NULL;This is caused by the fact that only errno.h and stdint.h are included. None of them officially define NULL. But they probably do in gcc. Also ptrdiff_t is undefined.sysmem.c can easily be modified by either:The easiest is not to use NULL but (void*)0. This is something ST could do. including stddef.h . This defines NULL and also ptrdiff_t.or modifying build to forcebly include such a file in all files (adding "-include stddef.h" to compile command). I did this so I wouldn't have to modify the file (which would get overwritten by regenerating code). But it's kind of a hack.Linker script:Linker scmemory region not defined: RAM[bui

Hi everyone,I’m using the new Arm Toolchain for Embedded with CMake. I generated a project template from STM32CubeMX using the following settings:Toolchain/IDE: CMakeDefault compiler/linker: Arm-ClangCompilation works fine, but when linking starts, I get this error:ld.lld: error: STM32G474XX_FLASH.ld:105: symbol not found: READONLYIt looks like the Arm toolchain (even the latest v21.1.1) doesn’t support the READONLY keyword. It’s not a big deal to adjust the .ld file - just remove the READONLY keywords - but it would be nice if the generated template worked out of the box.Am I missing a setting somewhere? Is there an alternative/better way to obtain the linker file?Edit: I'm using the STM32G474.

Hi!I have set up a project for the STM32H523RET6 using the following setup:CubeMX: 6.15.0STM32 vscode extension: 3.5.1The project is configured to use starm-clang and starm-clang++ and the STARM_PICOLIBC toolchainI have created a .cpp file () that i added to my build and updated my CMakeLists.txt with the following: # Setup compiler settings set(CMAKE_CXX_STANDARD 23) set(CMAKE_CXX_STANDARD_REQUIRED ON) set(CMAKE_CXX_EXTENSIONS ON) ... enable_language(C CXX ASM) ... # Add sources to executable target_sources(${CMAKE_PROJECT_NAME} PRIVATE Core/Src/myclass.cpp # Add user sources here ) When building the project, it compiles fine, but clangd reports the following problem: nsupported option '-mcpu=' for target 'x86_64-pc-windows-msvc'clang(drv_unsupported_opt_for_target) Unsupported option '-mfpu=' for target 'x86_64-pc-windows-msvc'clang(drv_unsupported_opt_for_target) Unsupported option '-mfloat-abi=' for target 'x86_64-pc-windows-msvc'clang(drv_unsupported_opt_for_t

I am trying to flash and debug on CM4 on stm32 H755 ZI-Q board.1. Firstly i tried debugging CM7 , with debug configurations --> Connecting under reset and halt all cores.2. Then i tried debugging CM4, with debug configuration -->None(no reset)3. Also added .elf files in startup while debugging CM7.4. I separately debugged cm7 and then cm4, but got only cm7 prints output.5. So then i tried in other way,I launch the group debug, adding cm4 and cm7 both elf files,, still the cm7 i could able to see prints,, cm4 no prints.. basically, I think cm4 is not able to get up from stop mode.So, provide me solution 1.How to run cm4 independently? If this is not possible,  2. How to run the code written in m4, along with m7 in stm32 h755 zi-q series board.

Hi! I have a board based on STM32U585CIU, and I'm trying to enable a USB CDC on it. The problem is: when I generate a project in CubeMX for MDK-ARM, the code I've attached works fine, USB CDC device is being recognized by PC, and I can see messages via COM port. However, once I generate it for CubeIDE or Makefile, it just goes to hard fault in MX_USBX_Device_Init. Considering it works in Keil uVision, seems like hardware is completely okay, so there is something in the difference between gcc-arm and arm compiler. 

I've been trying to build example projects (i.e. Ux_Host_HID) with the make tool chain config with stm32cubeMXSo when i run make, a bunch of stuff compiles, then i get this error.make: *** No rule to make target 'build/tx_initialize_low_level.o', needed by 'build/Ux_Host_HID.elf'.  Stop.  So when I search for that tx_initialize_low_level I find a .s assembly file in  Ux_Host/Core.SrcAlso, the make file seems to already be looking for assembly files to create the .o but for some reason, make is not recognizing it has a valid rule.  $(BUILD_DIR)/%.o: %.s Makefile | $(BUILD_DIR)$(AS) -c $(CFLAGS) $< -o $@$(BUILD_DIR)/%.o: %.S Makefile | $(BUILD_DIR)$(AS) -c $(CFLAGS) $< -o $@  Note: all files are generated from STM32CubeMX form their examples directory.

Hi,I've started working with the SBSFU example recently on an STMU585. I have noticed that flashing the sbsfu_app_init.bin file is much slower (10x maybe?) if TrustZone is enabled.My workaround has been to create a script to do a regression to RDP level 0 with TrustZone disabled, which will mass erase the device, and then reflash both sbsfu_app_init.bin and the bootloader each time before renabling TrustZone at the end.I was wondering what the specific root cause of the slower flashing is, and if there is a better workflow that would allow me to update just the sbsfu_app_init.bin, or potentially just the NS part, with speeds comparable to those I get when the device is blank.Thanks,Mike

Hello, I'm trying to implement a USBX host... and the code generates a lot of runtime errors.  I assume everything is setup properly from CubeMX, but the code always fails.  1- I see that the mutex was created from the debug session: (ux_system_mutex)2- the system fails when trying to initialize the UX system.3- The error came from _ux_utility_mutex_on() and generates an error code 0x04VOID _ux_utility_mutex_on(UX_MUTEX *mutex) { UINT status; /* Call ThreadX to get system mutex. */ status = tx_mutex_get(mutex, TX_WAIT_FOREVER); /* Check for status. */ if (status != UX_SUCCESS) { /* Error trap. */ _ux_system_error_handler(UX_SYSTEM_LEVEL_THREAD, UX_SYSTEM_CONTEXT_UTILITY, status); } /* Return to caller. */ return; }so, it is trying to get the mutex (ux_system_mutex) and then wait forever....4- If I look at the AzRTOS manual: 0x045- The base code was auto generated by CubeMX.  So basically it is trying to initialize memory using a mutex before starting

I run STM32CubeMX from the terminal and get an errror message in termianl.You can find the details in ‘error.txt'.

Hi! I'm working on setting up FDCAN communication using the STM32N6570X0-Q Nucleo Board. However, when I try to modify the bit timing parameters—such as the Nominal Prescaler, Nominal Time Seg1, or Seg2—the baud rate doesn't seem to change regardless of the values I set.On the receiving side, I'm using a STM32H533RE Nucleo Board to echo the messages, but it doesn't receive anything—so I suspect the issue may be with the timing configuration on the N6570X0-Q.Do you have any idea what might be causing this behavior?Thanks in advance! 

Hi;I recently started playing with CubeMonitor - it's fantastic. I am using it to troubleshoot/fine tune a motor control project I'm working on - specifically, I am graphing the values I'm calculating for Space Vector Modulation to make sure things are working as I expect. Something odd I notice are these strange "glitches" in my waveforms:There is nothing obvious I can find in my algorithm that might cause these - I am wondering instead if they might be coming either from 1) some nuance in C I might not be aware of, or 2) something I don't have configured properly in CubeMonitor. To check the former: in my SVM algorithm, I do something like:typedef struct { float theta_rad; float u, v, w; float alpha, beta; float d, q; } FOC_HandleTypeDef; void calculate_svm(FOC_HandleTypeDef *foc) { // ... some stuff happens before this, // but at one point I use some members of // FOC as 'inputs' float uvw_min = fminf(foc->u, fminf(foc->v, foc->w)); float uvw_max = fm

I have enable the hfmpi2c1 in STM32CubeMX tools, but I can find the FMPI2C_HandleTypeDef is undefined in the source code. Do we have a demo code to show how to setup and use hfmpi2c1. It's very confuse that the F446 only have one Fast Plus Mode IIC interface. 

I would like to ask how to program multiple STM32 MCUs connected in a JTAG daisy chain individually, and how to disable automatic program execution after flashing. I would like to achive the following goals simultaneously:Program each STM32 MCU individually while connected in a daisy-chain configuration.Prevent the program from automatically running after the flashing process.In my current setup, I have successfully programmed two STM32L476RG devices individually through a J-Link Base Compact using STM32CubeIDE by configuring the Debug Configuration settings (see the attached image for details).However, after programing, the application code starts running automatically. When using STM32CubeProgrammer, unchecking the "Run after programming" option successfully disables the automatic execution.However, I could not find a way to specify individual targets (e.g., by IR length or other parameters) for daisy-chained devices in STM32CubeProgrammer, either in the GUI or CLI versions

Hi,I create an empty project for Nucleo-L432KC with all default settings using STM32CubeIDE 3.6.4. on Win11.Default settings are gcc toolchain and Debug preset.Then, I switch to Release Preset (CMake: Select Configure Preset - as described in Development — STM32CubeIDE for Visual Studio Code Extensions Documentation 0.1.0 documentation).After building in Build/Release, the compile commands are like{ "directory": "C:/tmp/test2/build/Release", "command": "C:\\ST\\cube\\gnu-tools-for-stm32\\13.3.1+st.9\\bin\\arm-none-eabi-gcc.exe -DRTE_DEVICE_STARTUP_STM32L4XX -IC:/tmp/test2/Inc -IC:/tmp/test2 -fdata-sections -ffunction-sections -Wl,--gc-sections -Wno-comment -O3 -DNDEBUG -std=gnu11 -mcpu=cortex-m4 -mfpu=fpv4-sp-d16 -mfloat-abi=hard -mlittle-endian -mthumb -Wall -Wextra -Wpedantic -Wno-unused-parameter -Os -o CMakeFiles\\test2.dir\\Src\\sysmem.c.obj -c C:\\tmp\\test2\\Src\\sysmem.c", "file": "C:\\tmp\\test2\\Src\\sysmem.c", "output": "CMakeFiles\\test2.dir\\Sr

I'm a first time user of STM32 and trying to get my environment setup to build using Visual Studio Code.  I installed this extension STM32CubeIDE for Visual Studio Code (based on the suggestion from this thread), which ended up installing a number of dependencies.The user guide seems to suggest "Create empty project" is the way to go.  I had some difficulty with the main install and had to tweak my settings.json, but I think I finally got the paths correct.  But when I build a project, I'm getting this error: [rollbar] Unhandled exception: Unhandled Promise rejection: Post-folder-open Error: No index file found. [object Object] I get the same error when using creating a new project via STM32CubeMX with clang:[rollbar] Unhandled exception: Unhandled Promise rejection: Post-folder-open Error: No index file found. [object Object] However, using  STM32CubeMX with gcc, I get this error: [cmake] -- The C compiler identification is unknow

Hello,I downloaded the STM32CUBE-IDE 1.19 for linux RPM today.I use fedora43 on x86. It failed to install:sh-5.3# sudo /bin/shsh-5.3# ./st-stm32cubeide_1.19.0_25607_20250703_0907_amd64.rpm_bundle.shCreating directory makeself_dir_ZeOUueVerifying archive integrity... 100% All good.Uncompressing STM32CubeIDE installer 100% grep: warning: stray \ before /STMicroelectronics Software License AgreementSLA0048 Rev4/March 2018Please indicate your acceptance or NON-acceptance by selecting "I ACCEPT" or "I DO NOT ACCEPT" as indicated below in the media.BY INSTALLING COPYING, DOWNLOADING, ACCESSING OR OTHERWISE USING THIS SOFTWARE PACKAGE OR ANY PART THEREOF (AND THE RELATEDDOCUMENTATION) FROM STMICROELECTRONICS INTERNATIONAL N.V, SWISS BRANCH AND/OR ITS AFFILIATED COMPANIES (STMICROELECTRONICS), THE RECIPIENT, ON BEHALF OF HIMSELF OR HERSELF, OR ON BEHALF OF ANY ENTITY BY WHICH SUCH RECIPIENT IS EMPLOYED AND/OR ENGAGED AGREES TO BE BOUND BY THIS SOFTWARE PACKAGE LICENSE AGREEMENT.Under STMi

Hello guys, I'm new for developing  STM32N6's board. I got a NUCLEO-N657X0-Q board and tried it in VS code. When I was trying to "Run and debug", the UI always said that "SWO support is not available form tyhe probe when using the ST-Link GDB server. Disable SWO." and stop running and debugging. Is there anyone have the same problem??  How can I fix this problem?? Thanks for your reply. 

Colleagues, I’ve got a nitpick question.Where’s the best place to define USE_FULL_ASSERT ? It can be defined as a preprocessor symbol. Or, it can be #defined in the stm32xxx_hal_conf.h file. Or, is there a setting in the .ioc file?The stm32g4xx_hal_conf.h which STM32cubeIDE had generated for me has this block:/** * @brief Uncomment the line below to expanse the "assert_param" macro in the * HAL drivers code */ /* #define USE_FULL_ASSERT 1U */But it’s not in the the USER CODE block. I’m concerned that the automatic code generator would restore the #define USE_FULL_ASSERT back to the commented out line next time it re-generates stm32g4xx_hal_conf.h .Best regards,- Nick