STM32 MCUs Software development tools

Hi. I just ordered a brand new NUCLEO-N657X0-Q, and cant connect to it using STM32CubeProgrammmer. I tried using ST-Link in different modes and reset modes. I also tried using external loader, and discovery under debug authentication, got the error message in the screenshot.Would appreciate some help.

My application use lots of RAM, I'm concern about the stack memory. The RAM size is  128KB,from 0x20000000 to 0x20020000. Form the linker script (.ld) file, _Min_Heap_Size = 0x200; /* required amount of heap */ _Min_Stack_Size = 0x400; /* required amount of stack */ Use "Build Analyzer" to check Memory Details, it says "._user_heap_stack  Run Address  0x2000d160     1.5K ". But when I run the debugger, SP starting at 0x20020000; when call functions it goes to 0x2001ffe8; 0x2001ffe0... My question are: 1. The ._user_heap_stack  Run Address should be 0x2001FA00  (size 0x400 + 0x200 ==>0x600), end at 0x20020000.  2. During debugging, I can use SP to view the stack pointer address. How can I view Heap address? 3. Stack area should be from  0x2001FC01 to 0x20020000; and heap area is from 0x2001FA00 to 0x2001FC00?     If stack overflow, it will corrupt heap area data?  

how to setup manually freertos without errors ?

Hi,APB1 clock can't be enabled if HSI clock is chosen. It says that the driven clock must be 10M<=x<48M, so, without using the PLL it's impossible to get the 10MHz. Things get worse when I try to feed the PLL from the HSI: it says that the PLL should have HSE as an input!If I go on despite the recommendation, then I get this message: How can I enable the APB1 clock without using an externa crystal? What should I do if I want to choose both: not to use an external crystal and still using the USB peripheral? For now I can disable the USB port because I'm not in a rush with it, but this isn't a solution.May I overlooking something?BTW, this is a custom board. Thank you in advanced! 

Fresh install of stm32cube1.9.0 (onto a Windows 10 64-bit system). This older-than-current version needs to be used (at least at first) to avoid issues in resurrecting an old project (from when1.9.0 was current).The default install path (from the application's installer) was overridden (to C:\ST\STM32CubeIDE_1.9.0\).I believe I am having a problem with the internal (to STM32CUBEIDE) value assigned to it's PATH variable (see below), however, I do not know how to update this internal PATH value to help the IDE find make.exe (which was installed in the system during the installation of the IDE overall.)A small (C-language) test program was created and an attempt made to build it, resulting in the following error:-----------------------------------------------------------------------------------------12:06:22 **** Build of configuration Debug for project test1 ****make allCannot run program "make": Launching failed Error: Program "make" not found in PATHPATH=[C:/ST/STM32CubeIDE_1.9.0/

Hi, I am using the following environment:OS: Ubuntu 20.04 LTSSTM32CubeMX: 6.15.0Visual Studio Code: 1.103.2STM32Cube for Visual Studio Code : 3.5.1Board: NUCLEO-H563ZI I found a bug where the STM32H563XX_FLASH.ld file is not generated after generating code.This bug can be reproduced by following the steps below:Create any cubemx project for STM32H563xx.Set "The Toolchain / IDE" to CMake.Click "Generate Code" .After generating code, the files "STM32H563xx_FLASH.ld" and "STM32H563xx_RAM.ld" are generated.However, the CMake build environment requires only STM32H563XX_FLASH.ld. As a result, the CMake build fails and reports the error : "my_project/STM32H563XX_FLASH.ld: No such file or directory" On the other hand, I found that the "STM32H563XX_FLASH.ld" can be generated if "Apply Application Regions Settings to Linker Files: ON" is enabled in Memory Management Tool. 

Hello,   the STM32TrustedPackageCreator supports the obk-file-generation via CLI: STM32TrustedPackageCreator_CLI.exe -obk C:\OBKey.xml   This requires a xml-file. The gui-version allows the generation directly from a password. Is this also possible via CLI?   Best regards, Peter

Hey everyone,I'm currently trying to work with the NUCLEO-U386RG-Q development board.When I try to download the latest available MCU package (1.2.0) via STM32CubeIDE, I only get an error message stating that the requested MCU package is not available. This is obviously incorrect, as I can manually download and install it from the ST website. When I try to install one of the previous releases, the connection is established, but the download is extremely slow for some reason. Other MCU packages, like those for the STM32U5 series, download without any issues and complete within seconds.I'm using STM32CubeIDE version 1.19 and have already completely reinstalled it. A refresh of the package list also doesn't solve the problem.Is anyone else experiencing this or knows how to fix it?Best regards,Max 

I have TrueStudio and STM32CubeIDE 1.17.0 installed in OS Win10, it works fine for my current project.But our cooperation will update the OS to Win11. The TrueStudio and STM32CubeIDE 1.17.0 will automatically work for Win 11 OS? or should I reinstall it?

Quick question - how do I get STM32CUBE Live Watch to display values in hex? I have the output radix set to 16 so normal watch shows hex values but live watch is in decimal.Latest preview versions of VSCode extensions (STLink GDB Server v0.10.1)

i am trying to install cube-ide on my RHEL (Red Hat Enterprise Linux) ,whenever i try to install that i found that a folder is creating and once i go through the instructions and accepting licenses the folder is automatically deleting and causing an error like no matches found Can anyone tell me how to resolve this issue

I used STM32CubeMX to configure an STM32F401RETx USART6 for 115200 baud. I searched but could not find an example to start from. I enabled USART to register USART callbacks using the STM32MXCube Project Manager Advanced Settings. My project utilizes FreeRTOS, so I added that and created a simple default task to receive a character and output a message each time a character is received.I used an FTDI TTL-232-RG-VIP-WE cable connected to GND, 3V3, PC6, and PC7, and to a Windows PC set for 115200/8/N/1.At run time, I set a breakpoint where the character should appear in my receive buffer, and observed that I was getting garbage in. Also, I saw I was getting garbage out. I stepped the initialization, configuration, interrupts, and callbacks — all appeared to be working correctly.In desperation, I experimented with the baud rate and found that I could send and receive good characters with the PC set to 38400 baud while the STM32F401 remained configured for 115200. What is going on here

MCU: STM32G474xxIDE: VSCodeToolchain: ARM Compiler 6 (armclang, armlink)Question: I'm a newcomer to using armclang and armlink for compiling and linking a project for an STM32G4 MCU. I've successfully compiled a minimal project and am trying to link it using armlink. My scatter file is: LR_IROM1 0x08000000 0x00080000 { ; load region size_region ER_IROM1 0x08000000 0x00080000 { ; load address = execution address *.o (.isr_vector, +First) *(InRoot$$Sections) .ANY (+RO) .ANY (+XO) } RW_IRAM1 0x20000000 0x00020000 { ; RW data .ANY (+RW +ZI) } } I use a startup script in GNU syntax generated from CubeMX.The linker command I am using is: armlink --cpu=Cortex-M4.fp.sp *.o --scatter=STM32G474XX_FLASH.sct --map --info=summarysizes --load_addr_map_info --xref --callgraph --symbols --info=sizes --info=totals --info=unused --info=veneers This setup results in the following linking errors: Error: L6218E: Undefined symbol _sidata (referred from startup_stm32g474xx.o). Error: L6218E

Hi, I followed the instructions in this post to install the STM32Cube extension for Visual Studio Code. After completing the installation, I noticed that two icons appeared in the left panel: one labeled "STM32Cube" and the other "STM32 VS Code Extension".The difference is that under "STM32Cube", there is only a "Key Actions" menu instead of the Project Manager, so I can’t import my project. On the other hand, the "STM32 VS Code Extension" icon works correctly and allows me to import my CMake project.Which one should I keep?

Hello,I am working with the STM32N6570-DK board and using STM32CubeIDE (version 1.19.0) .I cloned the full STM32N6 repository from:  https://github.com/STMicroelectronics/STM32CubeN6 When I try to build and flash the Ux_Device_Video project, I face a problem: I right-click on the Ux_Device_Video→ Debug As → it flashes successfully, but the program never reaches  main(). (Normally in every program the debugger points to first line in main).When I press Run, the debugger only executes a few assembly lines like below and stays there:What I observed in Debug Configurations (screenshot attached):For all other example projects, the Debugger tab looks normal , and in this project even FSBL  flashing good and stops at main and behaves as expected.But for Ux_Device_Video, I see an additional “Load Mode” section (highlighted in my screenshot). Other projects don’t have this.Also, under STM32 C/C++ Application, I normally see two entries (e.g. FSBL and the

Morning all,I'm really having some issues getting the latest CubeIDE (1.18.1) to work.This problem started when I was trying out some different processors for a new product (a CPU upgrade) - everytime I selected the new CPU and attempted to start a new project, I was told that it was missing an XML file and that no new project could be started without it.After trying/retrying several times with searches for the XML file turning up nothing - I came to the conclusion that I needed a re-install of the tools as this would give me the XML file I needed for the new board.I downloaded and installed the latest 1.18.1. Everytime I started CubeIDE, I would get a popup window after the splash screen  saying this or that log would hold the details of what was going wrong. Reading the log file showed that there was no swt-win32 in the java library path. The path listed in the log is pointing to a .DLL, which exists at the location given.Assuming I had an old java install, I got and installed t

STM32CubeIDE 1.9.0 has been working fine, until tonite. I started it and after it asked which workspace it acted like it was loading but never went anywhere. I killed it, removed the .lock file and restarted and the same thing except it's now exiting. The .ide.log has this:2022-06-15 23:57:26,297 [INFO] PluginManager:213 - loadIPPluginJar : add gic 2022-06-15 23:57:26,301 [INFO] IPUIPlugin:75 - create IPUIPlugin 2022-06-15 23:57:26,301 [INFO] PluginManager:213 - loadIPPluginJar : add openamp 2022-06-15 23:57:26,624 [FATAL] Updater:297 - Updater called before beeing initialized 2022-06-15 23:57:26,784 [FATAL] Updater:297 - Updater called before beeing initialized 2022-06-15 23:57:26,795 [INFO] DbMcusAds:47 - JSON generation date=Thu Sep 30 02:55:09 EDT 2021 (1632984909247) 2022-06-15 23:57:26,795 [FATAL] Updater:297 - Updater called before beeing initialized... and it just keeps going. I tried re-installing but it just says "Nothing to install", prolly because it is installed.OS is Debi

Hi,Running STM32MX Tool 6.15.0 On Windows 10.Started a project for a STM32U535CCTx.I've enabled ADC and configured channel IN8. The software thinks this is connected to PA4, but the datasheet says it is connected to PA3. See screen caps below. Regards John     

Good evening,I’m reaching out to ask if anyone might be familiar with a solution to a hard fault issue we’re currently encountering.Here is the situation:We are using a custom-designed board that incorporates an STM32H7 microcontroller and external Flash memory from Winbond. All source code is stored in the external Flash. The code itself is relatively simple and primarily handles UART communication.While debugging with CubeIDE, we frequently encounter hard faults when using the Live Expression feature. The fault appears to be an alignment error.The debugger connection is via SWD using ST-Link. Interestingly, when debugging via JTAG using SEGGER’s J-Link, the hard fault does not occur—even when using Live Expression. However, in this case, the console displays a message saying “cannot read register map.”This leads me to believe that the occurrence of hard faults may be influenced by the debugging interface (SWD vs. JTAG).I would greatly appreciate it if anyone could share insights into

Hi All,I had 2 questions in my current journey learning stm32 software as well as my stm32f746g board. I generated code based off an .ioc file created by the AI Edge Dev Cloud (a human activity keras model). My issues are as follows: 1. I debugged the code under OpenGDB and received this as my console output (see attachment below) however all my STM32 does is flash the led red and green rapidly. Does that mean it was successful? 2. I went back to the STM32CubeMX software to validate my model on target after debugging, however all i get met with is this output and the validation being stuck at 99% indefinitely. Any guidance on this issue is greatly appreciated. Thank you :)-Ramy 

I just bought a Nucleo-G491RE board and installed STM32CUBE to start a new project. During the project setup, the IDE downloaded and unzipped stm32cube_fw_g4_v160.zip. Then, when it downloaded stm32cube_fw_g4_v161.zip, this error message popped up. (screenshot below) Error downloading the following filesstm32cube_fw_g4_v161.zip (HTTP Access Forbidden) I'd appreciate any help to resolve the error. Thanks.    

I have a simple program using Touchgfx and Freertos running on the following:STM32H735G-DKSTM32CubeIDE V 1.19.0STM32Cube_FW_H7_V1.12.1I have enabled SDMMC1 (in 1 bit mode, but 4 bit is the same), configured FATFS, and created a task to mount the SD card and open a file, from a button on a touchgfx screen. The card is formatted for FAT16, but format type makes no difference.f_mount fails every time with mostly file system not found, and sometimes disk error.After weeks of painstaking step by step debugging, I realized the problem occurs at line 274 in sd_diskio.c:if (!((uint32_t)buff & 0x3))This file is automatically generated by CubeMX as part of the generation process. Once generated, this file has two user options, which are both required when using the H7 to enable cache maintenance:#define ENABLE_SD_DMA_CACHE_MAINTENANCE 1#define ENABLE_SCRATCH_BUFFERWith the scratch buffer enabled, this line is supposed to check whether the fs-win buffer where the SD card sector will be loaded

Bug Report: CubeMX Code Generation Error - Incorrect DMA Link in HAL_XSPI_MspInit Function Environment The software environment is STM32CubeIDEVersion: 1.19.0 Build: 25607_20250703_0907 (UTC)and the FW used is STM32CubeH7RS Firmware Package V1.2.0 / 05-February-2025. Problem Description When CubeMX generates theHAL_XSPI_MspInitfunction, it incorrectly linkshandle_HPDMA1_Channel0, which is the transmit (TX) DMA channel, to the receive (RX) pointerhdmarxof the XSPI handle. This causes manually modified code to be overwritten with the incorrect code every time the.iocfile is saved and the code is regenerated.Incorrectly Generated Code:__HAL_LINKDMA(hxspi, hdmarx, handle_HPDMA1_Channel0);Correct Code:__HAL_LINKDMA(hxspi, hdmatx, handle_HPDMA1_Channel0);Cause There is an error in the CubeMX code generation logic that incorrectly links the TX DMA channel to the RX pointer within theHAL_XSPI_MspInitfunction. Solution UtilizeUSER CODEblocks to protect manual

I'm trying to get semihosting to work under STM32CubeIDE with ST-LINK(OpenOCD) debugging.After being able to connect to the target (an STM32H503), I cannot step over the function initialise_monitor_handles()  (see source code). A lot of traps occur but I can't get into the real code to debug. (the lines in the main.c while loop).Bear with me but the code looks a bit like swiss cheese. Tried various reads, scanf etc. But at the moment I cannot get across the "initialise_monitor_handles();" call.

I'm using STM32f407 board with Cubeide. When load new code, I want to work code after pressing reset button.How can I set it in STM32CubeIde.Thank's