Product forums

It seems Azure® RTOS might be able to run on STM32MP135AAE3, the "Introduction to Azure RTOS with STM32" says "All STM32MP13x lines are covered by default by Azure® RTOS middleware within the matching STM32Cube MPU packages" but I'm having trouble finding any examples or "getting started" guides. 

Hello there,I am trying to get my Nucleo WBA55 running with the new CubeMX 6.14 and the new MCU package 1.6.0.Generating a project with just the bare minimum BLE GATT Server application, CMake configuration.Importing into VS Code with extension 2.1.1, compiling "debug" version, I get several multiple definition errors.The file is included twice in the makefiles in two different sections it seems.[build] [86/86] Linking C executable LCU_Demo.elf [build] FAILED: LCU_Demo.elf [build] C:\Windows\system32\cmd.exe /C "cd . && C:\ST\STM32CubeCLT_1.17.0\GNU-tools-for-STM32\bin\arm-none-eabi-gcc.exe -mcpu=cortex-m33 -mfpu=fpv4-sp-d16 -mfloat-abi=hard -Wall -Wextra -Wpedantic -fdata-sections -ffunction-sections -mcpu=cortex-m33 -mfpu=fpv4-sp-d16 -mfloat-abi=hard -Wall -Wextra -Wpedantic -fdata-sections -ffunction-sections -mcpu=cortex-m33 -mfpu=fpv4-sp-d16 -mfloat-abi=hard -Wall -Wextra -Wpedantic -fdata-sections -ffunction-sections -O0 -g3 -mcpu=cortex-m33 -mfpu=fpv4-sp-d16 -mfloat-abi

I've inherited a custom board that contains (among other things) an STM32MP135AAE3, an eMMC, and an Ethernet controller (DP83826ERHBR).  No SD card, no HDMI, no keyboard.  For that matter, no JTAG (board error). I had thought about running Zephyr but it's looking like that's not currently possible (discussion here).The documentation for creating a Linux image for this chip seem to fall under two categories:1) If you are using one of the development kits, here are explicit step-by-step instructions that will make it a joy2) If not, here (thud) are the manuals for yocto, bitbake, repo, and really the entire Linux kernel.  Read them all, digest them completely, and the answer will be obvious.Am I missing something?  Are there instructions somewhere to help me create a Linux image?  I don't need to become a yocto expert, I just need to create one Linux image one time.Thanks,Chris 

Hello,I have a Nucleo-L496ZG-P evaluation board that has a STM32L496ZGT6PU processor installed on.The processor has 2 CAN bus peripherals, and I would like to run an example that uses either of those.The STM32CubeMX has CAN examples, but not for STM32L, just for STM32F family.Could anyone recommend how I can import the CAN examples from the STM32F into the nucleo board.Thanks, D 

The STM32MP135AAE3 board was designed incorrectly, they have pins for a JTAG but forgot to actually route them to the MPU.  The eMMC is not going to have a first-stage bootloader on it, and it's looking like it's not possible to flash an SPL to the eMMC over USB alone?  Management won't be happy to do another board spin but it's sure looking like that's the only path forward.

Hello everyone,   I am trying to get an IP for my controller using the DHCP i can able to detect the Link up and link down but i can't able to get the IP. I had refered the example code for the ethernet(lwip https server netconn rtos) But i can't able to integrate with my main code can someone please help to either identify the issue or help me to integrate the example code with the main file.                     Thank you.Output Of my code  LWIP Stack InitializedDHCP Started...DHCP Task StartedLink Task StartedDHCP State Changed: 255 -> 6DHCP State: UNKNOWN (6)Ethernet Cable: CONNECTED DHCP Restarted

There is this errata for STM32G0B12.2.10 Prefetch failure when branching across flash memory banksDescriptionIn rare cases, the code prefetch may fail upon branching and function calls across flash memory banks,regardless of the DUAL_BANK and nSWAP_BANK option byte settings. The failing prefetch then provides anincorrect data to the CPU, which causes code execution corruption and may lead to HardFault interrupt.The workaround is obvious, either disable prefetch when jumping to code in the other bank and enable it again afterwards or use a small "trampoline" function in RAM. We can live with that, as our software is split into separately compiled "core" and "application" parts loaded into separate banks, having only a handful of interface functions there would be e.g. no jumps into library functions in the other bank. What about exceptions and interrupts, how are they affected? The ARM literature generally refers to loading the exception handler address from the vector ta

Hey!I have a ResNet model that I have deployed on my NUCLEO-N657X0-Q, and I want to run a training step on the board itself. I'm fine even with training just the last 1-2 layers of my model on a single sample, so that the memory on the board will suffice. Is the NPU itself capable of it? is there a assisting tool for on-board training? Is it more likely to succeed by using the ARM cpu for that?Thank you,Amir

H745 startup files are located in "CM7/Core/startup/startup_stm32h745xx_CM7.s" or "CM4/Core/startup/startup_stm32h745xx_CM4.s".However, the build source path for the startup files in the mx-generated.cmake in each core's folder is ". /Startup/startup_stm32h745xx_CM7.s" or ". /Startup/startup_stm32h745xx_CM4.s".Therefore, the cmake file must be changed to build.

Good afternoon, I am currently investigating the FHSS in the STM32WLE5JC chip. Using the STM32 SubGHz radio stackversion 1.3.1 I am unable to use any of the settings with LoRa. Enabling the FHSS and settingthe period does not change anything and the messages are transmitted as if this was omitted. It is my understanding that the radio IP is from Semtech, I cannot find much detail on if thehardware is based exactly on a Semtech board (i.e, SX1262) or a custom implementation. Back to LoRa FHSS have looked through the other forums and in 2020 an ST Employee saidit was not possible with the current stack, and in 2025 that it could be done(but the example provided was for LR-FHSS which uses a different modulation scheme GMSK). My questions:Is the STM32WLE5JC hardware comparable to a specific Semtech chip (so I can understand thelimitations)? Are the FHSS settings in the STM32 radio driver enabled for LoRa (it describesthe settings are LoRa specific)? Thank You!

generally i use cubemx choose cubeF1,but the  hal driver in github is ahead of official website,for example lateset version of cubef1 is 1.8.6.however hal driver has updated later for bug fix. so if i want to use the newest hal driver,i need download form github,and replace it in local repository. it is ok?

Hello,Does the ST67W611M1 support BLE Multi Role? In other words, can the ST67W611M1 BLE operate as a central and peripheral simultaneously?Thanks,Matthew

HI,I´ve to design a solution with a custom 5 and 7 inch color displays for a more or less static graphics (just some values and maybe some charts).Which interface / MCU would you recommend? MIPI-DSI? Would be fine to have a quite small MCU (preferrable no BGA or WLCSP)Best regards,Achim

  Hi everyone, I am working on an STM32G4 project using the HRTIM1 peripheral. My goal is to generate a burst of exactly 3 PWM pulses on Timer A and Timer B simultaneously, triggered by an External Event (Sync Input / EEV). The Goal: System waits for an external trigger (e.g., from TIM3 via TRGO or physical pin). Upon trigger, HRTIM Timer A and Timer B start. They output exactly 3 PWM cycles. They stop and wait for the next trigger (Single-Shot Retriggerable). The Configuration: I am using STM32CubeIDE and HAL, but I also tried direct register access. Master Timer: Acts as a dummy/pass-through for synchronization. Timer A & B: Mode: Single-Shot Retriggerable. Prescaler: Div1. Period: 1000 ticks. Repetition Counter: 2 (Targeting $N+1 = 3$ pulses). Start on Sync: Enabled. Reset on Sync: Enabled. Preload Enable: Disabled (To force immediate register update). The Problem: Despite setting REPxR = 2, I consistently see only 1 pulse on the scope when t

I have questions:Which C++ std used when building code in TouchGFX Designer? c++11, c++14, c++17?Which C++ std used when building code in MSVC simulator? c++11, c++14, c++17?Which c++ std ST used to verify and release TouchGFX?Can user change and modify the std? Any risk/ side effects to consider?

Hello!I'm working with the flexButton tool and wanted to change the color via eye-dropper. But once I turned on that option, I was unable to turn it off, I can't disable the color picker and it continously changes the color even after deselecting.  Isolating the cause a bit I realized it only happens when I move my cursor off screen into another window and click, operating the tool normally (staying within touchGFX window) correctly turns off the tool after selecting. Just a bug I noticed,Thank you

Hello I have a problem compiling TouchGFX for Keil. My procedure:1/ Run TouchGFX Designer 4.252/ Create New project3/ Go to tab – By Partners – select board Riverdi STM32 121STM32H74/ Click to Create5/ Add Button ( for example)6/ Click Generate CODE7/ Go to project Folder and open ioc file Riverdi_121STM32H7_1_6.ioc STM32CubeMx v6.58/ Go to tab Middleware and Software pack Select FreeRTOS M7 Select tab Advanced settings USE_NEWLIB_REENTRAND ← disable !!! ( for keil it must be turned off )9/ Go to project manager Switch toolchain IDE to MDK keil 10/ Save and Generate CODE11/ After generate code by STM32CubeMx I'm going back to Touchgfx Designer and click generate code again TouchGfx generate code for keil.12/ Run project in Keil13/ Go to Option for target Go to tab OUTPUTand uncheck Browse information14/ Project build target15 / Keil linking problem Not enough information to list image symbols.Not enough information to list load addresses in the image map.Finished: 2 information, 0

hi,I am currently using BMP + rgb565 LTDC display for GUI icon creation without Touch GFX ( Everything is on "" C"" only and Manual preparation icon using hex converter ).Already known one is ! for a small size ,BMP is OK but large size BMP icon  will increase the size and processing time  .   Above Picture size is 6.30KB in BMP                                                Above Picture size is 2.03KB in JPEG      Visually i could not see any difference  BMP and JPEG in my laptop   .Now i am using internal memory for small icon  (Now working with stm32h723zgt6 and plan to upgrade to stm32u5 series )Please clarify my doubts. 0. Can i use Hardware Inbuilt JPEG decoder for JPEG  GUI icon creation with rgb565 format  ? . 1. Do you think i can achieve best frame rate by  Hardwa

Hello,after upgrading from STM32CubeIDE 1.9.0 to 2.0.0 (Eclipse 2024‑09), we encountered a critical issue with Git integration when using SSH and Azure DevOps.In version 1.9.0, Git operations (pull/push) worked correctly because the IDE allowed selecting an external SSH client (e.g., OpenSSH or PuTTY/Pageant). Azure DevOps recently deprecated ssh-rsa (SHA‑1) and now requires rsa-sha2-256 or rsa-sha2-512 for host key algorithms.In STM32CubeIDE 2.0.0:The IDE uses Apache MINA sshd internally for SSH.External SSH client configuration (GIT_SSH or core.sshCommand) is ignored.Apache MINA does not support RSA-SHA2 host key algorithms, so the handshake fails with:                 Unable to negotiate key exchange for server host key algorithms.                  Client offered: ssh-rsa, ecdsa-sha2-nistp256, ssh-ed25519                 Server requires: rsa-sha2-256, rsa

Hi, Does Spice Model of VIPERGAN65D available? If so from where can I download it?

Hi STM32 community, I have ordered those the Evalboards: NUCLEO-G0B1RE X-NUCLEO-SNK1M1 - USB Type-C Power Delivery Sink  I have followed all of your existing applications, but none of the had lead to an similar result as displayed by your application descriptions. Lets stick to the newest one:(24) How to build an USBPD Sink application using the X-Cube-TCPP software pack - YouTubeI followed every step accurately several times now, but the result is always the same, Contract -> DEFAULT 5V no current. What am I doing wrong? I have studied all of your AN, watched all Videos, read and did the Wiki but never got any positive result as described..... Im frustrated and hoping for your help. Thank you very much! Br,KUDAThe usbc_test_3.zip is the zipped project.

Hi together,I want to use the Video Widget with the STM32U599 and TouchGFX version 4.25.0, but I can't activate video mode in STM32CubeMX.Are there any new details about this problem?Or is video usage not supported with the STM32U599?  

GOAL To use Ethernet + LwIP on the STM32H743 chip without cache. Explanation We don't want to use the cache when using Ethernet because when the cache is enabled and data is sent and received via DMA on the SPI line, the correct data doesn't propagate to the bus due to a cache miss. Disabling the cache resolves this issue, which is why we prefer not to use the cache when using Ethernet. Cube Versions: STMCubeIDE v1.18.1  STMCubeMX v6.14.1 STMCubeFW_H7 v1.12.0 We used the settings in the following link to configure the project by @Adam BERLINGER 's project: https://community.st.com/t5/stm32-mcus/how-to-create-a-project-for-stm32h7-with-ethernet-and-lwip-stack/ta-p/49308. The CubeMX configurations are as follows: CORTEX_M7 configurations ETH configurations LWIP configurations Problems Encountered When Cache is Disabled 1. Usage Fault: The device enters a Usage Fault error when the program execution branches into the function SCB_InvalidateDCache_by_Addr() within the fo

  Subject: TouchGFX Custom Application FLASH Overflow Issue & QSPI Memory SetupGreetings Team,I’m new to STM32 TouchGFX and trying to create a custom TouchGFX application following How to integrate TouchGFX in a custom board (The long way round) - YouTubeAfter setting everything up, I encountered the following error during the build:`.text' will not fit in region `FLASH' region `FLASH' overflowed by 296512 bytes  I understand that my MCU has only 128KB of Flash, which is insufficient for TouchGFX. I want to configure TouchGFX to use only SDRAM instead of Flash. How does TouchGFX utilize SDRAM, and how can I correctly set it up for this purpose?Additionally, when comparing my custom application to the STM32H7 Dev Board example, I noticed different memory configurations in the Cortex-M7 section under System Core. Specifically, I don’t fully understand how QSPI Flash is mapped as two memory regions starting from 0x90000000.I've attached a screenshot for reference.

Autocomplete is not working on v2.0.0, also didn't work on v1.19.MacOS 15.5Macbook Pro 16"2,3 GHz 8-Core Intel Core i9Any suggestions?