STM32CubeMX (MCUs)

 Hi everyone,I've noticed a change in UART pin configuration behavior from CubeMX 6.10.0 to 6.16.0 that has caused a practical issue:Before (v6.10.0): It was possible to configure only the RX or TX pin for an asynchronous UART, enabling "receive-only" or "transmit-only" modes.Now (v6.16.0): Both TX and RX pins must be assigned simultaneously; otherwise, the UART peripheral cannot be enabled.Problem: This wastes a GPIO pin. My specific use case is an SBUS receiver (one-way communication, requiring only the RX pin). Now I'm forced to allocate an additional TX pin that will never actually be used.Questions:1. Is this a permanent new requirement in CubeMX, or is it a GUI bug in version 6.16.0?2. What is the technical reason behind this change?3. Is there any way to achieve a true single-pin UART configuration under the new version?Thanks for any insights.

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 

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.

I have been trying to get VS Code working with the latest STM32 VS Code plugins, and not having much luck. Latest plugins installed, on a nice fresh install of Fedora Desktop Linux.I have been using Red Hat/Fedora/CentOS, Ubuntu and versions of Mint and Arch as desktop OSes and development platforms since 1995...That being said, I was frustrated that I couldn't get this piece of the plugin puzzle working. The plugin could not find the Cube MX installation. So, I tried adding a .exe extension to it (yep, on Linux), and lo' and behold, there it was. Cube MX came to life.This tells me the plugins are coded to search for a .exe file extension by default.John

Hi, I'm trying to understand what filters I have to apply in CubeMX to find products that support 2 QUADSPI with independent clock. E.g. h7a3aii6q seems to be such part that allows me to have two independent  QUADSPI and there are parts like e.g. H743XIH6 that have QUADSPI Dual-Bank with two NCS but with only 1 Clock and there are also Parts with OCTOSPI, but they don't seem to have two independent interfaces most of them time.Any chance to make part search easier for me to find products that have features like these two independent QUADSPI?Thx

Hi there, im using Nucleo-F446RE , ST HAL and CubeMX, and im learning to configure the clockI use HSE Bypass mode (MCO from the st-link in the nucleo), everything perfect hereIn the RM i found out that, when using HSE BYPASS mode, PH1 (OSC_OU) should be left in HI-Z state What does HI-Z means in STM32 MCUs? does it mean configuring the pin as digital input? or it means configuring the pin as analog input?Or it depends?   When selecting HSE BYPASS MODE the chip image in the .ioc file gets set like this: But if i try to  set PH1 (OSC_OUT) as GPIO a conflict is generated, so, how can i set the state of PH1 as HI-Z?   Also, why does PH1 gets reserved and configured in the chip as OSC_OUT? is this to avoid configuring it as another thing?And finally, what it seems very weird to me, why the clock por GPIOH is enabled when CubeMX generates the code if i cant use PH1 as a normal GPIO? (im not using any other PHx PIN)   Thanks in advance and sorry for the english, not a n

In the H562 CubeMx setting the flash option bytes only allows flash number of sectors up to 7. If you look at the FLASH_EDATA_EraseProgram example in the H5 firmware bundle you will see it sets the number of sectors to 8 before calling FLASH_OB_EDATAConfig function. this function decrements by one before ORing the sector enable bits. Because CubeMx only allows up to seven when it generates code you do not get the full EDATA range. And if you set 1 in cubemx I guess you do not get any??? 

I have started a new project using STM32H573. This is using the following:Cube MX 6.16Cube IDE 2.0 - Build: 26820_20251114_1348 (UTC)The project has secure zone enabled. I have written bare metal code and shown I can run/debug code in the non-secure environment without issues, and I can invoke non-secure callable functions etc.My problem comes when I add Free RTOS and enable CMSIS RTOS v2 to the non-secure side.When I build, I get linker errors (see below).When I search the forums, there have been similar issues raised, but most are quite old posts or seem overly complex (which always worries me)``C:/Users/531023/git/TEML20/Monitor/NonSecure/Debug/../Core/Src/app_freertos.c:86:(.text.MX_FREERTOS_Init+0xa): undefined reference to `osThreadNew' C:/ST/STM32CubeIDE_1.18.0/STM32CubeIDE/plugins/com.st.stm32cube.ide.mcu.externaltools.gnu-tools-for-stm32.13.3.rel1.win32_1.0.100.202509120712/tools/bin/../lib/gcc/arm-none-eabi/13.3.1/../../../../arm-none-eabi/bin/ld.exe: ./Core/Src/app_free

So here is a weird thing I've noticed with a STM32U073... CubeMX shows that this CPU has the "Trace Asynchronous SW" option, but for this mcu it is no longer listed under SYS, but under its own category: "Trace and Debug" -> "DEBUG". Anyway, after enabling it, the signal shows up on the graphical pinout, with the name "DEBUG_JTDO-SWO". This is pretty much the same as i was used to with other STM32 mcu's. So when i was designing a PCB for this new STM32U073 MCU i assumed it had support for SWO (or SWV or whatever the name is). But now when I'm trying to debug this MCU, the debugger doesn't allow me to enable "Serial Wire Viewer (SWV)" in the debug configuration properties of STM32CubeIDE. The option is just greyed out. So I did some more digging in the datasheet for the STM32U0, and I'm starting to wonder if it even has support for SWO/SWV?  Can anyone enlighten me? Is this just a problem with the debug config in STM32CubeIDE, or does the STM32U0 have no SWO/

STM32Cube MCU Package for STM32U0 Series 1.3.0  + STM32CubeMX 6.15.0 + STM32U083RCT6You can configure the SWO output in the configurator (CubeMX), but in fact, SWO is not present in the microcontroller. 

Hi I am using STM32H7S78-DK.I don't understand how to use the compare tool in cube mx project.My project consists boot and application software, i.e., a total of four different software versions. Which software does the compare tool compare? How can I select the software?    

Hi all,I'm using STM32Cube IDE (version 1.13.1, build 17479_20230728_0839 ) to manage my company project based on a STM32U5 MCU.More specifically, I'm using linked list feature (for playing audio with SAI + DMA transfer) and I have noticed what appears to me as a bug in auto-generated code.Basically, I have created through CubeMx one list named SAIQueue, with one node named NodeTx and the auto-generated initialization code is the following:  /** * @brief DMA Linked-list SAIQueue configuration * @PAram None * @retval None */ HAL_StatusTypeDef MX_SAIQueue_Config(void) { HAL_StatusTypeDef ret = HAL_OK; /* DMA node configuration declaration */ DMA_NodeConfTypeDef pNodeConfig; /* Set node configuration ################################################*/ pNodeConfig.NodeType = DMA_GPDMA_LINEAR_NODE; pNodeConfig.Init.Request = GPDMA1_REQUEST_SAI1_B; pNodeConfig.Init.BlkHWRequest = DMA_BREQ_SINGLE_BURST; pNodeConfig.Init.Direction = DMA_MEMORY_TO_PERIPH; pNodeConfig.Init

1. Issues with Headless Execution and CLI CommandsGUI Opens with CLI Flags: Certain CLI flags designed for automation still launch the AWT/Swing GUI components, which causes failures in headless environments (like Docker containers or typical Linux servers).Specific Example: The -P flag (Package Manager) throws a java.awt.AWTError: Can't connect to X11 window server error because it requires a display.Expected Behavior: All package management and code generation functions should have a truly headless (no GUI dependencies) CLI option that works universally in server environments.Lack of Clear CLI Documentation: Documentation for reliable, headless CLI commands (like swmgr install) is sparse or non-existent in the official user manual (UM1718), leading to trial-and-error with internal commands. 2. Issues with Offline Package Management and Database IndexingDiscrepancy Between Repository and Database Paths: Manually placing firmware p

When attempting to initialize a single GPIO EXTI as disabled, the generated code still enables the interrupt. I did notice that if I uncheck all of the EXTI lines, then none of them are enabled in the generated code. So it seems it's all or nothing. Is this the expected behavior?Thanks,FrankThis:Generates: /* EXTI interrupt init*/ HAL_NVIC_SetPriority(EXTI0_IRQn, 5, 0); HAL_NVIC_EnableIRQ(EXTI0_IRQn);   HAL_NVIC_SetPriority(EXTI2_IRQn, 6, 0); HAL_NVIC_EnableIRQ(EXTI2_IRQn);   HAL_NVIC_SetPriority(EXTI3_IRQn, 6, 0); HAL_NVIC_EnableIRQ(EXTI3_IRQn);   HAL_NVIC_SetPriority(EXTI4_IRQn, 6, 0); HAL_NVIC_EnableIRQ(EXTI4_IRQn);However, this:Generates: /* EXTI interrupt init*/ HAL_NVIC_SetPriority(EXTI0_IRQn, 5, 0); HAL_NVIC_SetPriority(EXTI2_IRQn, 6, 0); HAL_NVIC_SetPriority(EXTI3_IRQn, 6, 0); HAL_NVIC_SetPriority(EXTI4_IRQn, 6, 0);

Hi, Is there anyway to generate codes in such way that we can customize our sub-project paths? E.g. Instead of having project/core/src, we have project/src... etc? Thanks in advance!

Hi,I'm using CubeMX 6.16.0 with the EWARM setting.When I click "Generate Code," it creates the IAR project with the default name "project.eww."I open the project with IAR and save it with a new name.Then I click the "Generate Code" button on CubeMX, and it regenerates the workspace name, again as "project.eww."Is it possible to set the IAR project name from CubeMX to something other than "project.eww"?Thanks

After (auto) migration the .ioc file to the actual CubeMX 6.15.0, the Dual Quad SPI mode got changed to Octo SPI mode and there seems to be no way to get back to Dual Quad SPI mode, since the mode selection entry "Dual-Quad SPI" is not shown as mode selection entry.After taking a closer look, I found out, that the Dual-Quad SPI mode is avail only for the OCTOSPI2 device, whereas the Dual-quad SPI option is not avail for the OCTOSPI1 device (tested with a "clean" configuration). Since my hardware was doing fine with Dual-quad SPI mode on OCTOSPI1 device there seems to be no reason for removing the Dual quad SPI mode from OCTOSPI1.I think, this behaviour is not intended ? Actually I found no workaround for this problem except for manually changing the auto-generated code so it would be nice to have that issue fixed. Even if changing the .ioc file manually the Dual Quad SPI mode option is removed on next load into the MX GUI again. Best regrads, Heino 

Hello ST team and community,I am working with STM32H755 and noticed a potential bug in the code generated by STM32CubeMX when handling the RCC_FLAG_D2CKRDY check during system initialization. CubeMX produces two variants of the loop:while((__HAL_RCC_GET_FLAG(RCC_FLAG_D2CKRDY) != RESET) && (timeout-- > 0)); if (timeout < 0) { Error_Handler(); }andwhile((__HAL_RCC_GET_FLAG(RCC_FLAG_D2CKRDY) == RESET) && (timeout-- > 0)); if (timeout < 0) { Error_Handler(); }Issue observedThe loop condition uses (timeout-- > 0).This means the loop exits when timeout reaches 0, but never decrements below zero.As a result, the check if (timeout < 0) will never be true, so Error_Handler() is never called — even if the D2 domain clock never becomes ready.Since STM32H755 relies on proper synchronization between the Cortex‑M7 and Cortex‑M4 domains, missing this error condition can mask serious startup failures. Developers may assume the system initialized correctly when

Hi,in the latest cube (6.15) the STM32H573LI is not available, I can't start a project on this MCU.why is that? is there going to be a fix soon?

The new version of CubeMx (6.15.0) nags me about enabling SMPS mode, and I would like to, but the tool seemingly does not have the capability to do so.Under PWR->Low Power there's a checkbox for "Power Savings Mode" which other forum posts led me to believe would need to be enabled. I have found two things about this checkbox. 1.) It is already enabled and 2.) You cannot disable it.Then, under "System Core" -> RCC ->Parameter Settings, there's a "Supply Source" parameter. It has only a "PWR_LDO_SUPPLY" option, and cannot be changed to any other value.Am I missing something here, or is the tool broken? 

Hello, I get the same error recently on 1.19.0 for my STM32U385RGTxQ. The only option available in Power Supply source is PWR_LDO_SUPPLY

From Cube 1.19.0, i keep getting the "For the best power consumption figure, enable SMPS" and although the Nucleo-U385RG-Q board uses a Q variant, I'm not able to switch to SMPS from CubeMX.

Hello,I am trying to understand the memory allocation for the FDCAN, I am using a nucleo H723ZG.With the latest CubeIDE 1.18.1 I seem to be able to configure 32 TX Buffers and 32 TX FIFOs:However the message RAM mapping in AN5348 - Rev 5 shows that there are only a maximum on 32 for both Buffers and FIFOs:Is there a mistake in the MX interface that should not allow more that 32 of Buffers + FIFOs combined, or the memory mapping has 2 sections of up to 32 elements for both buffers and FIFOs? Thanks,Marco

As alread reported a while ago, strange things happen when entering an expression in CubeMX for Prescaler and/or Autoreload. If, for example, 72-1 is entered for Prescaler and 6667-1 for Autoreload, the following code will be generated:STM32CubeMX: TIM Prescaler, Period bug   TIM_InitStruct.Prescaler = 72-LL_TIM_IC_FILTER_FDIV1_N2; ... TIM_InitStruct.Autoreload = 6667-LL_TIM_IC_FILTER_FDIV1_N2; LL_TIM_IC_FILTER_FDIV1_N2 does not make any sense here.Depending on the exact values, the Timer will be initialized with some different settings than desired.With simple constants (without calculations) the generated code will be correct.In contrast to the previous post, in my case the MCU was a STM32F303.CubeMX is Version: 6.15.0-RC7, Build: 20250701-1659 (UTC)

I created default project with FreeRTOS in STM32CubeMx for the board Nucleo-H563ZI. Nothing more added.Hard Fault is invoked at vRestoreContexOfFirstTask.I was looking for the solution in Copilot and Perplexity but without success. Only I can say it's probably no problem with stack overflow.Jan