STM32CubeIDE (MCUs)

Hi.How can I fold / unfold only selected lines ?Is it possible ? 

..

I mentioned this in my other recent thread: https://community.st.com/t5/stm32cubeide-mcus/start-3rd-party-gdb-server-or-any-script-from-debug-config/m-p/734462#M31799 and I thought I did have this working in the past, by perhaps I only used printf-style debugging via the Renode USART emulation - I don't recall for sure.If one uses a remote GDB server connection, it would seem to stand to reason that ST debugger hardware should not be expected or enforced to be present?The reason that this option is even there suggests one should be able to make this work - or is it an oversight / Eclipse CDT leftover?Can this be made to work? I.e. debug a firmware running on the Renode emulator (which starts a GDB server, but no ST-link or other debugger Hardware is involved), from CubeIDE.---EDIT: ok, so this is not about the ST-Link, but actually checking it's an ST MCU, because auf counterfeit chips?How unfortunate then that this wrecks debugging in an emulator with this IDE, especially in

Hello all,Whenever I try to update a Live Expressions value while the program is running, the debugger hangs (no update, no way to pause the execution).If I pause then update, it works (but of course this is no more a "Live" update).I use JLink. Any ideas?Thank you in advance.

I use STM32CubeIDE on Linux Ubuntu (and to a lesser extent on MacOS and Windows, but I don't recall having a problem on the last 2 operating systems). About once or twice a day, at random, several of the keys such as <backspace> and <enter> and <delete> stop working. All the alphabetical and number keys work as normal. If I switch to a different application all the keys are working, but when I switch back to Cube the problem continues.I solve the problem by saving my work, shutting and relaunching Cube. Is this a bug in the application?

We've designed a board with an H563 on it. Before I started the PCB layout I checked that FreeRTOS was supported for this device and saw it was.However, now I come to create a project in the Cube IDE it doesn't seem quite so straightforward.It [the H563 family] wasn't supported in version 1.9 of the IDE so I upgraded to 1.16.I'm concerned though that FreeRTOS no longer seems to be automatically in Cube's set up ands you seem to have to install this X-CUBE-FREERTOS  add on which seems to suggest though that TCP/IP is not supported. Is that correct?Also once you've downloaded this add on, how do you actually install it into Cube?If we need to run FreeRTOS with TCP/IP support is the H563 the wrong processor to be using?Thanks! :)  

Hi,In Cube 1.16.1, when I fill 'Log to File' field with the following absolute path using Cube variables, log file is not created.[Run > Debug Configurations > Debugger > Mics > Log To File] ${workspace_loc:/${ProjName}/dbg/log.txt} However, when the path is absolute all works Okay.Why the path with Cube variables does not work ?Where can I see what 'workspace_loc' or 'ProjName' expand to ?What other Cube variables there are apart from 'workspace_loc' and 'ProjName' ?Thank you 

Hi, i`m interested in using the mbedtls v1.3. I currently using the v1.2 and like to upgrade to v1.3 but could not see any v1.3 version in the configuration tool to choose. Is there any plan to upgrade?Thanks in advance

In CubeIDE, in the Debug Configurations window, there is the Debugger tab, where you can select "Autostart local GDB server" or "Connect to remote GDB server".Now, what I have been doing to debug code running on the Renode emulator, is manually start Renode & manually start its GDB server, and use the "remote GDB" option in the debug config.What I'd like to do is that when I start debugging / running from CubeIDE, that everything happens automatically,i.e. somehow Renode and its GDB are started, and _then_ (it needs to be mare sure to be running for connecting, of course) connecting to that GDB server, to load the binary & debug it.And optimally, close Renode down after it.Is there some trick to somehow achieve that?A follow-up question then would be: - is it possible somehow to have the simulated UART output, which is also seen from a SSH connection to Renode, from within CubeIDE?I remember to have seen setups where remove SSH sessions were started from within Eclipse CDT

Hi, I would like to port my Mbed Program (STM32 F103RB) to CubeIDE Project. This is due to the announcement from Mbed that they would stop supporting Mbed Studio from Aug 2026. Anyone can give me the advice on how to do the porting? Thanks a lot in advance.   

Hello everyone,I'm currently trying to create a nested project in CubeIDE. Unfortunately I can't find any useful information online.I'm probably searching for the wrong terms.I would like to create a Project structure something like this:Parent Project     - Bootloader Project     - Application ProjectIf the parent project is built, the bootloader and application projects should be built as well. Both the bootloader and application have the same target chip and are stored in different parts of the flash memory.I would highly appreciate any tips or suggestions on how to do this.Kind regardsDominik Schmid

I start the installation of STM32CubeIde on Debian 12, but it is impossible to continue the installation because it needs libpython2.7 and I have python3 installed on system. I can't downgrade the system.How can I do?Thank you

 Issue occurs in this build: To reproduce: In an STM32MCU project right click propertiesNavigate to C/C++ General -> Paths And SymbolsSelect the Symbols TabAdd a new Symbol and in the value field, reference an environment variable inside quotes e.g. "${PATH}"  Click save and the trailing double quote is removed The workaround is to edit the symbols elsewhere see below noting where I added the extra quote:  

Got a project that compiles on another computer (which I now cannot access to test whether it's still so), but now that I'm an another machine, same version of CubeIDE (1.12.1), I get the error quoted above. I'm not sure when I updated last & whether I tried to build then.Why would the compiler not find a type within library stuff? I haven't touched those files.Well, if, in sysmem.c, I add#include <sys/types.h>it compiles. I don't like messing with the vendor library files, though ;)EDIT:This was with toolchain GNU Tools for STM32 (11.3.rel1).When I chose Fixed: 10.3, it compiles as before.Why did I try that? Because, in a related project, I was suddenly also getting BS errors that prevented the build, e.g. about casting from a pointer type to a uint32 in a constexpr - this compiles just fine with 10.3, and it's essential, using unsigned values from a linker script, to address certain memory regions in the program.Note to self: never let IDE select toolchain willy nilly, and

After updating CubeIDE to V1.16.0, I get an error like "Data mismatch found at address 0x08001000 (byte = 0xFF instead of 0x0E)" and cannot complete the debugger. This error text was obtained from the Internet with similar symptoms, except for the address number, etc.I have tried all the solutions available on the Internet, but none of them worked. I also installed and tried CubeIDE released in the past, but it did not work.Before the update, I was able to complete the debugger reliably.Development Environment CubeIDE V1.15.1 or V1.16.0ST Link V3MINIE

Keil (among others) allows the choice of whether to do a full or partial erase before programming:Is it possible to do that in STM32CubeIDE ?If so, how ? 

Background: most of our development so far has been with AVR and cmd line based, using make files for testing. We only ever used IDE's for compiling for target and uploading: Arduino IDE for prototyping with bootloader and MpLab without bootloader for production hardware. We are now transitioning some hardware to STM32 and plan to do any new developments on that platform.The issue we are running into is that the integration of all hardware initialisation in the main.c file is different from the existing code: libraries with hardware interaction have all platform-specific code in separate c/h files. For example, a module that uses EE may have a writeEE and readEE function that are defined in 2 separate c files, e.g. atTiny402EE.c and atMegaEE.c. On an STM32 without internal EE, we will use external EE, so we will need functions that are in must cases only wrappers around HAL_I2C_Mem_Read(), etc.My question is: where should this live? inside the main.c file? : I prefer not, bec

hi, my computer is win11 and 64x so i thought there are no problem.   but, after clicked download start button from e mail, nothing happens.i try to click lastver and download button but my computer can't download stm32cubeide.  i didn't understand... stm32cubeide is impossible to download now....ummm i disable every firewall from google and computer settings, and i try a lot of times but the problems not solved now.  One thing is for sure: I signed up for membership and agreed to everything, but when I pressed the e-mailed download button, no downloads will run, and even if I keep trying, it won't download itself. This is regardless of the program's execution, but stm32cubeide won't download at all, so how do I fix this?and i am not talk about install. just download!! omg.... just download is not working!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! haaa...  hhhaaaaaaaaaaaaaaaaaaaaaaa..... got *** it... i tried it for almost 1 mont

Since it took a tremendous amount of effort in getting CMake working within STM32CubeIDE, I'd thought I'd share a method of getting it done. I've only tested this working in Windows. Many an hour could have been saved if ST, in their infinite wisdom, did not disable the Toolchain tree entry in the project C/C++ properties!Disclaimer: I'm not an experienced Eclipse user, so there is a very good chance that the steps below are not the correct way of doing things, and in future, ST might lock-down or hide more UI rendering these steps useless - you have been warned!There are a huge number of steps required, so this is not for the fainthearted!Prerequisite: You must have an existing CMake GNU Arm project up and running, if not, find out how set one up, or use CLion to create a CMake project from an existing STM32CubeMX .ioc.For the steps, I'm using the following folder layout (adjust to fit your preferences):Source (source code, folders, etc) build .. this where I set CMake to run the bu

Hi,I would like to compile some C++ code based on toolchain used e.g. CubeIDE(gcc-none-abi) vs uVision(clang) vs MSVC vs GxfDesigner.How can I uniquely identify CubeIDE's toolchain? How can I view the toolchain's intrinsic defines?Thanks

Hi,I've just installed CubeIDE 1.16 and without any change to code/source file, every press of Ctrl+B shows this dialog: Any idea why this message please ?   I did not have this with IDE1.14. Thanks

good morningI downloaded the package for debian yesterday but unfortunately it doesn't work on ubuntoso today I wanted to download the generic package but I access the site but the download doesn't start I tried several times nothing I accept the conditions I click download but nothing

STM32CubeIDE generates the hex file from elf file using below command.arm-none-eabi-objcopy -O ihex ${ProjName}.elf ${ProjName}.hexBut when i use the STM32CubeProgrammer, it generates the hex file which is smaller than that of STM32CubeIDE hex generate file. I wonder what arguments are used in "arm-none-eabi-objcopy" to match the hex file that generated by STM32CubeProgrammer. 

I'm operating an Infineon F-RAM CY15B116QSN at the OctoSPI port of an STM32H735 MCU with 100 MHz clock (F-RAM devices is specified up to 108MHz). The F-RAM datasheet requires a CS high time between commands of at least 145 ns, which requires 15 clock cycles @100MHz. The CubeMX tool limits the CS high time up to 8 cycles (3 bits wide); the same applies to the STM32H7xx HAL driver, which asserts, if this value is outside the range 1 to 8. In contrast to these limitations, the STM32H73x reference manual allows a CS high time of 1 to 64 (6 bits wide), which would suffer our needs ! Actually I added a copy of the 'HAL_OSPI_Init' function with different name to my project to get around this limitation, but it would be fine, if this bug will be fixed in future versions - or am I wrong at some point ?Best regards, Heino  

d64.deb'Note, selecting 'st-stlink-udev-rules' instead of './st-stlink-udev-rules-1.0.3-2-linux-all.deb'Note, selecting 'st-stm32cubeide-1.15.1' instead of './st-stm32cubeide-1.15.1-21094-20240412-1041_amd64.deb'Note, selecting 'segger-jlink-udev-rules' instead of './segger-jlink-udev-rules-7.94h-linux-all.deb'Some packages could not be installed. This may mean that you haverequested an impossible situation or if you are using the unstabledistribution that some required packages have not yet been createdor been moved out of Incoming.The following information may help to resolve the situation:The following packages have unmet dependencies:st-stm32cubeide-1.15.1 : Depends: libncurses5 but it is not installableDepends: libpython2.7 but it is not installableDepends: libwebkit2gtk-4.0-37 but it is not installableE: Unable to correct problems, you have held broken packages. Which STM32IDE we should download to fix this error