Ask questions and find answers on STM32 microprocessors and dedicated embedded software solutions, such as Yocto SDK, Qt, and OP-TEE.
Most recent activity
Hello We are trying to boot BareMetal application using Sd-card on STM32MP257F-DK Board. We have a few questions regarding this:1.In SD-Card Boot mode what is LOAD ADDRESS of image?2.At this address will it load only application or application with header?3.At entry point will it start executing the code or will it expect the vector table Thanks and regardsAkshitha Pattem
I have a STM32MP135 with a M24C64 64k bit I2C EEPROM. The EEPROM is divided into NVMEM cells, which can be written and read back via nodes in /sys/bus/nvmem/devices/... (NVMEM subsystem).The NVMEM cells do not behave in the same way as files: The length always corresponds to the size of the cell and the cell is always present (even if it has not yet been initialised).For certain applications it would be an advantage if the NVMEM cells behaved in the same way as a file.We could emulate the behaviour of a file by starting each NVMEM cell with a 12 byte header (4 byte magic, 4 byte size, 4 byte CRC):* if there is no magic, the ‘file’ does not exist.* the length information stores how many bytes have actually been written. This is also displayed as the file size.* a CRC secures the payloadThis principle corresponds to the u-boot-env layout (see https://github.com/torvalds/linux/blob/master/drivers/nvmem/layouts/u-boot-env.c)Unfortunately, the problem cannot be solved using the NVMEM API al
Hi St Teams , i followed this tutorial https://wiki.st.com/stm32mpu/wiki/How_to_build_kernel_for_Androidand i have a problem with linebuild_kernel [1/5]: Generate dtb binariesERROR: Not possible to generate the kernel imageother lines command work : build_kernel defaultconfig build_kernel vmlinux build_kernel modules build_kernel gpuand i did that :build_kernel --verbose=2 dtb[1/3]: Generate dtb binariesmake : on entre dans le répertoire « /home/linux/android_clean/device/stm/stm32mp2-kernel/linux-stm32mp2 »make[1] : on entre dans le répertoire « /home/linux/android_clean/out-bsp/stm32mp2/KERNEL_OBJ »make[1]: *** Aucune règle pour fabriquer la cible « dtbs ». Arrêt.make[1] : on quitte le répertoire « /home/linux/android_clean/out-bsp/stm32mp2/KERNEL_OBJ »make: *** [Makefile:256 : __sub-make] Erreur 2make : on quitte le répertoire « /home/linux/android_clean/device/stm/stm32mp2-kernel/linux-stm32mp2&nb
Hi,We are working on a product based on STM32MP157A MPU, As we thinking to update the kernel and QT version, we are looking for next OpenSTLinux 6.0 which supports scarthgap yocto and QT 6.5 LTS. When do we expect this release from ST?Also, What will be the LTS period of krikstone, mickledore and scarthgap?
Hi ST, I need to capture a picture on linux terminal. When I excute this cmd "v4l2-ctl -d /dev/video3 --set-fmt-video=width=2592,height=1940,pixelformat=YUYV --stream-mmap=3 --stream-skip=3 --stream-to=3.raw --stream-count=1 --stream-poll", the resolution of the picture is smaller than 2592 x 1940. I will attach the picture. What the reason is it? How could I get a full resulotion picture?
Hi all, I want to flash my EMMC using sd card? Do you have any guide or useful link for this?I have used Variscite SoM before and I was able to do it using yocto_install.sh script generated from yocto.How can I do this on STM32MP13X series? Regards
Hi,I am debugging the STM32MP257F-EV1 board with a debugger via JTAG/SWD. The kernel version is 6.1.82.While Linux executes, it loads the M33 firmware as confirmed in the prints below.----------------------------------------------------[ 15.673261] remoteproc remoteproc0: powering up m33[ 15.679697] remoteproc remoteproc0: Booting fw image USBPD_DRP_UCSI_CM33_NonSecure.elf, size 5402348[ 15.690327] rproc-virtio rproc-virtio.2.auto: assigned reserved memory node vdev0buffer@812fa000[ 15.698127] virtio_rpmsg_bus virtio0: rpmsg host is online[ 15.699120] rproc-virtio rproc-virtio.2.auto: registered virtio0 (type 7)[ 15.708049] remoteproc remoteproc0: remote processor m33 is now up----------------------------------------------------At the kernel prompt, I am able to break M33 core via the debugger, but soon after that, Linux freezes with the below message.----------------------------------------------------root@stm32mp
Hello STMicroelectronics Community: I read this wiki:https://wiki.st.com/stm32mpu/wiki/STM32CubeProgrammer_flashlayout. So I have some questions:1.Why are the offset addresses of fsbl1 and fsbl2 boot1 and boot2 in the following example? I noticed that except for emmc, this two values of other burning methods are a specific address.2.What is the meaning of metadatabin? Thanks,Ivy7
Hi everyone,I’m currently working with the STM32MP257F-EV1 evaluation board and would like to change the default UART used for debug output. By default, the EV1 board uses USART2 via the STLINK USB port for debug. However, for testing purposes, I’d like to reroute the debug console to USART6 so I can read debug messages from the header pins on the board. I’ve tried editing the device tree by modifying the chosen node and changing the parameter: stdout-path = "serial0:115200n8";to:stdout-path = "serial1:115200n8";According to my aliases, serial1 should correspond to USART6. However, after making this change, I encounter a watchdog panic very early during boot: m32_serc_handle_ilac:133 SERC exceptions [63:32]: 0x10 E/TC:0 stm32_serc_handle_ilac:139 SERC exception ID: 36 E/TC:0 Panic 'Watchdog' at /usr/src/debug/optee-os-stm32mp/4.0.0-stm32mp-r1/core/drivers/stm32_iwdg.c:228 <stm32_iwdg_it_handler>Based on this post I tried masking the st-m33firmware-
Good Morning, I don't know if anyone can help me or if this is right place to ask this question but I am having an issue when initializing the DDR on the STM32MP135F-DK board. I know the official advice is to run the DDR_init example prior to each run of my application. However for my final deployment of the code and for my ease of use id rather only have to build and run one program that initializes the DDR and then moves onto my program. I have tried dissecting the DDR_init example to no real luck. Especially around systemclock_config as in the DDR_init example the systemclock_config() only executes this if USE_DDR is not declared, yet the advice on the getting started with the stm32mpu lines states that for DDR use this must be added to the .cproject file. Has any one had any experience getting this working or know of any resources that might help? Many thanks, Stephen Kelleher
Hi ST, I have flash the Android images(st-android-13.0.0-2024-07-31-stm32mp257f-ev1-emmc-starter) on the STM32MP257-EV1. When booting the board(power supply by 5V3A adapter), the leds light as this status,is the board working normally? When the board booted and CN15 usb connected with PC , I found the device manager show the "Android aosp on stm32mp257f-ev1".When CN21 usb connected with PC, device manager shows up nothing new at allThe board was booting normally?If it is the normal status, how can I connect the terminal to the STM32MP257-EV1 on the host computer?
Hi everyone,I’m struggling with an OpenAMP / RPMsg problem on an STM32MP157 (M4 core).When I compile exactly the same application with STM32 CubeIDE (GCC) everything behaves as expected – two virtual‐UART channels (ttyRPMSG0/1) come up on the A7 and messages flow both directions.As soon as I switch the project over to IAR EWARM (v9.40): Linux still detects the vrings, creates /dev/ttyRPMSG0 and /dev/ttyRPMSG1, and the kernel log showsvirtio_rpmsg_bus virtio0: rpmsg host is onlinerpmsg_tty … new channel: 0x400 -> 0x0 : ttyRPMSG0 Writing from the A7 side blocks indefinitely:echo "hello" > /dev/ttyRPMSG0 never returns. On the M4 console I see no “Msg received…” printout – not even the IPCC interrupt fires (added a GPIO toggle inside HAL_IPCC_RX_IRQHandler for confirmation). What I have verified / triedInterrupt vectors – IPCC_RX1_IRQHandler and IPCC_TX1_IRQHandler are present in startup_stm32mp15xx.s, and the symbols appear in the EWARM map file.Mask
Hi,I'm trying to write a simple firmware for M33 core to test PWM outputs with TIM8 on STM32MP257FEV-1 board. It is possible to generate output for channels TIM8_CH1, TIM8_CH3 and TIM8_CH4, but as soon as I try to generate complementary outputs (TIM8_CH1N, TIM8_CH3N, TIM8_CH4N) to any of these timer channels, processor gets into HALT state. By this I mean that I've tried to map 6 GPIO pins to 6 complementary channels, because I need my firmware to drive a three-phase motor in the future. I've tried to do it as simple as possible by looking at some similar examples for other processors (e.g using HAL_TIMEx_PWMN_Start function to start complementary channel and so on), since I haven't found anything else for this processor. I'm also pretty aware that in my case there are 3 cores (cores of interest are A35 Linux core and M33 which will run this firmware). It seems that I have problems with Resource Manager and acquiring pins for channel complements, but I used them in
Dear Support Team,I am using the STM32MP257f-ev1 board. While the Starter Package boots Linux successfully, when I build the kernel using the Developer Package and copy the generated Image.gz, dtb, and lib/modules, the system reaches U-Boot but fails to boot the Linux kernel with the following error:```root '/dev/disk/by-partuuid/e91c4e10-16e6-4c0e-bd0e-77becf4a3582' doesn't exist.```I followed the official wiki instructions here:[https://wiki.stmicroelectronics.cn/stm32mpu/wiki/STM32MPU\_Developer\_Package](https://wiki.stmicroelectronics.cn/stm32mpu/wiki/STM32MPU_Developer_Package)using kernel version 6.6.48.I have some questions:1. The official README suggests running```make O="${OUTPUT_BUILD_DIR}" defconfig fragment*.config```but when I try this, zsh returns the error:```zsh: no matches found: fragment*.config```Therefore, I only ran```make O="${OUTPUT_BUILD_DIR}" defconfig```to proceed. Could this be the reason for the root partition recognition failure?2. How should the fragment\
I didn't change anything in the device tree just compiled the arm-trusted-firmware-v2.4-stm32mp-r2 instead of arm-trusted-firmware-v2.4-stm32mp-r1 and I get the following on UART:PANIC at PC : 0x2ffcfd2bWhat is different between the two versions?
Hi there, I followed the instructions as mentioned below.repo init -u https://github.com/STMicroelectronics/oe-manifest -b refs/tags/openstlinux-6.1-yocto-mickledore-mp1-v24.03.13 (v5.0.3)repo syncgot pyqt5 files from https://github.com/STMicroelectronics/meta-st-pkgrepo-updates/tree/openstlinux-ecosystem-5.0.0/recipes-python/pyqt5and copied them to layers/meta-qt5/recipes-python/pyqt5run DISTRO=openstlinux-weston MACHINE=stm32mp1 source layers/meta-st/scripts/envsetup.sh to get the conf/local.conf fileAll packages installed: “Check OK: all required packages are installed on host.”added IMAGE_INSTALL:append = " python3-pyqt5" to my image in build-openstlinuxweston-stm32mp1/conf/local.confbitbake st-weston image However, I'm encountering the following error:ERROR: Task (/home/abcd/DistributionPackage/layers/meta-openembedded/meta-oe/recipes-support/libiio/libiio_git.bb:do_configure) failed with exit code '1' when i checked error log in that it was mentioned
Hello ST team,our company is evaluating the STM32MP135F-DK to develop a new interface for our displays. This board was given to us by ST at the embedded world exhibition in Nuremberg, Germany. I am not familiar with MPUs and Yocto. I am having a lot of problems getting a simple PyQt5 GUI application (see mySimpleApp.txt) running on the board. I was able to successfully build the SDK and start a simple GTK application using the starter package.These are some of the steps I took:First, I used the starter package and installed there everything that i needed to run this application:apt install python3-pyqt5 qtwaylandcopied my application to /home/westonrun: su -l weston -c "env QT_QPA_PLATFORM=wayland-egl python3 /home/weston/mySimpleApp.py"These is the Error i get: [ 1] Failed to open device: No such file or directory, Try again...[ 2] Failed to open device: No such file or directory, Try again...[ 3] Failed to open device: No such file or directory, Try again...[ 4] Failed to
HelloI want to remove the vendorfs and userfs from the ev board, so I modified the tsv file just like this:after program the system of ev board, it looks like the rootfs cannot start up because the linux cannot recognize the rootfs partition:the uuid of rootfs-a partition used by linux is not correct, how to correct the uuid?
Hello, Im trying to follow the ST getting started guide for my STM32MP157F-DK2, currently at this part: https://wiki.st.com/stm32mpu/wiki/Getting_started/STM32MP1_boards/STM32MP157x-DK2/Develop_on_Arm%C2%AE_Cortex%C2%AE-A7/Modify,_rebuild_and_reload_the_Linux%C2%AE_kernelSo, I followed every instruction, but I just cant find the BSP file mentioned, the en.SOURCES-stm32mp1-openstlinux-6.6-yocto-scarthgap-mpu-v24.11.06.tar.gzI have ofcourse followed the link provided, but it leads me to this: https://www.st.com/en/embedded-software/stm32mp1dev.html#get-softwarewhich is the same address from which I downloaded the SDK in a previous section of the guide. I have looked for this file in the SDK, and in the Distribution package I later cloned from git. I can not find it, and I dont know how to continue with the guide. My knowlegde about Linux and mainly Yocto is limited, and I am mainly going through this to learn. I would love to discover my mistake is a silly one, but cu
Hello ST Community,I am currently working with the STM32MP157F-DK2 development kit and exploring RTOS options for the Cortex-M4 core. While FreeRTOS is officially supported through STM32CubeIDE, I would like to know if ThreadX (Azure RTOS)is officially supported and available for use on the Cortex-M4 core of the STM32MP157F-DK2.Specifically, I am looking for:Official support or examples of ThreadX running on the M4 core of this platformBSP or middleware integration through STM32CubeMX or STM32CubeIDEAny known limitations, configuration notes, or documentationI understand that ThreadX is well-integrated in STM32H7 and STM32U5 series, but it is unclear whether the same level of support is available for the STM32MP1 series, particularly the M4 core.If anyone has successfully used ThreadX on the STM32MP157F-DK2 (Cortex-M4), or if ST officially supports this configuration, I would appreciate your insights or pointers to relevant resources.Thank you in advance.
Hi,i want to use the secure storage feature of our custom STM32MP151CAA based platform. Therefore i have to write a HUK to the OTP as explained in the STM32MP1 wiki article. But how do i generate a HUK? What exactly do i have to program into the OTP?Thanks in advance,Börje
Hi,i want to enable optee-os for our STM32MP151CAA based platform. We use upstream TF-A, OPTEE and Barebox-Bootloader (very similar to u-boot), not from the yocto st layer. Yocto version is scarthgap, OPTEE version 4.1.I was able to switch from sp_min to optee for the AARCH32_SP and to let TF-A call OPTEE (BL32):NOTICE: CPU: STM32MP151CAA Rev.Z NOTICE: Model: xxx NOTICE: BL2: v2.10.4(release):devtool-patched-dirty NOTICE: BL2: Built : 16:23:33, Apr 26 2024 NOTICE: BL2: Booting BL32 I/TC: Early console on UART#4 I/TC: I/TC: Pager is enabled. Hashes: 2976 bytes I/TC: Pager pool size: 76kB I/TC: Non-secure external DT found I/TC: Embedded DTB found I/TC: OP-TEE version: 4.1.0-8-g8ab3c8212-dev (gcc version 13.3.0 (GCC)) #1 Fri Nov 15 13:08:46 UTC 2024 arm I/TC: WARNING: This OP-TEE configuration might be insecure! I/TC: WARNING: Please check https://optee.readthedocs.io/en/latest/architecture/porting_guidelines.html I/TC: Primary CPU initializing I/TC: RCC is non-secure I/TC: WARNING: All
Hi,I am using the stm32mp157-disco board and trying to mount one of my sd card partitions to my linux VFS at boot time with systemd. I have modified the file /etc/fstab, adding the following entry:# <file system> <mount pt> <type> <options> <dump> <pass>/dev/mmcblk0p8 /boot ext4 defaults,nofail 0 2but somehow systemd cannot mount the partition before the timeout. I've been unable to change the timeout for the disk mounting by adding x-systemd.mount-timeout under the <options> term of the entry. It seems that Openstlinux has been able to ignore the params in /etc/fstab.However, I can mount the same disk partition with command:mount /dev/mmcblk0p8 /bootafter the login. So I am not sure what I have missed. Can s
Hello We have custom board with STM32MP257DAI3 Processor.We would like to test it using Buildroot. Could you please advise whether it is feasible to use the STM32MP257F-EV1 Buildroot configuration for testing our STM32MP257DAI3-based custom board? Additionally, we would appreciate any recommendations or necessary modifications required to ensure compatibility. Thanks & RegardsAkshitha Pattem
I am working through the MPU Wiki guide Develop on the Cortex-M33 for STM32MP257FI am running the FlashLayout_sdcard_stm32mp257f-dk-ca35tdcid-ostl-m33-examples-optee.tsv image.I can successfully load and debug binaries for the M33 and the virtual UART behaves as expected. However, after adding the __LOG_TRACE_IO_ macro, any attempt to read from /sys/kernel/debug/remoteproc/remoteproc0/trace0 returns: Trace not availableFrom minicom, I can see that the log messages are correctly sent to /dev/ttyACM1I've observed the same behaviour for both the DK and EV kits.Having the log output to the remoteproc trace would be very useful during development. Any advice on getting this working would be great!Thanks in advance.
WATCH ON-DEMAND | STM32 Summit 2026
Already have an account? Login
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.