Ask questions and find answers on STM32 microprocessors and dedicated embedded software solutions, such as Yocto SDK, Qt, and OP-TEE.
Most recent activity
I am making a board using STM32mp151a.I want to make Linux OS for this machine.I refered to the web page below.https://wiki.st.com/stm32mpu/wiki/How_to_create_your_own_machineHowever, Error happened on bitbake process.Please tell me the cause and countermeasures.Error message************************************************** *************| ERROR: Input tree has errors, aborting (use -f to force output)| make[1]: *** [/home/chino/STM32MPU_workspace/yct/build-openstlinuxweston-stm32mp1-led_g_test2/tmp-glibc/work-shared/stm32mp1-led_g_test2/kernel-source/scripts/Makefile.lib:414 : arch/arm/boot/dts/stm32mp157f-led_g_test2-mx.dtb] Error 2| make: *** [/home/chino/STM32MPU_workspace/yct/build-openstlinuxweston-stm32mp1-led_g_test2/tmp-glibc/work-shared/stm32mp1-led_g_test2/kernel-source/Makefile:1471: stm32mp157f-led_g_test2-mx. dtb] Error 2| ERROR: oe_runmake failed| WARNING: exit code 1 from a shell command.ERROR: Task (/home/chino/STM32MPU_workspace/yct/layers/meta-st/meta-st-stm32mp/reci
Hello,I'm trying to configure an externall interrupt (for a digital signal) to be handle from the User Space. I have configured the following node in the device tree on the root node:/ {... exti_gpio_node { compatible = "test,button"; input-gpios = <&gpiod 15 (GPIO_ACTIVE_LOW | GPIO_PULL_UP)>; interrupts-extended = <&gpiod 15 IRQ_TYPE_EDGE_FALLING>; interrupt-names = "qwerty"; status = "okay"; }; ...}; I have read this STM documment about interrupts overview, but I'm not clear about how it's supose to be the handling of this interruption from the userSpace. My goal is to make a script that have some "callback()" function to be able to do something with the interrupt signal.I appreciate any information that can make me understand how the workflow should be, or even if you could give me an example about it.Juan P.@Tesla DeLorean @waclawek.jan @TDK @Peter BENSCH @Eleon BORLINI @Imen.D @Brian TIDAL @PatrickF @Olivier G
We are using STM32MP135-DK and we connect to SPI5 an SD Card Adapter.We modified the device tree as follows to use the mmc-spi-slot driver: &spi5{ pinctrl-names = "default", "sleep"; pinctrl-0 = <&spi5_pins_mx>; pinctrl-1 = <&spi5_sleep_pins_mx>; status = "okay"; /* USER CODE BEGIN spi5 */ spisdcard@0 { compatible = "mmc-spi-slot"; reg = <0>; gpios = <&gpiog 1 (GPIO_ACTIVE_LOW | GPIO_PULL_UP)>; voltage-ranges = <3300 3300>; spi-max-frequency = <2400000>; status = "okay"; disable-wp; }; &nb
To workaround path issue with depmod -a /lib/modules# cp -r 5.15.67 5.15.67-g8d25ddcc4f56# /sbin/depmod -adepmod: ERROR: could not open directory /lib/modules/5.15.67-g8d25ddcc4f56: No such file or directorydepmod: FATAL: could not search modules: No such file or directorythis error is because after compiling some kernel changes, mostly device tree changes, Makefile populates lib/modules/5.15.67 however, how to fix where the modules path is, or what the kernel version is ?# uname -r5.15.67-g8d25ddcc4f56
Dear All, en.flash-stm32mp1-openstlinux-6-1-yocto-mickledore-mp1-v23-06-21.tar.gzan en.FLASH-stm32mp1-openstlinux-5.15-yocto-kirkstone-mp1-v22.11.23.tar.gz look corrupted. I can't extract them. I need your help. Danke you, Sir.
Hi all,I have been trying to boot a custom ST32MP15 board, I get through TF-A and U-boot, but boot fails while Starting Kernel.I had some issues with external 24MHz crystal so I am using HSI to provide CPU clocks. On TF-A and U-Boot device trees I was able to find the nodes to fix that and used STM32CubeMX to help me calculate the PLL multipliers. Although, couldn't find similar nodes for kernel device tree.After U-Boot is loaded I can see that MPU and peripheral clocks have the values I set up (U-boot has some troubles reading values from STPMIC1 but TF-A was able to configure the PMIC)I am able to load zImage and dtb from SD Card and call boot command, but after that the device would hang on "Starting Kernel...". I followed the tutorial at https://wiki.st.com/stm32mpu/wiki/Dmesg_and_Linux_kernel_log# to enable earlyprintk, trying to boot after that gets past the "Starting Kernel..." message, but kernel just prints some garbled lines and reboots the device. I have copied the mess
u-boot yocto build fails with this messagetest -e arch/arm/dts/stm32mp157c-mydemo-mx.dtb || ( \| echo >&2; \| echo >&2 "Device Tree Source (arch/arm/dts/stm32mp157c-mydemo-mx.dtb) is not correctly specified."; \| echo >&2 "Please define 'CONFIG_DEFAULT_DEVICE_TREE'"; \| echo >&2 "or build with 'DEVICE_TREE=<device_tree>' argument"; \arch/arm/dts/stm32mp157c-mydemo-mx.dts existsarch/arm/dts/stm32mp157c-mydemo-mx-u-boot.dtsi existsstm32mp157c-mydemo.dtb has been added to the MakefileAny ideas?
Hello,I am trying to read an analog value of PA4 of the STM32MP157CAC with ADC1 but it does not work.I am using from Emtrion the emSTAMP-Argon as module.I also tried to read the internal temperature sensor instead with ADC2 but that failed as well.I do see AD_START to get cleared after being set; thus the basic ADC is working (at least that is my understanding). Nevertheless, I do not see any change of the ADC_DR register is always 0x8000.The ADC_PCSEL is modified to enable the channels and ADC_SQR1 is also configure.The ADC of STM32s are not new to me as I did use STM32F4, STM32F3 and STM32WB for some time. If I am switching from 16 bit to 12 bit e.g. I do see ADC_DR shifting. Any suggestions or hints would be great as I do not know what I am doing wrong.I might be wrong but in engineering mode, there shouldn't be any issues with TF-A etc. possible or?Thank youSteffen
I am making a board using STM32mp151a. I want to write Linux OS to the SD card. However, the command on the web page below only has STM32mp153 and STM32mp157 for the evaluation board. https://wiki.st.com/stm32mpu-ecosystem-v5/wiki/How_to_create_your_own_machine ./create_sdcard_from_flashlayout.sh ../flashlayout_<built-image>/<FlashLayout file> What do I have to do?
We are migrating from the Kirkstone eco-system to the Mickledore eco-system. We found that the extlinux.conf file that gets created is identical in the two builds. However, u-boot now forces a menu to the user, even though we have CONFIG_BOOTDELAY=-2 and only one boot label in the file. This seems to be an intentional change in u-boot, see commit c23bb03. Although other documentation seems to indicate that a menu with only one label boots that label immediately.The produced extlinux.conf is attached, renamed to extlinux.conf.log to allow upload.The u-boot output is: In: serial Out: serial Err: serial Retrieving file: /extlinux/extlinux.conf Retrieving file: /splash_landscape.bmp ** File not found /splash_landscape.bmp ** There is no valid bmp file at the given address 1: OpenSTLinux Enter choice: We find that if we modify the extlinux recipe to have `UBOOT_EXTLINUX_SPLASH = ""` that this causes the MENU token in extlinx.conf to be removed. Without the MENU
A particular STM32MP1 based board uses the USB OTG interface configured to the cortex A7. Is there any guidance how to configure the kernel so that it will use the USB device port with a gadget driver, so that the host sees it as keyboard, serial port, mouse, audio interface ? For example, as pictured, a snippet of usb otg in the .dts ( high speed). It uses dwc2 as driver for OTG function, How to assign the device descriptor to the otg/usb interface ? Full speed would be fine as well. In the kernel sources there is "legacy" and "function" code.
Hi :)Working on the EVA32MP257F, I am trying to generate the openstlinux for it, however, I see no target MACHINE in the yocto project for it. I guess that a private repo is required for this?Also, can I apply the RT patch provided for the MP1 on top of it?Adrian
Hello,I need to synchronize the time between the A7 and M4 cores. While I found some resources on the ST Community that accessing the STGEN timer from the A7 side.the article is belowhttps://community.st.com/t5/embedded-software-mpus/how-can-i-get-access-to-m4-timers-from-a7-linux-is-it-possible/m-p/151769So, I tried this on A7 too.for reading/writing peripheral registers from the A7 core is outlined in STWiki article. https://wiki.st.com/stm32mpu/wiki/How_to_read_or_write_peripheral_registersI think accessing /dev/mem offer a same capability. I tried and successfully employed this approach for UART4. However, I failed when it comes to reading the STGENR using this method.Could someone provide guidance on how I might access the STGEN timer from the A7 core? If there's a more efficient or recommended way to synchronize the time between the A7 and M4 cores, that information would be invaluable as well.Thank you for your assistance.
Hi everyone,I'm using STM32g491 and rs485 modules. I'm trying to transmit data from uart1 with DMA when I receive data. There is a dilemma it transmit data but also does not? When I try to send a buffer which defined on stm's code, it transmits but when I try to transmit data come through GUI, it does not transmit. There is no problem about data coming from GUI. I can see it when I debug. Here is my code and the data coming from gui on debug if(uartCh.BLUT == 1) { HAL_GPIO_WritePin(BLUT _EN_GPIO_Port, BLUT _EN_Pin, 1); uart1_return = HAL_UART_Transmit_DMA(&huart1, guiBLUT_gelen, 11); uartCh.BLUT = 0; } if(uartCh.GNS == 1) { HAL_GPIO_WritePin(GNS_EN_GPIO_Port, GNS_EN_Pin, 1); uart1_return = HAL_UART_Transmit_DMA(&huart2, guiGNS_gelen, 21); uartCh.GNS = 0; }
Hi,I am trying to set the rootfs read-only. With other yocto based environments I usually setEXTRA_IMAGE_FEATURES ?= "read-only-rootfs"in local.conf. Using OpenSTLinux this results in a build error: ERROR: st-image-userfs-1.0-r0 do_rootfs: Unable to remove packages. Command '['/home/jenkins/Development/SW_APP/nexio_tmp/ost-5.15/build/tmp-glibc/work/stm32mp1_nexio-ostl-linux-gnueabi/st-image-userfs/1.0-r0/recipe-sysroot-native/usr/bin/dpkg', '--admindir=/home/jenkins/Development/SW_APP/nexio_tmp/ost-5.15/build/tmp-glibc/work/stm32mp1_nexio-ostl-linux-gnueabi/st-image-userfs/1.0-r0/rootfs/var/lib/dpkg', '--instdir=/home/jenkins/Development/SW_APP/nexio_tmp/ost-5.15/build/tmp-glibc/work/stm32mp1_nexio-ostl-linux-gnueabi/st-image-userfs/1.0-r0/rootfs', '-P', '--force-depends', 'update-alternatives-opkg']' returned 2: dpkg: warning: 'ldconfig' not found in PATH or not executable dpkg: error: 1 expected program not found in PATH or not executable Note: root's PATH should usually co
I'm trying to build an SDK (Yocto Langdale) for core-image-minimal:bitbake -c populate_sdk core-image-minimal and get this error: nothing provides libpython2.7.so.1.0()(64bit) needed by nativesdk-gcc-arm-none-eabi-10-r0.x86_64_nativesdk Full error:ERROR: core-image-minimal-1.0-r0 do_populate_sdk: Could not invoke dnf. Command '/home/giz/projects/Stm32Prototype/yocto-project-langdale/build/tmp/work/stm32mp1-poky-linux-gnueabi/core-image-minimal/1.0-r0/recipe-sysroot-native/usr/bin/dnf -v --rpmverbosity=info -y -c /home/giz/projects/Stm32Prototype/yocto-project-langdale/build/tmp/work/stm32mp1-poky-linux-gnueabi/core-image-minimal/1.0-r0/sdk/image/etc/dnf/dnf.conf --setopt=reposdir=/home/giz/projects/Stm32Prototype/yocto-project-langdale/build/tmp/work/stm32mp1-poky-linux-gnueabi/core-image-minimal/1.0-r0/sdk/image/etc/yum.repos.d --installroot=/home/giz/projects/Stm32Prototype/yocto-project-langdale/build/tmp/work/stm32mp1-poky-linux-gnueabi/core-image-minimal/1.0-r0/sdk/image
STM32MPU ecosystem minor release V4.1.1 is online ! Dear STM32 MPU Community, We are please to announce that New MP1 Ecosystem minor release V4.1.1 is available. It contains : Update to Yocto Project 4.0.11 (2022-04.11-kirkstone) Update to Linux Kernel 5.15.118 BSP maturation with an amount of 49 fixes delivered BSP components (TF-A, u-Boot, OP-TEE and Linux Kernel) Please check out the STM32_MPU_ecosystem_release_note_-_v4.1.1 on STM32 Wiki Ecosystem-V4 to get the complete description and the new download links. This minor release integrates several fixes in (TF-A, u-Boot, OP-TEE and Linux Kernel)and all the updates included by switching from Linux Kernel 5.15.67 to 5.15.118. We recommend to everyone working on Ecosystem v4.1.0 to switch on this minor release. Of course, the current wiki V5 and the previous ecosystems are still available : Ecosystem-V5-Wiki Ecosystem-V3-Wiki Ec
flashlayout tsv file not generated using meta-st-stm32mp-addons but images created.What do I need to add to generate the Flashlayout for sdcard?
There are many questions around related to the DCMI and embedded syncs on various forums. ST's documentation is not very helpful in this regard and I could not find any meaningful solutions on the web. I had a similar issue using STM32MP157 - unfortunately discrete syncs for the DCMI are not available on our CPU board so I had to make it work no matter what. The good news is it does work and works well but there are some things to take note of. Here is a rundown on what needs to be done.The reference manual is a bit terse on what needs to be done but gives a hint or two.DCMI_ESUR=0xFFFFFFFF; DCMI_ESCR=0xB69D80AB;If you are happy with capturing just one field - set ESCR either to odd or even field codes as you desire. If you need both fields then set DCMI_ESUR=0xB0B0B0B0 so it ignores the field bit.Enable DCMI interrupts for HSync and VSync - you can ignore the rest. In your interrupt count the 0x18 and 0x10 states of MIR before you clear the flags. The sum of both gives you the nu
I am trying to port a FW which was running on STM32H7 controller to F4 core of STM32MP157.I get this errorsCM4.elf section `.text' will not fit in region `SRAM1_text' CM4 C/C++ Problemmake: *** [makefile:68: CM4.elf] Error 1 CM4 C/C++ Problemregion `SRAM1_text' overflowed by 1073768 bytes CM4 C/C++ Problemregion SRAM1_text overflowed with text and data CM4 C/C++ Problemsection .data VMA [0000000010020000,0000000010020327] overlaps section .text VMA [0000000010000000,0000000010124003] CM4 C/C++ ProblemHow to over come this problem, is it possible to change the allocated RAM for M4 code?
Hi,I am using the stm32mp157f-dk2, I'm using the core-image-minimal image and i'm trying to create a custom image. However, I would like to increase the rootfs size in order to add modules in the root filesystem once the kernel starts. I tried modifying the file /meta-st/meta-st-stm32mp/conf/machine/include/st-nachine-common-stm32mp.incIMAGE_ROOTFS_MAXSIZE ?= "2017660"Then, i tried to modify the /meta-st/meta-st-stm32mp/recipes-devtools/sdcard-raw-tools/sdcard-raw-tools/create_sdcard_from_flashlayout.sh file, just as the post DEFAULT_RAW_SIZE=4096 DEFAULT_ROOTFS_PARTITION_SIZE=2097152I also modified the local.conf fileIMAGE_ROOTFS_EXTRA_SPACE = "1008830"The problem i'm expecting is, the image built, but I see this message at the end of the build processWARNING: quoia-custom-1.0-r0 do_create_flashlayout_config: >>> Cannot generate FlashLayout_sdcard_stm32mp157f-dk2-trusted.tsv file: the st-image-userfs-poky-stm32mp1.ext4 binary size (1167261696) for userfs
I have STM32MP157A-DK1 boardI am using distribution package.I want to add c++ library based on cmake to my distribution.I crated the recipe using commanddevtool add --no-same-dir ../pass/to/mylibThen build usingbitbake mylibThe build is done but I get this error bu do_package_qa"do_package_qa: QA Issue: -dev package mylib-dev contains non-symlink .so '/usr/lib/mylib.so' [dev-elf]"what is the reason for such error and how to overcome it?ThanksHossam
Doing a build using yocto mickledore it fails with this error message.CMake Error at host/xtest/CMakeLists.txt:3 (include): include could not find requested file: ... /tmp-glibc/work/osd32mp1_brk-ostl-linux-gnueabi/optee-test/3.19.0+gitAUTOINC+ab9863cc18-r0/recipe-sysroot/usr/include/optee/export-user_ta/host_include/conf.cmakeThe kernel is built correctly but it fails on building optee-test.It seems that the directories and files in export-user_ta are not auto-generated. What bitbake configuration makes this happen?
I tried to add development tools to my openSTLinux using apt-get install, but couldn't find gcc.so I tried to create a distribution including gcc, make, cmake and git.When I search for available recipes for gcc usingfind ../layers/ -iname "*gcc*.bb"I get the following list of gcc tools../layers/meta-st/meta-st-stm32mp/recipes-devtools/gcc-arm-none-eabi/gcc-arm-none-eabi-native_11.bb../layers/meta-st/meta-st-stm32mp/recipes-devtools/gcc-arm-none-eabi/nativesdk-gcc-arm-none-eabi_11.bb../layers/openembedded-core/meta/recipes-devtools/gcc/gcc-cross-canadian_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/libgcc-initial_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/gcc-cross_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/gcc-crosssdk_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/gcc-sanitizers_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/gcc_12.2.bb../layers/openembedded-core/meta/recipes-devtools/gcc/libgcc_12.2.bb../
We are using STM32MP13 with Linux for the A7 and using HAL (no OS) for the M4 Coprocessor. In Linux we configured in device tree the I2S (codec and sound are also configured in device tree). &i2s1 { clocks = <&rcc SPI1>, <&rcc SPI1_K>, <&rcc PLL3_Q>, <&rcc PLL3_R>; clock-names = "pclk", "i2sclk", "x8k", "x11k"; pinctrl-names = "default", "sleep"; pinctrl-0 = <&i2s1_pins_audio &i2s1_pins_z_audio >; pinctrl-1 = <&i2s1_sleep_pins_audio &i2s1_sleep_pins_z_audio >; status = "okay"; dmas = <&dmamux1 38 0x400 0x01>; dma-names = "tx"; i2s1_port: port { i2s1_endpoint: endpoint { remote-endpoint = <&codec_corscience_endpoint>; format = "i2s"; }; }; }; At M4 this was done using CubeIDE for the SPI3 and SPI4. These ports are working but as soon we play a sound using aplay the SPI3/4 hangs.
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.