STM32 MPUs Embedded software and solutions

Hello,I'm trying to send a CAN frame with can-utils library on STM32MP157F-EVAL board.I read the following sites [1] and [2] and tried the commands:root@stm32mp1:~# ip link set can0 up type can bitrate 1000000 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 root@stm32mp1:~# cansend can0 055#11223344 write: No buffer space availableno messages are sent and after some trys I get the error: No buffer space available.root@stm32mp1:~# candump can0candump also receives no frames. The opposite station sends can messages and they are arriving the stm32 eval board. I have measu

Hello ST Teams!Question, again! I'd like to know if the image "st-image-core" or "st-image-weston" are compatible with the "read-only-rootfs" from yocto append? (thud or dunfell) Well I've already tried on the st-image-core and bitbake compilation failed, saying something about user-fs.Any hints would be welcome!Have a nice confined day!

Hi, I'm trying to make a headless system using the STM32MP153 and I would like to disable all graphical layers in the standard distribution.I'm using openstlinux-5.4-dunfell-mp1-20-06-24 (ECO V2 I believe)I have made my-own distro layer and was able to bitbake st-image-core( have added to my own_distro.conf file:DISTRO_FEATURES_remove = " wayland "DISTRO_FEATURES_remove = " x11 ")Next step is to remove the graphical layers starting with qt5If I do a bitbake-layers show-layers I don't want to find meta-qt5 (or meta-gnome, meta-xfce)Question: how do I do that?BTW I'm using the STM32MP157C evaluation board to get my headless distro up and running.

I am working with the STM32MP157 creating an SD card image using OpenSTLinux. I would like to use RAUC to perform updates but to do so I need to create a second rootfs partition. One active partition and the other for updates. I couldn't find in the wiki how to modify the partitions and general flash layout. Does anyone have any recommendations on where to start looking or what to modify in OpenSTLinux to modify the generated partition layout. Also any advice with RAUC is appreciated. https://github.com/rauc/rauc

What I want to achieve is starting the DFU ROM bootloader (like on a MCU) from a running Linux (maybe via a dedicated kernel module) or, if this helps, via re-configuring u-boot and doing a software reboot cycle.

Hello everybody,Here is my problem. I am using the buildroot distro instead of OpenSTLinux and want toconnect my STM32 via ethernet using SSH. I use MobaXterm as remote control software.P.s After a system boot I get the following response although the ethernet LEDs are not blinkingbuildroot login: [ 4.237860] stm32-dwmac 5800a000.ethernet eth0: Link is Up - 1Gbps/Full - flow control rx/tx [ 4.251707] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes readyAfter ifconfig -a I get the following response:eth0 Link encap:Ethernet HWaddr 00:80:E1:42:6B:16   inet addr:192.168.8.45 Bcast:192.168.8.255   Mask:255.255.255.0   inet6 addr: fe80::280:e1ff:fe42:6b16/64 Scope:Link   UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1   RX packets:107 errors:0 dropped:0 overruns:0 frame:0   TX packets:11 errors:0 dropped:0 overruns:0 carrier:0   collisions:0 txqueuelen:1000   RX bytes:10299 (10.0 KiB) TX bytes:866 (866.0 B)   Interrupt:60 Base addres

..

When building a distro for our custom board with a STM32MP153AAC, u-boot-stm32mp fails to build when UBOOT_CONFIG is set to "basic". We would like to do a ddr3 test, this is the reason for the config.Here is the relevant section from conf/machine/stm32mp1-raichu-v2.conf file########################################################################### # # User machine customization sections # ###########################################################################   # Boot Scheme # ========================================================================= BOOTSCHEME_LABELS += "trusted" # WORKAROUND to generate U-BOOT SPL for DDR Tuning tools usage UBOOT_CONFIG += "basic"It seems that the dsi block is defined in arch/arm/dts/stm32mp157.dtsi, but we have a stm32mp153, and the device tree generated by CubeMX only includes stm32mp153.dtsi. So obviously, this isn't going to work:https://github.com/STMicroelectronics/meta-st-stm32mp/blob/dunfell/recipes-bsp/u-boot/u-boot-stm32mp/0004-ARM

Hello,I am trying to figure out how to drive the USB traffic (any of the available ports acting as device) from/to the M4 core. I can see that the SoC interconnection allows for the data to flow between the peripheral and the M4 core but I cannot find a way to achieve this form of operation.The runtime operation of the USB-OTG peripheral is assigned to the A7 core but I assume there is a way to allow the endpoint data to be handled by the M4 core in some way (maybe some DMA-based scheme?). I would really appreciate if somebody can cast some light over this issue to get started evaluating this use case.The target application requires real-time operation for the data being transmitted over USB and this feature being handled by the M4 core is quasi-mandatory. I do not see a clear way of achieving the real-time performance by transmitting all the data over the IPCC interface, even using indirect more, as this would require a very high frequency polling of this memory region by the A7 core.

Hello ST Teams!Question again!I'd like to set up an access point on my st_image_core, SDK1.2i've added hotspot in layer append, and i can configure hostapd.confI've followed the howto :https://wiki.st.com/stm32mpu/wiki/How_to_configure_a_wlan_interface_on_hotspot_modeBut it seems that somethings goes wrong. Even if i try with your st-hotspot-wifi-service.sh.My question is:Do i need to add a dhcp server like udhcp server (udhcpd)?I've looked on the starter kit for DK2, and i cannot find any of these! What package from Starter kit I need to import to make things work?What am i missing?Seeya!

I've setup the distribution package, i'm able to build the default image, and have created a custom image that includes some additional recipes.Now I want to continue with creating my custom machine. I've created a custom device tree with cubeMX and placed it in meta-st-stm32mp-addons/mx/<projectname>. I followed the steps https://wiki.st.com/stm32mpu/wiki/How_to_create_your_own_machineMy .conf looks like this:#@TYPE: Machine #@NAME: stm32mp1-mx #@DESCRIPTION: Configuration for STM32CubeMX generated project #@NEEDED_BSPLAYERS: layers/meta-openembedded/meta-oe layers/meta-openembedded/meta-python   ########################################################################### # # Default machine configuration sections # ###########################################################################   include conf/machine/include/st-machine-common-stm32mp.inc include conf/machine/include/st-machine-providers-stm32mp.inc   # Define specific common machine name MACHINEOVERRID

Any plans to upstream support for STM32H757? Currently kernel tree seems to have stm32h743.

Found while building the openstlinux-weston DISTRO st-image-weston image.Missing ${S}/ in front of stm32wrapper4dbg. Line 13 should be:install -D -m 0755 ${S}/stm32wrapper4dbg -t ${D}${bindir}

Hi All, Please , I`m pretty new here but so far using standard layer it works correctly hover when trying to use the QT / Distro openstlinux-eglfs layers meta-st I`m getting error related to make-mod-scripts.ERROR: Task (/local/STM32MP15-Ecosystem-v2.0.0/Developer-Package/stm32mp1-openstlinux-5.4-dunfell-mp1-20-06-24/layers/openembedded-core/meta/recipes-kernel/make-mod-scripts/make-mod-scripts_1.0.bb:do_configure) failed with exit code '1'Can any good soul help-me on that ? Anything I can do here ?Thank You so Much !

I've assigned the UART8 device in the devicetree of the linux kernel (OpenSTLinux):&pinctrl { // ... uart8_pins: uart8-0 { pins1 { pinmux = <STM32_PINMUX('E', 0, AF8)>; /* UART8_RX */ bias-disable; }; pins2 { pinmux = <STM32_PINMUX('E', 1, AF8)>; /* UART8_TX */ bias-disable; drive-push-pull; slew-rate = <0>; }; };   uart8_sleep_pins: uart8-sleep-0 { pins { pinmux = <STM32_PINMUX('E', 0, ANALOG)>, /* UART8_RX */ <STM32_PINMUX('E', 1, ANALOG)>; /* UART8_TX */ }; }; // ... } // ... &uart8 { pinctrl-names = "default", "sleep"; pinctrl-0 = <&uart8_pins>; pinctrl-1 = <&uart8_sleep_pins>; /delete-property/dmas; /delete-property/dma-names; status = "okay"; };The uart8 (0x40019000) is detected upon boot and the device /dev/STM1 is created:[ 1.196012] stm32-usart 40010000.serial: interrupt mode for rx (no dma) [ 1.196036] stm32-usart 40010000.serial: interrupt mod

I'm trying to build the standard st-image-weston image following this guide:https://wiki.st.com/stm32mpu/wiki/STM32MP1_Distribution_PackageAfter starting the build with bitbake st-image-westonWhen it reaches around the 68% of the process I get the following error:ERROR: tf-a-stm32mp-2.2.r1-r0 do_compile: oe_runmake failed ERROR: tf-a-stm32mp-2.2.r1-r0 do_compile: Execution of '/home/osboxes/openstlinux-5.4-dunfell-mp1-20-06-24/build-openstlinuxweston-stm32mp1/tmp-glibc/work/stm32mp1-ostl-linux-gnueabi/tf-a-stm32mp/2.2.r1-r0/temp/run.do_compile.21431' failed with exit code 1: make: Entering directory '/home/osboxes/openstlinux-5.4-dunfell-mp1-20-06-24/build-openstlinuxweston-stm32mp1/tmp-glibc/work-shared/stm32mp1/tfa-source' make: *** No targets specified and no makefile found. Stop. make: Leaving directory '/home/osboxes/openstlinux-5.4-dunfell-mp1-20-06-24/build-openstlinuxweston-stm32mp1/tmp-glibc/work-shared/stm32mp1/tfa-source' WARNING: exit code 1 from a shell command.   ERR

Hi all, I am trying to switch to using FIT images and I am looking at the example .its file here:https://github.com/STMicroelectronics/u-boot/blob/0e6522cbd1bfccd7281acdf8ab5ebbd37d29eab1/board/st/stm32mp1/fit_kernel_dtb.itsI am trying to understand why 0xC0008000 is used for the load address/entry point. If I understand correctly:By default, U-boot will load the FIT image to address 0xC2000000Then it will copy the kernel image to load address 0xC0008000, and jump to itThe kernel image needs to decompress itself to final address 0xC0008000; however since the entry point of the zImage is 0xC0008000, it will need to relocate itself first to some address which is higher than (0xC0008000 + uncompressed kernel length)And then finally decompress to final location 0xC0008000Wouldn't it be possible to skip the relocation in step 3 by choosing a different address for the load address/entry point in the FIT image?Thanks,Guillermo

Hi all,The file include/configs/stm32mp1.h in U-boot defines CONFIG_LOADADDR to 0xc2000000. I understand this is the address in RAM where u-boot puts the uImage loaded from persistent storage (SD, NAND, etc).At the same time, according to [2], the kernel uImage is built with LOADADDR set to 0xc2000040. I understand this is the address where the kernel is relocated.Wouldn't this be a problem, as the address of the loaded uImage and the address used for relocation seem to overlap?Also, I think that common convention for ARM is to use <base RAM address> + 0x8000 for relocating the kernel; what is the reason to use 0xc2000040 here? [1]: https://github.com/STMicroelectronics/u-boot/blob/v2020.01-stm32mp/include/configs/stm32mp1.h [2]: https://github.com/STMicroelectronics/meta-st-stm32mp/blob/dunfell/recipes-kernel/linux/linux-stm32mp/README.HOW_TO.txt

..

Hello.I created a custom device tree in CubMH and added its instructions "How to create your own machine" (https://wiki.st.com/stm32mpu/wiki/How_to_create_your_own_machine)I wanted to run I2C5 in production mode. Running my code stops CM4 and CA9, which then leads to a reboot. At the same time, there are no signals on the pins.Running the example I2C_TwoBoards_ComIT (STM32CubeMP1 / Projects / STM32MP157C-DK2 / Examples / I2C / I2C_TwoBoards_ComIT) CM4 loops at the following code fragment:while (HAL_I2C_GetState (& hi2c5)! = HAL_I2C_STATE_READY)  {  }In this case, CA9 is working normally.In engineering mode, my code and example code work fine, signals appear on pins.I2c5 setup:m4_i2c5_pins_mx: m4_i2c5_mx-0 { pins { pinmux = <STM32_PINMUX('A', 11, RSVD)>, /* I2C5_SCL */ <STM32_PINMUX('A', 12, RSVD)>; /* I2C5_SDA */ };};&m4_i2c5{ pinctrl-names = "default"; pinctrl-0 = <&m4_i2c5_pins_mx>; status = "okay"; /* USER CODE BEGIN m4_i2c5 */

Hey Team, I'm trying to get through a board bring up with the STM32MP157 silicon ver 2001 marked with a Z. We are trying to boot the system from the on board eMMC (Insignis Technology Corp P/N: NSEC53K008-IT 8GB eMMC ) and are having issues. We have configured the boot pins correctly and also confirmed emmc hardware connections are correct by successfully reading and writing the emmc from linux while booting from the SD card. There seems to be a lack of documentation on how to configure the eMMC and partition it. Is there any specific partition scheme that needs to be followed for the MP1? And link to an app note that documents how to configure the eMMC via STM32Programmer or UBoot for booting would be extremely helpful.Jason

Hi,I am currently using the STM32MP1-DK1 board in engineering modeWhen I am debugging the program and stepping instruction by instruction in IAR, the debugging process is lost at a time during the execution.I can execute the program and and I can insert breakpoint wherever I want. But, before the instruction msr msp, r0 in the function vPortStartFirstTask in the file portasm.s has been executing, if we stop the program execution by inserting a breakpoint, the debugging process is lost with the error:A vector table read fault has occurred. The processor has escalated a configurable-priority exception to HardFault.   An MPU or Execute Never (XN) default memory map access violation has occurred on an instruction fetch (CFSR.IACCVIOL, MMFAR).   A precise data access error has occurred (CFSR.PRECISERR, BFAR) at data address 0x34020000. Exception occured at PC = 0x1000ced8, LR = 0x10006e9d

I'm trying to get the OpenSTLinux build environment up & running for the 1st time, following the instructions on the wiki step-by-step.So far I have:Built a Ubuntu 18 Virtual machine in Virtualbox - (18.04.5LTS, 4 cores, 8GbRAM, 500Gb HDD)Followed all the Starter package / Distribution package install & setup stepsProgrammed the DK2 and EVB with the Starter Packages (tip: you need Virtualbox Extensions enabling USB2.0 & add STMicroelectronics* as a USB filter)But now I am trying to actually build the whole project using BitBake and it keeps failing.I have not modified the OpenSTLinux at all, I have set it all up using the directory structure suggested in the Wiki, initialised the environment per the instructions, and invoked:bitbake -v st-image-weston 2>&1 | tee build.logAnd the process appears to have failed, which is rather dissappointing, although I can't tell if the error is a "real" error or just something that doesn't really matter.Here's errors from the log:ER

In our custom board we are using I2C1 to control and configure STPMIC1. I'd like to use I2C1 in TF-A so PMIC can be configured to power DDR during boot. I2C1 is not supported by default in TF-A, but looks like there is a way to do it as described in https://wiki.st.com/stm32mpu/wiki/PMIC_hardware_components#TF-A I followed the procedure described in the above link, but still when I include the reference to &i2c1 in my stm32mp153c-<project>.dts it throws the following build error| Building stm32mp1| Error: fdts/stm32mp153c-<project name>.dts:355.1-6 Label or path i2c1 not found| FATAL ERROR: Syntax error parsing input treeAm I missing something?Thanks,Nikhil

I'm trying to boot an STM32MP157 using the v5.4-stm32mp branch of ST's Linux kernel, and it doesn't appear that peripherals under the simple-bus node in the DTS file are getting probed. I copied the stm32mp157d-dk1.dts file as a starting point, but since my board doesn't have a PMIC, I commented out the PMIC element and instead pass a dummy regulator to the SDMMC peripherals. With no other changes, I got absolutely no output from the console. When I enabled earlyprintk support, I got output that stopped at "printk: console [tty0] enabled." I thought, "hmm, that's weird. Why isn't it loading my UART driver?" I edited the boot arguments to pass console=ttySTM0,115200 to the kernel (obviously it should be getting picked up from the device tree file), and this fixed the console printing issue, but I now have the attached log file.As you can see, the device tree is getting loaded, and the memory pools / CPU settings are handled. However, none of the STM32 peripheral drivers are getting load