Ask questions and find answers on STM32 microprocessors and dedicated embedded software solutions, such as Yocto SDK, Qt, and OP-TEE.
Most recent activity
Dear All,I would like to use MYD-LD25X development board, which uses the STM32MP257 I plan to run FreeRTOS on the M33 core and Linux on the A35 cores.Do you know if there are documents that I can follow to run FreeRTOS on M33 core and Linux on the A35 core? Thank you very much.
Heyho, until now I have only used the STM32 MCUs, from L01x, to G4, up to H7 (single core), always bare-metal. We're now thinking about scaling up to some MPU, for more processing power, and also for some faster interfaces (mainly gigabit ethernet). From my first investigations about the MPUs, I'm not sure if we can "afford" this on the personnel side, due to the complexity and very probably the necessity to use an OS (real-time is a must, at least "real" enough for multi-channel low latency audio). We have an overworked software department, and no "embedded SW" people, and I'm more of a hardware designer who can do some basic C stuff. So I'd like to know: which OSs are mostly used on the MPUs, or which are you using how's your experience with these what's the learning curve for newbees anybody working bare-metal with the MPUs, and how's that Thanks in advance!
I have a STM32MP257F custom board with a MT53E1G32D2FW-046 WT:C (4GB) memory.But when I'm booting the TF-A, then I get this message. Notice that this processor passes the DDR data(stm32mp_ddr_test_data_bus(void)) and DDR rw(stm32mp_ddr_test_rw_access(void)) test! So I don't think that the assembly was bad.ERROR: DDR addr bus test: can't access memory @ 0x100000000 BACKTRACE: START: stm32mp2_ddr_setupThis means that my CPU cannot access the address 0x100000000. This limit is exactly 4GB.If we looking at the code. That means that the offset variable must be 0x80000000, which is 2GB. The function stm32mp_ddr_test_addr_bus(size_t size) is beging called with size = 4294967296 = 4GB.uintptr_t stm32mp_ddr_test_addr_bus(size_t size) { size_t addressmask = size - 1U; size_t offset; size_t testoffset = 0U; /* Write the default pattern at each of the power-of-two offsets. */ for (offset = sizeof(u_register_t); (offset & addressmask) != 0
I am working on STM32MP157F-EV1 and would like your assistance in enabling SPI-NAND support within the OpenSTLinux (Yocto) environment.Overview:Development Board: STM32MP157F-EV1Storage Target: SPI-NAND (Serial NAND)Flash Part Number: GD5F4GM8UEYJGR (GigaDevice, 4 Gbit)Operating System: OpenSTLinux / Yocto-based BSPBoot Objective: Generate a complete bootable image for SPI-NAND (TF-A, U-Boot, Kernel, and RootFS)Currently, the official OpenSTLinux images support eMMC, SD card, NOR, and parallel NAND. However, our design requires SPI-NAND as the primary non-volatile memory. We would like to build and flash the image to SPI-NAND and eventually boot directly from it.Support Requested:Image Generation:How to configure Yocto (OpenSTLinux) to build an image for SPI-NAND on STM32MP157F-EV1?
Hello,I am working on the stm32mp257f-ev1 board and attempting to switch from the default USART2 early crash console output to USART6. I've successfully switch TF-A and U-BOOT console to USART6 but my OP-TEE console continuously outputs on USART2.I've made sure to change the serial0 alias from USART2 to USART6 in the OP-TEE device tree (as I've successfully done in the TF-A device tree) and I've also followed the directions to change the header file stm32mp2_def.h For USART6, I've changed the early crash console in stm32mp2_def.h to the following:#define STM32MP_DEBUG_USART_BASE USART6_BASE#define DEBUG_UART_TX_GPIO_BANK_ADDRESS GPIOF_BASE#define DEBUG_UART_TX_GPIO_BANK_CLK_REG RCC_GPIOFCFGR#define DEBUG_UART_TX_GPIO_BANK_CLK_EN RCC_GPIOxCFGR_GPIOxEN#define DEBUG_UART_TX_GPIO_PORT
I am porting Linux to a board that uses the STM32MP255F.I used the BSP from the STM32MP257F-DK board.For now, I was able to get the LCD panel to display, but the weston-graphical-session did not start.When I added the following line to /etc/default/weston, the weston-graphical-session started:OPTARGS=--use-pixmanThe demo menu appears, and I can launch the demos.However, the GPU demos do not run.As I understand it, the OPTARGS=--use-pixman setting is meant for when there is no GPU.Since the GPU demos also do not run, it seems that the GPU is not operating.Under what conditions does the GPU fail to operate?For reference, the Device Tree gpu section is configured as follows:&gpu {contiguous-area = <&gpu_reserved>;status = "okay";};Another thing I am concerned about is that the initialization around usb3dr does not match the custom board, so I have disabled it.&m33_rproc {mboxes = <&ipcc1 0x100>, <&ipcc1 0x101>, <&ipcc1 2>;mbox-names = "vq0",
In the STM32MP257F-DK Linux BSP, I believe that the initialization of the USB3DR peripheral devices (TCPP03-M20, CC1_EN, CC2_EN) is performed by the M33 core executing the following program:/usr/local/Cube-M33-examples/STM32MP257F-DK/Demonstrations/USBPD_DRP_UCSI/lib/firmware/USBPD_DRP_UCSI_CM33_NonSecure_stripped.elfWhere can I find the source code of this program?I have created a board using the STM32MP255F with a configuration similar to the STM32MP257F-DK board.The hardware configuration around USB3 is the same, but some pinctrl and I2C bus channels are different.Therefore, I would like to modify and use USBPD_DRP_UCSI_CM33_NonSecure_stripped.elf.
I am trying to build the OpenSTDroid kernel, I first acquired the "STM32MPU Distribution Package for Android", Then as mentioned in https://wiki.st.com/stm32mpu/wiki/How_to_build_kernel_for_Android, I loaded the kernel & Then tried to build the dtb binaries using:build_kernel dtbAnd got the following error:$ build_kernel dtb[1/1]: Generate dtb binariesERROR: Not possible to generate the kernel imageAfter commenting out the &>${redirect_out} part where make is generating the dtb files, I get the following output:$ build_kernel dtb[1/1]: Generate dtb binariesmake[1]: *** No rule to make target 'dtbs'. Stop.make: *** [Makefile:256: __sub-make] Error 2ERROR: Not possible to generate the kernel image
I have used HSEM on STM32H7 very well.Recently,I test the HSEM of STM32MP257 in development mode, by JTAG/SWD,I can lock the semaphore by reading the HSEM_RLRx register via CPU1~3, but after power reset,when I write the HSEM_RLRx register to lock or unlock semaphore, it doesn't work!the security /privilege / CID filtering of HSEM are all off after reset, and the cpu3 could access the HSEM_RLRx register correctly,but after cpu1/2 write the HSEM_RLRx register or HSEM_CnIER,both lock and unlock operations have no effect.pls give me some suggestions to check what I might have done wrong .thanks!
I opened a PR on https://github.com/STMicroelectronics/u-boot.git for a patch that is applicable only downstream, and since I did not receive any response, I contacted Patrice Chotard, who told me to contact you for submitting downstream patches.In addition to U-Boot, I have patches to submit to:https://github.com/STMicroelectronics/linux.githttps://github.com/STMicroelectronics/meta-st-openstlinux.githttps://github.com/STMicroelectronics/meta-st-stm32mpCould you please guide me on how I should proceed?Thanks and regards,Dario
Recently,I test STM32MP2 with STM32MP257F-EV1 board.I want to debug A35 core with STLINKV3 that on the STM32MP257F-EV1 board,does it support ?And which cores are the STLINK-V3EC can support,except cortex M.
Hi everyone, I am trying to bring up the DP83867 with RGMII on the STM32MP131DAG7. However I am getting the following error:[ 18.253959] stm32-dwmac 5800a000.ethernet end0: renamed from eth0 [ 23.695184] stm32-dwmac 5800a000.ethernet end0: Register MEM_TYPE_PAGE_POOL RxQ-0 [ 23.756147] stm32-dwmac 5800a000.ethernet end0: PHY [stmmac-0:00] driver [TI DP83867] (irq=POLL) [ 24.764629] stm32-dwmac 5800a000.ethernet: Failed to reset the dma [ 24.769435] stm32-dwmac 5800a000.ethernet end0: stmmac_hw_setup: DMA engine initialization failed [ 24.808705] stm32-dwmac 5800a000.ethernet end0: __stmmac_open: Hw setup failed I saw this is typically due to the 125MHz clock not being generated. In our schematic, we have the CLK_OUT from phy going to ETH_RGMII_CLK125. I noticed the phy was not generating a clock so using the "mii" tool in U-Boot, I tried to generate the clock by updating the 0x0170 register: Link I was unable to get any clock to generate even though the 25MHz clock
Hi,I’m working on STM32MP257F-DK (A35 + M33) using OpenAMP + RPMsg over IPCC and facing an issue with RPMsg endpoint creation on the M33 side.GoalSimple use case:Linux (A35) sends "START" / "STOP"M33 receives the message via RPMsgM33 toggles an LEDWhat is workingM33 firmware builds and boots correctly using remoteprocResource table is detectedVirtio RPMsg is created on Linux From Linux dmesg:virtio_rpmsg_bus virtio0: rpmsg host is online remoteproc remoteproc0: remote processor m33 is now upSo remoteproc and OpenAMP transport are working.ProblemWith the following M33 code, RPMsg endpoint is not getting created on Linux side:int rpmsg_recv_callback(struct rpmsg_endpoint *ept, void *data, size_t len, uint32_t src, void *priv) { if (len >= 5 && strncmp((char *)data, "START", 5) == 0) { led_run = 1; } else if (len >= 4 && strncmp((char *)data, "STOP", 4) == 0) { led_run = 0; HAL_GPIO_WritePin(GPIOZ, GPIO_PIN_0, GPIO_PIN_RESET); } return 0; } stru
hey currently i am working on stm32mp257dak3 on myir myc-ld257 board in which i want to access RCC of gpiog for i2c3, but i am enable to access it as soon as i try to access without resource manager it shows me "root@myd-ld25x:~# echo start > /sys/class/remoteproc/remoteproc0/state E/TC:1 stm32_iac_itr:192 IAC exceptions [159:128]: 0x10000000 E/TC:1 stm32_iac_itr:197 IAC exception ID: 156 E/TC:1 Panic at /usr/src/debug/optee-os-stm32mp/4.0.0-stm32mp-r1/core/drivers/firewall/stm32_iac.c:212 <stm32_iac_itr> E/TC:1 TEE load address @ 0x82000000 E/TC:1 Call stack: E/TC:1 0x82007ed8 E/TC:1 0x8203bb8c E/TC:1 0x82026308 E/TC:1 0x820396c0 E/TC:1 0x82013c0c E/TC:1 0x820017dc I/TC: Halting CPU 0" this error and my linux gets crash (on board linux not in my laptop) but when i uses resource manager " if (ResMgr_Request(RESMGR_RESOURCE
On the X-LINUX-QT Starter Package Wiki page it states "The X-LINUX-QT expansion package compatible with ecosystem release v6.1.0 provides only the Distribution package, which can be used to generate the Starter and Developer packages".Has anyone generated the QT starter package in v6.1.0?The image produced by "bitbake st-image-qt package-index" launches with the GTK starter package, not the QT starter package.
Hi,I am trying to use USART6 on STM32MP257F-DK development board. I have enabled it in the device tree, recompiled it and I now can see the /dev/ttySTM1 node. The problem is, that sometimes, after reboot, any access to the USART (even changing its baudrate using stty -F /dev/ttySTM1 command) causes SERC exception (with ID 36 which corresponds to USART6) to happen.This does not happen all the time, but most of the time. Once and it doesn't happen after boot it won't happen again until next reboot and the UART works just fine.here my device tree change:&usart6 {pinctrl-names = "default", "idle", "sleep";pinctrl-0 = <&usart6_pins_a>;pinctrl-1 = <&usart6_idle_pins_a>;pinctrl-2 = <&usart6_sleep_pins_a>;/delete-property/dmas;/delete-property/dma-names;status = "okay";}; it seemed first that disabling the DMA helps, but I am not so sure nowusart6_pins_a: usart6-0 {pins1 {pinmux = <STM32_PINMUX('F', 13, AF3)>; /* USART6_TX */bias-disable;drive-push-
Hello ST, We are trying to enable USART6 as a serial console, in the stm32mp257f-dk board, we have builded the SDK using Openstlinux "Distribution package", I had done the below changes in the stm32mp257f-dk.dts device tree present in the kernel-source and u-boot directory : aliases { serial1 = &usart6; }; chosen { stdout-path = "serial1:115200n8"; #address-cells = <2>; #size-cells = <2>; ranges; }; &usart2 { pinctrl-names = "default", "idle", "sleep"; pinctrl-0 = <&usart2_pins_a>; pinctrl-1 = <&usart2_idle_pins_a>; pinctrl-2 = <&usart2_sleep_pins_a>; /delete-property/dmas; /delete-property/dma-names; status = "okay"; }; &usart6 { pinctrl-names = "default", "idle", "sleep"; pinctrl-0 = <&usart6_pins_a>; pinctrl-1 = <&usart6_idle_pins_a>; pinctrl-2 = <&usart6_sleep_pins_a>; uart-has-rtscts; status = "okay"; }; And also modified the stm32mp257f-dk-
I'm working on a custom stm32mp133d board and Yocto (latest release), but something has me quite puzzled.Our hardware looks like this:- Kingston EMMC04G-MT32-01G10 (mmc2)- 4Gb DDR- STM32MP133d- STM32PMIC1most of it is very similar to STM32MP135F-DK. To configure our board, I'm using the stm32cube-mx approach. My issue is, that the generated FIPS image has somehow a misaligned BL33 (U-Boot) image, which results in a failed boot:NOTICE: CPU: STM32MP133D Rev.Y NOTICE: Model: STMicroelectronics custom STM32CubeMX board - openstlinux-6.6-yocto-scarthgap-mpu-v25.06.11 INFO: PMIC version = 0x21 NOTICE: Reset reason (0x35): INFO: Power-on Reset (rst_por) INFO: FCONF: Reading TB_FW firmware configuration file from: 0x2ffe0000 INFO: FCONF: Reading firmware configuration information for: stm32mp_fuse INFO: FCONF: Reading firmware configuration information for: stm32mp_io INFO: Using EMMC INFO: Instance 2 INFO: Boot used partition fsbl1 NOTICE: BL2: v2.10-stm32mp1-r2.0(debug):lts-v2.10.13-dir
Dear All,I am trying to minimize the u-boot and started to remove some features I don't need.I tried to remove whole video & LCD support, but run into issue that stm32mp1.c has a large patch, that applies LCD automatic recognition and extern reference to panel drivers. I am using Kirkstone, but it is also in the patch for Scarthgap.Patch: 0002-ARM-v2021.10-stm32mp-r1-BOARD.patchstm32mp1.c:extern U_BOOT_DRIVER(rm68200_panel);extern U_BOOT_DRIVER(hx8394_panel);(...)Here are feature I'd like to disable, but removing i.e. VIDEO_LCD_xxx features gives me a compile error for missing extern reference to rm68200_panel. CONFIG_CMD_ADC=yCONFIG_STM32_ADC=yCONFIG_FDT_SIMPLEFB=yCONFIG_DM_VIDEO=yCONFIG_BACKLIGHT_GPIO=yCONFIG_VIDEO_LCD_ORISETECH_OTM8009A=yCONFIG_VIDEO_LCD_RAYDIUM_RM68200=yCONFIG_VIDEO_LCD_ROCKTECH_HX8394=yCONFIG_VIDEO_STM32=yCONFIG_VIDEO_STM32_DSI=yI tried different things to remove, but neither works. Removing and/or modifing the whole patch is rather not worth th
Hello,I'm working with a STM32MP257D, which is using the I2C-driver "i2c-stm32f7.c" in the Linux kernel (compatible: st,stm32mp25-i2c). I'm using kernel version v6.6.78. I need to use I2C mangling, specifically I2C_M_NOSTART and I2C_M_STOP, to handle a receiving master scenario.My specific situation: I'm communicating with a device that sends the length of the data payload in the third byte of the transmission. This means I need to:1. Read the first 3 bytes to get the length information2. Continue reading the remaining data without sending a STOP condition after the initial 3-byte read3. Continue reading without sending a START condition when fetching the remaining dataCode Example:struct i2c_msg xfer[2]; xfer[0].addr = addr; xfer[0].flags = I2C_M_RD | I2C_M_STOP; xfer[0].len = 0x03; xfer[0].buf = buff; ret = i2c_transfer(i2c->adapter, xfer[0], 1); xfer[1].addr = addr; xfer[1].flags = I2C_M_RD | I2C_M_NOSTART; xfer[1].len = buff[2]; xfer[1].buf = &buff[3]; ret
Dear experts, why, under the same configuration for heterogeneous core communication, in the generated code, the debug section of the routine's initialization functions of IPPC and OPENAMP can be accessed, but the ones generated by ourselves cannot?The following picture was generated by myself.The following picture was generated by myself.Why does the routine configuration add an extra "remote" and "setting" in the file attributes?But there is none in the project that was generated by myself!Seeking help online, dear experts!
Hello everybody,For my project, I need to display (LVDS output) an HMI that contains external video streams.I plan to use LTDC layers in order to perform plane composition with the hardware block (for example, HMI on plane 1, RTP video on plane 2, and MIPI camera on plane 3).I use GStreamer to test H.264 decoding and to display video using kmssink.My problem is that I can display a video on each plane separately, but not on two or more planes:gst-launch-1.0 videotestsrc ! videoconvert ! kmssink plane-id=$PLANE_ID1 => okgst-launch-1.0 videotestsrc ! videoconvert ! kmssink plane-id=$PLANE_ID2 => okgst-launch-1.0 videotestsrc ! videoconvert ! kmssink plane-id=$PLANE_ID3 =>ok (I don't know why, but the size is smaller; maybe I can find the reason in the device tree)But if I do something like this:gst-launch-1.0 videotestsrc ! videoconvert ! kmssink plane-id=$PLANE_ID1 videotestsrc ! videoconvert ! kmssink plane-id=$PLANE_ID2the command fails.For my test I have a custom board
I'm trying to add the linux "gpio-poweroff" function to the distribution package. I've added the following to my device tree:gpio-poweroff { compatible = "gpio-poweroff"; gpios = <&gpioe 1 GPIO_ACTIVE_HIGH>; timeout-ms = <3000>; };The kernel/device tree compiles OK, but the poweroff function does not work. The logs give the following lines about the poweroff:[ 1.482136] poweroff-gpio gpio-poweroff: gpio_poweroff_probe: pm_power_off function already registered [ 1.482174] poweroff-gpio: probe of gpio-poweroff failed with error -16So I guess openSTLinux already has some handler attached? How do I disable this?
I'm bringing up a very simple board that uses STM32MP135 with LAN8742A-CZ-TR Ethernet PHY as on the eval board. Unlike the eval board, however, I clock the PHY directly from the SoC: PA11/ETH1_CLK from the STM32 is connected to XTAL1/CLKIN on the PHY.I obtained the drivers from the STM32CubeMP13 package (version 1.2.0, or GitHub commit 0ef689b06775da063006da84d952d8bd48f9e5e1). Since I'm not making use of the ETH1_RX_CLK, I tell the HAL to source the 50MHz clock directly from the RCC: eth_handle.Init.ClockSelection = HAL_ETH1_REF_CLK_RCC; The Ethernet driver in stm32mp13xx_hal_etc.c (correctly) makes use of this information to set a bit in the PMCSETR register: if (heth->Init.ClockSelection == HAL_ETH1_REF_CLK_RCC) { syscfg_config |= SYSCFG_PMCSETR_ETH1_REF_CLK_SEL; } HAL_SYSCFG_ETHInterfaceSelect(syscfg_config);However, later on, an assertion fails which does not expect the RCC bit to be set in this register:#define IS_SYSCFG_
Hi im working on stm32mp257f-ev1 and the "edt,etml0700z9ndha" display,Im unable to decode the logic why the back-light enable gpio only gets high when there is a display connected in the lvds port, what logic is used to detect the lvds display connection.Since i only see the display logs during boot only when there is a display connected, i need clarification regarding this from which driver this logic is implemented, and in the compatible string "edt,etml0700z9ndha" this is used but i found no driver with this compatible list.If in case i want to use the RVT101HVLNWCA0 what driver should i use and can i get a sample device tree for this display panel_lvds: panel-lvds { compatible = "edt,etml0700z9ndha", "panel-lvds"; enable-gpios = <&gpiog 15 GPIO_ACTIVE_HIGH>; backlight = <&panel_lvds_backlight>; power-supply = <&scmi_v3v3>; default-on; status = "okay"; width-mm = <156>; height-mm = <92>; data-mapping =
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.