Ask questions and find answers on STM32 microprocessors and dedicated embedded software solutions, such as Yocto SDK, Qt, and OP-TEE.
Most recent activity
STM32MP257F-EV1: SPI3 Slave Mode — NSS Pin (PB1) Fails to Claim Despite Correct ConfigurationSummaryOn the STM32MP257F-EV1 evaluation board, configuring SPI3 in slave mode (spi-slave;) works correctly for the SCK, MOSI, and MISO signals, but the NSS (chip select) pin — PB1, using the documented-correct AF1 alternate function — silently fails to be claimed by the pin controller. No error or warning is logged anywhere. All three other SPI3 pins on the same bus, using the same AF1 value, claim successfully.HardwareBoard: STM32MP257F-EV1 (MB1936) MPU: STM32MP257FAI3 OpenSTLinux version: 5.0.15-openstlinux-6.6-yocto-scarthgap-mpu-v26.02.18 Kernel: Linux 6.6.116-stm32mp-r3What we are trying to doConfigure SPI3 as an SPI slave (spi-slave; device tree property) so the STM32MP2 can receive data from an external SPI master over the same bus/pins already used for a sensor (previously configured as SPI master).Configuration usedCustom pinctrl group (added to stm32mp25-pinctrl.dtsi):spi3_slave_pins
Hello ST Support Team,We are working with the STM32MP257F-EV1 evaluation board and have identified a hardware/firmware-level restriction preventing us from using the SPI8 peripheral from the Linux (Cortex-A35) side, despite correct device tree configuration.Board: STM32MP257F-EV1OpenSTLinux version: 5.0.15-openstlinux-6.6-yocto-scarthgap-mpu-v26.02.18Kernel: Linux 6.6.116-stm32mp-r3The issue:When we enable the &spi8 node in our device tree (with correct pin configuration, matching the same pattern that successfully works for other SPI peripherals like SPI3), the driver never initializes at all. There is no error message, no warning, and no output anywhere in dmesg — the peripheral simply remains completely silent and unclaimed, as if the device tree node doesn't exist.What we've already investigated and ruled out:Device tree correctness — confirmed the node, pinctrl, and pin assignments are syntactically and semantically correct, following the identical working pattern used success
Hi all,I am bringing up the Ethernet switch (ETHSW) on an STM32MP257F-DK, bare-metal on the Cortex-A35 (no OS), following RM0457 Rev 5.Ping works with the switch bypassed (ETH1 directly to the PHY), and ping also works under Linux with the switch enabled on the same board. In bare-metal switch mode, ARP works but ICMP never arrives.SETUP- ETH1 -> Port0 (internal, no PHY), PHY (RTL8211F, RGMII) on Port2, negotiated 100 Mbps full duplex- SYSCFG_ETHSWCR = 0x6 SYSCFG_ETH1CR = 0x12 - Init order:ETHSW reset -> clocks -> SYSCFG_ETHSWCR -> release reset -> pulse ETHSW_FES_GEN_CTRLR INIT_REQ (bit 15) -> then ETH1 clocks/reset/DMA- PORT0 UNDER MANAGEMENT,MII,100 MBPS- PORT2 UNDER NORMAL, MII,100 MPBS WHAT WORKS- Broadcast/multicast from the PC reach ETH1 (ETH1 MMC broadcast counter increments)- The board's ARP reply (unicast) leaves ETH1, crosses Port0 -> Port2 (Port0 RX_UNICAST=1, Port2 TX_UNICAST=1) [confirm in Wireshark that the PC actually receives it]- Dynamic MAC tab
I am working with an STM32MP257F-EV1 Evaluation Board (board reference MB1936, target device STM32MP257FAI3) and am trying to debug a standalone bare-metal application running on the Cortex-M33 core directly over SWD using ST-Link.Current SetupBoard: STM32MP257F-EV1 Board reference: MB1936 Target: STM32MP257FAI3 Tools: STM32CubeIDE and STM32CubeProgrammer Boot configuration (SW1): Development boot Boot0 = 1 Boot1 = 1 Boot2 = 0 Boot3 = 0 JP3: No jumper fitted (STLINK-V3EC active) CubeIDE Debug Configuration: Load Mode = "thru JTAG/SWD link (Engineering mode)" Access Port = "8 - Cortex-M33" Debug interface = SWD I have also attached a screenshot of the debug error I am receiving.Current ObservationI am able to debug the Cortex-M33 successfully when the board is first booted from an SD card. After the SD-card boot completes, I can connect to and debug the M33 core from CubeIDE.However, my requirement is to debug the M33 without booting the board from an SD card, eMMC, or QSPI.Specif
Hi,I'm working on a custom STM32MP157C board with a Microchip LAN8742A PHY (RMII). Originally the board uses an external 50 MHz oscillator that feeds both the PHY and the SoC's PA1 (ETH1_RMII_REF_CLK) — the classic LAN8742A "REF_CLK In Mode" setup, same as ST's reference schematic. That works fine.I'm now trying to get rid of the external oscillator and have the SoC generate the 50 MHz REF_CLK itself instead, using RCC's CLK_ETH_PLL3Q — the setup ST documents in AN5031, Figure 45 ("RCC provides 50 MHz ETH_CLK"). For this test I desoldered the two series resistors that fed the oscillator signal in, isolated the oscillator, and bridged PA1 straight to the PHY's XTAL1/CLKIN through a wire + the same 22 Ω resistor as before. I also added st,eth-ref-clk-sel; to the ethernet node in both the kernel and U-Boot device trees, rebuilt and reflashed.Result: PHY never comes out of reset (phy_poll_reset failed: -110), and an oscilloscope on PA1 shows absolutely nothing — not even a weak or distorte
The STM32MP2 supports gstreamer. There is a version of Gstreamer for streaming through the WebRTC protocol.Another library for streaming though the WebRTC protocol would be LibWebRTC.My question is if the STM32MP2 have support for LibWebRTC as well? Thank you for your reply
Hey, Guys!I’m designing an application to record TDM data source with 48k x 16 slots x 32Bit Format based on STM32MP151.But from the datasheet, there is some confusion.It seems the SAI supports 32bit and 16 slots from the SAI main feature description. But the FRL register is limited to 256, which cannot meet the requirement: 32*16 = 512 > 256.I’m using an external CODEC to generate the TDM clock, so STM32MP151 is in slave mode.Does anyone have experience with an application like this?Thanks! Jampa.
Greetings,I’m working on a STM32MP13X based development board as a learning project. One of the things I’m trying to achieve is a BIOS-like program stored in a NOR flash from which the MPU will boot from every time.After that, the program will load details of any relevant peripherals, Like size & configuration of the installed DDR3 memory & any bootable media like USB drive or SD card. After which the program will boot the bootable media (if any).The aim is to keep the system flexible like a general purpose computer, i.e. no fixed DTS configurations like DDR size, etc in U-Boot or Linux, So that any Linux distro with kernel support for STM32MP1 can be booted off the media.Is this possible? If so, then can you give me an overview of how this may be achieved? I understand this is alot of work which I am willing to do if someone can point me in the right direction.Regards,Aditya M.
Hi everyone,I'm using STM32MP2 OpenSTLinux 6.6 Yocto Scarthgap MPU v24.11.06 with Linux kernel 6.6.48.I am trying to interface a Waveshare LVDS 10.1" FT5x06 touch display with the STM32MP257F-DK. We designed a custom PCB for the required pin mapping. 1. Display :Initially, the display was not working with the original compatible string. After changing the panel compatible to "panel-lvds" only then display started working. panel_lvds: panel-lvds { compatible = "panel-lvds";... /*compatible = "edt,etml0700z9ndha", "panel-lvds";*/ } 2. I2C Touch / HDMI Address :The FT5x06 touch controller uses I2C address 0x38.The ADV7535 HDMI device also uses 0x38 as one of its register-map addresses. Since I am not using HDMI for this test, I changed that ADV7535 address from 0x38 to 0x39:&i2c2 { adv7535: hdmi@3d { compatible = "adi,adv7535"; reg = <0x3d>, <0x3c>, <0x3f>, <0x39>; //<0x38>; } ili2511: ili2511@41 { compatible = "ilitek,ili251x
Can the STM32MP231 simultaneously capture or stream four 2-lane MIPI CSI-2 camera streams through a DS90UB960/TDES960 using VC0–VC3?I have reviewed ST’s documentation on “How to use one or multiple multi-stream MIPI CSI cameras.” But did not get complete clarity on simultaneous capture of four camera streams. The documentation example appears to show routing for up to three streams/channels through the DCMIPP pipes (Dump, Main, and Aux).
Hi,I have a SoM based on STM32MP135c (QSMP-1351) and I try to enable the secure boot using STM32_Programmer_CLI (STM32CubeProgrammer v2.22.0)Here is the flash layout file used before updating OTP:#opt Id Name Type Device Offset Binary- 0x01 fsbl-boot Binary none 0x0 tf-a-loader.stm32- 0x03 fip-boot FIP none 0x0 fip-loader.binI can write PKHTH without any issue:STM32_Programmer_CLI -c port=usb1 -otp write word=0x18 value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x19 value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1a value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1b value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1c value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1d value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1e value=0x????????STM32_Programmer_CLI -c port=usb1 -otp write word=0x1f value=0x????????When I want to lock device, I use the comm
Dear ST-Community,I was able to run QNX8 on stm32mp257f-dk (see it here: https://github.com/KozhinovAlexander/qnx8-bsp-stm32mp257f-dk)I have following boot-stage setup:#Opt Id Name Type IP Offset Binary- 0x01 fsbl-boot Binary none 0x0 arm-trusted-firmware/tf-a-stm32mp257f-dk-opteemin-programmer-usb.stm32- 0x02 fip-ddr FIP none 0x0 fip/fip-stm32mp257f-dk-ddr-opteemin-programmer-usb.bin- 0x03 fip-boot FIP none 0x0 fip/fip-stm32mp257f-dk-opteemin-programmer-usb.binP 0x04 fsbla1 Binary mmc0 0x00004400 arm-trusted-firmware/tf-a-stm32mp257f-dk-opteemin-sdcard.stm32P 0x05 fsbla2 Binary mmc0 0x00044400 arm-trusted-firmware/tf-a-stm32mp257f-dk-opteemin-sdcard.stm32PD 0x06 metadata1 FWU_MDATA mmc0 0x00084400 arm-trusted-firmware/metadata.binPD 0x07 metadata2 FWU_MDATA mmc0 0x000C4400 arm-trusted-firmware/metadata.binP 0x08 fip-a FIP mmc0 0x00104400 fip/fip-stm32mp257f-dk-opteemin-sdcard.binPED 0x09 fip-b FIP mmc0 0x00504400 nonePED 0x0A u-boot-env ENV mmc0 0x00904400 noneP 0x10 bootfs System mmc
Hi,I am working with the STM32MP257F-EV1 board and would like to debug the Cortex-M33 core using the debugger using ST-LINK.Currently, I am unable to debug the Cortex-M33 core directly and I receive an error during the debug connection. I have attached a screenshot of the debug error for reference.I noticed the following behavior: If I boot the Cortex-A35 core from the SD card and allow Linux to complete its boot process, I can then successfully debug the Cortex-M33 core. However, if I do not boot Linux on the A35 core, I am unable to connect to/debug the M33 core. I would like to know whether the Cortex-M33 on the STM32MP257F-EV1 can be debugged standalone, without booting the Cortex-A35/Linux, similar to debugging the Cortex-M4/M7 on a conventional STM32 MCU such as the STM32H745XI-DISCO.Specifically: Is booting Linux on the Cortex-A35 mandatory before the Cortex-M33 can be debugged? Is there a way to debug the Cortex-M33 standalone immediately after reset/power-up, without boot
Board: STM32MP257F-EV1Image: FLASH-stm32mp2-openstlinux-6.6-yocto-scarthgap-mpu-v26.06.10Kernel: Linux 6.6.129USB port: CN15 (USB_DRD, Type-C)I'm trying to run a USB gadget (UVC, via configfs/libcomposite) on CN15, andthe UDC never attaches, even though every layer up to dwc3's pullup steplooks correct:- Type-C/UCSI (M33 firmware USBPD_DRP_UCSI_CM33_NonSecure_stripped.elf, ucsi-stm32g0-i2c driver) correctly negotiates the port as UFP/device: /sys/class/typec/port0/data_role = host [device] /sys/class/typec/port0/power_role = source [sink] A partner is detected, and the UCSI power-supply object confirms VBUS at 5V. This all live-updates correctly on physical cable unplug/replug.- dwc3 correctly switches to device mode: /sys/kernel/debug/usb/48300000.usb/mode → "device"- But the pullup never actually happens: /sys/class/udc/48300000.usb/state stays "not attached" indefinitelyManually retrying via soft_connect just logs "UDC had already started" —the UDC core believes the connect s
I am developing an M33 application on STM32MP257F MPU, I would like to know the steps to import, build and debug secure project for Cortex-M33 in STM32CubeIDE for M33-TD flavor. I would appreciate if any one could tell the best and easy way to deploy custom M33 firmware in to this MPU and test it with or without linux?FYI, I am using STM32MP257F-DK (Discovery Kit) Board. Thanks in advance,Gupta
Hello, I’m trying to learn how to use the STM32MP135CAEx for a upcomming project.I have been looking online through tutorials and documentation but have a few questions.Project background:I need a device that has 2 ethernet interfaces, a LCD interface and a bunch of common peripherals.I’m a big fan of STM23 controllers and have a few years of experience with them so trying out the MP series would be a next logical step for me.1)My idea was to make a small linux image on run it on the MPU. That linux then should have access to a display with toutch.The display I had in mind is 400x800 pixels.I don’t know if you can run without DDR memory, I’m gone need it, when I look at the standards RAM available.But can you run without?2)Is there a “bring up guide”?I have been trying to set up my ioc but I keep having an error marker in my DDR tab.So I can have a indication of that I understood things correctly. 3)What I’m used to from the controllers is that one a peripheral is configured correctly
Dear ST teamWhy does ST team add the “negedge” flag in DT for the SDCARD timing initialization on ST eval DKs?Sébastien
Hi ST,I am working on STM32MP257 platform and I have read some of the previous topic and thread, seems that web multimedia is somewhat supported now by using Qt at its minimalist setting. But the previous claim of “ST is working on enabling chromium/ web browser, and expected to have it out by late 2025” hasn’t really got any update.Can we know if we really have a definitive answer or status update or is there any example or demo?
Hi! I tried to stream a 1080p (big buck Bunny) video with gstreamer. It seems that it was falling back to software and starts lagging.Are there any known issues with gstreamer for the MP2 i should be aware of? I have not used the latest OpenSTLinux version yet. I'm using v26.02.18. Still: kmscube is working perfectly in 1080p. So, I assume the the Vulkan/OpenGL ES is working properly as it should.
Hi,I am using an STM32H7S3L8 with D-Cache enabled and want to place DMA buffers in a dedicated non-cacheable SRAM region.My linker script places the DMA section in SRAM2:SRAMAHB (rw) : ORIGIN = 0x30000000, LENGTH = 16KSRAM_DMA (xrw) : ORIGIN = 0x30004000, LENGTH = 16K.dma_buffer (NOLOAD) :{ . = ALIGN(32); __dma_buffer_start__ = .; KEEP(*(.dma_buffer)) KEEP(*(.dma_buffer*)) . = ALIGN(32); __dma_buffer_end__ = .;} > SRAM_DMAThe DMA buffer is correctly linked at 0x30004000.I configured the MPU according to the non-cacheable DMA buffer strategy described in ST's article. The relevant configuration is: static void MPU_Config(void){ MPU_Region_InitTypeDef MPU_InitStruct = {0}; /* Disables the MPU */ HAL_MPU_Disable(); /* Disables all MPU regions */ for(uint8_t i=0; i<__MPU_REGIONCOUNT; i++) { HAL_MPU_DisableRegion(i); } /** Initializes and configures the Region and the memory to be protected */ MPU_InitStruct.Enable = MPU_REGION_ENABLE; MPU_InitStruct.
This account only shares restored community content posted by members between May 22 and June 9, 2026, who did not register for the new community.I've followed the build instructions for the STM32MP distribution package 6.6.116 and managed to build the boot sd card for MP135F-DK and that works fine.When i build for MP257F-EV1 board, the build works fine and the boot sd card appears ok, but during boot it looks like the kernel load resets the processor partway through : [ 4.201246] m_can_platform 402f0000.can: m_can device registered (irq=76, version=32)[ 4.204152] stm32-cryp 42030000.crypto: will run requests pump with realtime priority[ 4.204774] stm32-pcie 48400000.pcie: IB MEM 0x0080000000..0x00ffffffff -> 0x0080000000[ 4.221546] stm32-pcie 48400000.pcie: iATU: unroll T, 4 ob, 4 ib, align 64K, limit 4G[ 4.255481] stm32-cryp 42030000.crypto: Initialized[ 4.262966] Galcore version 6.4.21.1.1058597NOTICE: CPU: STM32MP257FAI Rev.YNOTICE: Model: STMicroelectronics STM32MP257F-EV1 Eval
HelloI am building Linux on a custom board using the STM32MP255F-AK.The following error occurs during kernel boot:[ 15.371350] stm32-dwmac 482c0000.eth1: Failed to reset the DMA[ 15.386447] stm32-dwmac 482c0000.eth1 end0: stmmac_hw_setup: DMA engine initialization failed[ 15.389570] stm32-dwmac 482c0000.eth1 end0: __stmmac_open: Hw setup failedAfter embedding printk in the kernel source code to investigate,I found that dwmac4_dma_reset() in /drivers/net/ethernet/stmicro/stmmac/dwmac4_lib.c was returning ETIMEDOUT(110).[dwmac4_lib.c]int dwmac4_dma_reset(void __iomem *ioaddr) {u32 value = readl(ioaddr + DMA_BUS_MODE);/* DMA SW reset */value |= DMA_BUS_MODE_SFT_RESET;writel(value, ioaddr + DMA_BUS_MODE);return readl_poll_timeout(ioaddr + DMA_BUS_MODE, value,!(value & DMA_BUS_MODE_SFT_RESET),10000, 1000000); }I expect 0x482c0000 to be stored in ioaddr, but different values such as 0x82108000, 0x82128000, and 0x82xxx000 were stored there each time.Ive looked into it from various angles
This account only shares restored community content posted by members between May 22 and June 9, 2026, who did not register for the new community.I wonder how I can enable ETH Clock Output for my Realtek RTL8211F Ethernet PHY. Right now, the crystal for RTL8211F won't start and I don't know why. I'm guessing that it's an unstable crystal. Anyway! To enable the ETH Clock Output for my Ethernet PHY, I've set the the ETH Clock Output:Then I've set the CK_KER_ETH1 to either 25 MHz or 125 MHz. I've chosen the 25 MHz option.According to the STM32MP257F reference manual RM0457, the ETH Clock Output can only be used for 125 MHz.But according to the STM32MPU Wiki, it's possible to use 25 MHz as ETH Clock Output:https://wiki.st.com/stm32mpu/wiki/Ethernet_device_tree_configuration#RGMII_with_25_MHz_on_ETH_CLK_-28no_PHY_Crystal-29--CLK125_from_PHY_-28Reference_clock_-28standard_RGMII_clock_name-29-29All you need to do Is to add the pinmuxeth1_clk_pins_mx: eth1_clk_mx-0 {pins {pinmux = <STM32_PI
Board: STM32MP257F-EV1 (also relevant to STM32MP257F-DK)Tool: ST Edge AI Developer Cloud, ST Edge AI Core 2.2.0The "Optimize" step (NBG generation) consistently fails with:"Error while generating optimized file. Generation does not contain any output."Model details:- Architecture: torchvision.models.mobilenet_v2(weights=None, num_classes=100)- Trained from scratch on CIFAR-100 (100 classes), PyTorch 2.5.1+cu121- Input: 3x224x224, output: 100 classes- Quantized on-platform to INT8, per-tensor (per-channel disabled)Attempts (all fail identically):1. Original export, opset_version=122. Simplified via onnx-simplifier (0.4.36) - only reduced redundant Constant nodes (176->108); Conv/Clip/Add/Gemm counts unchanged3. Re-exported with opset_version=174. All used random-value calibration (no .npz calibration dataset provided)No "Show terminal" or detailed log option appears for this specific error on the Optimize step.Control test (works fine):Imported mobilenet_a050_pt_224_qdq_int8_image_cl
Hello,I want to know what is the latest Chromium version supported by OpenST Linux on Scarthgap.Do we have GPU acceleration on it?Thank you in advance,Hiba @PPAGE.13@Billy OWEN @Gregory PLANCHON
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.