STM32 MCUs Embedded software

Hi,I am using an STM32F777BIT6 in USB to Serial mode.Everything works fine except for a random loss of data on the windows side.After weeks of frustration I found the cause of this problem: Windows 10 CDC driver for virtual com ports. I tried the program on a windows 7 PC and it works without any data loss. According to this link Changing the default CDC driver of windows 10 will solve the problem:https://electronics.stackexchange.com/questions/425133/stm32-usb-cdc-packet-loss-with-active-pc-usageHowever, I have no idea how to find an alternate CDC driver for windows 10 and Google didn't help either. Does anyone have a link or know of such a driver?Best wishes

Which is the correct way to use the L4 HAL driver to start a conversion using two ADCs (ADC1 and ADC2) configured in "regular simultaneous multi-mode with independent injected" and MDMA disabled (MDMA=0b00)?The HAL_ADCEx_MultiMode_Start_DMA() only supports a single DMA reading the common data register.Using two intependent calls to HAL_ADC_Start_DMA() gives an error, because the ADC instances are configured in multi-mode.Using HAL_ADCEx_MultiMode_Start_DMA() on the master and HAL_ADC_Start_DMA() on the slave won't work either, the slave start attempt returns with HAL_BUSY and, besides that, the master DMA would be set on the common data register...Thanks

lwip udp server does not send big data correctlyhelloI have implemented a udp server on stm32h7 using raw lwip.my function is reading data from sd memory and sending to pc, so I read a block(512bytes) of data from sd memory by f_read function of fatfs and send it using udp_send to remote host.everything is OK with reading data from sd memory, but when I want to send for example a 2kbyte file from sd to PC, I receive only 1636bytes on pc. I use udp terminal of matlab. when I check the received data it shows that only firs 512 received bytes are correct and other received bytes seems to be a peice of next 512byte read from sd memory.here is the code I wrote for ///////////////////////////////////////////////////////////////#define SD_read_buffer 512udp_server_send. struct pbuf *p_tx; struct pbuf *sb; char *c;

..

..

Hello everyone, I am working on a project in wich I want to use a nucleo-l476RG as a MIDI device. I saw on midi official website that the USB Audioclass was compliant with MIDI application (https://www.midi.org/midi-articles/basic-of-usb). However when I set the board as a Audio class usb device and I flash the program generated by CubeMx (without modifications), my board is still seen as a serial peripheral with StLink (even if I remove the jumpers as said on th St official schematic

Supposed I am using external memory via an 16 bit address / 8 bit data bus with FSMC (STM32F714). An RTOS is running.Is access to this memory atomic (in the sense of thread safety) when I read/write 32 bit wide data? The consideration is, for instance:bytes 1 and 2 are already written, then a task switch occurs, and another task uses the FSMC controller for a different read/write operation;worse: the second task accesses the same memory location, thus overriding all 4 bytes, and then, when the former task continues, it finishes writing bytes 3 and 4.Can the processor handle these operations atomically, or is usage of a mutex appropriate when accessing the same resource (external memory, or the FSMC infrastructure as a whole)?

Hi all,Bootloader selection for STM32G47xxxx described in AN2606 shows the different steps followed by the bootloader.Where can I find the delays available between steps ?For example, how long the booloader waits to receive the 0x7F value on USARTx ?BR

Hello, I am trying to start the bootloader on a STM32F427 to program the MCU via USB port in DFU mode. I am not new to this topic because I wrote some programs for STM32F072 where I can successfully start the bootloader and take externally control of the MCU via serial port (there I don't need to program the boards via USB). In that products my software calls correctly the bootloader that works without any problem.On this new design there is a bigger STM32F427; this time I need to program the MCU via the USB port and not via an UART. Here is the problem:If I start the bootloader using hardware signals (BOOT0 and RESET), all works as it should do: using STM32CubeProgrammer I can take control both via an UART or via the USB port.If I start the bootloader by software, STM32CubeProgrammer can take control only via the UART: the USB port doesn't start in DFU mode so the programmer doesn't find it.After many attemps, to have a clean s

I'm planning to use a cheap STM32 board (either with a F103 or F401/11) as a joystick/rotary encoder controller for arcade games. I already have a working proof of concept using the old STM32duino cores (not HAL based), but I prefer to use the HAL environment to avoid more abstraction levels.I used an existing example as a starting point (https://github.com/miniwinwm/BluePillDemo/tree/master/BluePillDemo_USB_HID), and I run into a problem trying to send data at high rates.I'm using a Windows PC with mouserate.exe (http://tscherwitschke.de/old/mouseratechecker.html) to check effective rates. Windows can handle 1000Hz rates for mice.First, I modified usbd_conf.h and changed#define HID_FS_BINTERVAL   0x1 // was 0x0Ato increase the reported polling rate to 1000HzThen I put the following code in my main.cint main(void) { /* USER CODE BEGIN 1 */ // in the mouse repor

Welcome everyoneI used Free ROS with two tasks, the first task was to blink the led and the second task was to receive from uart and then to resend the received message. The two tasks have the same priorityThe led flash works fine but the uart only receives and transmits one time then it doesn't workDoes anyone have an idea what is happening?

I didn't find a good tutorial or a legit example code to initialize the ethernet protocol. My project use UART communication at the moment but I need the ethernet. Is there anybody who could help me?Thanks a lot!

Hello!When trying to implement USB OTG HS Host, I am running into an issue. In file "stm32f4xx_ll_usb.c", function "USB_HC_StartXfer" there is a piece of code that enables the USB_OTG_GINTMSK_PTXFEM interrupt. However, it then proceeds to write the entire data to the FIFO without checking how much space is available. Additionally, the interrupt handler "HAL_HCD_IRQHandler" doesn't give a care about those interrupts, NPTXE isn't even cleared.Everything works fine for transfers that are smaller than the FIFO size and can fit in a single packet, but the second I am exceeding that I am running into the issue with this code.What am I missing here?This person https://community.st.com/s/question/0D53W00000CfE0ASAV/stm32f407-usb-related-questions also ran into this on STM32F4.

hello everyoneI'm working on a USB MSC application. Standalone works fine. When I create code with FreeRtos, the hal_delay () functions are left in the usbh_conf.c file. Replacing the hal_delay codes generated by cubemx with the following codes, there is no problem. I would be glad if you can make the necessary changes in the next firmware update. #if (USBH_USE_OS == 1)  osDelay (Delay);#else  HAL_Delay (Delay);#endifolder posthello everyoneI'm trying to make a USB MSC application with stm32f407 standalone works fine. I haven't been able to run it with freertos yet, but I noticed that CubeMx created the USBH_Delay function called from within the operating system as follows.void USBH_Delay (uint32_t Delay){  HAL_Delay (Delay);}When I replace it with the code below, the freertos stopped freezing when I plug in usb

Hi,these defines are defined very strange/misleading in STM32H743xx.h (from current version 1.7: STM32Cube_FW_H7_V1.7):#define LPTIM_CR_SNGSTRT_Msk (0x40001UL << LPTIM_CR_SNGSTRT_Pos) /*!< 0x00080002 */According to RefMan Desc of SNGSTRT, this should be only 1 Bit (Only Bit 1, so number 0x00000002 ... but here they defeine 2 bits, 0x00080002... is this some "undocumented feature" or some error in the header file?)... I also checked STM32L053 header file for comparison, there this bit is defined correctly as 0x00000002... .Further the following mask values I think are wrong / very misleading:#define TIM_CCMR1_OC1M_Msk (0x1007UL << TIM_CCMR1_OC1M_Pos) /*!< 0x00010070 */ #define TIM_CCMR1_OC2M_Msk (0x1007UL << TIM_CCMR1_OC2M_Pos) /*!< 0x01007000 */ #define TIM_CCMR2_OC3M_Msk (0x7UL << TIM_CCMR2_OC3M_Pos) /*!< 0x00000070 */ #define TIM_CCMR2_OC4M_Msk (0x7UL <&lt

I can get my USB CDC on Nucleo-144 recognized from Device Manager. It will appear like COM8 and Terminal Emulator can connect to it. I can even echo back data received.Just when I add some minor functionality in the while loop in main() (like check for pressed button or incrementing some variables) it can not be connected any more from Terminal Emulator. It can still be seen in the Device Manager as COM8, just connection attempts fail. It seems as if the MCU had resource problems. Running it on a F439 on sysclk of 168 MHz will allow me incrementing a variable in the endless loop and the serial port will be stilll functional. Adding some more code (checking user button) and it can not be connected any more.On a slower F412 running on 72 MHz even this variable increment is too much and will prevent connecting as serial port.The lineCDC_Transmit_FS(Buf, *Len) ;added to CDC_Receive_FS(uint8_t* Buf, uint32_t *Len) will run an

I believe there is a bug with the L496 series and the serial number the DFU firmware returns. Whenever I unpower and repower my board I get a different 12 digit hex number, for example this:05FA03B042C5becomes F0F30783F95Cafter next power cylce and so on. Anyone noticed this before? I couldn't find any note of this anywhere. Any suggestions for a work around?I am writing a windows application to load a testing firmware through DFU, then do some testing and then load another app via DFU again and was planning on keeping track via this serial number. Alas, I cannot.

Hi!I created a simple test project based on the example "I2C_OneBoard_AdvCommunication_DMAAndIT" found in the STM32Cube_FW_F4_V1.25.2 package.The project basically works, but I have to take the number of data to be sent one more than the amount to be actually sent.So when I want to read from the EEPROM and have to send out the start address, I have to set the DMA transmission to 3 bytes instead of two.It's just a quick and dirty project to help me understanding the basics...Thanks for any help!My main.c (I've deleted irrelevant parts...):uint8_t rd_value[20] = {0}; uint8_t wr_value[20] = {0x14,0x13,0x12,0x11,0x10, 0x0F,0x0E,0x0D,0x0C,0x0B, 0x0A,0x09,0x08,0x07,0x06, 0x05,0x04,0x03,0x02,0x01};   uint8_t tx_buffer[32] = {}; uint8_t* tx_data_buffer = tx_buffer + 2; uint8_t rx_buffer[34] = {};   __IO uint8_t ubMasterRequestDirection = 0; __IO uint8_t ubMasterNbDataToReceive = 0; __IO

STM32F1_3.5.0 has bit description of FLASH_ACR_PRFTBESTM32CUBEF1_1.8.1 has no bit description of FLASH_ACR_PRFTBE

What is the polynomial in CRC16?Crc16-ccitt, CRC16-IBM or something else?I tried everything I could, but it didn't work?Can someone help me?Here's how I set up this register:uint32_t wakeup_filter[8] = {0x00ffffff, 0x00000000, 0x00000000, 0x00000000,                0x00000001, 0x00000013, 0x0000f40e, 0x00000000};thank you!

HelloI'm using the USB CDC class on my STM32H7.Currently, I am using ST's VID (0x0483) and PID (0x5740).I have four questions.1.Can I use ST's VID and PID in our final product version?2.If so, can I redistribute ST's Virtual COM Port driver to our customers?3.How much does it cost to use ST's VID (0x0483) and unique PID?4.We have a USB VID, but how much does it cost to change the VCP driver if we use our VID?Best Regards,

Hello,i work with STM32G070KBT and i jump from my code to the buitin ST bootloader. Then i use SPI to upload my new code/binary and send the Go command. I receive an ACK but nothing happen. My code is not executet or stops working early.If the bootloader starts after a full chips erase (done with STM32CubeProgrammer) the Go command works as expected.My jump code looks like:#define SYSTEM_MEMORY_ADDRESS 0x1FFF0000   void jump_to_bootloader() // https://github.com/markusgritsch/SilF4ware/blob/master/SilF4ware/drv_reset.c // https://stm32f4-discovery.net/2017/04/tutorial-jump-system-memory-software-stm32/ { HAL_NVIC_DisableIRQ(

I am using STM32Cube_FW_F7_V1.15.0 with a STM32F723IET6 implementing the internal HS USB as device. I have implemented the CDC VCP device. I plug the USB cable into a host computer, running either Windows or linux. If I am running a terminal program on the host computer, I am able to send characters to the host. If I am not running any receiver on the host (but still plugged in), if I try to send characters to the host, CDC_Transmit_HS() returns USBD_BUSY. Since I need to check for this case when the connection is up and I am sending characters normally, my program hangs on this condition if there is no listener running on the host.What I really need is to detect the condition of whether the host has opened the device, so I can buffer my data until the device is opened on the host. In testing with a Windows 10 host, I found that hUsbDeviceHS.dev_state == USBD_STATE_CONFIGURED is true after the device is configured, whether or not the connection is there. I also found t

Hello, I had problems with setting up HAL_I2C_Mem_Read_DMA() function. Firstly I only defined the Rx DMA channel, but the program was still ending in HardFault_Handler(). After a long debugging I found bug in the "static void I2C_DMAXferCplt(DMA_HandleTypeDef *hdma) " (stm32f1xx_hal_i2c.c) line 6661 (14,15 in the snippet ). The pointer to the undefined (Tx) DMA channel has value NULL -> write operation to this address cause HardFault. Is there any better way how to avoid the hard fault, then the creation of the Tx DMA?static void I2C_DMAXferCplt(DMA_HandleTypeDef *hdma) { I2C_HandleTypeDef *hi2c = (I2C_HandleTypeDef *)((DMA_HandleTypeDef *)hdma)->Parent; /* Derogation MISRAC2012-Rule-11.5 */   /* Declaration of temporary variable to prevent undefined behavior of volatile usage */ HAL_I2C_StateTypeDef CurrentState = hi2c->State; HAL_I2C_ModeTypeDef CurrentMode = hi2c->Mode; uint32_t CurrentXferOptions = hi2c->XferOptions; &

Hello,I'm new to ST. What MPU families does ST have for Cortex-M4F with FreeRTOS and support including demo/eval boards with demo projects, running in IAR, for FreeRTOS?  This would include porting and FreeRTOSconfig files etc.?I am especially interested in IoT.Thanks George