search

hideakitai / ESP32DMASPI

SPI library for ESP32 which use DMA buffer to send/receive transactions
MIT License
170 stars 36 forks source link

For every 1 master transaction I got 8-10 slave transactions #4

Closed bakwc closed 3 years ago

bakwc commented 3 years ago

Master code:

int n = 0;
void loopMaster() {
    n += 1;
    spi_master_tx_buf[0] = n % 100;

    master->transfer(spi_master_tx_buf, spi_master_rx_buf, BUFFER_SIZE);

    printf("RX BUF[0] = %d\n", (int)spi_master_rx_buf[0]);

    delay(1000);
}

Slave code:

void loopSlave() {
    // set buffer (reply to master) data here

    spi_slave_tx_buf[0] = (spi_slave_rx_buf[0] * 2) % 100;

    // if there is no transaction in queue, add transaction
    if (slave->remained() == 0)
        slave->queue(spi_slave_rx_buf, spi_slave_tx_buf, BUFFER_SIZE);

    // if transaction has completed from master,
    // available() returns size of results of transaction,
    // and buffer is automatically updated

    while (slave->available())
    {
        printf("Request from master: %d\n", (int)spi_slave_rx_buf[0]);
        slave->pop();
    }
}

setup is the same as in your README example

Slave output:

20:35:39.502 -> Request from master: 164
20:35:43.576 -> Request from master: 2
20:35:43.576 -> Request from master: 2
20:35:43.576 -> Request from master: 2
20:35:43.576 -> Request from master: 2
20:35:43.576 -> Request from master: 8
20:35:43.576 -> Request from master: 8
20:35:43.576 -> Request from master: 8
20:35:43.576 -> Request from master: 8
20:35:43.576 -> Request from master: 8
20:35:43.576 -> Request from master: 8
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 4
20:35:44.583 -> Request from master: 141
20:35:44.583 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:45.586 -> Request from master: 103
20:35:46.582 -> Request from master: 8
20:35:46.582 -> Request from master: 8
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143
20:35:46.582 -> Request from master: 143

Expected behavior: got 1 printf on slave per second Actual behavior: got 8-10 printf on slave per second

hideakitai commented 3 years ago

Hi @bakwc , thank you for reporting.

Have you tried master_simple and slave_simple_polling examples? Could you try if they work and let me know about your environment, please?

If they won't work, please give me your minimum reproducible whole code.

bakwc commented 3 years ago

I tried them too -> same behavior. I receive 0 0 0 0 0 0 0 0 ... on slave, 5 time per second. The code is excatly master_simple and slave_simple_polling. Board - esp32cam.

bakwc commented 3 years ago

It was a hardware issue, i used 20cm wires with plastic connectors (no soldering). I took 3cm wires and soldered them, and everything works without any issues. Thank you for support and for the library!