SPI是串行外设接口(Serial Peripheral Interface)的缩写。是 Motorola 公司推出的一 种同步串行接口技术，是一种高速的，全双工，同步的通信总线。主要应用在EEPROM、Flash、实时时钟(RTC)、数模转换器(ADC)、网络控制器、MCU、数字信号处理器(DSP)以及数字信号解码器之间。

MicroPython

from machine import SPI

spi = SPI(0)
spi = SPI(0, 100_000)
spi = SPI(0, 100_000, polarity=1, phase=1)

spi.write('test')
spi.read(5)

buf = bytearray(3)
spi.write_readinto('out', buf)

C/C++

以bme280温度压力传感器为例。

bme280_spi_bb.png
#include <stdio.h>
#include <string.h>
#include "pico/stdlib.h"
#include "hardware/spi.h"

#define PIN_MISO 16
#define PIN_CS   17
#define PIN_SCK  18
#define PIN_MOSI 19

#define SPI_PORT spi0
#define READ_BIT 0x80

int32_t t_fine;

uint16_t dig_T1;
int16_t dig_T2, dig_T3;
uint16_t dig_P1;
int16_t dig_P2, dig_P3, dig_P4, dig_P5, dig_P6, dig_P7, dig_P8, dig_P9;
uint8_t dig_H1, dig_H3;
int8_t dig_H6;
int16_t dig_H2, dig_H4, dig_H5;

/* The following compensation functions are required to convert from the raw ADC
data from the chip to something usable. Each chip has a different set of
compensation parameters stored on the chip at point of manufacture, which are
read from the chip at startup and used inthese routines.
*/
int32_t compensate_temp(int32_t adc_T) {
    int32_t var1, var2, T;
    var1 = ((((adc_T >> 3) - ((int32_t) dig_T1 << 1))) * ((int32_t) dig_T2)) >> 11;
    var2 = (((((adc_T >> 4) - ((int32_t) dig_T1)) * ((adc_T >> 4) - ((int32_t) dig_T1))) >> 12) * ((int32_t) dig_T3))
            >> 14;

    t_fine = var1 + var2;
    T = (t_fine * 5 + 128) >> 8;
    return T;
}

uint32_t compensate_pressure(int32_t adc_P) {
    int32_t var1, var2;
    uint32_t p;
    var1 = (((int32_t) t_fine) >> 1) - (int32_t) 64000;
    var2 = (((var1 >> 2) * (var1 >> 2)) >> 11) * ((int32_t) dig_P6);
    var2 = var2 + ((var1 * ((int32_t) dig_P5)) << 1);
    var2 = (var2 >> 2) + (((int32_t) dig_P4) << 16);
    var1 = (((dig_P3 * (((var1 >> 2) * (var1 >> 2)) >> 13)) >> 3) + ((((int32_t) dig_P2) * var1) >> 1)) >> 18;
    var1 = ((((32768 + var1)) * ((int32_t) dig_P1)) >> 15);
    if (var1 == 0)
        return 0;

    p = (((uint32_t) (((int32_t) 1048576) - adc_P) - (var2 >> 12))) * 3125;
    if (p < 0x80000000)
        p = (p << 1) / ((uint32_t) var1);
    else
        p = (p / (uint32_t) var1) * 2;

    var1 = (((int32_t) dig_P9) * ((int32_t) (((p >> 3) * (p >> 3)) >> 13))) >> 12;
    var2 = (((int32_t) (p >> 2)) * ((int32_t) dig_P8)) >> 13;
    p = (uint32_t) ((int32_t) p + ((var1 + var2 + dig_P7) >> 4));

    return p;
}

uint32_t compensate_humidity(int32_t adc_H) {
    int32_t v_x1_u32r;
    v_x1_u32r = (t_fine - ((int32_t) 76800));
    v_x1_u32r = (((((adc_H << 14) - (((int32_t) dig_H4) << 20) - (((int32_t) dig_H5) * v_x1_u32r)) +
                   ((int32_t) 16384)) >> 15) * (((((((v_x1_u32r * ((int32_t) dig_H6)) >> 10) * (((v_x1_u32r *
                                                                                                  ((int32_t) dig_H3))
            >> 11) + ((int32_t) 32768))) >> 10) + ((int32_t) 2097152)) *
                                                 ((int32_t) dig_H2) + 8192) >> 14));
    v_x1_u32r = (v_x1_u32r - (((((v_x1_u32r >> 15) * (v_x1_u32r >> 15)) >> 7) * ((int32_t) dig_H1)) >> 4));
    v_x1_u32r = (v_x1_u32r < 0 ? 0 : v_x1_u32r);
    v_x1_u32r = (v_x1_u32r > 419430400 ? 419430400 : v_x1_u32r);

    return (uint32_t) (v_x1_u32r >> 12);
}

static inline void cs_select() {
    asm volatile("nop \n nop \n nop");
    gpio_put(PIN_CS, 0);  // Active low
    asm volatile("nop \n nop \n nop");
}

static inline void cs_deselect() {
    asm volatile("nop \n nop \n nop");
    gpio_put(PIN_CS, 1);
    asm volatile("nop \n nop \n nop");
}

static void write_register(uint8_t reg, uint8_t data) {
    uint8_t buf[2];
    buf[0] = reg & 0x7f;  // remove read bit as this is a write
    buf[1] = data;
    cs_select();
    spi_write_blocking(SPI_PORT, buf, 2);
    cs_deselect();
    sleep_ms(10);
}

static void read_registers(uint8_t reg, uint8_t *buf, uint16_t len) {
    // For this particular device, we send the device the register we want to read
    // first, then subsequently read from the device. The register is auto incrementing
    // so we don't need to keep sending the register we want, just the first.
    reg |= READ_BIT;
    cs_select();
    spi_write_blocking(SPI_PORT, &reg, 1);
    sleep_ms(10);
    spi_read_blocking(SPI_PORT, 0, buf, len);
    cs_deselect();
    sleep_ms(10);
}

/* This function reads the manufacturing assigned compensation parameters from the device */
void read_compensation_parameters() {
    uint8_t buffer[26];

    read_registers(0x88, buffer, 24);

    dig_T1 = buffer[0] | (buffer[1] << 8);
    dig_T2 = buffer[2] | (buffer[3] << 8);
    dig_T3 = buffer[4] | (buffer[5] << 8);

    dig_P1 = buffer[6] | (buffer[7] << 8);
    dig_P2 = buffer[8] | (buffer[9] << 8);
    dig_P3 = buffer[10] | (buffer[11] << 8);
    dig_P4 = buffer[12] | (buffer[13] << 8);
    dig_P5 = buffer[14] | (buffer[15] << 8);
    dig_P6 = buffer[16] | (buffer[17] << 8);
    dig_P7 = buffer[18] | (buffer[19] << 8);
    dig_P8 = buffer[20] | (buffer[21] << 8);
    dig_P9 = buffer[22] | (buffer[23] << 8);

    dig_H1 = buffer[25];

    read_registers(0xE1, buffer, 8);

    dig_H2 = buffer[0] | (buffer[1] << 8);
    dig_H3 = (int8_t) buffer[2];
    dig_H4 = buffer[3] << 4 | (buffer[4] & 0xf);
    dig_H5 = (buffer[5] >> 4) | (buffer[6] << 4);
    dig_H6 = (int8_t) buffer[7];
}

static void bme280_read_raw(int32_t *humidity, int32_t *pressure, int32_t *temperature) {
    uint8_t buffer[8];

    read_registers(0xF7, buffer, 8);
    *pressure = ((uint32_t) buffer[0] << 12) | ((uint32_t) buffer[1] << 4) | (buffer[2] >> 4);
    *temperature = ((uint32_t) buffer[3] << 12) | ((uint32_t) buffer[4] << 4) | (buffer[5] >> 4);
    *humidity = (uint32_t) buffer[6] << 8 | buffer[7];
}

int main() {
    stdio_init_all();

    printf("Hello, bme280! Reading raw data from registers via SPI...\n");

    // This example will use SPI0 at 0.5MHz.
    spi_init(SPI_PORT, 500 * 1000);
    gpio_set_function(PIN_MISO, GPIO_FUNC_SPI);
    gpio_set_function(PIN_SCK, GPIO_FUNC_SPI);
    gpio_set_function(PIN_MOSI, GPIO_FUNC_SPI);

    // Chip select is active-low, so we'll initialise it to a driven-high state
    gpio_init(PIN_CS);
    gpio_set_dir(PIN_CS, GPIO_OUT);
    gpio_put(PIN_CS, 1);

    // See if SPI is working - interrograte the device for its I2C ID number, should be 0x60
    uint8_t id;
    read_registers(0xD0, &id, 1);
    printf("Chip ID is 0x%x\n", id);

    read_compensation_parameters();

    write_register(0xF2, 0x1); // Humidity oversampling register - going for x1
    write_register(0xF4, 0x27);// Set rest of oversampling modes and run mode to normal

    int32_t humidity, pressure, temperature;

    while (1) {
        bme280_read_raw(&humidity, &pressure, &temperature);

        // These are the raw numbers from the chip, so we need to run through the
        // compensations to get human understandable numbers
        pressure = compensate_pressure(pressure);
        temperature = compensate_temp(temperature);
        humidity = compensate_humidity(humidity);

        printf("Humidity = %.2f%%\n", humidity / 1024.0);
        printf("Pressure = %dPa\n", pressure);
        printf("Temp. = %.2fC\n", temperature / 100.0);

        sleep_ms(1000);
    }

    return 0;
}

CMakeList.txt:

add_executable(bme280_spi
        bme280_spi.c
        )

# Pull in our (to be renamed) simple get you started dependencies
target_link_libraries(bme280_spi pico_stdlib hardware_spi)

# create map/bin/hex file etc.
pico_add_extra_outputs(bme280_spi)

# add url via pico_set_program_url
example_auto_set_url(bme280_spi)