RT-Thread-请教SPI初始化SD卡RT-Thread问答社区

请教SPI初始化SD卡

发布于 2019-01-25 23:22:51 浏览：3197 订阅该版

```/*
  * Copyright (c) 2006-2018, RT-Thread Development Team
 *
 * SPDX-License-Identifier: Apache-2.0
 *
 * Change Logs:
 * Date           Author       Notes
 * 2018-09-25     misonyo      first edition.
 */
/*
 * 程序清单：这是一个SD卡设备的使用例程
 * 例程导出了 sd_sample 命令到控制终端
 * 命令调用格式：sd_sample sd0
 * 命令解释：命令第二个参数是要使用的SD设备的名称，为空则使用例程默认的SD设备。
 * 程序功能：程序会产生一个块大小的随机数，然后写入SD卡中，然后在读取这部分写入的数据。
 *            对比写入和读出的数据是否一致，一致则表示程序运行正确。
*/

#include <rtthread.h>
#include <rtdevice.h>
#include <stdlib.h>

#define SD_DEVICE_NAME    "sd0"

extern rt_err_t msd_init(const char *sd_device_name, const char *spi_device_name);

void fill_buffer(rt_uint8_t *buff, rt_uint32_t buff_length)
{
 rt_uint32_t index;
 /* 往缓冲区填充随机数 */
 for (index = 0; index < buff_length; index++)
 {
 buff[index] = ((rt_uint8_t)rand()) & 0xff;
 }
}

static int sd_sample(int argc, char *argv[])
{
 rt_err_t ret;
 rt_device_t sd_device;
 char sd_name[RT_NAME_MAX];
 rt_uint8_t *write_buff, *read_buff;
 struct rt_device_blk_geometry geo;
 rt_uint8_t block_num;
 /* 判断命令行参数是否给定了设备名称 */
 if (argc == 2)
 {
 rt_strncpy(sd_name, argv[1], RT_NAME_MAX);
 }
 else
 {
 rt_strncpy(sd_name, SD_DEVICE_NAME, RT_NAME_MAX);
 }
 /* 查找设备获取设备句柄 */
 sd_device = rt_device_find(sd_name);
 if (sd_device == RT_NULL)
 {
 rt_kprintf("find device %s failed! ", sd_name);
 return RT_ERROR;
 }
 /* 打开设备 */
 ret = rt_device_open(sd_device, RT_DEVICE_OFLAG_RDWR);
 if (ret != RT_EOK)
 {
 rt_kprintf("open device %s failed! ", sd_name);
 return ret;
 }

rt_memset(&geo, 0, sizeof(geo));
 /* 获取块设备信息 */
 ret = rt_device_control(sd_device, RT_DEVICE_CTRL_BLK_GETGEOME, &geo);
 if (ret != RT_EOK)
 {
 rt_kprintf("control device %s failed! ", sd_name);
 return ret;
 }
 rt_kprintf("device information: ");
 rt_kprintf("sector size : %d byte ", geo.bytes_per_sector);
 rt_kprintf("sector count : %d ", geo.sector_count);
 rt_kprintf("block size : %d byte ", geo.block_size);
 /* 准备读写缓冲区空间，大小为一个块 */
 read_buff = rt_malloc(geo.block_size);
 if (read_buff == RT_NULL)
 {
 rt_kprintf("no memory for read buffer! ");
 return RT_ERROR;
 }
 write_buff = rt_malloc(geo.block_size);
 if (write_buff == RT_NULL)
 {
 rt_kprintf("no memory for write buffer! ");
 rt_free(read_buff);
 return RT_ERROR;
 }

/* 填充写数据缓冲区，为写操作做准备 */
    fill_buffer(write_buff, geo.block_size);

/* 把写数据缓冲的数据写入SD卡中，大小为一个块，size参数以块为单位 */
 block_num = rt_device_write(sd_device, 0, write_buff, 1);
 if (1 != block_num)
 {
 rt_kprintf("write device %s failed! ", sd_name);
 }

/* 从SD卡中读出数据，并保存在读数据缓冲区中 */
 block_num = rt_device_read(sd_device, 0, read_buff, 1);
 if (1 != block_num)
 {
 rt_kprintf("read %s device failed! ", sd_name);
 }

/* 比较写数据缓冲区和读数据缓冲区的内容是否完全一致 */
 if (rt_memcmp(write_buff, read_buff, geo.block_size) == 0)
 {
 rt_kprintf("Block test OK! ");
 }
 else
 {
 rt_kprintf("Block test Fail! ");
 }
 /* 释放缓冲区空间 */
 rt_free(read_buff);
 rt_free(write_buff);

return RT_EOK;
}
/* 导出到 msh 命令列表中 */
MSH_CMD_EXPORT(sd_sample, sd device sample);

int main()
{
	msd_init(SD_DEVICE_NAME, "spi1");
	return 0;
}
```大家好，我在做SPI实验时遇到了图中的问题，请问一下该如何解决？
我使用的是原子mini stm32f103rct6开发板，SD卡的使能引脚是PA3；
在main线程中调用了msd_init()函数（不知道这样使用是否正确、规范？）。
编译运行后list_device，见上图。
运行sd_sample的时候就出现问题了，是使能引脚有误么？该在哪里修改呢...:)

![sd_sample_error.png](https://oss-club.rt-thread.org/uploads/201901/25/230319lswe4vzris2mtryg.png)