一、基本概念
1,为什么需要文件系统?
在没有文件系统的存储器中,一般通过手工记录,那些变量存放在那些位置,很难有系统的管理
难以记录有效数据的位置 、难以确定存储介质的剩余空间、 不明确应以何种格式来解读数据
加入文件系统后,就可以引导区、目录等对内部数据进行管控,并且对于数据可以命名,方便以各种格式对应的解析数据;
2,文件系统是什么?
1-嵌入式中,最常用的就是文件系统就是FATFS,是一种小型文件系统;
2-FATFS在初步搭建在外部flash或者内存中时,第一次需要进格式化(格式化的本质就是利用spi的读写功能,在存储介质建立了一些组织结构,这些结构包括操作系统引导区、目录和文件)
3,加入文件系统后的读写过程?
1- 数据都以文件的形式存储;
2- 写入新文件时,先在目录中创建一个文件索引,它指示了文件存放的物理地址,再把数据存储到该地址中;
3- 当需要读取数据时,可以从目录中找到该文件的索引,进而在相应的地址中读取出数据;
4,有文件系统和没有文件系统最大的差异
1-没有文件系统,写入数据和读取数据,需要指定明确的物理地址
2-有文件系统后,文件系统的存在使存取数据时,不再是简单地向某物理地址直接读写,而是要遵循它的读写格式。如经过逻辑转换,一个完整的文件可能被分开成多段存储到不连续的物理地址,使用目录或链表的方式来获知下一段的位置。
二、文件系统示意图说明
1,文件分配表,存放A.TXT B.TXT C.TXT
2,对应的目录分配表
主要参数:1:文件名 ;2,开始簇;3,文件大小;4,创建时间;5,读取属性;
3, 实际文件分配表:
文件 a.txt 我们根据目录 项中指定的 a.txt 的首簇为 2,然后找到文件分配表的第 2 簇记录,上面登记 的是 3,就能确定下一簇是 3。找到文件分配表的第 3 簇记录,上面登记的 是 4,就能确定下一簇是 4......直到指到第 11 簇,发现下一个指向是 FF,就是结束。文件便读取完毕。
在内存是删除B文件后,创建D.TXT后
文件D,在物理内存中被分为两块,目录示意图如下:
实际文件分配如下:
至此文件系统的基本内容完毕,下面介绍FATFS文件系统及移植过程
三、FATFS文件管理系统介绍
1,FATFS系统在系统中的位置
APP接口----FATFS----底层接口
文件系统本质上,可以理解为一个函数B,外部可以通过函数A调用函数B,而函数B不接触底层操作,而是通过调用函数C对底层存储进行操作。
2,FATFS中重要的文件及其作用
1 - integer.h:文件中包含了一些数值类型定义,对一些类型进行typedef。
2 - diskio.c:包含底层存储介质的操作函数,这些函数需要用户自己实现,主要添加底层驱动函数。我们主要的底层接口都要在这边调用执行
3 - ff.c: FatFs核心文件,文件管理的实现方法。该文件独立于底层介质操作文件的函数,利用这些函数实现文件的读写,在主函数中要包含ff.h,从而执行fatfs文件系统
4- cc936.c:本文件在option目录下,是简体中文支持所需要添加的文件,包含了简体中文的GBK和Unicode相互转换功能函数。
5-ffconf.h:这个头文件包含了对FatFs功能配置的宏定义,通过修改这些宏定义就可以裁剪FatFs的功能。如需要支持简体中文,需要把ffconf.h中的_CODE_PAGE 的宏改成936并把上面的cc936.c文件加入到工程之中。
至此,文件系统的基本介绍到此结束,下面开始移植
四、FATFS文件系统移植
1,文件系统移植,本质上就是把文件系统对于下层的硬件操作接口,与文件系统本身的操作给建立连接
2,移植需要借用spi-flash控制的源码,在spi-flash控制的源码的基础上,添加文件系统
3,主要的文件需要包含上述的6个c文件或者h文件,在文件文件中,需要调整主要是diskio.c \ ffconf.h
下面开始逐步说明
diskio.c的底层硬件操作的移植
1-头文件,包含flash的底层头文件
#include "diskio.h" /* FatFs lower layer API */
#include "./flash/bsp_spi_flash.h"
2-宏定义,设备名称
#define SDCARD 0
#define FLASH_SPI 1
/*-----------------------------------------------------------------------*/
/* Get Drive Status */
/*-----------------------------------------------------------------------*/
//diskio是底层硬件和FATFs操作的连接渠道,主要的配置都在这边执行
3-设备状态确认
实际上是通过SPI_FLASH_ReadID()==sFLASH_ID 来判定设备是否正常
DSTATUS disk_status (
BYTE pdrv /* Physical drive nmuber to identify the drive */
)
{
DSTATUS stat;
// int result;
switch (pdrv) {
case SDCARD :
case FLASH_SPI :
//通过读取到flash id来确认状态
if(SPI_FLASH_ReadID()==sFLASH_ID)
{
stat= 0;
}
else
{
stat= STA_NOINIT;
}
// translate the reslut code here
return stat;
}
return STA_NOINIT;
}
4-设备初始化,实际上调用SPI_FLASH_Init();对GPIO接口、SPI1模块进行底层的初始化操作
DSTATUS disk_initialize (
BYTE pdrv /* Physical drive nmuber to identify the drive */
)
{
// DSTATUS stat;
// int result;
switch (pdrv) {
case SDCARD :
case FLASH_SPI:
SPI_FLASH_Init();
return disk_status(FLASH_SPI);
}
return STA_NOINIT;
}
5-实际上是调用SPI_FLASH_BufferRead(buff, sector*4096, count*4096);
在 disk_read 中的参数要进行转变,每个扇区是4096个字节,所以物理初始地址=扇区号*4096;字节数=扇区数*4096
DRESULT disk_read (
BYTE pdrv, /* 型号:Flash 还是SD */
BYTE *buff, /* 指针传递回数据 */
DWORD sector, /* 扇区号 */
UINT count /* 要读取的扇区数 */
)
{
DRESULT res;
// int result;
switch (pdrv) {
case SDCARD :
case FLASH_SPI:
SPI_FLASH_BufferRead(buff, sector*4096, count*4096);
res= RES_OK;
return res;
}
return RES_PARERR;
}
6-本质上是调用了 、
SPI_FLASH_SectorErase(sector*4096);
SPI_FLASH_BufferWrite((BYTE *)buff, sector*4096, count*4096);
在flash中,写入数据要先擦除再写入
ps:在写入时,建议加一个等待busy,(具体看我上一篇 )这样更稳妥。但是我看野火源码中没有加,似乎问题不大?
#if _USE_WRITE
DRESULT disk_write (
BYTE pdrv, /* Physical drive nmuber to identify the drive */
const BYTE *buff, /* Data to be written */
DWORD sector, /* Sector address in LBA */
UINT count /* Number of sectors to write */
)
{
DRESULT res;
// int result;
switch (pdrv) {
case SDCARD :
case FLASH_SPI:
SPI_FLASH_SectorErase(sector*4096);
SPI_FLASH_BufferWrite((BYTE *)buff, sector*4096, count*4096);
res= RES_OK;
return res;
}
return RES_PARERR;
}
#endif
7-这边主要时杂项指令,调用这个函数,通过buff指针传递回去;
这边void*指针的数据传递,需关注
*(DWORD*) buff=2048;
*(WORD*) buff=4096;
/*-----------------------------------------------------------------------*/
/* Miscellaneous Functions */
/*-----------------------------------------------------------------------*/
#if _USE_IOCTL
DRESULT disk_ioctl (
BYTE pdrv, /* Physical drive nmuber (0..) */
BYTE cmd, /* Control code */
void *buff /* Buffer to send/receive control data */
)
{
DRESULT res;
// int result;
switch (pdrv) {
case SDCARD :
case FLASH_SPI:
switch(cmd)
{
//扇区数量
case GET_SECTOR_COUNT:
*(DWORD*) buff=2048;
break;
//每个扇区大小
case GET_SECTOR_SIZE:
*(WORD*) buff=4096;
break;
//每次擦除扇区的最小个数
case GET_BLOCK_SIZE:
*(DWORD*) buff=1;
break;
}
res= RES_OK;
return res;
}
return RES_PARERR;
}
#endif
8-时间项,理论上要关联rtc,但是这边直接return 0
DWORD get_fattime (void)
{
return 0;
}
ffconf.h的底层硬件操作的移植
看英文注释即可
/*---------------------------------------------------------------------------/
/ FatFs - FAT file system module configuration file R0.11a (C)ChaN, 2015
/---------------------------------------------------------------------------*/
#define _FFCONF 64180 /* Revision ID */
/*---------------------------------------------------------------------------/
/ Function Configurations
/---------------------------------------------------------------------------*/
#define _FS_READONLY 0
/* This option switches read-only configuration. (0:Read/Write or 1:Read-only)
/ Read-only configuration removes writing API functions, f_write(), f_sync(),
/ f_unlink(), f_mkdir(), f_chmod(), f_rename(), f_truncate(), f_getfree()
/ and optional writing functions as well. */
#define _FS_MINIMIZE 0
/* This option defines minimization level to remove some basic API functions.
/
/ 0: All basic functions are enabled.
/ 1: f_stat(), f_getfree(), f_unlink(), f_mkdir(), f_chmod(), f_utime(),
/ f_truncate() and f_rename() function are removed.
/ 2: f_opendir(), f_readdir() and f_closedir() are removed in addition to 1.
/ 3: f_lseek() function is removed in addition to 2. */
#define _USE_STRFUNC 1
/* This option switches string functions, f_gets(), f_putc(), f_puts() and
/ f_printf().
/
/ 0: Disable string functions.
/ 1: Enable without LF-CRLF conversion.
/ 2: Enable with LF-CRLF conversion. */
#define _USE_FIND 0
/* This option switches filtered directory read feature and related functions,
/ f_findfirst() and f_findnext(). (0:Disable or 1:Enable) */
#define _USE_MKFS 1 //在格式化时刻,需要配置成1
/* This option switches f_mkfs() function. (0:Disable or 1:Enable) */
#define _USE_FASTSEEK 0
/* This option switches fast seek feature. (0:Disable or 1:Enable) */
#define _USE_LABEL 0
/* This option switches volume label functions, f_getlabel() and f_setlabel().
/ (0:Disable or 1:Enable) */
#define _USE_FORWARD 0
/* This option switches f_forward() function. (0:Disable or 1:Enable)
/ To enable it, also _FS_TINY need to be set to 1. */
/*---------------------------------------------------------------------------/
/ Locale and Namespace Configurations
/---------------------------------------------------------------------------*/
#define _CODE_PAGE 932
/* This option specifies the OEM code page to be used on the target system.
/ Incorrect setting of the code page can cause a file open failure.
/
/ 1 - ASCII (No extended character. Non-LFN cfg. only)
/ 437 - U.S.
/ 720 - Arabic
/ 737 - Greek
/ 771 - KBL
/ 775 - Baltic
/ 850 - Latin 1
/ 852 - Latin 2
/ 855 - Cyrillic
/ 857 - Turkish
/ 860 - Portuguese
/ 861 - Icelandic
/ 862 - Hebrew
/ 863 - Canadian French
/ 864 - Arabic
/ 865 - Nordic
/ 866 - Russian
/ 869 - Greek 2
/ 932 - Japanese (DBCS)
/ 936 - Simplified Chinese (DBCS)
/ 949 - Korean (DBCS)
/ 950 - Traditional Chinese (DBCS)
*/
#define _USE_LFN 0
#define _MAX_LFN 255
/* The _USE_LFN option switches the LFN feature.
/
/ 0: Disable LFN feature. _MAX_LFN has no effect.
/ 1: Enable LFN with static working buffer on the BSS. Always NOT thread-safe.
/ 2: Enable LFN with dynamic working buffer on the STACK.
/ 3: Enable LFN with dynamic working buffer on the HEAP.
/
/ When enable the LFN feature, Unicode handling functions (option/unicode.c) must
/ be added to the project. The LFN working buffer occupies (_MAX_LFN + 1) * 2 bytes.
/ When use stack for the working buffer, take care on stack overflow. When use heap
/ memory for the working buffer, memory management functions, ff_memalloc() and
/ ff_memfree(), must be added to the project. */
#define _LFN_UNICODE 0
/* This option switches character encoding on the API. (0:ANSI/OEM or 1:Unicode)
/ To use Unicode string for the path name, enable LFN feature and set _LFN_UNICODE
/ to 1. This option also affects behavior of string I/O functions. */
#define _STRF_ENCODE 3
/* When _LFN_UNICODE is 1, this option selects the character encoding on the file to
/ be read/written via string I/O functions, f_gets(), f_putc(), f_puts and f_printf().
/
/ 0: ANSI/OEM
/ 1: UTF-16LE
/ 2: UTF-16BE
/ 3: UTF-8
/
/ When _LFN_UNICODE is 0, this option has no effect. */
#define _FS_RPATH 0
/* This option configures relative path feature.
/
/ 0: Disable relative path feature and remove related functions.
/ 1: Enable relative path feature. f_chdir() and f_chdrive() are available.
/ 2: f_getcwd() function is available in addition to 1.
/
/ Note that directory items read via f_readdir() are affected by this option. */
/*---------------------------------------------------------------------------/
/ Drive/Volume Configurations
/---------------------------------------------------------------------------*/
#define _VOLUMES 2
/* Number of volumes (logical drives) to be used. */
#define _STR_VOLUME_ID 0
#define _VOLUME_STRS "RAM","NAND","CF","SD1","SD2","USB1","USB2","USB3"
/* _STR_VOLUME_ID option switches string volume ID feature.
/ When _STR_VOLUME_ID is set to 1, also pre-defined strings can be used as drive
/ number in the path name. _VOLUME_STRS defines the drive ID strings for each
/ logical drives. Number of items must be equal to _VOLUMES. Valid characters for
/ the drive ID strings are: A-Z and 0-9. */
#define _MULTI_PARTITION 0
/* This option switches multi-partition feature. By default (0), each logical drive
/ number is bound to the same physical drive number and only an FAT volume found on
/ the physical drive will be mounted. When multi-partition feature is enabled (1),
/ each logical drive number is bound to arbitrary physical drive and partition
/ listed in the VolToPart[]. Also f_fdisk() funciton will be available. */
#define _MIN_SS 512
#define _MAX_SS 4096
/* These options configure the range of sector size to be supported. (512, 1024,
/ 2048 or 4096) Always set both 512 for most systems, all type of memory cards and
/ harddisk. But a larger value may be required for on-board flash memory and some
/ type of optical media. When _MAX_SS is larger than _MIN_SS, FatFs is configured
/ to variable sector size and GET_SECTOR_SIZE command must be implemented to the
/ disk_ioctl() function. */
#define _USE_TRIM 0
/* This option switches ATA-TRIM feature. (0:Disable or 1:Enable)
/ To enable Trim feature, also CTRL_TRIM command should be implemented to the
/ disk_ioctl() function. */
#define _FS_NOFSINFO 0
/* If you need to know correct free space on the FAT32 volume, set bit 0 of this
/ option, and f_getfree() function at first time after volume mount will force
/ a full FAT scan. Bit 1 controls the use of last allocated cluster number.
/
/ bit0=0: Use free cluster count in the FSINFO if available.
/ bit0=1: Do not trust free cluster count in the FSINFO.
/ bit1=0: Use last allocated cluster number in the FSINFO if available.
/ bit1=1: Do not trust last allocated cluster number in the FSINFO.
*/
/*---------------------------------------------------------------------------/
/ System Configurations
/---------------------------------------------------------------------------*/
#define _FS_TINY 0
/* This option switches tiny buffer configuration. (0:Normal or 1:Tiny)
/ At the tiny configuration, size of the file object (FIL) is reduced _MAX_SS
/ bytes. Instead of private sector buffer eliminated from the file object,
/ common sector buffer in the file system object (FATFS) is used for the file
/ data transfer. */
#define _FS_NORTC 0
#define _NORTC_MON 1
#define _NORTC_MDAY 1
#define _NORTC_YEAR 2015
/* The _FS_NORTC option switches timestamp feature. If the system does not have
/ an RTC function or valid timestamp is not needed, set _FS_NORTC to 1 to disable
/ the timestamp feature. All objects modified by FatFs will have a fixed timestamp
/ defined by _NORTC_MON, _NORTC_MDAY and _NORTC_YEAR.
/ When timestamp feature is enabled (_FS_NORTC == 0), get_fattime() function need
/ to be added to the project to read current time form RTC. _NORTC_MON,
/ _NORTC_MDAY and _NORTC_YEAR have no effect.
/ These options have no effect at read-only configuration (_FS_READONLY == 1). */
#define _FS_LOCK 0
/* The _FS_LOCK option switches file lock feature to control duplicated file open
/ and illegal operation to open objects. This option must be 0 when _FS_READONLY
/ is 1.
/
/ 0: Disable file lock feature. To avoid volume corruption, application program
/ should avoid illegal open, remove and rename to the open objects.
/ >0: Enable file lock feature. The value defines how many files/sub-directories
/ can be opened simultaneously under file lock control. Note that the file
/ lock feature is independent of re-entrancy. */
#define _FS_REENTRANT 0
#define _FS_TIMEOUT 1000
#define _SYNC_t HANDLE
/* The _FS_REENTRANT option switches the re-entrancy (thread safe) of the FatFs
/ module itself. Note that regardless of this option, file access to different
/ volume is always re-entrant and volume control functions, f_mount(), f_mkfs()
/ and f_fdisk() function, are always not re-entrant. Only file/directory access
/ to the same volume is under control of this feature.
/
/ 0: Disable re-entrancy. _FS_TIMEOUT and _SYNC_t have no effect.
/ 1: Enable re-entrancy. Also user provided synchronization handlers,
/ ff_req_grant(), ff_rel_grant(), ff_del_syncobj() and ff_cre_syncobj()
/ function, must be added to the project. Samples are available in
/ option/syscall.c.
/
/ The _FS_TIMEOUT defines timeout period in unit of time tick.
/ The _SYNC_t defines O/S dependent sync object type. e.g. HANDLE, ID, OS_EVENT*,
/ SemaphoreHandle_t and etc.. A header file for O/S definitions needs to be
/ included somewhere in the scope of ff.c. */
#define _WORD_ACCESS 0
/* The _WORD_ACCESS option is an only platform dependent option. It defines
/ which access method is used to the word data on the FAT volume.
/
/ 0: Byte-by-byte access. Always compatible with all platforms.
/ 1: Word access. Do not choose this unless under both the following conditions.
/
/ * Address misaligned memory access is always allowed to ALL instructions.
/ * Byte order on the memory is little-endian.
/
/ If it is the case, _WORD_ACCESS can also be set to 1 to reduce code size.
/ Following table shows allowable settings of some type of processors.
/
/ ARM7TDMI 0 *2 ColdFire 0 *1 V850E 0 *2
/ Cortex-M3 0 *3 Z80 0/1 V850ES 0/1
/ Cortex-M0 0 *2 x86 0/1 TLCS-870 0/1
/ AVR 0/1 RX600(LE) 0/1 TLCS-900 0/1
/ AVR32 0 *1 RL78 0 *2 R32C 0 *2
/ PIC18 0/1 SH-2 0 *1 M16C 0/1
/ PIC24 0 *2 H8S 0 *1 MSP430 0 *2
/ PIC32 0 *1 H8/300H 0 *1 8051 0/1
/
/ *1:Big-endian.
/ *2:Unaligned memory access is not supported.
/ *3:Some compilers generate LDM/STM for mem_cpy function.
*/
下面可以在mian中操作:
1 - 包含对应头文件,fatfs包含#include "ff.H"
#include "stm32f10x.h" // Device header
#include "./led/bsp_led.h"
#include "BSP_USART.H"
#include "./flash/bsp_spi_flash.h"
#include "ff.H"
2 - 相关变量定义
1-一个文件系统对应一个系统对象FATFS。(这边不能定义在main函数中,容易栈溢出)
例如SD卡和flash中都要用FATFS,则需要定义两个系统对象
2- 文件对象FIL,具体的文件类型;
3- 状态,文件操作结果FRESULT
FATFS fs_falsh; /* FatFs文件系统对象 */
uint8_t writebuf[]="这是fatfs写入测试";
uint8_t readbuf[30]="";
FRESULT state; /* 文件操作结果 */
FIL fp; /* 文件对象 */
UINT write_size=0;
UINT read_size=0;
3-文件启动
1-启动首先第一步,是挂载系统。如果系统存在,挂载结果为FR_OK;如果系统不存在FR_NO_FILESYSTEM,则格式化处理;
2-格式化完成后,需要取消挂载一次,在重新挂载;
printf("测试开始:\n");
//启动文件系统
state=f_mount (&fs_falsh,"1:",1);
if(state==FR_OK)
{
printf("系统挂载完成:\n");
}
else
{
printf("系统挂载失败 ,%d:\n",state);
}
if(state==FR_NO_FILESYSTEM)
{
printf("格式化处理\n");
state=f_mkfs("1:",0,0);
if(state==FR_OK)
{
//格式化后,先取消挂载再重新挂载
printf("》FLASH已成功格式化文件系统。\r\n");
f_mount (NULL,"1:",1); //取消
f_mount (&fs_falsh,"1:",1); //重新挂栽
LED_GREEN;
}
else
{
printf("《《格式化失败。》》\r\n");
LED_RED;
}
//格式化后,先取消挂载再重新挂载
f_mount (NULL,"1:",1); //取消
f_mount (&fs_falsh,"1:",1); //重新挂栽
}
4,文件的读取,写入操作
1- 写入和读取文件,要先open,才能进行操作
2- 如果写入完成,立即读取,则需要调用lseek进行重新定位,才能读取,这是由文件光标位置决定,相当于退后光标;
3-如果先f_close,再进行读取,则无需lseek
4-文章来源:https://www.toymoban.com/news/detail-775692.html
//打开文件fopen
state=f_open (&fp,"1:abc.txt",FA_OPEN_ALWAYS|FA_WRITE|FA_READ );
if(state==FR_OK)
{
printf("open abc ok\n");
state=f_write(&fp,writebuf,sizeof(writebuf),&write_size);
if(state==FR_OK)
{
printf("写入完成 %d 字节",write_size);
printf("写入后立马读取\r\n");//写入后立即读取,需要lseek重新定位
f_lseek(&fp,4);
state=f_read(&fp,readbuf,f_size(&fp),&read_size);
if(state==FR_OK)
{
printf("读取完成 %d 字节\n",read_size);
printf("内容是:%s",readbuf);
LED_RED;
}
}
f_close(&fp);
}
else
{
printf("not open");
}
printf("****** 即将进行文件读取测试... ******\r\n");
state=f_open (&fp,"1:abc.txt",FA_OPEN_ALWAYS|FA_READ);
if(state==FR_OK)
{
printf("open abc ok with read\n");
state=f_read(&fp,readbuf,f_size(&fp),&read_size);
if(state==FR_OK)
{
printf("读取完成 %d 字节\n",read_size);
printf("内容是:%s",readbuf);
LED_RED;
}
}
以上,文件系统FATFS移植,主体部分完成。文章来源地址https://www.toymoban.com/news/detail-775692.html
到了这里,关于STM32-基本知识梳理8-FATFS文件系统移植的文章就介绍完了。如果您还想了解更多内容,请在右上角搜索TOY模板网以前的文章或继续浏览下面的相关文章,希望大家以后多多支持TOY模板网!