您的位置:首页 > 运维架构 > Linux

转载_Linux procfs开发指南(部分翻译并做了精简)

2013-08-09 14:24 330 查看
Linux Kernel Procfs Guide

Chapter 1. Introduction

/proc文件系统(procfs)在linux内核中是一个特殊的文件系统,是一个虚拟的文件系统,只存在于内存中

注:proc/sys是sysctl文件,不属于procfs

Chapter 2. Managing procfs entries

要使用procfs,需包含#include <linux/proc_fs.h>

1.Creating a regular file

struct proc_dir_entry* create_proc_entry(const char* name,mode_t mode, struct proc_dir_entry* parent);

parent为NULL时,表示文件建立在/proc目录下,失败返回NULL

create_proc_read_entry:创建并初始化一个procfs入口in one single call

2. Creating a symlink

struct proc_dir_entry* proc_symlink(const char* name, struct proc_dir_entry* parent, const char* dest);

在parent目录下创建一个从name到dest的符号链接,像:ln -s dest name

3. Creating a device

struct proc_dir_entry* proc_mknod(const char* name, mode_t mode,struct proc_dir_entry* parent, kdev_t rdev);

创建一个名为name设备文件,mode参数必须包含S_IFBLK或者S_IFCHR,像:mknod --mode=mode name rdev

4. Creating a directory

struct proc_dir_entry* proc_mkdir(const char* name, struct proc_dir_entry* parent);

在parent目录下创建名为name的目录

5.Removing an entry

void remove_proc_entry(const char* name, struct proc_dir_entry* parent);

从procfs中移除parent目录下的name入口。注意它不会递归移除

Chapter 3. Communicating with userland

procfs对文件的读写并不是直接的,而是通过call back函数来操作read_proc and/or write_proc

在create_proc_entry函数返回一个proc_dir_entry* entry,然后设置

entry->read_proc = read_proc_foo;

entry->write_proc = write_proc_foo;

1. Reading data

kernel向用户返回数据

int read_func(char* page, char** start, off_t off, int count,int* eof, void* data);

从start处的off偏移处开始写最多count个字节的内容到page中。

2. Writing data

用户写数据到kernel

int write_func(struct file* file, const char* buffer, unsigned long count, void* data);

从buffer中读count字节,buffer不存在于kernel memory空间,所以需要使用copy_from_user函数,file常NULL

3.A single call back for many files

struct proc_dir_entry* entry;

struct my_file_data *file_data;

file_data = kmalloc(sizeof(struct my_file_data), GFP_KERNEL);

entry->data = file_data;

int foo_read_func(char *page, char **start, off_t off,

int count, int *eof, void *data)

{

int len;

if(data == file_data) {

/* special case for this file */

} else {

/* normal processing */

}

return len;

}

在 remove_proc_entry是要free掉data

Chapter 4. Tips and tricks

1. Convenience functions

struct proc_dir_entry* create_proc_read_entry(const char* name,mode_t mode, struct proc_dir_entry* parent, read_proc_t* read_proc, void* data);

将只需要读操作的create_proc_entry简化掉了

2. Modules

struct proc_dir_entry* entry;

entry->owner = THIS_MODULE;

目前发现没有这个owner参数了,不用关注。

3. Mode and ownership

设置模式和权限

struct proc_dir_entry* entry;

entry->mode = S_IWUSR |S_IRUSR | S_IRGRP | S_IROTH;

entry->uid = 0;

entry->gid = 100;

Chapter 5. Example

board (http://www.lart.tudelft.nl/), which is

* sponsored by the Mobile Multi-media Communications

* (http://www.mmc.tudelft.nl/) and Ubiquitous Communications

* (http://www.ubicom.tudelft.nl/) projects.

*

* The author can be reached at:

*

* Erik Mouw

* Information and Communication Theory Group

* Faculty of Information Technology and Systems

* Delft University of Technology

* P.O. Box 5031

* 2600 GA Delft

* The Netherlands

*

*

* This program is free software; you can redistribute

* it and/or modify it under the terms of the GNU General

* Public License as published by the Free Software

* Foundation; either version 2 of the License, or (at your

* option) any later version.

*

* This program is distributed in the hope that it will be

* useful, but WITHOUT ANY WARRANTY; without even the implied

* warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR

* PURPOSE. See the GNU General Public License for more

* details.

*

* You should have received a copy of the GNU General Public

* License along with this program; if not, write to the

* Free Software Foundation, Inc., 59 Temple Place,* Suite 330, Boston, MA 02111-1307 USA

*

*/

#include <linux/module.h>

#include <linux/kernel.h>

#include <linux/init.h>

#include <linux/proc_fs.h>

#include <linux/sched.h>

#include <asm/uaccess.h>

#define MODULE_VERSION "1.0"

#define MODULE_NAME "procfs_example"

#define FOOBAR_LEN 8

struct fb_data_t {

char name[FOOBAR_LEN + 1];

char value[FOOBAR_LEN + 1];

};

static struct proc_dir_entry *example_dir, *foo_file,

*bar_file, *jiffies_file, *tty_device, *symlink;

struct fb_data_t foo_data, bar_data;

static int proc_read_jiffies(char *page, char **start,

off_t off, int count,

int *eof, void *data)

{

int len;

MOD_INC_USE_COUNT;

len = sprintf(page, "jiffies = %ld\n",

jiffies);

MOD_DEC_USE_COUNT;

return len;

}

static int proc_read_foobar(char *page, char **start,

off_t off, int count,

int *eof, void *data)

{

int len;

struct fb_data_t *fb_data = (struct fb_data_t *)data;

MOD_INC_USE_COUNT;

len = sprintf(page, "%s = ’%s’\n",

fb_data->name, fb_data->value);

MOD_DEC_USE_COUNT;

return len;

}

static int proc_write_foobar(struct file *file,

const char *buffer,

unsigned long count,

void *data)

{

int len;

struct fb_data_t *fb_data = (struct fb_data_t *)data;

MOD_INC_USE_COUNT;

if(count > FOOBAR_LEN)

len = FOOBAR_LEN;

else

len = count;

if(copy_from_user(fb_data->value, buffer, len)) {

MOD_DEC_USE_COUNT;

return -EFAULT;

}

fb_data->value[len] = ’\0’;

MOD_DEC_USE_COUNT;

return len;

}

static int __init init_procfs_example(void)

{

int rv = 0;

/* create directory */

example_dir = proc_mkdir(MODULE_NAME, NULL);

if(example_dir == NULL) {

rv = -ENOMEM;

goto out;

}

example_dir->owner = THIS_MODULE;

/* create jiffies using convenience function */

jiffies_file = create_proc_read_entry("jiffies",

0444, example_dir,

proc_read_jiffies,

NULL);

if(jiffies_file == NULL) {

rv = -ENOMEM;

goto no_jiffies;

}

jiffies_file->owner = THIS_MODULE;

/* create foo and bar files using same callback

* functions

*/

foo_file = create_proc_entry("foo", 0644, example_dir);

if(foo_file == NULL) {

rv = -ENOMEM;

goto no_foo;

}

strcpy(foo_data.name, "foo");

strcpy(foo_data.value, "foo");

foo_file->data = &foo_data;

foo_file->read_proc = proc_read_foobar;

foo_file->write_proc = proc_write_foobar;

foo_file->owner = THIS_MODULE;

bar_file = create_proc_entry("bar", 0644, example_dir);

if(bar_file == NULL) {

rv = -ENOMEM;

goto no_bar;

}

strcpy(bar_data.name, "bar");

strcpy(bar_data.value, "bar");

bar_file->data = &bar_data;

bar_file->read_proc = proc_read_foobar;

bar_file->write_proc = proc_write_foobar;

bar_file->owner = THIS_MODULE;

/* create tty device */

tty_device = proc_mknod("tty", S_IFCHR | 0666,

example_dir, MKDEV(5, 0));

if(tty_device == NULL) {

rv = -ENOMEM;

goto no_tty;

}

tty_device->owner = THIS_MODULE;

/* create symlink */

symlink = proc_symlink("jiffies_too", example_dir,

"jiffies");

if(symlink == NULL) {

rv = -ENOMEM;

goto no_symlink;

}

symlink->owner = THIS_MODULE;

/* everything OK */

printk(KERN_INFO "%s %s initialised\n",

MODULE_NAME, MODULE_VERSION);

return 0;

no_symlink:

remove_proc_entry("tty", example_dir);

no_tty:

remove_proc_entry("bar", example_dir);

no_bar:

remove_proc_entry("foo", example_dir);

no_foo:

remove_proc_entry("jiffies", example_dir);

no_jiffies:

remove_proc_entry(MODULE_NAME, NULL);

out:

return rv;

}

static void __exit cleanup_procfs_example(void)

{

remove_proc_entry("jiffies_too", example_dir);

remove_proc_entry("tty", example_dir);

remove_proc_entry("bar", example_dir);

remove_proc_entry("foo", example_dir);

remove_proc_entry("jiffies", example_dir);

remove_proc_entry(MODULE_NAME, NULL);

printk(KERN_INFO "%s %s removed\n",

MODULE_NAME, MODULE_VERSION);

}

module_init(init_procfs_example);

module_exit(cleanup_procfs_example);

MODULE_AUTHOR("Erik Mouw");

MODULE_DESCRIPTION("procfs examples");

EXPORT_NO_SYMBOLS;

The Linux Kernel Module Programming Guide

Chapter 5. The /proc File System

1. The /proc File System

proc_register_dynamic防止kernel节点号冲突

1.定义一个文件结构

struct proc_dir_entry *OurProcFile;

2.创建一个proc文件

OurProcFile = create_proc_entry(PROCFS_NAME,0644,NULL);

3.给proc文件结构赋值

OurProcFile->read_proc = procfile_read; //指定相应读函数

OurProcFile->write_proc = procfile_write; //指定相应写函数

OurProcFile->mode = S_IFREG | S_IRUGO;

OurProcFile->uid = 0;

OurProcFile->gid = 0;

OurProcFile->size = 37;

初始化完成。

2. Read and Write a /proc File

从用户空间到kernel空间数据传递使用copy_from_user or get_user,因为linux内存是分段的,不是一个独立的地址,每一个进程都有一个独立的地址空间。比如:A有0x1234,B也有0x1234;

系统会自动处理kernel的访问和写。

int procfile_read(char *buffer,char **buffer_location,off_t offset, int buffer_length, int *eof, void *data)

{

........................

if (offset > 0) /* we have finished to read, return 0 */

{

ret = 0;

} else /* fill the buffer, return the buffer size */

{

memcpy(buffer, procfs_buffer, procfs_buffer_size);

ret = procfs_buffer_size;

}

return ret;

}

int procfile_write(struct file *file, const char *buffer, unsigned long count,void *data)

{

...............

if ( copy_from_user(procfs_buffer, buffer, procfs_buffer_size) )

{

return ?EFAULT;

}

return procfs_buffer_size;

}

3. Manage /proc file with standard filesystem

两个结构体比较重要

struct inode_operations 包含了struct file_operations指针

static struct inode_operations Inode_Ops_4_Our_Proc_File =

{

.permission = module_permission, /* check for permissions */

};

int init_module()

{

.............

Our_Proc_File?>proc_iops = &Inode_Ops_4_Our_Proc_File;

Our_Proc_File?>mode = S_IFREG | S_IRUGO | S_IWUSR;

.............

}

static int module_permission(struct inode *inode, int op, struct nameidata *foo)

{

/*

* We allow everybody to read from our module, but

* only root (uid 0) may write to it

*/

if (op == 4 || (op == 2 && current?>euid == 0))

return 0;

/*

* If it's anything else, access is denied

*/

return ?EACCES;

}

4. Manage /proc file with seq_file

使用seq_file库来管理proc

头文件:#include <linux/seq_file.h> /* for seq_file */

static struct seq_operations my_seq_ops = {

.start = my_seq_start,

.next = my_seq_next,

.stop = my_seq_stop,

.show = my_seq_show

};

static struct file_operations my_file_ops = {

.owner = THIS_MODULE,

.open = my_open,

.read = seq_read,

.llseek = seq_lseek,

.release = seq_release

};

static int my_open(struct inode *inode, struct file *file)

{

return seq_open(file, &my_seq_ops);

}

static int my_seq_show(struct seq_file *s, void *v)

{

loff_t *spos = (loff_t *) v;

seq_printf(s, "%Ld\n", *spos);

return 0;

}

static void *my_seq_next(struct seq_file *s, void *v, loff_t *pos)

{

unsigned long *tmp_v = (unsigned long *)v;

(*tmp_v)++;

(*pos)++;

return NULL;

}

static void *my_seq_start(struct seq_file *s, loff_t *pos)

{

static unsigned long counter = 0;

/* beginning a new sequence ? */

if ( *pos == 0 )

{

/* yes => return a non null value to begin the sequence */

return &counter;

}

else

{

/* no => it's the end of the sequence, return end to stop reading */

*pos = 0;

return NULL;

}

}

Essential Linux Device Drivers 附录C翻译

当故障的原因很难确定时,监控和分析由procfs提供的数据可能会提供帮助。

但是,当数据量很大时,procfs的read()实现变得很复杂。Seq 文件接口是一种内核

提供的帮助机制用来简化这样的实现。Seq文件接口使得procfs操作干净又整洁。

让我们先介绍一个逐步变得复杂的procfs read()实现,然后看看Seq文件接口如何将

这个复杂的实现变得优雅。我们也会逐步跟新2.6内核中还没有采用seq file的驱动.

Seq 文件接口的优点

通过一个例子来看看Seq的优点吧。同通常的设备驱动一样,假设你有某一个数据结构的链表,

每一个节点都包含一个字符串域(称为info)。一下代码C1通过一个/proc/readme文件将其输出

到用户空间。当用户读这个文件,procfs的read()方法的具体实现,readme_proc将被调用,它将

遍历链表,将所有节点的info域传递给文件系统缓冲区。

C1:

#include <linux/kernel.h>

#include <linux/init.h>

#include <linux/module.h>

#include <linux/proc_fs.h>

MODULE_DESCRIPTION("test proc");

MODULE_LICENSE("GPL");

/* Private Data structure */

struct _mydrv_struct

{

/* ... */

struct list_head list; /* Link to the next node */

char info[10]; /* Info to pass via the procfs file */

/* ... */

};

static LIST_HEAD(mydrv_list); /* List Head */

static int readme_proc(char *page, char **start, off_t offset, int count, int *eof, void *data)

{

off_t thischunk_len = 0;

struct _mydrv_struct *p;

printk( KERN_DEBUG "offset is %d\n",offset );

/* Traverse the list and copy info into the supplied buffer */

list_for_each_entry(p, &mydrv_list, list)

{

thischunk_len += sprintf(page+thischunk_len, p->info);

}

*eof = 1; /* Indicate completion */

return thischunk_len;

}

static int test_proc_init_module(void)

{

int i;

static struct proc_dir_entry *entry = NULL ;

struct _mydrv_struct * mydrv_new;

/* Create /proc/readme */

entry = create_proc_entry("readme", S_IWUSR, NULL);

if (entry)

{

entry->read_proc = readme_proc;

}

/* Handcraft mydrv_list for testing purpose.

In the real world, device driver logic

maintains the list and populates the 'info' field */

for (i=0;i<100;i++)

{

mydrv_new = kmalloc(sizeof(struct _mydrv_struct), GFP_ATOMIC);

sprintf(mydrv_new->info, "Node No: %d\n", i);

list_add_tail(&mydrv_new->list, &mydrv_list);

}

return 0;

}

static void test_proc_exit_module(void)

{

remove_proc_entry("readme",0);

printk( KERN_DEBUG "Module test_proc exit\n" );

}

module_init(test_proc_init_module);

module_exit(test_proc_exit_module);

使用如下命令

bash> cat /proc/readme

Node No: 0

Node No: 1

...

Node No: 99

当read()方法被调用,它将提供一页的内存(译者注:通常是4k)用来传递信息到用户空间。

正如你所看到的,readme_proc的第一个参数是一个指向1页大小的缓冲区的指针。

第二个参数start用于当数据量超过1页时read方法的实现,在例子C2中我们可以明确它的使用方法。

接下来的两个参数非别用于指定当前读请求的偏移和要读的byte数量

*eof用来告诉调用者是否还需要读更多的数据。如果你注释掉了上面对*eof置一的语句,

readme_proc将被再次调用,且offset被指定为1190(节点0-节点99中info域最终显示的字符串数量)

readme_proc返回此次写入缓冲区的有效字符数。

在例子C1里,procfs read方法的返回被限定在1页以内。如果你将节点数量从100增加到500,将看到以下输出:
bash> cat /proc/readmeNode No: 0Node No: 1...Node No: 322proc_file_read: Apparent buffer overflow!

正如你所看到的,当超过4096byte的ascii字符被制造出来,产生了一个缓冲区溢出

为了解决这个问题,你需要修改c1的代码,使用以前提到的start变量

这将导致代码有点复杂,这样的实现基于以下原理:

1、readme_proc被调用多次,每次调用都将改变两个参数:每次要地区的最大数据量count和相对于

文件开始的偏移。每次的count值必须小于一页

2、每次调用结束后,offset增加readme_proc的返回值的大小

3、readme_proc的eof信号仅仅当总共产生的数据量小于等于本次请求的count+offset才发出。

如果eof没有被设置,readme_proc将再次被调用,传递给它的offset将增加上一次调用readme_proc的

返回值

4、每次调用结束后,只有start指向的数据是有效的并被传递给调用者

为了更好地理解操作过程,

C2代码将打印每次被调用的参数: start、offset、count、page

通过以下修改,你的代码可以摆脱限制,输出大量的信息
bash> cat /proc/readmeNode No: 0Node No: 1...Node No: 499

C2:

#include <linux/kernel.h>

#include <linux/init.h>

#include <linux/module.h>

#include <linux/proc_fs.h>

MODULE_DESCRIPTION("test proc");

MODULE_AUTHOR("root ()");

MODULE_LICENSE("GPL");

/* Private Data structure */

struct _mydrv_struct

{

/* ... */

struct list_head list; /* Link to the next node */

char info[10]; /* Info to pass via the procfs file */

/* ... */

};

static LIST_HEAD(mydrv_list); /* List Head */

static int readme_proc(char *page, char **start, off_t offset, int count, int *eof, void *data)

{

int i = 0;

off_t thischunk_start = 0;

off_t thischunk_len = 0;

struct _mydrv_struct *p;

/* Loop thru the list collecting device info */

list_for_each_entry(p, &mydrv_list, list) {

thischunk_len += sprintf(page+thischunk_len, p->info);

/* Advance thischunk_start only to the extent that the next

* read will not result in total bytes more than (offset+count)

*/

if (thischunk_start + thischunk_len < offset) {

thischunk_start += thischunk_len;

thischunk_len = 0;

} else if (thischunk_start + thischunk_len > offset+count) {

break;

} else {

continue;

}

}

/* Actual start */

*start = page + (offset - thischunk_start);

/* Calculate number of written bytes */

thischunk_len -= (offset - thischunk_start);

if (thischunk_len > count) {

thischunk_len = count;

} else {

*eof = 1;

}

return thischunk_len;

}

static int test_proc_init_module(void)

{

int i;

static struct proc_dir_entry *entry = NULL ;

struct _mydrv_struct * mydrv_new;

/* Create /proc/readme */

entry = create_proc_entry("readme", S_IWUSR, NULL);

if (entry)

{

entry->read_proc = readme_proc;

}

/* Handcraft mydrv_list for testing purpose.

In the real world, device driver logic

maintains the list and populates the 'info' field */

for (i=0;i<500;i++)

{

mydrv_new = kmalloc(sizeof(struct _mydrv_struct), GFP_ATOMIC);

sprintf(mydrv_new->info, "Node No: %d\n", i);

list_add_tail(&mydrv_new->list, &mydrv_list);

}

return 0;

}

static void test_proc_exit_module(void)

{

remove_proc_entry("readme",0);

printk( KERN_DEBUG "Module test_proc exit\n" );

}

module_init(test_proc_init_module);

module_exit(test_proc_exit_module);

当你面对C2中复杂的read实现感到前途黯淡时,Seq 文件接口前来救驾。

顾名思义,seq file interface 将procfs文件内容看做是序列化的对象

(Seq)编程接口提供的是这个序列化对象的迭代器

因此你的Seq代码必须实现以下的迭代器方法

1、start()将被seq interface首先调用,它初始化迭代子对象的位置并且返回插入的第一个迭代子对象

2、next()增加迭代器的位置,返回指向下一个迭代子的指针。这个函数不知道迭代子的内核结构,

将其看做黑盒

3、show()当用户读procfs的文件时,解释传递给它的迭代子,转换成要输出给用户的字符串。

这个方法充分利用了内核提供的 seq_printf(), seq_putc(), and seq_puts() 来格式化输出

4、stop()最后被调用用来清理

响应用户对相关proc文件的操作时,seq file interface会自动调用以上迭代器方法去产生输出。

你不用再去担心页大小和eof信号

C3

#include <linux/kernel.h>

#include <linux/init.h>

#include <linux/module.h>

#include <linux/proc_fs.h>

#include <linux/seq_file.h>

MODULE_DESCRIPTION("test proc");

MODULE_AUTHOR("root ()");

MODULE_LICENSE("GPL");

/* Private Data structure */

struct _mydrv_struct

{

/* ... */

struct list_head list; /* Link to the next node */

char info[10]; /* Info to pass via the procfs file */

/* ... */

};

static LIST_HEAD(mydrv_list); /* List Head */

/* start() method */

static void *

mydrv_seq_start(struct seq_file *seq, loff_t *pos)

{

struct _mydrv_struct *p;

loff_t off = 0;

/* The iterator at the requested offset */

list_for_each_entry(p, &mydrv_list, list) {

if (*pos == off++) return p;

}

return NULL;

}

/* next() method */

static void *

mydrv_seq_next(struct seq_file *seq, void *v, loff_t *pos)

{

/* 'v' is a pointer to the iterator returned by start() or

by the previous invocation of next() */

struct list_head *n = ((struct _mydrv_struct *)v)->list.next;

++*pos; /* Advance position */

/* Return the next iterator, which is the next node in the list */

return(n != &mydrv_list) ?

list_entry(n, struct _mydrv_struct, list) : NULL;

}

/* show() method */

static int

mydrv_seq_show(struct seq_file *seq, void *v)

{

const struct _mydrv_struct *p = v;

/* Interpret the iterator, 'v' */

seq_printf(seq, p->info);

return 0;

}

/* stop() method */

static void mydrv_seq_stop(struct seq_file *seq, void *v)

{

/* No cleanup needed in this example */

}

/* Define iterator operations */

static struct seq_operations mydrv_seq_ops = {

.start = mydrv_seq_start,

.next = mydrv_seq_next,

.stop = mydrv_seq_stop,

.show = mydrv_seq_show,

};

static int

mydrv_seq_open(struct inode *inode, struct file *file)

{

/* Register the operators */

return seq_open(file, &mydrv_seq_ops);

}

static struct file_operations mydrv_proc_fops = {

.owner = THIS_MODULE,

.open = mydrv_seq_open, /* User supplied */

.read = seq_read, /* Built-in helper function */

.llseek = seq_lseek, /* Built-in helper function */

.release = seq_release, /* Built-in helper funciton */

};

static int test_proc_init_module(void)

{

int i;

static struct proc_dir_entry *entry = NULL ;

struct _mydrv_struct * mydrv_new;

/* Create /proc/readme */

entry = create_proc_entry("readme", S_IWUSR, NULL);

if (entry)

{

entry->proc_fops = &mydrv_proc_fops;

}

/* Handcraft mydrv_list for testing purpose.

In the real world, device driver logic

maintains the list and populates the 'info' field */

for (i=0;i<500;i++)

{

mydrv_new = kmalloc(sizeof(struct _mydrv_struct), GFP_ATOMIC);

sprintf(mydrv_new->info, "Node No: %d\n", i);

list_add_tail(&mydrv_new->list, &mydrv_list);

}

return 0;

}

static void test_proc_exit_module(void)

{

remove_proc_entry("readme",0);

printk( KERN_DEBUG "Module test_proc exit\n" );

}

module_init(test_proc_init_module);

module_exit(test_proc_exit_module);

注意:如果不需要实现迭代输出的功能,使用single_open函数即可,具体可以看 /fs/proc 目录下的version.c文件,有实现的范例。
内容来自用户分享和网络整理,不保证内容的准确性,如有侵权内容,可联系管理员处理 点击这里给我发消息
标签: