Linux driver Development What to know (iii)------CONTAINER_OF, timers and system calls

CONTAINER_OF Macro

Defined:

/**

* Container_of-cast a member of a structure out to the containing structure

* @ptr: The pointer to the member.

* @type: The type of the container struct this was embedded in.

* @member: The name of the member within the struct.

*

*/

#define CONTAINER_OF (PTR, type, member) ({\

Const typeof (((type *) 0)->member) * __mptr = (PTR); \

(Type *) ((char *) __mptr-offsetof (type, member)); })

#endif

The first step:

Const typeof ((type *) 0->member) * __mptr= (PTR)

Struct list_head *__mptr=(ptr)

Step Two:

(Type *) (char *) __mptr-offsetof (Type,menber)

0x***-offset

Step Three:offsetof (Type,member)

((TYPE *) 0)->member

& (TYPE *) 0)->menber

Kernel timers

Clock interrupt

HZ: constant determines the frequency at which the clock occurs

Tick: Time interval at which time breaks occur tick=1/hz

Jiffies: Core variables,

struct timer_list

{

unsigned long expires; Timeout period

void (*function) (unsigned long0) processing function

unsigned long data; The parameters passed to it when the kernel call time-out handler function

}

Static initialization

Timer_initalizer (_function,_expires,_data)

Define_time (_name,_function,_expires,_data);

_name The name of the kernel timer variable you want to define

Void init_timer (struct timer_list *timer);

Modify Timer

650) this.width=650; "Src=" Http://s5.51cto.com/wyfs02/M02/8A/BE/wKiom1g6MEviN7l2AACB4QttqMw102.png-wh_500x0-wm_3 -wmp_4-s_1419104650.png "title=" image 1.png "alt=" Wkiom1g6mevin7l2aacb4qttqmw102.png-wh_50 "/>

Init_timer (time) initialize timer, pass address

#define Bark_timer 5

#deifne Bark_data 100

Xxx.expires = jiffies+hz*bark_timer;//bark_timer=5 Five seconds

Xxx.function=xxxxxx

Xxxx.data=bark_data;

Add_timer (&xxxxx

Del_timer ()

2 system calls

ANSI c fopen( library function for standard C)

UNIX C Open (System call)

User-state programs generally do not have access to kernel-state resources, only through interrupts or system calls to enter the kernel state from the user state

2.6.35 has a 366 Kernel system call for application services.

The principle of system invocation:

The application first populates the register with the appropriate values and then calls a special instruction to jump to a fixed location in the kernel, which finds the appropriate function execution based on the fixed values populated by the application.

appropriate value:arch/arm/asm/unistd.h

Call Special instruction Arm->swi(soft interrupt)

X86 0x80

fixed position: In the arm system, the application jumps to Entry-conmmon->vector_swi and uses the sys_call_table (Calls. S)

User space Programming fd=open("aa.txt"... )

5-> Register

Swi

____________________

Vector_swi

SYS_CALL_TABLE[5]

Sys_open

SYSCALL_DEFINR3 Stitching

1) Add a new kernel function

Vim ARCH/ARM/KERNEL/SYS_ARM.C

asmlinkage int sys_add (int x,int y)

{

PRINTK ("entry sys_add!\n");

Return X+y;

}

2) Update header file arch/arm/include/asm/unistd.h

#define __NR_ADD (__nr_syscall_base+366)

3) update the system call table (calls. S)

Call (Sys_add)

The final make can be

This article is from the "Mao Scattered People" blog, please be sure to keep this source http://songmao.blog.51cto.com/11700139/1876960

Linux driver Development What to know (iii)------CONTAINER_OF, timers and system calls