Working of System Calls in Operating Systems

-
System call is used in user application to request some hardware/Operating system service access.
When user call a system call it actually call some user level library (glibc/ulibc) which is in between User and Operating System.They are implemented as short pieces of assembly code named as system call stub.Which actually place system call number associated to the particular system call to a register (For x86 EAX) , place the arguments other registers (for x86 EBX,ECX etc.)  and issue a trap instruction ( in X86 calling assembly call INT 0x80, "sc" for powerpc) to switch from user mode to kernel privileged mode.


In trap handler kernel code copies arguments to kernel stack and call corresponding system call handler.


According to system call dispatcher a new system call may be attached to kernel in compile time or run time.


After running system code handler it return by putting return value in some register. (In PPC r3 register)


Note: While passing arguments from user can pass args in several way.
Some implementation in x86
1) push to stack
open:
    push    dword mode
    push    dword flags
    push    dword path
    mov eax, 5
    push    eax     ; Or any other dword
    int 80h
    add esp, byte 16
 
2) linux way - Put in registers
open:
    mov eax, 5
    mov ebx, path
    mov ecx, flags
    mov edx, mode
    int 80h



System call implementation in i386
http://www.tldp.org/HOWTO/html_single/Implement-Sys-Call-Linux-2.6-i386/
http://cse.yeditepe.edu.tr/~moergin/courses/cse331/addsystemcall.html


On ARM:
Its similar to x86/PPC ,It passes arguments through r0-r6,r7 is used for passing system call number.We call 'SWI' instruction to similar to "INT 80" in x86.
movl r7,#378
mov  r0,#25
mov r1,#35
SWI 0
mov r7,#1 
SWI  0 
Dynamic attachment
http://linux.die.net/lkmpg/x978.html


Update: in x86 __asmlinkage modifier is used in system side system call function implementation because we tell the function to use stack instead of registers for arguments of the function.
Post P4 processor use sysenter to enter in syscall mode instead of int 80,

 In very detail :
http://12000.org/my_notes/system_calls_in_linux/system_call_in_linux/index.htm

Comments

Post a Comment

Popular posts from this blog

Airtel Digital tv remote factory reset and reprogram

-
Following is the way to reset Airtel Digial tv universal remote reset procedure. 1. Press red button and ok button together untill red light blinks twice. 2. Release the buttons and press 980 buttons one by one (not together)   3. Red light will blink again two times, now press ok button. Reprogramming There are 5 basic functions which your new Airtel remote needs to learn about your TV 1. TV Power code 2. TV Mute code 3. AV code 4. TV VOL + code 5. TV VOL – code Steps for Universal remote setup: Step 1:  Bring Airtel digital TV remote in learning mode. Press and hold ‘’OK + Numeric 2” key for 3 seconds. The LED blinks twice to indicate that your Airtel remote is ready to learn. Step 2 :  Hold the source remote and Airtel remote in front of each other ( 3cm-6m apart). Step 3:  Press & release the key (First on Airtel remote and then on Source remote) which needs to be programmed in Airtel remote. Example: Press and release the TV Power key on both remotes and release. The A
Read more

Tracking Linux kworker threads

-
How to find out which part of kernel/module has created this workqueue. How to track a kworker-thread named for example ''kworker/0:3 to its origin in kernel-space? I found  this thread on lkml  that answers your question a little. (It seems even Linus himself was puzzled as to how to find out the origin of those threads.) Basically, there are two ways of doing this: $ echo workqueue:workqueue_queue_work > /sys/kernel/debug/tracing/set_event $ cat /sys/kernel/debug/tracing/trace_pipe > out.txt (wait a few secs) For this you will need  ftrace  to be compiled in your kernel, and to enable it with: mount -t debugfs nodev /sys/kernel/debug More information on the function tracer facilities of Linux is available in the  ftrace.txt documentation . This will output what threads are all doing, and is useful for tracing multiple small jobs. cat /proc/THE_OFFENDING_KWORKER/stack This will output the stack of a single thread doing a lot of work. It may allow you
Read more

Asynchronus I/O (AIO) in vxworks and Linux

-
In generic Asynchronus I/O (AIO) can be designed using User lavel or in Kernel level. In user level 1) Process needs async callback/notification create another thread and process which goes to sleep on blocking calls (read/write/ioctls/select) 2) One Async master handler (process/thread) keeps track.All the process needs the service submit job to them.The master handler handles the async fd in Blocking mode and reply to requesting processes (or call the callback function) using some async notificatin like signal. In Kernel Level 1) Call some system call and submit some job to kernel workQ .After work is complete kernel notify using signal to user space.   In VxWorks :  aioPxLib provide the POSIX 1003.1b standard.Which is task implementation of aio service. functions available: 1) aio_read() - initiate an asyn. read. 2) aio_write() - initiate an syn write 3) aio_listio() - Initiate a aio list upto LIO_MAX 4) aio_error() - Get an aio error status. 5) aio_return() - Get a
Read more