This is the third part of the [chapter](http://0xax.gitbooks.io/linux-insides/content/SysCall/index.html) that describes system calls in the Linux kernel and we saw preparations after a system call caused by an userspace application and process of handling of a system call in the previous [part](http://0xax.gitbooks.io/linux-insides/content/SysCall/syscall-2.html). In this part we will look at two concepts that are very close to the system call concept, they are called `vsyscall` and `vdso`.
We already know what is it a `system call`. This is special routine in the Linux kernel which userspace application asks to do privileged task, like to read or to write to a file, to open a socket and etc. As you maybe know, invoking of a system call is expensive operation in the Linux kernel, because processor must interrupt currently executing task and switch context to the kernel mode and to jump again into userspace after a system call handler will finish its work. These two mechanisms - `vsyacall` and `vdso` designed to speed up this process for the certain system calls and in this part we will try to understand how these mechanisms are arranged.
We already know what is it a `system call`. This is special routine in the Linux kernel which userspace application asks to do privileged task, like to read or to write to a file, to open a socket and etc. As you maybe know, invoking of a system call is expensive operation in the Linux kernel, because processor must interrupt currently executing task and switch context to the kernel mode and to jump again into userspace after a system call handler will finish its work. These two mechanisms - `vsyscall` and `vdso` designed to speed up this process for the certain system calls and in this part we will try to understand how these mechanisms are arranged.
zhanxw
9 years ago