2010/11/18 Parmenides <mobile.parmenides@xxxxxxxxx>
Hi,
 According to ULK 3rd edition, the kernel stack is located in user
space, such as a linear address of 0x015fa000ã
Â
Not Quite. For each process, there are 2 stacks - User Stack & a Kernel Stack . This Per-Process kernel stack is located in the data segment of the kernel. It is not located in the user space as you understood. When a process switches to kernel mode, the hardware segmentation registers cs(code segment) & ds(data segment) are set to __KERNEL_CS & __KERNEL_DS respectively. Hence, depending on whether the process is executing in the user mode or kernel mode, the user stack or kernel stack is found according to the current value of the DS segmentation register.
As far as this situation
is concerned, I have several questions.
1. Now that the kernel stack is used by the kernel code, why isn't it
allocated in the kernel space?
Question is based on a wrong assumption of yours. Kernel stack is in the kernel data segment.
2. For the kernel code, is it feasible to the use the user stack? Why
do we bother to switch to the kernel stack?
The answer is Yes, you could. But it would be pretty messy & inconvenient. We just don't do it in the linux kernel atleast on x86. Kernel Data Segment & User Data Segment is different. I guess you could just map the user space stack in the kernel address space too & use it. Using two seperate stacks is just more efficient & convenient.
3. What's the difference between the user space and the kernel space on earth?
Well, i guess this question should have been at the top. Before you can understand the difference between user space stack & kernel space stack, you have to know the difference between user space & kernel space. How could you ask questions 1 & 2 without knowing 3?? Well, i am not sure about the "earth" but i can tell you the difference in a computer. :))) .
x86 processor supports multiple privilege levels. The kernel runs in ring 0 (most privilaged) and the user space runs in ring 3(least privileged). There are certain instructions which you can execute only in the privileged(kernel) mode. O'wise that instruction will result in a trap. "cli" is an example of such instruction (used for disabling interrupts). So, applications programs run in unprivilaged mode & hence only execute instructions which no other process needs to know about. For ex. if you concat two strings, nobody else needs to know about it. On the other hand, if you want to execute any instruction like "cli" or reading from file or using a device, you have to goto the kernel which acts as the mediator for all processes. Thats why have Operating systems -Â Thats why we have user mode & kernel mode.
Venkatram Tummala
x86 processor supports multiple privilege levels. The kernel runs in ring 0 (most privilaged) and the user space runs in ring 3(least privileged). There are certain instructions which you can execute only in the privileged(kernel) mode. O'wise that instruction will result in a trap. "cli" is an example of such instruction (used for disabling interrupts). So, applications programs run in unprivilaged mode & hence only execute instructions which no other process needs to know about. For ex. if you concat two strings, nobody else needs to know about it. On the other hand, if you want to execute any instruction like "cli" or reading from file or using a device, you have to goto the kernel which acts as the mediator for all processes. Thats why have Operating systems -Â Thats why we have user mode & kernel mode.
Venkatram Tummala
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ
