Hello.... > For security and general sanity of the system, kernel and user code > can't use the same stack. So each process has to have 2 stacks -- one > is in kernel memory, used by kernel code, and one is in user memory, > used by the user code. it is important to note that kernel stack is located on memory area with privilege level 0, so it can only be accessed by code which has privilege level 0 too. Conversely, user stack has privilege level 3, so technically it can be accessed by user mode and kernel mode. This memory area's privilege is defined by the related segment descritp During the switch between user and kernel mode (e.g by issuing system call), Stack Segment selector (SS) is also updated to point on related segment descriptor. IIRC, on user mode, SS is equivalent to __USER_DS, while on kernel mode, it is __KERNEL_DS. Both segment actually overlaps (both starts on address 0 and ends at oxFFFFFFFF) , so something else is needed to protect these two segment. If you read on Understanding Linux Kernel 2nd ed chapter 2, you will realize that the protection is doubled by paging mechanism. Page frame containing kernel stack has Supervisor flag set as 0. This means it can only be accessed by kernel mode codes Perhaps the final question is "how can I determine the code is in user mode or kernel mode?" remember that Linux kernel operates in protected mode, thus initially codes are referenced by logical address, composed by segment selector and offset. Inside the segment selector, there is an RPL (requestor privilege level) field (2 bit length) which defines CPL (current privilege level). Kernel codes always issue logical address which has RPL=0, whereas user mode code has RPL=3. I guess it is also the work of kernel itself which is doing switching between segment selectors hope it helps regards Mulyadi -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
