Based on my understanding, each process accesses memory through virtual addresses and not physical addresses, and it is the responsibility of the CPU to translate these virtual addresses to physical addresses through the MMU unit, and two or more processes can have the same virtual address.
So let's say Process A is trying to access the virtual address 12345, and also Process B is trying to access the virtual address 12345.
How will the MMU translate the virtual address of each process into a physical address, does it have a mapping table for each process that maps virtual addresses into physical addresses (because I thought that the CPU does not even know what a "process" is, and its only responsibility is to execute instructions blindly without caring which instruction belongs to which process, and a "process" is only an OS concept)?
Best Answer
In Linux, the kernel maintains a three-level page table (regardless of the CPU’s capabilities). The top level is the page global directory, and each process has its own directory,
pgdinmm_struct. Thus each process can have its own mappings, so address 12345 in different processes points to different physical addresses.CPUs aren’t really aware of processes, but they do tend to have features to support them. On x86-style CPUs, there are various task-related features, but they actually tend to be ignored. Since process scheduling is managed by the kernel, it can keep track of page table changes itself, and update whatever CPU state is required to switch to a new process’s page table when it switches tasks. On x86 PCs, that involves updating the CR3 control register which points to the page directory.
The Page Table Management chapter in Mel Gorman’s Understanding the Linux Virtual Memory Manager book gives a good overview.