> Please help me to figure out some basic concepts in MM. > From the books, i learned that VMA to PA translation consists of traversing > the full page directory, which consists of Global Directory, Middle > Directory and Page Table. > I have also read that, VMA to PA translation is done using a macro > virt_to_page defined as given below. > #define virt_to_page(kaddr) (mem_map + (__pa(kaddr) >> PAGE_SHIFT)) > If > there is a macro, why we need code for traversing all the page > directories ?. This macro is a simple math, which points to an index > in global mem_map array, which contains all the pages in the system.
That's a noble question. But if you think about it, the processor's MMU is what walks these directories to find the page table entry (containing the physical address) corresponding to a particular virtual memory address. The kernel sets these tables up for the processor. Also the mem_map array contains "page descriptors". This is the structure that describes a particular page frame in physical memory. (http://lxr.linux.no/#linux+v2.6.31/include/linux/mm_types.h#L40) The macro you're talking about doesn't translate virtual to physical addresses, it just fetches the page descriptor of the physical page which the virtual memory address translates to, from the mem_map array. > So my doubts are, > Is the macro only for ZONE_NORMAL (upto 896M, which is directly mapped by > kernel) memory pages ?. > Is mem_map array contains pages upto ZONE_NORMAL ? No, all physical memory is described. > Is page traversing happens only for HIGH_MEM ?. No, page transversing done by the CPU happens for all virtual addresses. Further NORMAL HIGH AND LOW MEM are physical memory areas, not virtual. About low / high mem, what I understand is for 32 bit x86 architectures, the kernel couldn't directly access physical memory (using virtual addresses) above 896MB. Only 0 - 896MB was directly mapped - so the kernel image had to be loaded in this directly mapped memory and couldn't be loaded in physical memory above 896MB. The only way the kernel could access physical memory above 896MB was by indirectly setting up temporary mappings and cycling these mappings between requests. In 64 bit architectures, the kernel doesn't use HIGH_MEM at all. Hope the above makes sense. Let me know if I went wrong anywhere.. -Joel -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecar...@nl.linux.org Please read the FAQ at http://kernelnewbies.org/FAQ
