Hello, I was wondering - kmalloc directly uses the kmem_cache caches (i.e. the geometrically distributed caches). These specific caches include caches of 4096bytes (a page of memory), 8192bytes (2 pages), all the way up to 131072bytes (32 pages). Now the caches get their memory slabs from get_free_pages. get_free_pages already has functionality to allocate a page (4096bytes), 2 pages, 4 pages, all the way up to 512 pages. To me there is a redundancy of allocating physically contiguous memory sizes of 1 page to 32 pages. Both kmalloc and get_free_pages can do it, but calling get_free_pages is faster than calling kmalloc to get the same amount of memory. I can think of some reasons to have it this way but they basically all come down to convenience. Can anyone tell me if there is a specific reason for this? TIA -Peter __________________________________ Do you Yahoo!? Free Pop-Up Blocker - Get it now http://companion.yahoo.com/ -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
