From: Gioh Kim Date: Mon, 7 Apr 2014 22:37:37 +0000 (-0700) Subject: mm/vmalloc.c: enhance vm_map_ram() comment X-Git-Tag: firefly_0821_release~176^2~4111^2~121 X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=3643763834b935208b067db9f4a239aba9dbe28d;p=firefly-linux-kernel-4.4.55.git mm/vmalloc.c: enhance vm_map_ram() comment vm_map_ram() has a fragmentation problem when it cannot purge a chunk(ie, 4M address space) if there is a pinning object in that addresss space. So it could consume all VMALLOC address space easily. We can fix the fragmentation problem by using vmap instead of vm_map_ram() but vmap() is known to be slow compared to vm_map_ram(). Minchan said vm_map_ram is 5 times faster than vmap in his tests. So I thought we should fix fragment problem of vm_map_ram because our proprietary GPU driver has used it heavily. On second thought, it's not an easy because we should reuse freed space for solving the problem and it could make more IPI and bitmap operation for searching hole. It could mitigate API's goal which is very fast mapping. And even fragmentation problem wouldn't show in 64 bit machine. Another option is that the user should separate long-life and short-life object and use vmap for long-life but vm_map_ram for short-life. If we inform the user about the characteristic of vm_map_ram the user can choose one according to the page lifetime. Let's add some notice messages to user. [akpm@linux-foundation.org: tweak comment text] Signed-off-by: Gioh Kim Reviewed-by: Zhang Yanfei Cc: Minchan Kim Cc: Johannes Weiner Cc: Joonsoo Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- diff --git a/mm/vmalloc.c b/mm/vmalloc.c index a7b522f4851d..bf233b283319 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1085,6 +1085,12 @@ EXPORT_SYMBOL(vm_unmap_ram); * @node: prefer to allocate data structures on this node * @prot: memory protection to use. PAGE_KERNEL for regular RAM * + * If you use this function for less than VMAP_MAX_ALLOC pages, it could be + * faster than vmap so it's good. But if you mix long-life and short-life + * objects with vm_map_ram(), it could consume lots of address space through + * fragmentation (especially on a 32bit machine). You could see failures in + * the end. Please use this function for short-lived objects. + * * Returns: a pointer to the address that has been mapped, or %NULL on failure */ void *vm_map_ram(struct page **pages, unsigned int count, int node, pgprot_t prot)