On 07/16, Benjamin LaHaise wrote: > > On Thu, Jul 16, 2015 at 09:05:03PM +0200, Oleg Nesterov wrote: > > Thanks! > ... > > but the problem looks clear: CONFIG_MMU is not set, so we need > > a dummy filemap_page_mkwrite() along with generic_file_mmap() and > > generic_file_readonly_mmap(). > > > > I'll send the fix, but... > > > > Benjamin, Jeff, shouldn't AIO depend on MMU? Or it can actually work even > > if CONFIG_MMU=n? > > It should work when CONFIG_MMU=n, Really? I am just curious. alloc_anon_inode() doesn't set S_IFREG, it seems that nommu.c:do_mmap_pgoff() should just fail in validate_mmap_request() ? Even if not, it should fail because of MAP_SHARED && !NOMMU_MAP_DIRECT? I am just trying to understand, could you explain? Oleg. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>