Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 1 | Direct Access for files |
| 2 | ----------------------- |
| 3 | |
| 4 | Motivation |
| 5 | ---------- |
| 6 | |
| 7 | The page cache is usually used to buffer reads and writes to files. |
| 8 | It is also used to provide the pages which are mapped into userspace |
| 9 | by a call to mmap. |
| 10 | |
| 11 | For block devices that are memory-like, the page cache pages would be |
| 12 | unnecessary copies of the original storage. The DAX code removes the |
| 13 | extra copy by performing reads and writes directly to the storage device. |
| 14 | For file mappings, the storage device is mapped directly into userspace. |
| 15 | |
| 16 | |
| 17 | Usage |
| 18 | ----- |
| 19 | |
| 20 | If you have a block device which supports DAX, you can make a filesystem |
Matthew Wilcox | 44f4c05 | 2015-07-03 10:40:38 -0400 | [diff] [blame] | 21 | on it as usual. The DAX code currently only supports files with a block |
| 22 | size equal to your kernel's PAGE_SIZE, so you may need to specify a block |
| 23 | size when creating the filesystem. When mounting it, use the "-o dax" |
| 24 | option on the command line or add 'dax' to the options in /etc/fstab. |
Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 25 | |
| 26 | |
| 27 | Implementation Tips for Block Driver Writers |
| 28 | -------------------------------------------- |
| 29 | |
| 30 | To support DAX in your block driver, implement the 'direct_access' |
| 31 | block device operation. It is used to translate the sector number |
| 32 | (expressed in units of 512-byte sectors) to a page frame number (pfn) |
| 33 | that identifies the physical page for the memory. It also returns a |
| 34 | kernel virtual address that can be used to access the memory. |
| 35 | |
| 36 | The direct_access method takes a 'size' parameter that indicates the |
| 37 | number of bytes being requested. The function should return the number |
| 38 | of bytes that can be contiguously accessed at that offset. It may also |
| 39 | return a negative errno if an error occurs. |
| 40 | |
| 41 | In order to support this method, the storage must be byte-accessible by |
| 42 | the CPU at all times. If your device uses paging techniques to expose |
| 43 | a large amount of memory through a smaller window, then you cannot |
| 44 | implement direct_access. Equally, if your device can occasionally |
| 45 | stall the CPU for an extended period, you should also not attempt to |
| 46 | implement direct_access. |
| 47 | |
| 48 | These block devices may be used for inspiration: |
| 49 | - axonram: Axon DDR2 device driver |
| 50 | - brd: RAM backed block device driver |
| 51 | - dcssblk: s390 dcss block device driver |
| 52 | |
| 53 | |
| 54 | Implementation Tips for Filesystem Writers |
| 55 | ------------------------------------------ |
| 56 | |
| 57 | Filesystem support consists of |
| 58 | - adding support to mark inodes as being DAX by setting the S_DAX flag in |
| 59 | i_flags |
| 60 | - implementing the direct_IO address space operation, and calling |
| 61 | dax_do_io() instead of blockdev_direct_IO() if S_DAX is set |
| 62 | - implementing an mmap file operation for DAX files which sets the |
Matthew Wilcox | 844f35d | 2015-09-08 14:58:57 -0700 | [diff] [blame] | 63 | VM_MIXEDMAP and VM_HUGEPAGE flags on the VMA, and setting the vm_ops to |
| 64 | include handlers for fault, pmd_fault and page_mkwrite (which should |
| 65 | probably call dax_fault(), dax_pmd_fault() and dax_mkwrite(), passing the |
| 66 | appropriate get_block() callback) |
Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 67 | - calling dax_truncate_page() instead of block_truncate_page() for DAX files |
Matthew Wilcox | 25726bc | 2015-02-16 15:59:35 -0800 | [diff] [blame] | 68 | - calling dax_zero_page_range() instead of zero_user() for DAX files |
Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 69 | - ensuring that there is sufficient locking between reads, writes, |
| 70 | truncates and page faults |
| 71 | |
| 72 | The get_block() callback passed to the DAX functions may return |
| 73 | uninitialised extents. If it does, it must ensure that simultaneous |
| 74 | calls to get_block() (for example by a page-fault racing with a read() |
| 75 | or a write()) work correctly. |
| 76 | |
| 77 | These filesystems may be used for inspiration: |
| 78 | - ext2: the second extended filesystem, see Documentation/filesystems/ext2.txt |
Ross Zwisler | 923ae0f | 2015-02-16 15:59:38 -0800 | [diff] [blame] | 79 | - ext4: the fourth extended filesystem, see Documentation/filesystems/ext4.txt |
Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 80 | |
| 81 | |
| 82 | Shortcomings |
| 83 | ------------ |
| 84 | |
| 85 | Even if the kernel or its modules are stored on a filesystem that supports |
| 86 | DAX on a block device that supports DAX, they will still be copied into RAM. |
| 87 | |
Matthew Wilcox | d92576f | 2015-02-16 15:59:44 -0800 | [diff] [blame] | 88 | The DAX code does not work correctly on architectures which have virtually |
| 89 | mapped caches such as ARM, MIPS and SPARC. |
| 90 | |
Matthew Wilcox | 95ec8da | 2015-02-16 15:59:09 -0800 | [diff] [blame] | 91 | Calling get_user_pages() on a range of user memory that has been mmaped |
| 92 | from a DAX file will fail as there are no 'struct page' to describe |
| 93 | those pages. This problem is being worked on. That means that O_DIRECT |
| 94 | reads/writes to those memory ranges from a non-DAX file will fail (note |
| 95 | that O_DIRECT reads/writes _of a DAX file_ do work, it is the memory |
| 96 | that is being accessed that is key here). Other things that will not |
| 97 | work include RDMA, sendfile() and splice(). |