1 Direct Access for files
2 -----------------------
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
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.
20 If you have a block device which supports DAX, you can make a filesystem
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.
27 Implementation Tips for Block Driver Writers
28 --------------------------------------------
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.
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.
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.
48 These block devices may be used for inspiration:
49 - brd: RAM backed block device driver
50 - dcssblk: s390 dcss block device driver
51 - pmem: NVDIMM persistent memory driver
54 Implementation Tips for Filesystem Writers
55 ------------------------------------------
57 Filesystem support consists of
58 - adding support to mark inodes as being DAX by setting the S_DAX flag in
60 - implementing ->read_iter and ->write_iter operations which use dax_iomap_rw()
61 when inode has S_DAX flag set
62 - implementing an mmap file operation for DAX files which sets the
63 VM_MIXEDMAP and VM_HUGEPAGE flags on the VMA, and setting the vm_ops to
64 include handlers for fault, pmd_fault, page_mkwrite, pfn_mkwrite. These
65 handlers should probably call dax_iomap_fault() passing the appropriate
66 fault size and iomap operations.
67 - calling iomap_zero_range() passing appropriate iomap operations instead of
68 block_truncate_page() for DAX files
69 - ensuring that there is sufficient locking between reads, writes,
70 truncates and page faults
72 The iomap handlers for allocating blocks must make sure that allocated blocks
73 are zeroed out and converted to written extents before being returned to avoid
74 exposure of uninitialized data through mmap.
76 These filesystems may be used for inspiration:
77 - ext2: see Documentation/filesystems/ext2.txt
78 - ext4: see Documentation/filesystems/ext4.txt
79 - xfs: see Documentation/filesystems/xfs.txt
85 The libnvdimm subsystem stores a record of known media error locations for
86 each pmem block device (in gendisk->badblocks). If we fault at such location,
87 or one with a latent error not yet discovered, the application can expect
88 to receive a SIGBUS. Libnvdimm also allows clearing of these errors by simply
89 writing the affected sectors (through the pmem driver, and if the underlying
90 NVDIMM supports the clear_poison DSM defined by ACPI).
92 Since DAX IO normally doesn't go through the driver/bio path, applications or
93 sysadmins have an option to restore the lost data from a prior backup/inbuilt
94 redundancy in the following ways:
96 1. Delete the affected file, and restore from a backup (sysadmin route):
97 This will free the file system blocks that were being used by the file,
98 and the next time they're allocated, they will be zeroed first, which
99 happens through the driver, and will clear bad sectors.
101 2. Truncate or hole-punch the part of the file that has a bad-block (at least
102 an entire aligned sector has to be hole-punched, but not necessarily an
103 entire filesystem block).
105 These are the two basic paths that allow DAX filesystems to continue operating
106 in the presence of media errors. More robust error recovery mechanisms can be
107 built on top of this in the future, for example, involving redundancy/mirroring
108 provided at the block layer through DM, or additionally, at the filesystem
109 level. These would have to rely on the above two tenets, that error clearing
110 can happen either by sending an IO through the driver, or zeroing (also through
117 Even if the kernel or its modules are stored on a filesystem that supports
118 DAX on a block device that supports DAX, they will still be copied into RAM.
120 The DAX code does not work correctly on architectures which have virtually
121 mapped caches such as ARM, MIPS and SPARC.
123 Calling get_user_pages() on a range of user memory that has been mmaped
124 from a DAX file will fail when there are no 'struct page' to describe
125 those pages. This problem has been addressed in some device drivers
126 by adding optional struct page support for pages under the control of
127 the driver (see CONFIG_NVDIMM_PFN in drivers/nvdimm for an example of
128 how to do this). In the non struct page cases O_DIRECT reads/writes to
129 those memory ranges from a non-DAX file will fail (note that O_DIRECT
130 reads/writes _of a DAX file_ do work, it is the memory that is being
131 accessed that is key here). Other things that will not work in the
132 non struct page case include RDMA, sendfile() and splice().