• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1Direct Access for files
2-----------------------
3
4Motivation
5----------
6
7The page cache is usually used to buffer reads and writes to files.
8It is also used to provide the pages which are mapped into userspace
9by a call to mmap.
10
11For block devices that are memory-like, the page cache pages would be
12unnecessary copies of the original storage.  The DAX code removes the
13extra copy by performing reads and writes directly to the storage device.
14For file mappings, the storage device is mapped directly into userspace.
15
16
17Usage
18-----
19
20If you have a block device which supports DAX, you can make a filesystem
21on it as usual.  The DAX code currently only supports files with a block
22size equal to your kernel's PAGE_SIZE, so you may need to specify a block
23size when creating the filesystem.
24
25Currently 3 filesystems support DAX: ext2, ext4 and xfs.  Enabling DAX on them
26is different.
27
28Enabling DAX on ext2
29-----------------------------
30
31When mounting the filesystem, use the "-o dax" option on the command line or
32add 'dax' to the options in /etc/fstab.  This works to enable DAX on all files
33within the filesystem.  It is equivalent to the '-o dax=always' behavior below.
34
35
36Enabling DAX on xfs and ext4
37----------------------------
38
39Summary
40-------
41
42 1. There exists an in-kernel file access mode flag S_DAX that corresponds to
43    the statx flag STATX_ATTR_DAX.  See the manpage for statx(2) for details
44    about this access mode.
45
46 2. There exists a persistent flag FS_XFLAG_DAX that can be applied to regular
47    files and directories. This advisory flag can be set or cleared at any
48    time, but doing so does not immediately affect the S_DAX state.
49
50 3. If the persistent FS_XFLAG_DAX flag is set on a directory, this flag will
51    be inherited by all regular files and subdirectories that are subsequently
52    created in this directory. Files and subdirectories that exist at the time
53    this flag is set or cleared on the parent directory are not modified by
54    this modification of the parent directory.
55
56 4. There exist dax mount options which can override FS_XFLAG_DAX in the
57    setting of the S_DAX flag.  Given underlying storage which supports DAX the
58    following hold:
59
60    "-o dax=inode"  means "follow FS_XFLAG_DAX" and is the default.
61
62    "-o dax=never"  means "never set S_DAX, ignore FS_XFLAG_DAX."
63
64    "-o dax=always" means "always set S_DAX ignore FS_XFLAG_DAX."
65
66    "-o dax"        is a legacy option which is an alias for "dax=always".
67		    This may be removed in the future so "-o dax=always" is
68		    the preferred method for specifying this behavior.
69
70    NOTE: Modifications to and the inheritance behavior of FS_XFLAG_DAX remain
71    the same even when the filesystem is mounted with a dax option.  However,
72    in-core inode state (S_DAX) will be overridden until the filesystem is
73    remounted with dax=inode and the inode is evicted from kernel memory.
74
75 5. The S_DAX policy can be changed via:
76
77    a) Setting the parent directory FS_XFLAG_DAX as needed before files are
78       created
79
80    b) Setting the appropriate dax="foo" mount option
81
82    c) Changing the FS_XFLAG_DAX flag on existing regular files and
83       directories.  This has runtime constraints and limitations that are
84       described in 6) below.
85
86 6. When changing the S_DAX policy via toggling the persistent FS_XFLAG_DAX flag,
87    the change in behaviour for existing regular files may not occur
88    immediately.  If the change must take effect immediately, the administrator
89    needs to:
90
91    a) stop the application so there are no active references to the data set
92       the policy change will affect
93
94    b) evict the data set from kernel caches so it will be re-instantiated when
95       the application is restarted. This can be achieved by:
96
97       i. drop-caches
98       ii. a filesystem unmount and mount cycle
99       iii. a system reboot
100
101
102Details
103-------
104
105There are 2 per-file dax flags.  One is a persistent inode setting (FS_XFLAG_DAX)
106and the other is a volatile flag indicating the active state of the feature
107(S_DAX).
108
109FS_XFLAG_DAX is preserved within the filesystem.  This persistent config
110setting can be set, cleared and/or queried using the FS_IOC_FS[GS]ETXATTR ioctl
111(see ioctl_xfs_fsgetxattr(2)) or an utility such as 'xfs_io'.
112
113New files and directories automatically inherit FS_XFLAG_DAX from
114their parent directory _when_ _created_.  Therefore, setting FS_XFLAG_DAX at
115directory creation time can be used to set a default behavior for an entire
116sub-tree.
117
118To clarify inheritance, here are 3 examples:
119
120Example A:
121
122mkdir -p a/b/c
123xfs_io -c 'chattr +x' a
124mkdir a/b/c/d
125mkdir a/e
126
127	dax: a,e
128	no dax: b,c,d
129
130Example B:
131
132mkdir a
133xfs_io -c 'chattr +x' a
134mkdir -p a/b/c/d
135
136	dax: a,b,c,d
137	no dax:
138
139Example C:
140
141mkdir -p a/b/c
142xfs_io -c 'chattr +x' c
143mkdir a/b/c/d
144
145	dax: c,d
146	no dax: a,b
147
148
149The current enabled state (S_DAX) is set when a file inode is instantiated in
150memory by the kernel.  It is set based on the underlying media support, the
151value of FS_XFLAG_DAX and the filesystem's dax mount option.
152
153statx can be used to query S_DAX.  NOTE that only regular files will ever have
154S_DAX set and therefore statx will never indicate that S_DAX is set on
155directories.
156
157Setting the FS_XFLAG_DAX flag (specifically or through inheritance) occurs even
158if the underlying media does not support dax and/or the filesystem is
159overridden with a mount option.
160
161
162
163Implementation Tips for Block Driver Writers
164--------------------------------------------
165
166To support DAX in your block driver, implement the 'direct_access'
167block device operation.  It is used to translate the sector number
168(expressed in units of 512-byte sectors) to a page frame number (pfn)
169that identifies the physical page for the memory.  It also returns a
170kernel virtual address that can be used to access the memory.
171
172The direct_access method takes a 'size' parameter that indicates the
173number of bytes being requested.  The function should return the number
174of bytes that can be contiguously accessed at that offset.  It may also
175return a negative errno if an error occurs.
176
177In order to support this method, the storage must be byte-accessible by
178the CPU at all times.  If your device uses paging techniques to expose
179a large amount of memory through a smaller window, then you cannot
180implement direct_access.  Equally, if your device can occasionally
181stall the CPU for an extended period, you should also not attempt to
182implement direct_access.
183
184These block devices may be used for inspiration:
185- brd: RAM backed block device driver
186- dcssblk: s390 dcss block device driver
187- pmem: NVDIMM persistent memory driver
188
189
190Implementation Tips for Filesystem Writers
191------------------------------------------
192
193Filesystem support consists of
194- adding support to mark inodes as being DAX by setting the S_DAX flag in
195  i_flags
196- implementing ->read_iter and ->write_iter operations which use dax_iomap_rw()
197  when inode has S_DAX flag set
198- implementing an mmap file operation for DAX files which sets the
199  VM_MIXEDMAP and VM_HUGEPAGE flags on the VMA, and setting the vm_ops to
200  include handlers for fault, pmd_fault, page_mkwrite, pfn_mkwrite. These
201  handlers should probably call dax_iomap_fault() passing the appropriate
202  fault size and iomap operations.
203- calling iomap_zero_range() passing appropriate iomap operations instead of
204  block_truncate_page() for DAX files
205- ensuring that there is sufficient locking between reads, writes,
206  truncates and page faults
207
208The iomap handlers for allocating blocks must make sure that allocated blocks
209are zeroed out and converted to written extents before being returned to avoid
210exposure of uninitialized data through mmap.
211
212These filesystems may be used for inspiration:
213- ext2: see Documentation/filesystems/ext2.rst
214- ext4: see Documentation/filesystems/ext4/
215- xfs:  see Documentation/admin-guide/xfs.rst
216
217
218Handling Media Errors
219---------------------
220
221The libnvdimm subsystem stores a record of known media error locations for
222each pmem block device (in gendisk->badblocks). If we fault at such location,
223or one with a latent error not yet discovered, the application can expect
224to receive a SIGBUS. Libnvdimm also allows clearing of these errors by simply
225writing the affected sectors (through the pmem driver, and if the underlying
226NVDIMM supports the clear_poison DSM defined by ACPI).
227
228Since DAX IO normally doesn't go through the driver/bio path, applications or
229sysadmins have an option to restore the lost data from a prior backup/inbuilt
230redundancy in the following ways:
231
2321. Delete the affected file, and restore from a backup (sysadmin route):
233   This will free the filesystem blocks that were being used by the file,
234   and the next time they're allocated, they will be zeroed first, which
235   happens through the driver, and will clear bad sectors.
236
2372. Truncate or hole-punch the part of the file that has a bad-block (at least
238   an entire aligned sector has to be hole-punched, but not necessarily an
239   entire filesystem block).
240
241These are the two basic paths that allow DAX filesystems to continue operating
242in the presence of media errors. More robust error recovery mechanisms can be
243built on top of this in the future, for example, involving redundancy/mirroring
244provided at the block layer through DM, or additionally, at the filesystem
245level. These would have to rely on the above two tenets, that error clearing
246can happen either by sending an IO through the driver, or zeroing (also through
247the driver).
248
249
250Shortcomings
251------------
252
253Even if the kernel or its modules are stored on a filesystem that supports
254DAX on a block device that supports DAX, they will still be copied into RAM.
255
256The DAX code does not work correctly on architectures which have virtually
257mapped caches such as ARM, MIPS and SPARC.
258
259Calling get_user_pages() on a range of user memory that has been mmaped
260from a DAX file will fail when there are no 'struct page' to describe
261those pages.  This problem has been addressed in some device drivers
262by adding optional struct page support for pages under the control of
263the driver (see CONFIG_NVDIMM_PFN in drivers/nvdimm for an example of
264how to do this). In the non struct page cases O_DIRECT reads/writes to
265those memory ranges from a non-DAX file will fail (note that O_DIRECT
266reads/writes _of a DAX file_ do work, it is the memory that is being
267accessed that is key here).  Other things that will not work in the
268non struct page case include RDMA, sendfile() and splice().
269