Blame view

Documentation/nommu-mmap.txt 12.4 KB
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
1
  =============================
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
2
  No-MMU memory mapping support
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
3
  =============================
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
  
  The kernel has limited support for memory mapping under no-MMU conditions, such
  as are used in uClinux environments. From the userspace point of view, memory
  mapping is made use of in conjunction with the mmap() system call, the shmat()
  call and the execve() system call. From the kernel's point of view, execve()
  mapping is actually performed by the binfmt drivers, which call back into the
  mmap() routines to do the actual work.
  
  Memory mapping behaviour also involves the way fork(), vfork(), clone() and
  ptrace() work. Under uClinux there is no fork(), and clone() must be supplied
  the CLONE_VM flag.
  
  The behaviour is similar between the MMU and no-MMU cases, but not identical;
  and it's also much more restricted in the latter case:
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
18
   (#) Anonymous mapping, MAP_PRIVATE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
19
20
21
22
23
24
  
  	In the MMU case: VM regions backed by arbitrary pages; copy-on-write
  	across fork.
  
  	In the no-MMU case: VM regions backed by arbitrary contiguous runs of
  	pages.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
25
   (#) Anonymous mapping, MAP_SHARED
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
26
27
28
29
30
  
  	These behave very much like private mappings, except that they're
  	shared across fork() or clone() without CLONE_VM in the MMU case. Since
  	the no-MMU case doesn't support these, behaviour is identical to
  	MAP_PRIVATE there.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
31
   (#) File, MAP_PRIVATE, PROT_READ / PROT_EXEC, !PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
32
33
34
35
36
37
38
39
40
41
42
  
  	In the MMU case: VM regions backed by pages read from file; changes to
  	the underlying file are reflected in the mapping; copied across fork.
  
  	In the no-MMU case:
  
           - If one exists, the kernel will re-use an existing mapping to the
             same segment of the same file if that has compatible permissions,
             even if this was created by another process.
  
           - If possible, the file mapping will be directly on the backing device
b4caecd48   Christoph Hellwig   fs: introduce f_o...
43
             if the backing device has the NOMMU_MAP_DIRECT capability and
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
44
45
46
47
             appropriate mapping protection capabilities. Ramfs, romfs, cramfs
             and mtd might all permit this.
  
  	 - If the backing device device can't or won't permit direct sharing,
b4caecd48   Christoph Hellwig   fs: introduce f_o...
48
             but does have the NOMMU_MAP_COPY capability, then a copy of the
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
49
50
51
52
53
54
             appropriate bit of the file will be read into a contiguous bit of
             memory and any extraneous space beyond the EOF will be cleared
  
  	 - Writes to the file do not affect the mapping; writes to the mapping
  	   are visible in other processes (no MMU protection), but should not
  	   happen.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
55
   (#) File, MAP_PRIVATE, PROT_READ / PROT_EXEC, PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
56
57
58
59
60
61
62
63
  
  	In the MMU case: like the non-PROT_WRITE case, except that the pages in
  	question get copied before the write actually happens. From that point
  	on writes to the file underneath that page no longer get reflected into
  	the mapping's backing pages. The page is then backed by swap instead.
  
  	In the no-MMU case: works much like the non-PROT_WRITE case, except
  	that a copy is always taken and never shared.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
64
   (#) Regular file / blockdev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
65
66
67
68
69
70
  
  	In the MMU case: VM regions backed by pages read from file; changes to
  	pages written back to file; writes to file reflected into pages backing
  	mapping; shared across fork.
  
  	In the no-MMU case: not supported.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
71
   (#) Memory backed regular file, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
72
73
74
75
76
77
78
79
80
  
  	In the MMU case: As for ordinary regular files.
  
  	In the no-MMU case: The filesystem providing the memory-backed file
  	(such as ramfs or tmpfs) may choose to honour an open, truncate, mmap
  	sequence by providing a contiguous sequence of pages to map. In that
  	case, a shared-writable memory mapping will be possible. It will work
  	as for the MMU case. If the filesystem does not provide any such
  	support, then the mapping request will be denied.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
81
   (#) Memory backed blockdev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
82
83
84
85
86
87
88
  
  	In the MMU case: As for ordinary regular files.
  
  	In the no-MMU case: As for memory backed regular files, but the
  	blockdev must be able to provide a contiguous run of pages without
  	truncate being called. The ramdisk driver could do this if it allocated
  	all its memory as a contiguous array upfront.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
89
   (#) Memory backed chardev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
90
91
92
93
94
95
96
97
  
  	In the MMU case: As for ordinary regular files.
  
  	In the no-MMU case: The character device driver may choose to honour
  	the mmap() by providing direct access to the underlying device if it
  	provides memory or quasi-memory that can be accessed directly. Examples
  	of such are frame buffers and flash devices. If the driver does not
  	provide any such support, then the mapping request will be denied.
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
98
  Further notes on no-MMU MMAP
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
99
  ============================
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
100
   (#) A request for a private mapping of a file may return a buffer that is not
8feae1311   David Howells   NOMMU: Make VMAs ...
101
102
       page-aligned.  This is because XIP may take place, and the data may not be
       paged aligned in the backing store.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
103
   (#) A request for an anonymous mapping will always be page aligned.  If
8feae1311   David Howells   NOMMU: Make VMAs ...
104
105
106
107
       possible the size of the request should be a power of two otherwise some
       of the space may be wasted as the kernel must allocate a power-of-2
       granule but will only discard the excess if appropriately configured as
       this has an effect on fragmentation.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
108
   (#) The memory allocated by a request for an anonymous mapping will normally
ea6376395   Jie Zhang   nommu: fix malloc...
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
       be cleared by the kernel before being returned in accordance with the
       Linux man pages (ver 2.22 or later).
  
       In the MMU case this can be achieved with reasonable performance as
       regions are backed by virtual pages, with the contents only being mapped
       to cleared physical pages when a write happens on that specific page
       (prior to which, the pages are effectively mapped to the global zero page
       from which reads can take place).  This spreads out the time it takes to
       initialize the contents of a page - depending on the write-usage of the
       mapping.
  
       In the no-MMU case, however, anonymous mappings are backed by physical
       pages, and the entire map is cleared at allocation time.  This can cause
       significant delays during a userspace malloc() as the C library does an
       anonymous mapping and the kernel then does a memset for the entire map.
  
       However, for memory that isn't required to be precleared - such as that
       returned by malloc() - mmap() can take a MAP_UNINITIALIZED flag to
       indicate to the kernel that it shouldn't bother clearing the memory before
       returning it.  Note that CONFIG_MMAP_ALLOW_UNINITIALIZED must be enabled
       to permit this, otherwise the flag will be ignored.
  
       uClibc uses this to speed up malloc(), and the ELF-FDPIC binfmt uses this
       to allocate the brk and stack region.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
133
   (#) A list of all the private copy and anonymous mappings on the system is
8feae1311   David Howells   NOMMU: Make VMAs ...
134
       visible through /proc/maps in no-MMU mode.
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
135

c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
136
   (#) A list of all the mappings in use by a process is visible through
dbf8685c8   David Howells   [PATCH] NOMMU: Im...
137
       /proc/<pid>/maps in no-MMU mode.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
138
   (#) Supplying MAP_FIXED or a requesting a particular mapping address will
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
139
       result in an error.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
140
   (#) Files mapped privately usually have to have a read method provided by the
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
141
142
143
144
       driver or filesystem so that the contents can be read into the memory
       allocated if mmap() chooses not to map the backing device directly. An
       error will result if they don't. This is most likely to be encountered
       with character device files, pipes, fifos and sockets.
6fa5f80bc   David Howells   [PATCH] NOMMU: Ma...
145

853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
146
  Interprocess shared memory
0112c4c64   David Howells   [PATCH] NOMMU: Ad...
147
148
149
150
151
  ==========================
  
  Both SYSV IPC SHM shared memory and POSIX shared memory is supported in NOMMU
  mode.  The former through the usual mechanism, the latter through files created
  on ramfs or tmpfs mounts.
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
152
  Futexes
930e652a2   David Howells   [PATCH] NOMMU: Ma...
153
154
155
156
157
158
  =======
  
  Futexes are supported in NOMMU mode if the arch supports them.  An error will
  be given if an address passed to the futex system call lies outside the
  mappings made by a process or if the mapping in which the address lies does not
  support futexes (such as an I/O chardev mapping).
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
159
  No-MMU mremap
6fa5f80bc   David Howells   [PATCH] NOMMU: Ma...
160
161
162
  =============
  
  The mremap() function is partially supported.  It may change the size of a
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
163
  mapping, and may move it [#]_ if MREMAP_MAYMOVE is specified and if the new size
6fa5f80bc   David Howells   [PATCH] NOMMU: Ma...
164
165
166
167
168
169
170
171
172
173
174
175
176
  of the mapping exceeds the size of the slab object currently occupied by the
  memory to which the mapping refers, or if a smaller slab object could be used.
  
  MREMAP_FIXED is not supported, though it is ignored if there's no change of
  address and the object does not need to be moved.
  
  Shared mappings may not be moved.  Shareable mappings may not be moved either,
  even if they are not currently shared.
  
  The mremap() function must be given an exact match for base address and size of
  a previously mapped object.  It may not be used to create holes in existing
  mappings, move parts of existing mappings or resize parts of mappings.  It must
  act on a complete mapping.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
177
  .. [#] Not currently supported.
6fa5f80bc   David Howells   [PATCH] NOMMU: Ma...
178

853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
179
  Providing shareable character device support
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
  ============================================
  
  To provide shareable character device support, a driver must provide a
  file->f_op->get_unmapped_area() operation. The mmap() routines will call this
  to get a proposed address for the mapping. This may return an error if it
  doesn't wish to honour the mapping because it's too long, at a weird offset,
  under some unsupported combination of flags or whatever.
  
  The driver should also provide backing device information with capabilities set
  to indicate the permitted types of mapping on such devices. The default is
  assumed to be readable and writable, not executable, and only shareable
  directly (can't be copied).
  
  The file->f_op->mmap() operation will be called to actually inaugurate the
  mapping. It can be rejected at that point. Returning the ENOSYS error will
b4caecd48   Christoph Hellwig   fs: introduce f_o...
195
  cause the mapping to be copied instead if NOMMU_MAP_COPY is specified.
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
196
197
198
199
200
201
202
203
204
205
206
  
  The vm_ops->close() routine will be invoked when the last mapping on a chardev
  is removed. An existing mapping will be shared, partially or not, if possible
  without notifying the driver.
  
  It is permitted also for the file->f_op->get_unmapped_area() operation to
  return -ENOSYS. This will be taken to mean that this operation just doesn't
  want to handle it, despite the fact it's got an operation. For instance, it
  might try directing the call to a secondary driver which turns out not to
  implement it. Such is the case for the framebuffer driver which attempts to
  direct the call to the device-specific driver. Under such circumstances, the
b4caecd48   Christoph Hellwig   fs: introduce f_o...
207
  mapping request will be rejected if NOMMU_MAP_COPY is not specified, and a
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
208
  copy mapped otherwise.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
209
  .. important::
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
210
211
212
213
214
215
216
217
218
219
220
  
  	Some types of device may present a different appearance to anyone
  	looking at them in certain modes. Flash chips can be like this; for
  	instance if they're in programming or erase mode, you might see the
  	status reflected in the mapping, instead of the data.
  
  	In such a case, care must be taken lest userspace see a shared or a
  	private mapping showing such information when the driver is busy
  	controlling the device. Remember especially: private executable
  	mappings may still be mapped directly off the device under some
  	circumstances!
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
221
  Providing shareable memory-backed file support
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
222
223
224
225
226
227
228
229
230
231
232
233
234
235
  ==============================================
  
  Provision of shared mappings on memory backed files is similar to the provision
  of support for shared mapped character devices. The main difference is that the
  filesystem providing the service will probably allocate a contiguous collection
  of pages and permit mappings to be made on that.
  
  It is recommended that a truncate operation applied to such a file that
  increases the file size, if that file is empty, be taken as a request to gather
  enough pages to honour a mapping. This is required to support POSIX shared
  memory.
  
  Memory backed devices are indicated by the mapping's backing device info having
  the memory_backed flag set.
853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
236
  Providing shareable block device support
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
237
238
239
240
241
  ========================================
  
  Provision of shared mappings on block device files is exactly the same as for
  character devices. If there isn't a real device underneath, then the driver
  should allocate sufficient contiguous memory to honour any supported mapping.
dd8632a12   Paul Mundt   NOMMU: Make mmap ...
242

853afb719   Mauro Carvalho Chehab   nommu-mmap.txt: d...
243
  Adjusting page trimming behaviour
dd8632a12   Paul Mundt   NOMMU: Make mmap ...
244
245
246
247
248
249
250
251
252
  =================================
  
  NOMMU mmap automatically rounds up to the nearest power-of-2 number of pages
  when performing an allocation.  This can have adverse effects on memory
  fragmentation, and as such, is left configurable.  The default behaviour is to
  aggressively trim allocations and discard any excess pages back in to the page
  allocator.  In order to retain finer-grained control over fragmentation, this
  behaviour can either be disabled completely, or bumped up to a higher page
  watermark where trimming begins.
c49e51a53   Mauro Carvalho Chehab   nommu-mmap.txt: s...
253
  Page trimming behaviour is configurable via the sysctl ``vm.nr_trim_pages``.