Documentation/nommu-mmap.txt

   1                          =============================
   2                          NO-MMU MEMORY MAPPING SUPPORT
   3                          =============================
   4
   5 The kernel has limited support for memory mapping under no-MMU conditions, such
   6 as are used in uClinux environments. From the userspace point of view, memory
   7 mapping is made use of in conjunction with the mmap() system call, the shmat()
   8 call and the execve() system call. From the kernel's point of view, execve()
   9 mapping is actually performed by the binfmt drivers, which call back into the
  10 mmap() routines to do the actual work.
  11
  12 Memory mapping behaviour also involves the way fork(), vfork(), clone() and
  13 ptrace() work. Under uClinux there is no fork(), and clone() must be supplied
  14 the CLONE_VM flag.
  15
  16 The behaviour is similar between the MMU and no-MMU cases, but not identical;
  17 and it's also much more restricted in the latter case:
  18
  19  (*) Anonymous mapping, MAP_PRIVATE
  20
  21         In the MMU case: VM regions backed by arbitrary pages; copy-on-write
  22         across fork.
  23
  24         In the no-MMU case: VM regions backed by arbitrary contiguous runs of
  25         pages.
  26
  27  (*) Anonymous mapping, MAP_SHARED
  28
  29         These behave very much like private mappings, except that they're
  30         shared across fork() or clone() without CLONE_VM in the MMU case. Since
  31         the no-MMU case doesn't support these, behaviour is identical to
  32         MAP_PRIVATE there.
  33
  34  (*) File, MAP_PRIVATE, PROT_READ / PROT_EXEC, !PROT_WRITE
  35
  36         In the MMU case: VM regions backed by pages read from file; changes to
  37         the underlying file are reflected in the mapping; copied across fork.
  38
  39         In the no-MMU case:
  40
  41          - If one exists, the kernel will re-use an existing mapping to the
  42            same segment of the same file if that has compatible permissions,
  43            even if this was created by another process.
  44
  45          - If possible, the file mapping will be directly on the backing device
  46            if the backing device has the BDI_CAP_MAP_DIRECT capability and
  47            appropriate mapping protection capabilities. Ramfs, romfs, cramfs
  48            and mtd might all permit this.
  49
  50          - If the backing device device can't or won't permit direct sharing,
  51            but does have the BDI_CAP_MAP_COPY capability, then a copy of the
  52            appropriate bit of the file will be read into a contiguous bit of
  53            memory and any extraneous space beyond the EOF will be cleared
  54
  55          - Writes to the file do not affect the mapping; writes to the mapping
  56            are visible in other processes (no MMU protection), but should not
  57            happen.
  58
  59  (*) File, MAP_PRIVATE, PROT_READ / PROT_EXEC, PROT_WRITE
  60
  61         In the MMU case: like the non-PROT_WRITE case, except that the pages in
  62         question get copied before the write actually happens. From that point
  63         on writes to the file underneath that page no longer get reflected into
  64         the mapping's backing pages. The page is then backed by swap instead.
  65
  66         In the no-MMU case: works much like the non-PROT_WRITE case, except
  67         that a copy is always taken and never shared.
  68
  69  (*) Regular file / blockdev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
  70
  71         In the MMU case: VM regions backed by pages read from file; changes to
  72         pages written back to file; writes to file reflected into pages backing
  73         mapping; shared across fork.
  74
  75         In the no-MMU case: not supported.
  76
  77  (*) Memory backed regular file, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
  78
  79         In the MMU case: As for ordinary regular files.
  80
  81         In the no-MMU case: The filesystem providing the memory-backed file
  82         (such as ramfs or tmpfs) may choose to honour an open, truncate, mmap
  83         sequence by providing a contiguous sequence of pages to map. In that
  84         case, a shared-writable memory mapping will be possible. It will work
  85         as for the MMU case. If the filesystem does not provide any such
  86         support, then the mapping request will be denied.
  87
  88  (*) Memory backed blockdev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
  89
  90         In the MMU case: As for ordinary regular files.
  91
  92         In the no-MMU case: As for memory backed regular files, but the
  93         blockdev must be able to provide a contiguous run of pages without
  94         truncate being called. The ramdisk driver could do this if it allocated
  95         all its memory as a contiguous array upfront.
  96
  97  (*) Memory backed chardev, MAP_SHARED, PROT_READ / PROT_EXEC / PROT_WRITE
  98
  99         In the MMU case: As for ordinary regular files.
 100
 101         In the no-MMU case: The character device driver may choose to honour
 102         the mmap() by providing direct access to the underlying device if it
 103         provides memory or quasi-memory that can be accessed directly. Examples
 104         of such are frame buffers and flash devices. If the driver does not
 105         provide any such support, then the mapping request will be denied.
 106
 107
 108 ============================
 109 FURTHER NOTES ON NO-MMU MMAP
 110 ============================
 111
 112  (*) A request for a private mapping of a file may return a buffer that is not
 113      page-aligned.  This is because XIP may take place, and the data may not be
 114      paged aligned in the backing store.
 115
 116  (*) A request for an anonymous mapping will always be page aligned.  If
 117      possible the size of the request should be a power of two otherwise some
 118      of the space may be wasted as the kernel must allocate a power-of-2
 119      granule but will only discard the excess if appropriately configured as
 120      this has an effect on fragmentation.
 121
 122  (*) A list of all the private copy and anonymous mappings on the system is
 123      visible through /proc/maps in no-MMU mode.
 124
 125  (*) A list of all the mappings in use by a process is visible through
 126      /proc/<pid>/maps in no-MMU mode.
 127
 128  (*) Supplying MAP_FIXED or a requesting a particular mapping address will
 129      result in an error.
 130
 131  (*) Files mapped privately usually have to have a read method provided by the
 132      driver or filesystem so that the contents can be read into the memory
 133      allocated if mmap() chooses not to map the backing device directly. An
 134      error will result if they don't. This is most likely to be encountered
 135      with character device files, pipes, fifos and sockets.
 136
 137
 138 ==========================
 139 INTERPROCESS SHARED MEMORY
 140 ==========================
 141
 142 Both SYSV IPC SHM shared memory and POSIX shared memory is supported in NOMMU
 143 mode.  The former through the usual mechanism, the latter through files created
 144 on ramfs or tmpfs mounts.
 145
 146
 147 =======
 148 FUTEXES
 149 =======
 150
 151 Futexes are supported in NOMMU mode if the arch supports them.  An error will
 152 be given if an address passed to the futex system call lies outside the
 153 mappings made by a process or if the mapping in which the address lies does not
 154 support futexes (such as an I/O chardev mapping).
 155
 156
 157 =============
 158 NO-MMU MREMAP
 159 =============
 160
 161 The mremap() function is partially supported.  It may change the size of a
 162 mapping, and may move it[*] if MREMAP_MAYMOVE is specified and if the new size
 163 of the mapping exceeds the size of the slab object currently occupied by the
 164 memory to which the mapping refers, or if a smaller slab object could be used.
 165
 166 MREMAP_FIXED is not supported, though it is ignored if there's no change of
 167 address and the object does not need to be moved.
 168
 169 Shared mappings may not be moved.  Shareable mappings may not be moved either,
 170 even if they are not currently shared.
 171
 172 The mremap() function must be given an exact match for base address and size of
 173 a previously mapped object.  It may not be used to create holes in existing
 174 mappings, move parts of existing mappings or resize parts of mappings.  It must
 175 act on a complete mapping.
 176
 177 [*] Not currently supported.
 178
 179
 180 ============================================
 181 PROVIDING SHAREABLE CHARACTER DEVICE SUPPORT
 182 ============================================
 183
 184 To provide shareable character device support, a driver must provide a
 185 file->f_op->get_unmapped_area() operation. The mmap() routines will call this
 186 to get a proposed address for the mapping. This may return an error if it
 187 doesn't wish to honour the mapping because it's too long, at a weird offset,
 188 under some unsupported combination of flags or whatever.
 189
 190 The driver should also provide backing device information with capabilities set
 191 to indicate the permitted types of mapping on such devices. The default is
 192 assumed to be readable and writable, not executable, and only shareable
 193 directly (can't be copied).
 194
 195 The file->f_op->mmap() operation will be called to actually inaugurate the
 196 mapping. It can be rejected at that point. Returning the ENOSYS error will
 197 cause the mapping to be copied instead if BDI_CAP_MAP_COPY is specified.
 198
 199 The vm_ops->close() routine will be invoked when the last mapping on a chardev
 200 is removed. An existing mapping will be shared, partially or not, if possible
 201 without notifying the driver.
 202
 203 It is permitted also for the file->f_op->get_unmapped_area() operation to
 204 return -ENOSYS. This will be taken to mean that this operation just doesn't
 205 want to handle it, despite the fact it's got an operation. For instance, it
 206 might try directing the call to a secondary driver which turns out not to
 207 implement it. Such is the case for the framebuffer driver which attempts to
 208 direct the call to the device-specific driver. Under such circumstances, the
 209 mapping request will be rejected if BDI_CAP_MAP_COPY is not specified, and a
 210 copy mapped otherwise.
 211
 212 IMPORTANT NOTE:
 213
 214         Some types of device may present a different appearance to anyone
 215         looking at them in certain modes. Flash chips can be like this; for
 216         instance if they're in programming or erase mode, you might see the
 217         status reflected in the mapping, instead of the data.
 218
 219         In such a case, care must be taken lest userspace see a shared or a
 220         private mapping showing such information when the driver is busy
 221         controlling the device. Remember especially: private executable
 222         mappings may still be mapped directly off the device under some
 223         circumstances!
 224
 225
 226 ==============================================
 227 PROVIDING SHAREABLE MEMORY-BACKED FILE SUPPORT
 228 ==============================================
 229
 230 Provision of shared mappings on memory backed files is similar to the provision
 231 of support for shared mapped character devices. The main difference is that the
 232 filesystem providing the service will probably allocate a contiguous collection
 233 of pages and permit mappings to be made on that.
 234
 235 It is recommended that a truncate operation applied to such a file that
 236 increases the file size, if that file is empty, be taken as a request to gather
 237 enough pages to honour a mapping. This is required to support POSIX shared
 238 memory.
 239
 240 Memory backed devices are indicated by the mapping's backing device info having
 241 the memory_backed flag set.
 242
 243
 244 ========================================
 245 PROVIDING SHAREABLE BLOCK DEVICE SUPPORT
 246 ========================================
 247
 248 Provision of shared mappings on block device files is exactly the same as for
 249 character devices. If there isn't a real device underneath, then the driver
 250 should allocate sufficient contiguous memory to honour any supported mapping.
 251
 252
 253 =================================
 254 ADJUSTING PAGE TRIMMING BEHAVIOUR
 255 =================================
 256
 257 NOMMU mmap automatically rounds up to the nearest power-of-2 number of pages
 258 when performing an allocation.  This can have adverse effects on memory
 259 fragmentation, and as such, is left configurable.  The default behaviour is to
 260 aggressively trim allocations and discard any excess pages back in to the page
 261 allocator.  In order to retain finer-grained control over fragmentation, this
 262 behaviour can either be disabled completely, or bumped up to a higher page
 263 watermark where trimming begins.
 264
 265 Page trimming behaviour is configurable via the sysctl `vm.nr_trim_pages'.