|MMAP(2)||System Calls Manual||MMAP(2)|
LIBRARY¶Standard C Library (libc, -lc)
*addr, size_t len,
mmap() system call causes the pages starting at addr and continuing for at most len bytes to be mapped from the object described by fd, starting at byte offset offset. If len is not a multiple of the pagesize, the mapped region may extend past the specified range. Any such extension beyond the end of the mapped object will be zero-filled.
If addr is non-zero, it is used as a hint to the system. (As a convenience to the system, the actual address of the region may differ from the address supplied.) If addr is zero, an address will be selected by the system. The actual starting address of the region is returned. A successful mmap deletes any previous mapping in the allocated address range.
The protections (region accessibility) are specified in the prot argument by or'ing the following values:
- Pages may not be accessed.
- Pages may be read.
- Pages may be written.
- Pages may be executed.
The flags argument specifies the type of the mapped object, mapping options and whether modifications made to the mapped copy of the page are private to the process or are to be shared with other references. Sharing, mapping type and options are specified in the flags argument by or'ing the following values:
- Request a region in the first 2GB of the current process's address space.
If a suitable region cannot be found,
mmap() will fail. This flag is only available on 64-bit platforms.
- Align the region on a requested boundary. If a suitable region cannot be
mmap() will fail. The n argument specifies the binary logarithm of the desired alignment.
- Align the region to maximize the potential use of large
(“super”) pages. If a suitable region cannot be found,
mmap() will fail. The system will choose a suitable page size based on the size of mapping. The page size used as well as the alignment of the region may both be affected by properties of the file being mapped. In particular, the physical address of existing pages of a file may require a specific alignment. The region is not guaranteed to be aligned on any specific boundary.
- Map anonymous memory not associated with any specific file. The file
descriptor used for creating
MAP_ANONmust be -1. The offset argument must be 0.
- This flag is identical to
MAP_ANONand is provided for compatibility.
- This flag can only be used in combination with
MAP_FIXED. Please see the definition of
MAP_FIXEDfor the description of its effect.
- Do not permit the system to select a different address than the one
specified. If the specified address cannot be used,
mmap() will fail. If
MAP_FIXEDis specified, addr must be a multiple of the pagesize. If
MAP_EXCLis not specified, a successful
MAP_FIXEDrequest replaces any previous mappings for the process' pages in the range from addr to addr + len. In contrast, if
MAP_EXCLis specified, the request will fail if a mapping already exists within the range.
- Notify the kernel that the region may contain semaphores and that special handling may be necessary.
- Region is not included in a core file.
- Causes data dirtied via this VM map to be flushed to physical media only
when necessary (usually by the pager) rather than gratuitously. Typically
this prevents the update daemons from flushing pages dirtied through such
maps and thus allows efficient sharing of memory across unassociated
processes using a file-backed shared memory map. Without this option any
VM pages you dirty may be flushed to disk every so often (every 30-60
seconds usually) which can create performance problems if you do not need
that to occur (such as when you are using shared file-backed mmap regions
for IPC purposes). Note that VM/file system coherency is maintained
whether you use
MAP_NOSYNCor not. This option is not portable across UNIX platforms (yet), though some may implement the same behavior by default.
WARNING! Extending a file with ftruncate(2), thus creating a big hole, and then filling the hole by modifying a shared
mmap() can lead to severe file fragmentation. In order to avoid such fragmentation you should always pre-allocate the file's backing store by
write()ing zero's into the newly extended area prior to modifying the area via your
mmap(). The fragmentation problem is especially sensitive to
MAP_NOSYNCpages, because pages may be flushed to disk in a totally random order.
The same applies when using
MAP_NOSYNCto implement a file-based shared memory store. It is recommended that you create the backing store by
write()ing zero's to the backing file rather than
ftruncate()ing it. You can test file fragmentation by observing the KB/t (kilobytes per transfer) results from an “
iostat 1” while reading a large file sequentially, e.g., using “
dd if=filename of=/dev/null bs=32k”.
The fsync(2) system call will flush all dirty data and metadata associated with a file, including dirty NOSYNC VM data, to physical media. The sync(8) command and sync(2) system call generally do not flush dirty NOSYNC VM data. The msync(2) system call is usually not needed since BSD implements a coherent file system buffer cache. However, it may be used to associate dirty VM pages with file system buffers and thus cause them to be flushed to physical media sooner rather than later.
- Immediately update the calling process's lowest-level virtual address
translation structures, such as its page table, so that every memory
resident page within the region is mapped for read access. Ordinarily
these structures are updated lazily. The effect of this option is to
eliminate any soft faults that would otherwise occur on the initial read
accesses to the region. Although this option does not preclude
prot from including
PROT_WRITE, it does not eliminate soft faults on the initial write accesses to the region.
- Modifications are private.
- Modifications are shared.
MAP_ANON, and offset of 0. The fd argument must be -1 and prot must include at least
PROT_WRITE. This option creates a memory region that grows to at most len bytes in size, starting from the stack top and growing down. The stack top is the starting address returned by the call, plus len bytes. The bottom of the stack at maximum growth is the starting address returned by the call.
NOTES¶Although this implementation does not impose any alignment restrictions on the offset argument, a portable program must only use page-aligned values.
Large page mappings require that the pages backing an object be
aligned in matching blocks in both the virtual address space and RAM. The
system will automatically attempt to use large page mappings when mapping an
object that is already backed by large pages in RAM by aligning the mapping
request in the virtual address space to match the alignment of the large
physical pages. The system may also use large page mappings when mapping
portions of an object that are not yet backed by pages in RAM. The
MAP_ALIGNED_SUPER flag is an optimization that will
align the mapping request to the size of a large page similar to
MAP_ALIGNED, except that the system will override
this alignment if an object already uses large pages so that the mapping
will be consistent with the existing large pages. This flag is mostly useful
for maximizing the use of large pages on the first mapping of objects that
do not yet have pages present in RAM.
RETURN VALUES¶Upon successful completion,
mmap() returns a pointer to the mapped region. Otherwise, a value of
MAP_FAILEDis returned and errno is set to indicate the error.
mmap() system call will fail if:
- The flag
PROT_READwas specified as part of the prot argument and fd was not open for reading. The flags
PROT_WRITEwere specified as part of the flags and prot argument and fd was not open for writing.
- The fd argument is not a valid open file descriptor.
- An invalid value was passed in the prot argument.
- An undefined option was set in the flags argument.
- None of
MAP_STACKwas specified. At least one of these flags must be included.
MAP_FIXEDwas specified and the addr argument was not page aligned, or part of the desired address space resides out of the valid address space for a user process.
MAP_32BITwere specified and part of the desired address space resides outside of the first 2GB of user address space.
- The len argument was equal to zero.
MAP_ALIGNEDwas specified and the desired alignment was either larger than the virtual address size of the machine or smaller than a page.
MAP_ANONwas specified and the fd argument was not -1.
MAP_ANONwas specified and the offset argument was not 0.
MAP_EXCLwere specified, but the requested region is already used by a mapping.
MAP_EXCLwas specified, but
MAP_ANONhas not been specified and fd did not reference a regular or character special file.
MAP_FIXEDwas specified and the addr argument was not available.
MAP_ANONwas specified and insufficient memory was available.
SEE ALSO¶madvise(2), mincore(2), minherit(2), mlock(2), mprotect(2), msync(2), munlock(2), munmap(2), getpagesize(3), getpagesizes(3)
|February 18, 2015||Linux 4.9.0-9-amd64|