mpage_readpages(9) [centos man page]

MPAGE_READPAGES(9)						   The Linux VFS						MPAGE_READPAGES(9)

NAME

       mpage_readpages - populate an address space with some pages & start reads against them

SYNOPSIS

       int mpage_readpages(struct address_space * mapping, struct list_head * pages, unsigned nr_pages, get_block_t get_block);

ARGUMENTS

       mapping
	   the address_space

       pages
	   The address of a list_head which contains the target pages. These pages have their ->index populated and are otherwise uninitialised.
	   The page at pages->prev has the lowest file offset, and reads should be issued in pages->prev to pages->next order.

       nr_pages
	   The number of pages at *pages

       get_block
	   The filesystem's block mapper function.

DESCRIPTION

       This function walks the pages and the blocks within each page, building and emitting large BIOs.

       If anything unusual happens, such as:

       - encountering a page which has buffers - encountering a page which has a non-hole after a hole - encountering a page with non-contiguous
       blocks

       then this code just gives up and calls the buffer_head-based read function. It does handle a page which has holes at the end - that is a
       common case: the end-of-file on blocksize < PAGE_CACHE_SIZE setups.

BH_BOUNDARY EXPLANATION
       There is a problem. The mpage read code assembles several pages, gets all their disk mappings, and then submits them all. That's fine, but
       obtaining the disk mappings may require I/O. Reads of indirect blocks, for example.

       So an mpage read of the first 16 blocks of an ext2 file will cause I/O to be

SUBMITTED IN THE FOLLOWING ORDER

       12 0 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16

       because the indirect block has to be read to get the mappings of blocks 13,14,15,16. Obviously, this impacts performance.

       So what we do it to allow the filesystem's get_block function to set BH_Boundary when it maps block 11. BH_Boundary says: mapping of the
       block after this one will require I/O against a block which is probably close to this one. So you should push what I/O you have currently
       accumulated.

       This all causes the disk requests to be issued in the correct order.

COPYRIGHT

Kernel Hackers Manual 3.10					     June 2014							MPAGE_READPAGES(9)

MLOCK(2)						      BSD System Calls Manual							  MLOCK(2)

NAME

     mlock, munlock -- lock (unlock) physical pages in memory

SYNOPSIS

     #include <sys/types.h>
     #include <sys/mman.h>

     int
     mlock(caddr_t addr, size_t len);

     int
     munlock(caddr_t addr, size_t len);

DESCRIPTION

     The mlock system call locks into memory the physical pages associated with the virtual address range starting at addr for len bytes.  The
     munlock call unlocks pages previously locked by one or more mlock calls.  For both, the addr parameter should be aligned to a multiple of the
     page size.  If the len parameter is not a multiple of the page size, it will be rounded up to be so.  The entire range must be allocated.

     After an mlock call, the indicated pages will cause neither a non-resident page nor address-translation fault until they are unlocked.  They
     may still cause protection-violation faults or TLB-miss faults on architectures with software-managed TLBs.  The physical pages remain in
     memory until all locked mappings for the pages are removed.  Multiple processes may have the same physical pages locked via their own virtual
     address mappings.	A single process may likewise have pages multiply-locked via different virtual mappings of the same pages or via nested
     mlock calls on the same address range.  Unlocking is performed explicitly by munlock or implicitly by a call to munmap which deallocates the
     unmapped address range.  Locked mappings are not inherited by the child process after a fork(2).

     Since physical memory is a potentially scarce resource, processes are limited in how much they can lock down.  A single process can mlock the
     minimum of a system-wide ``wired pages'' limit and the per-process RLIMIT_MEMLOCK resource limit.

RETURN VALUES

     A return value of 0 indicates that the call succeeded and all pages in the range have either been locked or unlocked.  A return value of -1
     indicates an error occurred and the locked status of all pages in the range remains unchanged.  In this case, the global location errno is
     set to indicate the error.

ERRORS

     Mlock() will fail if:

     [EINVAL]		The address given is not page aligned or the length is negative.

     [EAGAIN]		Locking the indicated range would exceed either the system or per-process limit for locked memory.

     [ENOMEM]		Some portion of the indicated address range is not allocated.  There was an error faulting/mapping a page.
     Munlock() will fail if:

     [EINVAL]		The address given is not page aligned or the length is negative.

     [ENOMEM]		Some portion of the indicated address range is not allocated.  Some portion of the indicated address range is not locked.

SEE ALSO

     fork(2), mincore(2), minherit(2), mmap(2), munmap(2), setrlimit(2), getpagesize(3)

BUGS

     Unlike The Sun implementation, multiple mlock calls on the same address range require the corresponding number of munlock calls to actually
     unlock the pages, i.e.  mlock nests.  This should be considered a consequence of the implementation and not a feature.

     The per-process resource limit is a limit on the amount of virtual memory locked, while the system-wide limit is for the number of locked
     physical pages.  Hence a process with two distinct locked mappings of the same physical page counts as 2 pages against the per-process limit
     and as only a single page in the system limit.

HISTORY

     The mlock() and munlock() functions first appeared in 4.4BSD.

BSD
								   June 2, 1993 							       BSD