aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2009-01-09 17:00:58 -0500
committerLinus Torvalds <torvalds@linux-foundation.org>2009-01-09 17:00:58 -0500
commitc40f6f8bbc4cbd2902671aacd587400ddca62627 (patch)
treea991e5521e10943f4457fb7f494e00aec75cc7df /Documentation
parent1a7d0f0bec4be078ce2cfb11538c0f4ffbbed8e5 (diff)
parentcb6ff208076b5f434db1b8c983429269d719cef5 (diff)
Merge git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-2.6-nommu
* git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-2.6-nommu: NOMMU: Support XIP on initramfs NOMMU: Teach kobjsize() about VMA regions. FLAT: Don't attempt to expand the userspace stack to fill the space allocated FDPIC: Don't attempt to expand the userspace stack to fill the space allocated NOMMU: Improve procfs output using per-MM VMAs NOMMU: Make mmap allocation page trimming behaviour configurable. NOMMU: Make VMAs per MM as for MMU-mode linux NOMMU: Delete askedalloc and realalloc variables NOMMU: Rename ARM's struct vm_region NOMMU: Fix cleanup handling in ramfs_nommu_get_umapped_area()
Diffstat (limited to 'Documentation')
-rw-r--r--Documentation/nommu-mmap.txt31
-rw-r--r--Documentation/sysctl/vm.txt18
2 files changed, 44 insertions, 5 deletions
diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 7714f57caad5..b565e8279d13 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -109,12 +109,18 @@ and it's also much more restricted in the latter case:
109FURTHER NOTES ON NO-MMU MMAP 109FURTHER NOTES ON NO-MMU MMAP
110============================ 110============================
111 111
112 (*) A request for a private mapping of less than a page in size may not return 112 (*) A request for a private mapping of a file may return a buffer that is not
113 a page-aligned buffer. This is because the kernel calls kmalloc() to 113 page-aligned. This is because XIP may take place, and the data may not be
114 allocate the buffer, not get_free_page(). 114 paged aligned in the backing store.
115 115
116 (*) A list of all the mappings on the system is visible through /proc/maps in 116 (*) A request for an anonymous mapping will always be page aligned. If
117 no-MMU mode. 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.
118 124
119 (*) A list of all the mappings in use by a process is visible through 125 (*) A list of all the mappings in use by a process is visible through
120 /proc/<pid>/maps in no-MMU mode. 126 /proc/<pid>/maps in no-MMU mode.
@@ -242,3 +248,18 @@ PROVIDING SHAREABLE BLOCK DEVICE SUPPORT
242Provision of shared mappings on block device files is exactly the same as for 248Provision of shared mappings on block device files is exactly the same as for
243character devices. If there isn't a real device underneath, then the driver 249character devices. If there isn't a real device underneath, then the driver
244should allocate sufficient contiguous memory to honour any supported mapping. 250should allocate sufficient contiguous memory to honour any supported mapping.
251
252
253=================================
254ADJUSTING PAGE TRIMMING BEHAVIOUR
255=================================
256
257NOMMU mmap automatically rounds up to the nearest power-of-2 number of pages
258when performing an allocation. This can have adverse effects on memory
259fragmentation, and as such, is left configurable. The default behaviour is to
260aggressively trim allocations and discard any excess pages back in to the page
261allocator. In order to retain finer-grained control over fragmentation, this
262behaviour can either be disabled completely, or bumped up to a higher page
263watermark where trimming begins.
264
265Page trimming behaviour is configurable via the sysctl `vm.nr_trim_pages'.
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index cd05994a49e6..a3415070bcac 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -38,6 +38,7 @@ Currently, these files are in /proc/sys/vm:
38- numa_zonelist_order 38- numa_zonelist_order
39- nr_hugepages 39- nr_hugepages
40- nr_overcommit_hugepages 40- nr_overcommit_hugepages
41- nr_trim_pages (only if CONFIG_MMU=n)
41 42
42============================================================== 43==============================================================
43 44
@@ -348,3 +349,20 @@ Change the maximum size of the hugepage pool. The maximum is
348nr_hugepages + nr_overcommit_hugepages. 349nr_hugepages + nr_overcommit_hugepages.
349 350
350See Documentation/vm/hugetlbpage.txt 351See Documentation/vm/hugetlbpage.txt
352
353==============================================================
354
355nr_trim_pages
356
357This is available only on NOMMU kernels.
358
359This value adjusts the excess page trimming behaviour of power-of-2 aligned
360NOMMU mmap allocations.
361
362A value of 0 disables trimming of allocations entirely, while a value of 1
363trims excess pages aggressively. Any value >= 1 acts as the watermark where
364trimming of allocations is initiated.
365
366The default value is 1.
367
368See Documentation/nommu-mmap.txt for more information.