Commit c24b7201 authored by David Howells's avatar David Howells Committed by Linus Torvalds

mm: reformat the Unevictable-LRU documentation

Do a bit of reformatting on the Unevictable-LRU documentation.
Signed-off-by: default avatarDavid Howells <dhowells@redhat.com>
Acked-by: default avatarLee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 5a52edde
==============================
This document describes the Linux memory management "Unevictable LRU" UNEVICTABLE LRU INFRASTRUCTURE
infrastructure and the use of this infrastructure to manage several types ==============================
of "unevictable" pages. The document attempts to provide the overall
rationale behind this mechanism and the rationale for some of the design ========
decisions that drove the implementation. The latter design rationale is CONTENTS
discussed in the context of an implementation description. Admittedly, one ========
can obtain the implementation details--the "what does it do?"--by reading the
code. One hopes that the descriptions below add value by provide the answer (*) The Unevictable LRU
to "why does it do that?".
- The unevictable page list.
Unevictable LRU Infrastructure: - Memory control group interaction.
- Marking address spaces unevictable.
The Unevictable LRU adds an additional LRU list to track unevictable pages - Detecting Unevictable Pages.
and to hide these pages from vmscan. This mechanism is based on a patch by - vmscan's handling of unevictable pages.
Larry Woodman of Red Hat to address several scalability problems with page
(*) mlock()'d pages.
- History.
- Basic management.
- mlock()/mlockall() system call handling.
- Filtering special vmas.
- munlock()/munlockall() system call handling.
- Migrating mlocked pages.
- mmap(MAP_LOCKED) system call handling.
- munmap()/exit()/exec() system call handling.
- try_to_unmap().
- try_to_munlock() reverse map scan.
- Page reclaim in shrink_*_list().
============
INTRODUCTION
============
This document describes the Linux memory manager's "Unevictable LRU"
infrastructure and the use of this to manage several types of "unevictable"
pages.
The document attempts to provide the overall rationale behind this mechanism
and the rationale for some of the design decisions that drove the
implementation. The latter design rationale is discussed in the context of an
implementation description. Admittedly, one can obtain the implementation
details - the "what does it do?" - by reading the code. One hopes that the
descriptions below add value by provide the answer to "why does it do that?".
===================
THE UNEVICTABLE LRU
===================
The Unevictable LRU facility adds an additional LRU list to track unevictable
pages and to hide these pages from vmscan. This mechanism is based on a patch
by Larry Woodman of Red Hat to address several scalability problems with page
reclaim in Linux. The problems have been observed at customer sites on large reclaim in Linux. The problems have been observed at customer sites on large
memory x86_64 systems. For example, a non-numal x86_64 platform with 128GB memory x86_64 systems.
of main memory will have over 32 million 4k pages in a single zone. When a
large fraction of these pages are not evictable for any reason [see below], To illustrate this with an example, a non-NUMA x86_64 platform with 128GB of
vmscan will spend a lot of time scanning the LRU lists looking for the small main memory will have over 32 million 4k pages in a single zone. When a large
fraction of pages that are evictable. This can result in a situation where fraction of these pages are not evictable for any reason [see below], vmscan
all cpus are spending 100% of their time in vmscan for hours or days on end, will spend a lot of time scanning the LRU lists looking for the small fraction
with the system completely unresponsive. of pages that are evictable. This can result in a situation where all CPUs are
spending 100% of their time in vmscan for hours or days on end, with the system
The Unevictable LRU infrastructure addresses the following classes of completely unresponsive.
unevictable pages:
The unevictable list addresses the following classes of unevictable pages:
+ page owned by ramfs
+ page mapped into SHM_LOCKed shared memory regions (*) Those owned by ramfs.
+ page mapped into VM_LOCKED [mlock()ed] vmas
(*) Those mapped into SHM_LOCK'd shared memory regions.
The infrastructure might be able to handle other conditions that make pages
(*) Those mapped into VM_LOCKED [mlock()ed] VMAs.
The infrastructure may also be able to handle other conditions that make pages
unevictable, either by definition or by circumstance, in the future. unevictable, either by definition or by circumstance, in the future.
The Unevictable LRU List THE UNEVICTABLE PAGE LIST
-------------------------
The Unevictable LRU infrastructure consists of an additional, per-zone, LRU list The Unevictable LRU infrastructure consists of an additional, per-zone, LRU list
called the "unevictable" list and an associated page flag, PG_unevictable, to called the "unevictable" list and an associated page flag, PG_unevictable, to
indicate that the page is being managed on the unevictable list. The indicate that the page is being managed on the unevictable list.
PG_unevictable flag is analogous to, and mutually exclusive with, the PG_active
flag in that it indicates on which LRU list a page resides when PG_lru is set. The PG_unevictable flag is analogous to, and mutually exclusive with, the
The unevictable LRU list is source configurable based on the UNEVICTABLE_LRU PG_active flag in that it indicates on which LRU list a page resides when
Kconfig option. PG_lru is set. The unevictable list is compile-time configurable based on the
UNEVICTABLE_LRU Kconfig option.
The Unevictable LRU infrastructure maintains unevictable pages on an additional The Unevictable LRU infrastructure maintains unevictable pages on an additional
LRU list for a few reasons: LRU list for a few reasons:
1) We get to "treat unevictable pages just like we treat other pages in the (1) We get to "treat unevictable pages just like we treat other pages in the
system, which means we get to use the same code to manipulate them, the system - which means we get to use the same code to manipulate them, the
same code to isolate them (for migrate, etc.), the same code to keep track same code to isolate them (for migrate, etc.), the same code to keep track
of the statistics, etc..." [Rik van Riel] of the statistics, etc..." [Rik van Riel]
2) We want to be able to migrate unevictable pages between nodes--for memory (2) We want to be able to migrate unevictable pages between nodes for memory
defragmentation, workload management and memory hotplug. The linux kernel defragmentation, workload management and memory hotplug. The linux kernel
can only migrate pages that it can successfully isolate from the lru lists. can only migrate pages that it can successfully isolate from the LRU
If we were to maintain pages elsewise than on an lru-like list, where they lists. If we were to maintain pages elsewhere than on an LRU-like list,
can be found by isolate_lru_page(), we would prevent their migration, unless where they can be found by isolate_lru_page(), we would prevent their
we reworked migration code to find the unevictable pages. migration, unless we reworked migration code to find the unevictable pages
itself.
The unevictable list does not differentiate between file-backed and anonymous,
swap-backed pages. This differentiation is only important while the pages are,
in fact, evictable.
The unevictable LRU list does not differentiate between file backed and swap The unevictable list benefits from the "arrayification" of the per-zone LRU
backed [anon] pages. This differentiation is only important while the pages lists and statistics originally proposed and posted by Christoph Lameter.
are, in fact, evictable.
The unevictable LRU list benefits from the "arrayification" of the per-zone The unevictable list does not use the LRU pagevec mechanism. Rather,
LRU lists and statistics originally proposed and posted by Christoph Lameter. unevictable pages are placed directly on the page's zone's unevictable list
under the zone lru_lock. This allows us to prevent the stranding of pages on
the unevictable list when one task has the page isolated from the LRU and other
tasks are changing the "evictability" state of the page.
The unevictable list does not use the lru pagevec mechanism. Rather,
unevictable pages are placed directly on the page's zone's unevictable
list under the zone lru_lock. The reason for this is to prevent stranding
of pages on the unevictable list when one task has the page isolated from the
lru and other tasks are changing the "evictability" state of the page.
MEMORY CONTROL GROUP INTERACTION
--------------------------------
Unevictable LRU and Memory Controller Interaction The unevictable LRU facility interacts with the memory control group [aka
memory controller; see Documentation/cgroups/memory.txt] by extending the
lru_list enum.
The memory controller data structure automatically gets a per-zone unevictable
list as a result of the "arrayification" of the per-zone LRU lists (one per
lru_list enum element). The memory controller tracks the movement of pages to
and from the unevictable list.
The memory controller data structure automatically gets a per zone unevictable
lru list as a result of the "arrayification" of the per-zone LRU lists. The
memory controller tracks the movement of pages to and from the unevictable list.
When a memory control group comes under memory pressure, the controller will When a memory control group comes under memory pressure, the controller will
not attempt to reclaim pages on the unevictable list. This has a couple of not attempt to reclaim pages on the unevictable list. This has a couple of
effects. Because the pages are "hidden" from reclaim on the unevictable list, effects:
the reclaim process can be more efficient, dealing only with pages that have
a chance of being reclaimed. On the other hand, if too many of the pages (1) Because the pages are "hidden" from reclaim on the unevictable list, the
charged to the control group are unevictable, the evictable portion of the reclaim process can be more efficient, dealing only with pages that have a
working set of the tasks in the control group may not fit into the available chance of being reclaimed.
memory. This can cause the control group to thrash or to oom-kill tasks.
(2) On the other hand, if too many of the pages charged to the control group
are unevictable, the evictable portion of the working set of the tasks in
Unevictable LRU: Detecting Unevictable Pages the control group may not fit into the available memory. This can cause
the control group to thrash or to OOM-kill tasks.
The function page_evictable(page, vma) in vmscan.c determines whether a
page is evictable or not. For ramfs pages and pages in SHM_LOCKed regions,
page_evictable() tests a new address space flag, AS_UNEVICTABLE, in the page's MARKING ADDRESS SPACES UNEVICTABLE
address space using a wrapper function. Wrapper functions are used to set, ----------------------------------
clear and test the flag to reduce the requirement for #ifdef's throughout the
source code. AS_UNEVICTABLE is set on ramfs inode/mapping when it is created. For facilities such as ramfs none of the pages attached to the address space
This flag remains for the life of the inode. may be evicted. To prevent eviction of any such pages, the AS_UNEVICTABLE
address space flag is provided, and this can be manipulated by a filesystem
For shared memory regions, AS_UNEVICTABLE is set when an application using a number of wrapper functions:
successfully SHM_LOCKs the region and is removed when the region is
SHM_UNLOCKed. Note that shmctl(SHM_LOCK, ...) does not populate the page (*) void mapping_set_unevictable(struct address_space *mapping);
tables for the region as does, for example, mlock(). So, we make no special
effort to push any pages in the SHM_LOCKed region to the unevictable list. Mark the address space as being completely unevictable.
Vmscan will do this when/if it encounters the pages during reclaim. On
SHM_UNLOCK, shmctl() scans the pages in the region and "rescues" them from the (*) void mapping_clear_unevictable(struct address_space *mapping);
unevictable list if no other condition keeps them unevictable. If a SHM_LOCKed
region is destroyed, the pages are also "rescued" from the unevictable list in Mark the address space as being evictable.
the process of freeing them.
(*) int mapping_unevictable(struct address_space *mapping);
page_evictable() detects mlock()ed pages by testing an additional page flag,
PG_mlocked via the PageMlocked() wrapper. If the page is NOT mlocked, and a Query the address space, and return true if it is completely
non-NULL vma is supplied, page_evictable() will check whether the vma is unevictable.
These are currently used in two places in the kernel:
(1) By ramfs to mark the address spaces of its inodes when they are created,
and this mark remains for the life of the inode.
(2) By SYSV SHM to mark SHM_LOCK'd address spaces until SHM_UNLOCK is called.
Note that SHM_LOCK is not required to page in the locked pages if they're
swapped out; the application must touch the pages manually if it wants to
ensure they're in memory.
DETECTING UNEVICTABLE PAGES
---------------------------
The function page_evictable() in vmscan.c determines whether a page is
evictable or not using the query function outlined above [see section "Marking
address spaces unevictable"] to check the AS_UNEVICTABLE flag.
For address spaces that are so marked after being populated (as SHM regions
might be), the lock action (eg: SHM_LOCK) can be lazy, and need not populate
the page tables for the region as does, for example, mlock(), nor need it make
any special effort to push any pages in the SHM_LOCK'd area to the unevictable
list. Instead, vmscan will do this if and when it encounters the pages during
a reclamation scan.
On an unlock action (such as SHM_UNLOCK), the unlocker (eg: shmctl()) must scan
the pages in the region and "rescue" them from the unevictable list if no other
condition is keeping them unevictable. If an unevictable region is destroyed,
the pages are also "rescued" from the unevictable list in the process of
freeing them.
page_evictable() also checks for mlocked pages by testing an additional page
flag, PG_mlocked (as wrapped by PageMlocked()). If the page is NOT mlocked,
and a non-NULL VMA is supplied, page_evictable() will check whether the VMA is
VM_LOCKED via is_mlocked_vma(). is_mlocked_vma() will SetPageMlocked() and VM_LOCKED via is_mlocked_vma(). is_mlocked_vma() will SetPageMlocked() and
update the appropriate statistics if the vma is VM_LOCKED. This method allows update the appropriate statistics if the vma is VM_LOCKED. This method allows
efficient "culling" of pages in the fault path that are being faulted in to efficient "culling" of pages in the fault path that are being faulted in to
VM_LOCKED vmas. VM_LOCKED VMAs.
Unevictable Pages and Vmscan [shrink_*_list()] VMSCAN'S HANDLING OF UNEVICTABLE PAGES
--------------------------------------
If unevictable pages are culled in the fault path, or moved to the unevictable If unevictable pages are culled in the fault path, or moved to the unevictable
list at mlock() or mmap() time, vmscan will never encounter the pages until list at mlock() or mmap() time, vmscan will not encounter the pages until they
they have become evictable again, for example, via munlock() and have been have become evictable again (via munlock() for example) and have been "rescued"
"rescued" from the unevictable list. However, there may be situations where we from the unevictable list. However, there may be situations where we decide,
decide, for the sake of expediency, to leave a unevictable page on one of the for the sake of expediency, to leave a unevictable page on one of the regular
regular active/inactive LRU lists for vmscan to deal with. Vmscan checks for active/inactive LRU lists for vmscan to deal with. vmscan checks for such
such pages in all of the shrink_{active|inactive|page}_list() functions and pages in all of the shrink_{active|inactive|page}_list() functions and will
will "cull" such pages that it encounters--that is, it diverts those pages to "cull" such pages that it encounters: that is, it diverts those pages to the
the unevictable list for the zone being scanned. unevictable list for the zone being scanned.
There may be situations where a page is mapped into a VM_LOCKED vma, but the There may be situations where a page is mapped into a VM_LOCKED VMA, but the
page is not marked as PageMlocked. Such pages will make it all the way to page is not marked as PG_mlocked. Such pages will make it all the way to
shrink_page_list() where they will be detected when vmscan walks the reverse shrink_page_list() where they will be detected when vmscan walks the reverse
map in try_to_unmap(). If try_to_unmap() returns SWAP_MLOCK, shrink_page_list() map in try_to_unmap(). If try_to_unmap() returns SWAP_MLOCK,
will cull the page at that point. shrink_page_list() will cull the page at that point.
To "cull" an unevictable page, vmscan simply puts the page back on the lru To "cull" an unevictable page, vmscan simply puts the page back on the LRU list
list using putback_lru_page()--the inverse operation to isolate_lru_page()-- using putback_lru_page() - the inverse operation to isolate_lru_page() - after
after dropping the page lock. Because the condition which makes the page dropping the page lock. Because the condition which makes the page unevictable
unevictable may change once the page is unlocked, putback_lru_page() will may change once the page is unlocked, putback_lru_page() will recheck the
recheck the unevictable state of a page that it places on the unevictable lru unevictable state of a page that it places on the unevictable list. If the
list. If the page has become unevictable, putback_lru_page() removes it from page has become unevictable, putback_lru_page() removes it from the list and
the list and retries, including the page_unevictable() test. Because such a retries, including the page_unevictable() test. Because such a race is a rare
race is a rare event and movement of pages onto the unevictable list should be event and movement of pages onto the unevictable list should be rare, these
rare, these extra evictabilty checks should not occur in the majority of calls extra evictabilty checks should not occur in the majority of calls to
to putback_lru_page(). putback_lru_page().
Mlocked Page: Prior Work =============
MLOCKED PAGES
=============
The "Unevictable Mlocked Pages" infrastructure is based on work originally The unevictable page list is also useful for mlock(), in addition to ramfs and
SYSV SHM. Note that mlock() is only available in CONFIG_MMU=y situations; in
NOMMU situations, all mappings are effectively mlocked.
HISTORY
-------
The "Unevictable mlocked Pages" infrastructure is based on work originally
posted by Nick Piggin in an RFC patch entitled "mm: mlocked pages off LRU". posted by Nick Piggin in an RFC patch entitled "mm: mlocked pages off LRU".
Nick posted his patch as an alternative to a patch posted by Christoph Nick posted his patch as an alternative to a patch posted by Christoph Lameter
Lameter to achieve the same objective--hiding mlocked pages from vmscan. to achieve the same objective: hiding mlocked pages from vmscan.
In Nick's patch, he used one of the struct page lru list link fields as a count
of VM_LOCKED vmas that map the page. This use of the link field for a count In Nick's patch, he used one of the struct page LRU list link fields as a count
prevented the management of the pages on an LRU list. Thus, mlocked pages were of VM_LOCKED VMAs that map the page. This use of the link field for a count
not migratable as isolate_lru_page() could not find them and the lru list link prevented the management of the pages on an LRU list, and thus mlocked pages
field was not available to the migration subsystem. Nick resolved this by were not migratable as isolate_lru_page() could not find them, and the LRU list
putting mlocked pages back on the lru list before attempting to isolate them, link field was not available to the migration subsystem.
thus abandoning the count of VM_LOCKED vmas. When Nick's patch was integrated
with the Unevictable LRU work, the count was replaced by walking the reverse Nick resolved this by putting mlocked pages back on the lru list before
map to determine whether any VM_LOCKED vmas mapped the page. More on this attempting to isolate them, thus abandoning the count of VM_LOCKED VMAs. When
below. Nick's patch was integrated with the Unevictable LRU work, the count was
replaced by walking the reverse map to determine whether any VM_LOCKED VMAs
mapped the page. More on this below.
Mlocked Pages: Basic Management
Mlocked pages--pages mapped into a VM_LOCKED vma--represent one class of BASIC MANAGEMENT
unevictable pages. When such a page has been "noticed" by the memory ----------------
management subsystem, the page is marked with the PG_mlocked [PageMlocked()]
flag. A PageMlocked() page will be placed on the unevictable LRU list when mlocked pages - pages mapped into a VM_LOCKED VMA - are a class of unevictable
it is added to the LRU. Pages can be "noticed" by memory management in pages. When such a page has been "noticed" by the memory management subsystem,
several places: the page is marked with the PG_mlocked flag. This can be manipulated using the
PageMlocked() functions.
1) in the mlock()/mlockall() system call handlers.
2) in the mmap() system call handler when mmap()ing a region with the A PG_mlocked page will be placed on the unevictable list when it is added to
MAP_LOCKED flag, or mmap()ing a region in a task that has called the LRU. Such pages can be "noticed" by memory management in several places:
mlockall() with the MCL_FUTURE flag. Both of these conditions result
in the VM_LOCKED flag being set for the vma. (1) in the mlock()/mlockall() system call handlers;
3) in the fault path, if mlocked pages are "culled" in the fault path,
and when a VM_LOCKED stack segment is expanded. (2) in the mmap() system call handler when mmapping a region with the
4) as mentioned above, in vmscan:shrink_page_list() when attempting to MAP_LOCKED flag;
reclaim a page in a VM_LOCKED vma via try_to_unmap().
(3) mmapping a region in a task that has called mlockall() with the MCL_FUTURE
Mlocked pages become unlocked and rescued from the unevictable list when: flag
1) mapped in a range unlocked via the munlock()/munlockall() system calls. (4) in the fault path, if mlocked pages are "culled" in the fault path,
2) munmapped() out of the last VM_LOCKED vma that maps the page, including and when a VM_LOCKED stack segment is expanded; or
unmapping at task exit.
3) when the page is truncated from the last VM_LOCKED vma of an mmap()ed file. (5) as mentioned above, in vmscan:shrink_page_list() when attempting to
4) before a page is COWed in a VM_LOCKED vma. reclaim a page in a VM_LOCKED VMA via try_to_unmap()
all of which result in the VM_LOCKED flag being set for the VMA if it doesn't
Mlocked Pages: mlock()/mlockall() System Call Handling already have it set.
mlocked pages become unlocked and rescued from the unevictable list when:
(1) mapped in a range unlocked via the munlock()/munlockall() system calls;
(2) munmap()'d out of the last VM_LOCKED VMA that maps the page, including
unmapping at task exit;
(3) when the page is truncated from the last VM_LOCKED VMA of an mmapped file;
or
(4) before a page is COW'd in a VM_LOCKED VMA.
mlock()/mlockall() SYSTEM CALL HANDLING
---------------------------------------
Both [do_]mlock() and [do_]mlockall() system call handlers call mlock_fixup() Both [do_]mlock() and [do_]mlockall() system call handlers call mlock_fixup()
for each vma in the range specified by the call. In the case of mlockall(), for each VMA in the range specified by the call. In the case of mlockall(),
this is the entire active address space of the task. Note that mlock_fixup() this is the entire active address space of the task. Note that mlock_fixup()
is used for both mlock()ing and munlock()ing a range of memory. A call to is used for both mlocking and munlocking a range of memory. A call to mlock()
mlock() an already VM_LOCKED vma, or to munlock() a vma that is not VM_LOCKED an already VM_LOCKED VMA, or to munlock() a VMA that is not VM_LOCKED is
is treated as a no-op--mlock_fixup() simply returns. treated as a no-op, and mlock_fixup() simply returns.
If the vma passes some filtering described in "Mlocked Pages: Filtering Vmas" If the VMA passes some filtering as described in "Filtering Special Vmas"
below, mlock_fixup() will attempt to merge the vma with its neighbors or split below, mlock_fixup() will attempt to merge the VMA with its neighbors or split
off a subset of the vma if the range does not cover the entire vma. Once the off a subset of the VMA if the range does not cover the entire VMA. Once the
vma has been merged or split or neither, mlock_fixup() will call VMA has been merged or split or neither, mlock_fixup() will call
__mlock_vma_pages_range() to fault in the pages via get_user_pages() and __mlock_vma_pages_range() to fault in the pages via get_user_pages() and to
to mark the pages as mlocked via mlock_vma_page(). mark the pages as mlocked via mlock_vma_page().
Note that the vma being mlocked might be mapped with PROT_NONE. In this case, Note that the VMA being mlocked might be mapped with PROT_NONE. In this case,
get_user_pages() will be unable to fault in the pages. That's OK. If pages get_user_pages() will be unable to fault in the pages. That's okay. If pages
do end up getting faulted into this VM_LOCKED vma, we'll handle them in the do end up getting faulted into this VM_LOCKED VMA, we'll handle them in the
fault path or in vmscan. fault path or in vmscan.
Also note that a page returned by get_user_pages() could be truncated or Also note that a page returned by get_user_pages() could be truncated or
migrated out from under us, while we're trying to mlock it. To detect migrated out from under us, while we're trying to mlock it. To detect this,
this, __mlock_vma_pages_range() tests the page_mapping after acquiring __mlock_vma_pages_range() checks page_mapping() after acquiring the page lock.
the page lock. If the page is still associated with its mapping, we'll If the page is still associated with its mapping, we'll go ahead and call
go ahead and call mlock_vma_page(). If the mapping is gone, we just mlock_vma_page(). If the mapping is gone, we just unlock the page and move on.
unlock the page and move on. Worse case, this results in page mapped In the worst case, this will result in a page mapped in a VM_LOCKED VMA
in a VM_LOCKED vma remaining on a normal LRU list without being remaining on a normal LRU list without being PageMlocked(). Again, vmscan will
PageMlocked(). Again, vmscan will detect and cull such pages. detect and cull such pages.
mlock_vma_page(), called with the page locked [N.B., not "mlocked"], will mlock_vma_page() will call TestSetPageMlocked() for each page returned by
TestSetPageMlocked() for each page returned by get_user_pages(). We use get_user_pages(). We use TestSetPageMlocked() because the page might already
TestSetPageMlocked() because the page might already be mlocked by another be mlocked by another task/VMA and we don't want to do extra work. We
task/vma and we don't want to do extra work. We especially do not want to especially do not want to count an mlocked page more than once in the
count an mlocked page more than once in the statistics. If the page was statistics. If the page was already mlocked, mlock_vma_page() need do nothing
already mlocked, mlock_vma_page() is done. more.
If the page was NOT already mlocked, mlock_vma_page() attempts to isolate the If the page was NOT already mlocked, mlock_vma_page() attempts to isolate the
page from the LRU, as it is likely on the appropriate active or inactive list page from the LRU, as it is likely on the appropriate active or inactive list
at that time. If the isolate_lru_page() succeeds, mlock_vma_page() will at that time. If the isolate_lru_page() succeeds, mlock_vma_page() will put
putback the page--putback_lru_page()--which will notice that the page is now back the page - by calling putback_lru_page() - which will notice that the page
mlocked and divert the page to the zone's unevictable LRU list. If is now mlocked and divert the page to the zone's unevictable list. If
mlock_vma_page() is unable to isolate the page from the LRU, vmscan will handle mlock_vma_page() is unable to isolate the page from the LRU, vmscan will handle
it later if/when it attempts to reclaim the page. it later if and when it attempts to reclaim the page.
Mlocked Pages: Filtering Special Vmas FILTERING SPECIAL VMAS
----------------------
mlock_fixup() filters several classes of "special" vmas: mlock_fixup() filters several classes of "special" VMAs:
1) vmas with VM_IO|VM_PFNMAP set are skipped entirely. The pages behind 1) VMAs with VM_IO or VM_PFNMAP set are skipped entirely. The pages behind
these mappings are inherently pinned, so we don't need to mark them as these mappings are inherently pinned, so we don't need to mark them as
mlocked. In any case, most of the pages have no struct page in which to mlocked. In any case, most of the pages have no struct page in which to so
so mark the page. Because of this, get_user_pages() will fail for these mark the page. Because of this, get_user_pages() will fail for these VMAs,
vmas, so there is no sense in attempting to visit them. so there is no sense in attempting to visit them.
2) vmas mapping hugetlbfs page are already effectively pinned into memory. 2) VMAs mapping hugetlbfs page are already effectively pinned into memory. We
We don't need nor want to mlock() these pages. However, to preserve the neither need nor want to mlock() these pages. However, to preserve the
prior behavior of mlock()--before the unevictable/mlock changes-- prior behavior of mlock() - before the unevictable/mlock changes -
mlock_fixup() will call make_pages_present() in the hugetlbfs vma range mlock_fixup() will call make_pages_present() in the hugetlbfs VMA range to
to allocate the huge pages and populate the ptes. allocate the huge pages and populate the ptes.
3) vmas with VM_DONTEXPAND|VM_RESERVED are generally user space mappings of 3) VMAs with VM_DONTEXPAND or VM_RESERVED are generally userspace mappings of
kernel pages, such as the vdso page, relay channel pages, etc. These pages kernel pages, such as the VDSO page, relay channel pages, etc. These pages
are inherently unevictable and are not managed on the LRU lists. are inherently unevictable and are not managed on the LRU lists.
mlock_fixup() treats these vmas the same as hugetlbfs vmas. It calls mlock_fixup() treats these VMAs the same as hugetlbfs VMAs. It calls
make_pages_present() to populate the ptes. make_pages_present() to populate the ptes.
Note that for all of these special vmas, mlock_fixup() does not set the Note that for all of these special VMAs, mlock_fixup() does not set the
VM_LOCKED flag. Therefore, we won't have to deal with them later during VM_LOCKED flag. Therefore, we won't have to deal with them later during
munlock() or munmap()--for example, at task exit. Neither does mlock_fixup() munlock(), munmap() or task exit. Neither does mlock_fixup() account these
account these vmas against the task's "locked_vm". VMAs against the task's "locked_vm".
Mlocked Pages: Downgrading the Mmap Semaphore.
munlock()/munlockall() SYSTEM CALL HANDLING
mlock_fixup() must be called with the mmap semaphore held for write, because -------------------------------------------
it may have to merge or split vmas. However, mlocking a large region of
memory can take a long time--especially if vmscan must reclaim pages to The munlock() and munlockall() system calls are handled by the same functions -
satisfy the regions requirements. Faulting in a large region with the mmap do_mlock[all]() - as the mlock() and mlockall() system calls with the unlock vs
semaphore held for write can hold off other faults on the address space, in lock operation indicated by an argument. So, these system calls are also
the case of a multi-threaded task. It can also hold off scans of the task's handled by mlock_fixup(). Again, if called for an already munlocked VMA,
address space via /proc. While testing under heavy load, it was observed that mlock_fixup() simply returns. Because of the VMA filtering discussed above,
the ps(1) command could be held off for many minutes while a large segment was VM_LOCKED will not be set in any "special" VMAs. So, these VMAs will be
mlock()ed down.
To address this issue, and to make the system more responsive during mlock()ing
of large segments, mlock_fixup() downgrades the mmap semaphore to read mode
during the call to __mlock_vma_pages_range(). This works fine. However, the
callers of mlock_fixup() expect the semaphore to be returned in write mode.
So, mlock_fixup() "upgrades" the semphore to write mode. Linux does not
support an atomic upgrade_sem() call, so mlock_fixup() must drop the semaphore
and reacquire it in write mode. In a multi-threaded task, it is possible for
the task memory map to change while the semaphore is dropped. Therefore,
mlock_fixup() looks up the vma at the range start address after reacquiring
the semaphore in write mode and verifies that it still covers the original
range. If not, mlock_fixup() returns an error [-EAGAIN]. All callers of
mlock_fixup() have been changed to deal with this new error condition.
Note: when munlocking a region, all of the pages should already be resident--
unless we have racing threads mlocking() and munlocking() regions. So,
unlocking should not have to wait for page allocations nor faults of any kind.
Therefore mlock_fixup() does not downgrade the semaphore for munlock().
Mlocked Pages: munlock()/munlockall() System Call Handling
The munlock() and munlockall() system calls are handled by the same functions--
do_mlock[all]()--as the mlock() and mlockall() system calls with the unlock
vs lock operation indicated by an argument. So, these system calls are also
handled by mlock_fixup(). Again, if called for an already munlock()ed vma,
mlock_fixup() simply returns. Because of the vma filtering discussed above,
VM_LOCKED will not be set in any "special" vmas. So, these vmas will be
ignored for munlock. ignored for munlock.
If the vma is VM_LOCKED, mlock_fixup() again attempts to merge or split off If the VMA is VM_LOCKED, mlock_fixup() again attempts to merge or split off the
the specified range. The range is then munlocked via the function specified range. The range is then munlocked via the function
__mlock_vma_pages_range()--the same function used to mlock a vma range-- __mlock_vma_pages_range() - the same function used to mlock a VMA range -
passing a flag to indicate that munlock() is being performed. passing a flag to indicate that munlock() is being performed.
Because the vma access protections could have been changed to PROT_NONE after Because the VMA access protections could have been changed to PROT_NONE after
faulting in and mlocking pages, get_user_pages() was unreliable for visiting faulting in and mlocking pages, get_user_pages() was unreliable for visiting
these pages for munlocking. Because we don't want to leave pages mlocked(), these pages for munlocking. Because we don't want to leave pages mlocked,
get_user_pages() was enhanced to accept a flag to ignore the permissions when get_user_pages() was enhanced to accept a flag to ignore the permissions when
fetching the pages--all of which should be resident as a result of previous fetching the pages - all of which should be resident as a result of previous
mlock()ing. mlocking.
For munlock(), __mlock_vma_pages_range() unlocks individual pages by calling For munlock(), __mlock_vma_pages_range() unlocks individual pages by calling
munlock_vma_page(). munlock_vma_page() unconditionally clears the PG_mlocked munlock_vma_page(). munlock_vma_page() unconditionally clears the PG_mlocked
flag using TestClearPageMlocked(). As with mlock_vma_page(), munlock_vma_page() flag using TestClearPageMlocked(). As with mlock_vma_page(),
use the Test*PageMlocked() function to handle the case where the page might munlock_vma_page() use the Test*PageMlocked() function to handle the case where
have already been unlocked by another task. If the page was mlocked, the page might have already been unlocked by another task. If the page was
munlock_vma_page() updates that zone statistics for the number of mlocked mlocked, munlock_vma_page() updates that zone statistics for the number of
pages. Note, however, that at this point we haven't checked whether the page mlocked pages. Note, however, that at this point we haven't checked whether
is mapped by other VM_LOCKED vmas. the page is mapped by other VM_LOCKED VMAs.
We can't call try_to_munlock(), the function that walks the reverse map to check We can't call try_to_munlock(), the function that walks the reverse map to
for other VM_LOCKED vmas, without first isolating the page from the LRU. check for other VM_LOCKED VMAs, without first isolating the page from the LRU.
try_to_munlock() is a variant of try_to_unmap() and thus requires that the page try_to_munlock() is a variant of try_to_unmap() and thus requires that the page
not be on an lru list. [More on these below.] However, the call to not be on an LRU list [more on these below]. However, the call to
isolate_lru_page() could fail, in which case we couldn't try_to_munlock(). isolate_lru_page() could fail, in which case we couldn't try_to_munlock(). So,
So, we go ahead and clear PG_mlocked up front, as this might be the only chance we go ahead and clear PG_mlocked up front, as this might be the only chance we
we have. If we can successfully isolate the page, we go ahead and have. If we can successfully isolate the page, we go ahead and
try_to_munlock(), which will restore the PG_mlocked flag and update the zone try_to_munlock(), which will restore the PG_mlocked flag and update the zone
page statistics if it finds another vma holding the page mlocked. If we fail page statistics if it finds another VMA holding the page mlocked. If we fail
to isolate the page, we'll have left a potentially mlocked page on the LRU. to isolate the page, we'll have left a potentially mlocked page on the LRU.
This is fine, because we'll catch it later when/if vmscan tries to reclaim the This is fine, because we'll catch it later if and if vmscan tries to reclaim
page. This should be relatively rare. the page. This should be relatively rare.
Mlocked Pages: Migrating Them...
MIGRATING MLOCKED PAGES
A page that is being migrated has been isolated from the lru lists and is -----------------------
held locked across unmapping of the page, updating the page's mapping
[address_space] entry and copying the contents and state, until the A page that is being migrated has been isolated from the LRU lists and is held
page table entry has been replaced with an entry that refers to the new locked across unmapping of the page, updating the page's address space entry
page. Linux supports migration of mlocked pages and other unevictable and copying the contents and state, until the page table entry has been
pages. This involves simply moving the PageMlocked and PageUnevictable states replaced with an entry that refers to the new page. Linux supports migration
from the old page to the new page. of mlocked pages and other unevictable pages. This involves simply moving the
PG_mlocked and PG_unevictable states from the old page to the new page.
Note that page migration can race with mlocking or munlocking of the same
page. This has been discussed from the mlock/munlock perspective in the Note that page migration can race with mlocking or munlocking of the same page.
respective sections above. Both processes [migration, m[un]locking], hold This has been discussed from the mlock/munlock perspective in the respective
the page locked. This provides the first level of synchronization. Page sections above. Both processes (migration and m[un]locking) hold the page
migration zeros out the page_mapping of the old page before unlocking it, locked. This provides the first level of synchronization. Page migration
so m[un]lock can skip these pages by testing the page mapping under page zeros out the page_mapping of the old page before unlocking it, so m[un]lock
lock. can skip these pages by testing the page mapping under page lock.
When completing page migration, we place the new and old pages back onto the To complete page migration, we place the new and old pages back onto the LRU
lru after dropping the page lock. The "unneeded" page--old page on success, after dropping the page lock. The "unneeded" page - old page on success, new
new page on failure--will be freed when the reference count held by the page on failure - will be freed when the reference count held by the migration
migration process is released. To ensure that we don't strand pages on the process is released. To ensure that we don't strand pages on the unevictable
unevictable list because of a race between munlock and migration, page list because of a race between munlock and migration, page migration uses the
migration uses the putback_lru_page() function to add migrated pages back to putback_lru_page() function to add migrated pages back to the LRU.
the lru.
mmap(MAP_LOCKED) SYSTEM CALL HANDLING
Mlocked Pages: mmap(MAP_LOCKED) System Call Handling -------------------------------------
In addition the the mlock()/mlockall() system calls, an application can request In addition the the mlock()/mlockall() system calls, an application can request
that a region of memory be mlocked using the MAP_LOCKED flag with the mmap() that a region of memory be mlocked supplying the MAP_LOCKED flag to the mmap()
call. Furthermore, any mmap() call or brk() call that expands the heap by a call. Furthermore, any mmap() call or brk() call that expands the heap by a
task that has previously called mlockall() with the MCL_FUTURE flag will result task that has previously called mlockall() with the MCL_FUTURE flag will result
in the newly mapped memory being mlocked. Before the unevictable/mlock changes, in the newly mapped memory being mlocked. Before the unevictable/mlock
the kernel simply called make_pages_present() to allocate pages and populate changes, the kernel simply called make_pages_present() to allocate pages and
the page table. populate the page table.
To mlock a range of memory under the unevictable/mlock infrastructure, the To mlock a range of memory under the unevictable/mlock infrastructure, the
mmap() handler and task address space expansion functions call mmap() handler and task address space expansion functions call
mlock_vma_pages_range() specifying the vma and the address range to mlock. mlock_vma_pages_range() specifying the vma and the address range to mlock.
mlock_vma_pages_range() filters vmas like mlock_fixup(), as described above in mlock_vma_pages_range() filters VMAs like mlock_fixup(), as described above in
"Mlocked Pages: Filtering Vmas". It will clear the VM_LOCKED flag, which will "Filtering Special VMAs". It will clear the VM_LOCKED flag, which will have
have already been set by the caller, in filtered vmas. Thus these vma's need already been set by the caller, in filtered VMAs. Thus these VMA's need not be
not be visited for munlock when the region is unmapped. visited for munlock when the region is unmapped.
For "normal" vmas, mlock_vma_pages_range() calls __mlock_vma_pages_range() to For "normal" VMAs, mlock_vma_pages_range() calls __mlock_vma_pages_range() to
fault/allocate the pages and mlock them. Again, like mlock_fixup(), fault/allocate the pages and mlock them. Again, like mlock_fixup(),
mlock_vma_pages_range() downgrades the mmap semaphore to read mode before mlock_vma_pages_range() downgrades the mmap semaphore to read mode before
attempting to fault/allocate and mlock the pages; and "upgrades" the semaphore attempting to fault/allocate and mlock the pages and "upgrades" the semaphore
back to write mode before returning. back to write mode before returning.
The callers of mlock_vma_pages_range() will have already added the memory The callers of mlock_vma_pages_range() will have already added the memory range
range to be mlocked to the task's "locked_vm". To account for filtered vmas, to be mlocked to the task's "locked_vm". To account for filtered VMAs,
mlock_vma_pages_range() returns the number of pages NOT mlocked. All of the mlock_vma_pages_range() returns the number of pages NOT mlocked. All of the
callers then subtract a non-negative return value from the task's locked_vm. callers then subtract a non-negative return value from the task's locked_vm. A
A negative return value represent an error--for example, from get_user_pages() negative return value represent an error - for example, from get_user_pages()
attempting to fault in a vma with PROT_NONE access. In this case, we leave attempting to fault in a VMA with PROT_NONE access. In this case, we leave the
the memory range accounted as locked_vm, as the protections could be changed memory range accounted as locked_vm, as the protections could be changed later
later and pages allocated into that region. and pages allocated into that region.
Mlocked Pages: munmap()/exit()/exec() System Call Handling munmap()/exit()/exec() SYSTEM CALL HANDLING
-------------------------------------------
When unmapping an mlocked region of memory, whether by an explicit call to When unmapping an mlocked region of memory, whether by an explicit call to
munmap() or via an internal unmap from exit() or exec() processing, we must munmap() or via an internal unmap from exit() or exec() processing, we must
munlock the pages if we're removing the last VM_LOCKED vma that maps the pages. munlock the pages if we're removing the last VM_LOCKED VMA that maps the pages.
Before the unevictable/mlock changes, mlocking did not mark the pages in any Before the unevictable/mlock changes, mlocking did not mark the pages in any
way, so unmapping them required no processing. way, so unmapping them required no processing.
To munlock a range of memory under the unevictable/mlock infrastructure, the To munlock a range of memory under the unevictable/mlock infrastructure, the
munmap() hander and task address space tear down function call munmap() handler and task address space call tear down function
munlock_vma_pages_all(). The name reflects the observation that one always munlock_vma_pages_all(). The name reflects the observation that one always
specifies the entire vma range when munlock()ing during unmap of a region. specifies the entire VMA range when munlock()ing during unmap of a region.
Because of the vma filtering when mlocking() regions, only "normal" vmas that Because of the VMA filtering when mlocking() regions, only "normal" VMAs that
actually contain mlocked pages will be passed to munlock_vma_pages_all(). actually contain mlocked pages will be passed to munlock_vma_pages_all().
munlock_vma_pages_all() clears the VM_LOCKED vma flag and, like mlock_fixup() munlock_vma_pages_all() clears the VM_LOCKED VMA flag and, like mlock_fixup()
for the munlock case, calls __munlock_vma_pages_range() to walk the page table for the munlock case, calls __munlock_vma_pages_range() to walk the page table
for the vma's memory range and munlock_vma_page() each resident page mapped by for the VMA's memory range and munlock_vma_page() each resident page mapped by
the vma. This effectively munlocks the page, only if this is the last the VMA. This effectively munlocks the page, only if this is the last
VM_LOCKED vma that maps the page. VM_LOCKED VMA that maps the page.
Mlocked Page: try_to_unmap()
[Note: the code changes represented by this section are really quite small try_to_unmap()
compared to the text to describe what happening and why, and to discuss the --------------
implications.]
Pages can, of course, be mapped into multiple vmas. Some of these vmas may Pages can, of course, be mapped into multiple VMAs. Some of these VMAs may
have VM_LOCKED flag set. It is possible for a page mapped into one or more have VM_LOCKED flag set. It is possible for a page mapped into one or more
VM_LOCKED vmas not to have the PG_mlocked flag set and therefore reside on one VM_LOCKED VMAs not to have the PG_mlocked flag set and therefore reside on one
of the active or inactive LRU lists. This could happen if, for example, a of the active or inactive LRU lists. This could happen if, for example, a task
task in the process of munlock()ing the page could not isolate the page from in the process of munlocking the page could not isolate the page from the LRU.
the LRU. As a result, vmscan/shrink_page_list() might encounter such a page As a result, vmscan/shrink_page_list() might encounter such a page as described
as described in "Unevictable Pages and Vmscan [shrink_*_list()]". To in section "vmscan's handling of unevictable pages". To handle this situation,
handle this situation, try_to_unmap() has been enhanced to check for VM_LOCKED try_to_unmap() checks for VM_LOCKED VMAs while it is walking a page's reverse
vmas while it is walking a page's reverse map. map.
try_to_unmap() is always called, by either vmscan for reclaim or for page try_to_unmap() is always called, by either vmscan for reclaim or for page
migration, with the argument page locked and isolated from the LRU. BUG_ON() migration, with the argument page locked and isolated from the LRU. Separate
assertions enforce this requirement. Separate functions handle anonymous and functions handle anonymous and mapped file pages, as these types of pages have
mapped file pages, as these types of pages have different reverse map different reverse map mechanisms.
mechanisms.
(*) try_to_unmap_anon()
try_to_unmap_anon()
To unmap anonymous pages, each VMA in the list anchored in the anon_vma
To unmap anonymous pages, each vma in the list anchored in the anon_vma must be must be visited - at least until a VM_LOCKED VMA is encountered. If the
visited--at least until a VM_LOCKED vma is encountered. If the page is being page is being unmapped for migration, VM_LOCKED VMAs do not stop the
unmapped for migration, VM_LOCKED vmas do not stop the process because mlocked process because mlocked pages are migratable. However, for reclaim, if
pages are migratable. However, for reclaim, if the page is mapped into a the page is mapped into a VM_LOCKED VMA, the scan stops.
VM_LOCKED vma, the scan stops. try_to_unmap() attempts to acquire the mmap
semphore of the mm_struct to which the vma belongs in read mode. If this is try_to_unmap_anon() attempts to acquire in read mode the mmap semphore of
successful, try_to_unmap() will mlock the page via mlock_vma_page()--we the mm_struct to which the VMA belongs. If this is successful, it will
wouldn't have gotten to try_to_unmap() if the page were already mlocked--and mlock the page via mlock_vma_page() - we wouldn't have gotten to
will return SWAP_MLOCK, indicating that the page is unevictable. If the try_to_unmap_anon() if the page were already mlocked - and will return
mmap semaphore cannot be acquired, we are not sure whether the page is really SWAP_MLOCK, indicating that the page is unevictable.
unevictable or not. In this case, try_to_unmap() will return SWAP_AGAIN.
If the mmap semaphore cannot be acquired, we are not sure whether the page
try_to_unmap_file() -- linear mappings is really unevictable or not. In this case, try_to_unmap_anon() will
return SWAP_AGAIN.
Unmapping of a mapped file page works the same, except that the scan visits
all vmas that maps the page's index/page offset in the page's mapping's (*) try_to_unmap_file() - linear mappings
reverse map priority search tree. It must also visit each vma in the page's
mapping's non-linear list, if the list is non-empty. As for anonymous pages, Unmapping of a mapped file page works the same as for anonymous mappings,
on encountering a VM_LOCKED vma for a mapped file page, try_to_unmap() will except that the scan visits all VMAs that map the page's index/page offset
attempt to acquire the associated mm_struct's mmap semaphore to mlock the page, in the page's mapping's reverse map priority search tree. It also visits
returning SWAP_MLOCK if this is successful, and SWAP_AGAIN, if not. each VMA in the page's mapping's non-linear list, if the list is
non-empty.
try_to_unmap_file() -- non-linear mappings
As for anonymous pages, on encountering a VM_LOCKED VMA for a mapped file
If a page's mapping contains a non-empty non-linear mapping vma list, then page, try_to_unmap_file() will attempt to acquire the associated
try_to_un{map|lock}() must also visit each vma in that list to determine mm_struct's mmap semaphore to mlock the page, returning SWAP_MLOCK if this
whether the page is mapped in a VM_LOCKED vma. Again, the scan must visit is successful, and SWAP_AGAIN, if not.
all vmas in the non-linear list to ensure that the pages is not/should not be
mlocked. If a VM_LOCKED vma is found in the list, the scan could terminate. (*) try_to_unmap_file() - non-linear mappings
However, there is no easy way to determine whether the page is actually mapped
in a given vma--either for unmapping or testing whether the VM_LOCKED vma If a page's mapping contains a non-empty non-linear mapping VMA list, then
actually pins the page. try_to_un{map|lock}() must also visit each VMA in that list to determine
whether the page is mapped in a VM_LOCKED VMA. Again, the scan must visit
So, try_to_unmap_file() handles non-linear mappings by scanning a certain all VMAs in the non-linear list to ensure that the pages is not/should not
number of pages--a "cluster"--in each non-linear vma associated with the page's be mlocked.
mapping, for each file mapped page that vmscan tries to unmap. If this happens
to unmap the page we're trying to unmap, try_to_unmap() will notice this on If a VM_LOCKED VMA is found in the list, the scan could terminate.
return--(page_mapcount(page) == 0)--and return SWAP_SUCCESS. Otherwise, it However, there is no easy way to determine whether the page is actually
will return SWAP_AGAIN, causing vmscan to recirculate this page. We take mapped in a given VMA - either for unmapping or testing whether the
advantage of the cluster scan in try_to_unmap_cluster() as follows: VM_LOCKED VMA actually pins the page.
For each non-linear vma, try_to_unmap_cluster() attempts to acquire the mmap try_to_unmap_file() handles non-linear mappings by scanning a certain
semaphore of the associated mm_struct for read without blocking. If this number of pages - a "cluster" - in each non-linear VMA associated with the
attempt is successful and the vma is VM_LOCKED, try_to_unmap_cluster() will page's mapping, for each file mapped page that vmscan tries to unmap. If
retain the mmap semaphore for the scan; otherwise it drops it here. Then, this happens to unmap the page we're trying to unmap, try_to_unmap() will
for each page in the cluster, if we're holding the mmap semaphore for a locked notice this on return (page_mapcount(page) will be 0) and return
vma, try_to_unmap_cluster() calls mlock_vma_page() to mlock the page. This SWAP_SUCCESS. Otherwise, it will return SWAP_AGAIN, causing vmscan to
call is a no-op if the page is already locked, but will mlock any pages in recirculate this page. We take advantage of the cluster scan in
the non-linear mapping that happen to be unlocked. If one of the pages so try_to_unmap_cluster() as follows:
mlocked is the page passed in to try_to_unmap(), try_to_unmap_cluster() will
return SWAP_MLOCK, rather than the default SWAP_AGAIN. This will allow vmscan For each non-linear VMA, try_to_unmap_cluster() attempts to acquire the
to cull the page, rather than recirculating it on the inactive list. Again, mmap semaphore of the associated mm_struct for read without blocking.
if try_to_unmap_cluster() cannot acquire the vma's mmap sem, it returns
SWAP_AGAIN, indicating that the page is mapped by a VM_LOCKED vma, but If this attempt is successful and the VMA is VM_LOCKED,
couldn't be mlocked. try_to_unmap_cluster() will retain the mmap semaphore for the scan;
otherwise it drops it here.
Mlocked pages: try_to_munlock() Reverse Map Scan Then, for each page in the cluster, if we're holding the mmap semaphore
for a locked VMA, try_to_unmap_cluster() calls mlock_vma_page() to
TODO/FIXME: a better name might be page_mlocked()--analogous to the mlock the page. This call is a no-op if the page is already locked,
page_referenced() reverse map walker. but will mlock any pages in the non-linear mapping that happen to be
unlocked.
When munlock_vma_page()--see "Mlocked Pages: munlock()/munlockall()
System Call Handling" above--tries to munlock a page, it needs to If one of the pages so mlocked is the page passed in to try_to_unmap(),
determine whether or not the page is mapped by any VM_LOCKED vma, without try_to_unmap_cluster() will return SWAP_MLOCK, rather than the default
actually attempting to unmap all ptes from the page. For this purpose, the SWAP_AGAIN. This will allow vmscan to cull the page, rather than
unevictable/mlock infrastructure introduced a variant of try_to_unmap() called recirculating it on the inactive list.
try_to_munlock().
Again, if try_to_unmap_cluster() cannot acquire the VMA's mmap sem, it
returns SWAP_AGAIN, indicating that the page is mapped by a VM_LOCKED
VMA, but couldn't be mlocked.
try_to_munlock() REVERSE MAP SCAN
---------------------------------
[!] TODO/FIXME: a better name might be page_mlocked() - analogous to the
page_referenced() reverse map walker.
When munlock_vma_page() [see section "munlock()/munlockall() System Call
Handling" above] tries to munlock a page, it needs to determine whether or not
the page is mapped by any VM_LOCKED VMA without actually attempting to unmap
all PTEs from the page. For this purpose, the unevictable/mlock infrastructure
introduced a variant of try_to_unmap() called try_to_munlock().
try_to_munlock() calls the same functions as try_to_unmap() for anonymous and try_to_munlock() calls the same functions as try_to_unmap() for anonymous and
mapped file pages with an additional argument specifing unlock versus unmap mapped file pages with an additional argument specifing unlock versus unmap
processing. Again, these functions walk the respective reverse maps looking processing. Again, these functions walk the respective reverse maps looking
for VM_LOCKED vmas. When such a vma is found for anonymous pages and file for VM_LOCKED VMAs. When such a VMA is found for anonymous pages and file
pages mapped in linear VMAs, as in the try_to_unmap() case, the functions pages mapped in linear VMAs, as in the try_to_unmap() case, the functions
attempt to acquire the associated mmap semphore, mlock the page via attempt to acquire the associated mmap semphore, mlock the page via
mlock_vma_page() and return SWAP_MLOCK. This effectively undoes the mlock_vma_page() and return SWAP_MLOCK. This effectively undoes the
pre-clearing of the page's PG_mlocked done by munlock_vma_page. pre-clearing of the page's PG_mlocked done by munlock_vma_page.
If try_to_unmap() is unable to acquire a VM_LOCKED vma's associated mmap If try_to_unmap() is unable to acquire a VM_LOCKED VMA's associated mmap
semaphore, it will return SWAP_AGAIN. This will allow shrink_page_list() semaphore, it will return SWAP_AGAIN. This will allow shrink_page_list() to
to recycle the page on the inactive list and hope that it has better luck recycle the page on the inactive list and hope that it has better luck with the
with the page next time. page next time.
For file pages mapped into non-linear vmas, the try_to_munlock() logic works For file pages mapped into non-linear VMAs, the try_to_munlock() logic works
slightly differently. On encountering a VM_LOCKED non-linear vma that might slightly differently. On encountering a VM_LOCKED non-linear VMA that might
map the page, try_to_munlock() returns SWAP_AGAIN without actually mlocking map the page, try_to_munlock() returns SWAP_AGAIN without actually mlocking the
the page. munlock_vma_page() will just leave the page unlocked and let page. munlock_vma_page() will just leave the page unlocked and let vmscan deal
vmscan deal with it--the usual fallback position. with it - the usual fallback position.
Note that try_to_munlock()'s reverse map walk must visit every vma in a pages' Note that try_to_munlock()'s reverse map walk must visit every VMA in a page's
reverse map to determine that a page is NOT mapped into any VM_LOCKED vma. reverse map to determine that a page is NOT mapped into any VM_LOCKED VMA.
However, the scan can terminate when it encounters a VM_LOCKED vma and can However, the scan can terminate when it encounters a VM_LOCKED VMA and can
successfully acquire the vma's mmap semphore for read and mlock the page. successfully acquire the VMA's mmap semphore for read and mlock the page.
Although try_to_munlock() can be called many [very many!] times when Although try_to_munlock() might be called a great many times when munlocking a
munlock()ing a large region or tearing down a large address space that has been large region or tearing down a large address space that has been mlocked via
mlocked via mlockall(), overall this is a fairly rare event. mlockall(), overall this is a fairly rare event.
Mlocked Page: Page Reclaim in shrink_*_list()
PAGE RECLAIM IN shrink_*_list()
shrink_active_list() culls any obviously unevictable pages--i.e., -------------------------------
!page_evictable(page, NULL)--diverting these to the unevictable lru
list. However, shrink_active_list() only sees unevictable pages that shrink_active_list() culls any obviously unevictable pages - i.e.
made it onto the active/inactive lru lists. Note that these pages do not !page_evictable(page, NULL) - diverting these to the unevictable list.
have PageUnevictable set--otherwise, they would be on the unevictable list and However, shrink_active_list() only sees unevictable pages that made it onto the
shrink_active_list would never see them. active/inactive lru lists. Note that these pages do not have PageUnevictable
set - otherwise they would be on the unevictable list and shrink_active_list
would never see them.
Some examples of these unevictable pages on the LRU lists are: Some examples of these unevictable pages on the LRU lists are:
1) ramfs pages that have been placed on the lru lists when first allocated. (1) ramfs pages that have been placed on the LRU lists when first allocated.
2) SHM_LOCKed shared memory pages. shmctl(SHM_LOCK) does not attempt to (2) SHM_LOCK'd shared memory pages. shmctl(SHM_LOCK) does not attempt to
allocate or fault in the pages in the shared memory region. This happens allocate or fault in the pages in the shared memory region. This happens
when an application accesses the page the first time after SHM_LOCKing when an application accesses the page the first time after SHM_LOCK'ing
the segment. the segment.
3) Mlocked pages that could not be isolated from the lru and moved to the (3) mlocked pages that could not be isolated from the LRU and moved to the
unevictable list in mlock_vma_page(). unevictable list in mlock_vma_page().
3) Pages mapped into multiple VM_LOCKED vmas, but try_to_munlock() couldn't (4) Pages mapped into multiple VM_LOCKED VMAs, but try_to_munlock() couldn't
acquire the vma's mmap semaphore to test the flags and set PageMlocked. acquire the VMA's mmap semaphore to test the flags and set PageMlocked.
munlock_vma_page() was forced to let the page back on to the normal munlock_vma_page() was forced to let the page back on to the normal LRU
LRU list for vmscan to handle. list for vmscan to handle.
shrink_inactive_list() also diverts any unevictable pages that it finds on the
inactive lists to the appropriate zone's unevictable list.
shrink_inactive_list() also culls any unevictable pages that it finds on shrink_inactive_list() should only see SHM_LOCK'd pages that became SHM_LOCK'd
the inactive lists, again diverting them to the appropriate zone's unevictable after shrink_active_list() had moved them to the inactive list, or pages mapped
lru list. shrink_inactive_list() should only see SHM_LOCKed pages that became into VM_LOCKED VMAs that munlock_vma_page() couldn't isolate from the LRU to
SHM_LOCKed after shrink_active_list() had moved them to the inactive list, or recheck via try_to_munlock(). shrink_inactive_list() won't notice the latter,
pages mapped into VM_LOCKED vmas that munlock_vma_page() couldn't isolate from but will pass on to shrink_page_list().
the lru to recheck via try_to_munlock(). shrink_inactive_list() won't notice
the latter, but will pass on to shrink_page_list().
shrink_page_list() again culls obviously unevictable pages that it could shrink_page_list() again culls obviously unevictable pages that it could
encounter for similar reason to shrink_inactive_list(). Pages mapped into encounter for similar reason to shrink_inactive_list(). Pages mapped into
VM_LOCKED vmas but without PG_mlocked set will make it all the way to VM_LOCKED VMAs but without PG_mlocked set will make it all the way to
try_to_unmap(). shrink_page_list() will divert them to the unevictable list try_to_unmap(). shrink_page_list() will divert them to the unevictable list
when try_to_unmap() returns SWAP_MLOCK, as discussed above. when try_to_unmap() returns SWAP_MLOCK, as discussed above.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment