About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Documentation / vm / unevictable-lru.txt




Custom Search

Based on kernel version 3.13. Page generated on 2014-01-20 22:05 EST.

1				==============================
2				UNEVICTABLE LRU INFRASTRUCTURE
3				==============================
4	
5	========
6	CONTENTS
7	========
8	
9	 (*) The Unevictable LRU
10	
11	     - The unevictable page list.
12	     - Memory control group interaction.
13	     - Marking address spaces unevictable.
14	     - Detecting Unevictable Pages.
15	     - vmscan's handling of unevictable pages.
16	
17	 (*) mlock()'d pages.
18	
19	     - History.
20	     - Basic management.
21	     - mlock()/mlockall() system call handling.
22	     - Filtering special vmas.
23	     - munlock()/munlockall() system call handling.
24	     - Migrating mlocked pages.
25	     - mmap(MAP_LOCKED) system call handling.
26	     - munmap()/exit()/exec() system call handling.
27	     - try_to_unmap().
28	     - try_to_munlock() reverse map scan.
29	     - Page reclaim in shrink_*_list().
30	
31	
32	============
33	INTRODUCTION
34	============
35	
36	This document describes the Linux memory manager's "Unevictable LRU"
37	infrastructure and the use of this to manage several types of "unevictable"
38	pages.
39	
40	The document attempts to provide the overall rationale behind this mechanism
41	and the rationale for some of the design decisions that drove the
42	implementation.  The latter design rationale is discussed in the context of an
43	implementation description.  Admittedly, one can obtain the implementation
44	details - the "what does it do?" - by reading the code.  One hopes that the
45	descriptions below add value by provide the answer to "why does it do that?".
46	
47	
48	===================
49	THE UNEVICTABLE LRU
50	===================
51	
52	The Unevictable LRU facility adds an additional LRU list to track unevictable
53	pages and to hide these pages from vmscan.  This mechanism is based on a patch
54	by Larry Woodman of Red Hat to address several scalability problems with page
55	reclaim in Linux.  The problems have been observed at customer sites on large
56	memory x86_64 systems.
57	
58	To illustrate this with an example, a non-NUMA x86_64 platform with 128GB of
59	main memory will have over 32 million 4k pages in a single zone.  When a large
60	fraction of these pages are not evictable for any reason [see below], vmscan
61	will spend a lot of time scanning the LRU lists looking for the small fraction
62	of pages that are evictable.  This can result in a situation where all CPUs are
63	spending 100% of their time in vmscan for hours or days on end, with the system
64	completely unresponsive.
65	
66	The unevictable list addresses the following classes of unevictable pages:
67	
68	 (*) Those owned by ramfs.
69	
70	 (*) Those mapped into SHM_LOCK'd shared memory regions.
71	
72	 (*) Those mapped into VM_LOCKED [mlock()ed] VMAs.
73	
74	The infrastructure may also be able to handle other conditions that make pages
75	unevictable, either by definition or by circumstance, in the future.
76	
77	
78	THE UNEVICTABLE PAGE LIST
79	-------------------------
80	
81	The Unevictable LRU infrastructure consists of an additional, per-zone, LRU list
82	called the "unevictable" list and an associated page flag, PG_unevictable, to
83	indicate that the page is being managed on the unevictable list.
84	
85	The PG_unevictable flag is analogous to, and mutually exclusive with, the
86	PG_active flag in that it indicates on which LRU list a page resides when
87	PG_lru is set.
88	
89	The Unevictable LRU infrastructure maintains unevictable pages on an additional
90	LRU list for a few reasons:
91	
92	 (1) We get to "treat unevictable pages just like we treat other pages in the
93	     system - which means we get to use the same code to manipulate them, the
94	     same code to isolate them (for migrate, etc.), the same code to keep track
95	     of the statistics, etc..." [Rik van Riel]
96	
97	 (2) We want to be able to migrate unevictable pages between nodes for memory
98	     defragmentation, workload management and memory hotplug.  The linux kernel
99	     can only migrate pages that it can successfully isolate from the LRU
100	     lists.  If we were to maintain pages elsewhere than on an LRU-like list,
101	     where they can be found by isolate_lru_page(), we would prevent their
102	     migration, unless we reworked migration code to find the unevictable pages
103	     itself.
104	
105	
106	The unevictable list does not differentiate between file-backed and anonymous,
107	swap-backed pages.  This differentiation is only important while the pages are,
108	in fact, evictable.
109	
110	The unevictable list benefits from the "arrayification" of the per-zone LRU
111	lists and statistics originally proposed and posted by Christoph Lameter.
112	
113	The unevictable list does not use the LRU pagevec mechanism. Rather,
114	unevictable pages are placed directly on the page's zone's unevictable list
115	under the zone lru_lock.  This allows us to prevent the stranding of pages on
116	the unevictable list when one task has the page isolated from the LRU and other
117	tasks are changing the "evictability" state of the page.
118	
119	
120	MEMORY CONTROL GROUP INTERACTION
121	--------------------------------
122	
123	The unevictable LRU facility interacts with the memory control group [aka
124	memory controller; see Documentation/cgroups/memory.txt] by extending the
125	lru_list enum.
126	
127	The memory controller data structure automatically gets a per-zone unevictable
128	list as a result of the "arrayification" of the per-zone LRU lists (one per
129	lru_list enum element).  The memory controller tracks the movement of pages to
130	and from the unevictable list.
131	
132	When a memory control group comes under memory pressure, the controller will
133	not attempt to reclaim pages on the unevictable list.  This has a couple of
134	effects:
135	
136	 (1) Because the pages are "hidden" from reclaim on the unevictable list, the
137	     reclaim process can be more efficient, dealing only with pages that have a
138	     chance of being reclaimed.
139	
140	 (2) On the other hand, if too many of the pages charged to the control group
141	     are unevictable, the evictable portion of the working set of the tasks in
142	     the control group may not fit into the available memory.  This can cause
143	     the control group to thrash or to OOM-kill tasks.
144	
145	
146	MARKING ADDRESS SPACES UNEVICTABLE
147	----------------------------------
148	
149	For facilities such as ramfs none of the pages attached to the address space
150	may be evicted.  To prevent eviction of any such pages, the AS_UNEVICTABLE
151	address space flag is provided, and this can be manipulated by a filesystem
152	using a number of wrapper functions:
153	
154	 (*) void mapping_set_unevictable(struct address_space *mapping);
155	
156		Mark the address space as being completely unevictable.
157	
158	 (*) void mapping_clear_unevictable(struct address_space *mapping);
159	
160		Mark the address space as being evictable.
161	
162	 (*) int mapping_unevictable(struct address_space *mapping);
163	
164		Query the address space, and return true if it is completely
165		unevictable.
166	
167	These are currently used in two places in the kernel:
168	
169	 (1) By ramfs to mark the address spaces of its inodes when they are created,
170	     and this mark remains for the life of the inode.
171	
172	 (2) By SYSV SHM to mark SHM_LOCK'd address spaces until SHM_UNLOCK is called.
173	
174	     Note that SHM_LOCK is not required to page in the locked pages if they're
175	     swapped out; the application must touch the pages manually if it wants to
176	     ensure they're in memory.
177	
178	
179	DETECTING UNEVICTABLE PAGES
180	---------------------------
181	
182	The function page_evictable() in vmscan.c determines whether a page is
183	evictable or not using the query function outlined above [see section "Marking
184	address spaces unevictable"] to check the AS_UNEVICTABLE flag.
185	
186	For address spaces that are so marked after being populated (as SHM regions
187	might be), the lock action (eg: SHM_LOCK) can be lazy, and need not populate
188	the page tables for the region as does, for example, mlock(), nor need it make
189	any special effort to push any pages in the SHM_LOCK'd area to the unevictable
190	list.  Instead, vmscan will do this if and when it encounters the pages during
191	a reclamation scan.
192	
193	On an unlock action (such as SHM_UNLOCK), the unlocker (eg: shmctl()) must scan
194	the pages in the region and "rescue" them from the unevictable list if no other
195	condition is keeping them unevictable.  If an unevictable region is destroyed,
196	the pages are also "rescued" from the unevictable list in the process of
197	freeing them.
198	
199	page_evictable() also checks for mlocked pages by testing an additional page
200	flag, PG_mlocked (as wrapped by PageMlocked()), which is set when a page is
201	faulted into a VM_LOCKED vma, or found in a vma being VM_LOCKED.
202	
203	
204	VMSCAN'S HANDLING OF UNEVICTABLE PAGES
205	--------------------------------------
206	
207	If unevictable pages are culled in the fault path, or moved to the unevictable
208	list at mlock() or mmap() time, vmscan will not encounter the pages until they
209	have become evictable again (via munlock() for example) and have been "rescued"
210	from the unevictable list.  However, there may be situations where we decide,
211	for the sake of expediency, to leave a unevictable page on one of the regular
212	active/inactive LRU lists for vmscan to deal with.  vmscan checks for such
213	pages in all of the shrink_{active|inactive|page}_list() functions and will
214	"cull" such pages that it encounters: that is, it diverts those pages to the
215	unevictable list for the zone being scanned.
216	
217	There may be situations where a page is mapped into a VM_LOCKED VMA, but the
218	page is not marked as PG_mlocked.  Such pages will make it all the way to
219	shrink_page_list() where they will be detected when vmscan walks the reverse
220	map in try_to_unmap().  If try_to_unmap() returns SWAP_MLOCK,
221	shrink_page_list() will cull the page at that point.
222	
223	To "cull" an unevictable page, vmscan simply puts the page back on the LRU list
224	using putback_lru_page() - the inverse operation to isolate_lru_page() - after
225	dropping the page lock.  Because the condition which makes the page unevictable
226	may change once the page is unlocked, putback_lru_page() will recheck the
227	unevictable state of a page that it places on the unevictable list.  If the
228	page has become unevictable, putback_lru_page() removes it from the list and
229	retries, including the page_unevictable() test.  Because such a race is a rare
230	event and movement of pages onto the unevictable list should be rare, these
231	extra evictabilty checks should not occur in the majority of calls to
232	putback_lru_page().
233	
234	
235	=============
236	MLOCKED PAGES
237	=============
238	
239	The unevictable page list is also useful for mlock(), in addition to ramfs and
240	SYSV SHM.  Note that mlock() is only available in CONFIG_MMU=y situations; in
241	NOMMU situations, all mappings are effectively mlocked.
242	
243	
244	HISTORY
245	-------
246	
247	The "Unevictable mlocked Pages" infrastructure is based on work originally
248	posted by Nick Piggin in an RFC patch entitled "mm: mlocked pages off LRU".
249	Nick posted his patch as an alternative to a patch posted by Christoph Lameter
250	to achieve the same objective: hiding mlocked pages from vmscan.
251	
252	In Nick's patch, he used one of the struct page LRU list link fields as a count
253	of VM_LOCKED VMAs that map the page.  This use of the link field for a count
254	prevented the management of the pages on an LRU list, and thus mlocked pages
255	were not migratable as isolate_lru_page() could not find them, and the LRU list
256	link field was not available to the migration subsystem.
257	
258	Nick resolved this by putting mlocked pages back on the lru list before
259	attempting to isolate them, thus abandoning the count of VM_LOCKED VMAs.  When
260	Nick's patch was integrated with the Unevictable LRU work, the count was
261	replaced by walking the reverse map to determine whether any VM_LOCKED VMAs
262	mapped the page.  More on this below.
263	
264	
265	BASIC MANAGEMENT
266	----------------
267	
268	mlocked pages - pages mapped into a VM_LOCKED VMA - are a class of unevictable
269	pages.  When such a page has been "noticed" by the memory management subsystem,
270	the page is marked with the PG_mlocked flag.  This can be manipulated using the
271	PageMlocked() functions.
272	
273	A PG_mlocked page will be placed on the unevictable list when it is added to
274	the LRU.  Such pages can be "noticed" by memory management in several places:
275	
276	 (1) in the mlock()/mlockall() system call handlers;
277	
278	 (2) in the mmap() system call handler when mmapping a region with the
279	     MAP_LOCKED flag;
280	
281	 (3) mmapping a region in a task that has called mlockall() with the MCL_FUTURE
282	     flag
283	
284	 (4) in the fault path, if mlocked pages are "culled" in the fault path,
285	     and when a VM_LOCKED stack segment is expanded; or
286	
287	 (5) as mentioned above, in vmscan:shrink_page_list() when attempting to
288	     reclaim a page in a VM_LOCKED VMA via try_to_unmap()
289	
290	all of which result in the VM_LOCKED flag being set for the VMA if it doesn't
291	already have it set.
292	
293	mlocked pages become unlocked and rescued from the unevictable list when:
294	
295	 (1) mapped in a range unlocked via the munlock()/munlockall() system calls;
296	
297	 (2) munmap()'d out of the last VM_LOCKED VMA that maps the page, including
298	     unmapping at task exit;
299	
300	 (3) when the page is truncated from the last VM_LOCKED VMA of an mmapped file;
301	     or
302	
303	 (4) before a page is COW'd in a VM_LOCKED VMA.
304	
305	
306	mlock()/mlockall() SYSTEM CALL HANDLING
307	---------------------------------------
308	
309	Both [do_]mlock() and [do_]mlockall() system call handlers call mlock_fixup()
310	for each VMA in the range specified by the call.  In the case of mlockall(),
311	this is the entire active address space of the task.  Note that mlock_fixup()
312	is used for both mlocking and munlocking a range of memory.  A call to mlock()
313	an already VM_LOCKED VMA, or to munlock() a VMA that is not VM_LOCKED is
314	treated as a no-op, and mlock_fixup() simply returns.
315	
316	If the VMA passes some filtering as described in "Filtering Special Vmas"
317	below, mlock_fixup() will attempt to merge the VMA with its neighbors or split
318	off a subset of the VMA if the range does not cover the entire VMA.  Once the
319	VMA has been merged or split or neither, mlock_fixup() will call
320	__mlock_vma_pages_range() to fault in the pages via get_user_pages() and to
321	mark the pages as mlocked via mlock_vma_page().
322	
323	Note that the VMA being mlocked might be mapped with PROT_NONE.  In this case,
324	get_user_pages() will be unable to fault in the pages.  That's okay.  If pages
325	do end up getting faulted into this VM_LOCKED VMA, we'll handle them in the
326	fault path or in vmscan.
327	
328	Also note that a page returned by get_user_pages() could be truncated or
329	migrated out from under us, while we're trying to mlock it.  To detect this,
330	__mlock_vma_pages_range() checks page_mapping() after acquiring the page lock.
331	If the page is still associated with its mapping, we'll go ahead and call
332	mlock_vma_page().  If the mapping is gone, we just unlock the page and move on.
333	In the worst case, this will result in a page mapped in a VM_LOCKED VMA
334	remaining on a normal LRU list without being PageMlocked().  Again, vmscan will
335	detect and cull such pages.
336	
337	mlock_vma_page() will call TestSetPageMlocked() for each page returned by
338	get_user_pages().  We use TestSetPageMlocked() because the page might already
339	be mlocked by another task/VMA and we don't want to do extra work.  We
340	especially do not want to count an mlocked page more than once in the
341	statistics.  If the page was already mlocked, mlock_vma_page() need do nothing
342	more.
343	
344	If the page was NOT already mlocked, mlock_vma_page() attempts to isolate the
345	page from the LRU, as it is likely on the appropriate active or inactive list
346	at that time.  If the isolate_lru_page() succeeds, mlock_vma_page() will put
347	back the page - by calling putback_lru_page() - which will notice that the page
348	is now mlocked and divert the page to the zone's unevictable list.  If
349	mlock_vma_page() is unable to isolate the page from the LRU, vmscan will handle
350	it later if and when it attempts to reclaim the page.
351	
352	
353	FILTERING SPECIAL VMAS
354	----------------------
355	
356	mlock_fixup() filters several classes of "special" VMAs:
357	
358	1) VMAs with VM_IO or VM_PFNMAP set are skipped entirely.  The pages behind
359	   these mappings are inherently pinned, so we don't need to mark them as
360	   mlocked.  In any case, most of the pages have no struct page in which to so
361	   mark the page.  Because of this, get_user_pages() will fail for these VMAs,
362	   so there is no sense in attempting to visit them.
363	
364	2) VMAs mapping hugetlbfs page are already effectively pinned into memory.  We
365	   neither need nor want to mlock() these pages.  However, to preserve the
366	   prior behavior of mlock() - before the unevictable/mlock changes -
367	   mlock_fixup() will call make_pages_present() in the hugetlbfs VMA range to
368	   allocate the huge pages and populate the ptes.
369	
370	3) VMAs with VM_DONTEXPAND are generally userspace mappings of kernel pages,
371	   such as the VDSO page, relay channel pages, etc. These pages
372	   are inherently unevictable and are not managed on the LRU lists.
373	   mlock_fixup() treats these VMAs the same as hugetlbfs VMAs.  It calls
374	   make_pages_present() to populate the ptes.
375	
376	Note that for all of these special VMAs, mlock_fixup() does not set the
377	VM_LOCKED flag.  Therefore, we won't have to deal with them later during
378	munlock(), munmap() or task exit.  Neither does mlock_fixup() account these
379	VMAs against the task's "locked_vm".
380	
381	
382	munlock()/munlockall() SYSTEM CALL HANDLING
383	-------------------------------------------
384	
385	The munlock() and munlockall() system calls are handled by the same functions -
386	do_mlock[all]() - as the mlock() and mlockall() system calls with the unlock vs
387	lock operation indicated by an argument.  So, these system calls are also
388	handled by mlock_fixup().  Again, if called for an already munlocked VMA,
389	mlock_fixup() simply returns.  Because of the VMA filtering discussed above,
390	VM_LOCKED will not be set in any "special" VMAs.  So, these VMAs will be
391	ignored for munlock.
392	
393	If the VMA is VM_LOCKED, mlock_fixup() again attempts to merge or split off the
394	specified range.  The range is then munlocked via the function
395	__mlock_vma_pages_range() - the same function used to mlock a VMA range -
396	passing a flag to indicate that munlock() is being performed.
397	
398	Because the VMA access protections could have been changed to PROT_NONE after
399	faulting in and mlocking pages, get_user_pages() was unreliable for visiting
400	these pages for munlocking.  Because we don't want to leave pages mlocked,
401	get_user_pages() was enhanced to accept a flag to ignore the permissions when
402	fetching the pages - all of which should be resident as a result of previous
403	mlocking.
404	
405	For munlock(), __mlock_vma_pages_range() unlocks individual pages by calling
406	munlock_vma_page().  munlock_vma_page() unconditionally clears the PG_mlocked
407	flag using TestClearPageMlocked().  As with mlock_vma_page(),
408	munlock_vma_page() use the Test*PageMlocked() function to handle the case where
409	the page might have already been unlocked by another task.  If the page was
410	mlocked, munlock_vma_page() updates that zone statistics for the number of
411	mlocked pages.  Note, however, that at this point we haven't checked whether
412	the page is mapped by other VM_LOCKED VMAs.
413	
414	We can't call try_to_munlock(), the function that walks the reverse map to
415	check for other VM_LOCKED VMAs, without first isolating the page from the LRU.
416	try_to_munlock() is a variant of try_to_unmap() and thus requires that the page
417	not be on an LRU list [more on these below].  However, the call to
418	isolate_lru_page() could fail, in which case we couldn't try_to_munlock().  So,
419	we go ahead and clear PG_mlocked up front, as this might be the only chance we
420	have.  If we can successfully isolate the page, we go ahead and
421	try_to_munlock(), which will restore the PG_mlocked flag and update the zone
422	page statistics if it finds another VMA holding the page mlocked.  If we fail
423	to isolate the page, we'll have left a potentially mlocked page on the LRU.
424	This is fine, because we'll catch it later if and if vmscan tries to reclaim
425	the page.  This should be relatively rare.
426	
427	
428	MIGRATING MLOCKED PAGES
429	-----------------------
430	
431	A page that is being migrated has been isolated from the LRU lists and is held
432	locked across unmapping of the page, updating the page's address space entry
433	and copying the contents and state, until the page table entry has been
434	replaced with an entry that refers to the new page.  Linux supports migration
435	of mlocked pages and other unevictable pages.  This involves simply moving the
436	PG_mlocked and PG_unevictable states from the old page to the new page.
437	
438	Note that page migration can race with mlocking or munlocking of the same page.
439	This has been discussed from the mlock/munlock perspective in the respective
440	sections above.  Both processes (migration and m[un]locking) hold the page
441	locked.  This provides the first level of synchronization.  Page migration
442	zeros out the page_mapping of the old page before unlocking it, so m[un]lock
443	can skip these pages by testing the page mapping under page lock.
444	
445	To complete page migration, we place the new and old pages back onto the LRU
446	after dropping the page lock.  The "unneeded" page - old page on success, new
447	page on failure - will be freed when the reference count held by the migration
448	process is released.  To ensure that we don't strand pages on the unevictable
449	list because of a race between munlock and migration, page migration uses the
450	putback_lru_page() function to add migrated pages back to the LRU.
451	
452	
453	mmap(MAP_LOCKED) SYSTEM CALL HANDLING
454	-------------------------------------
455	
456	In addition the the mlock()/mlockall() system calls, an application can request
457	that a region of memory be mlocked supplying the MAP_LOCKED flag to the mmap()
458	call.  Furthermore, any mmap() call or brk() call that expands the heap by a
459	task that has previously called mlockall() with the MCL_FUTURE flag will result
460	in the newly mapped memory being mlocked.  Before the unevictable/mlock
461	changes, the kernel simply called make_pages_present() to allocate pages and
462	populate the page table.
463	
464	To mlock a range of memory under the unevictable/mlock infrastructure, the
465	mmap() handler and task address space expansion functions call
466	mlock_vma_pages_range() specifying the vma and the address range to mlock.
467	mlock_vma_pages_range() filters VMAs like mlock_fixup(), as described above in
468	"Filtering Special VMAs".  It will clear the VM_LOCKED flag, which will have
469	already been set by the caller, in filtered VMAs.  Thus these VMA's need not be
470	visited for munlock when the region is unmapped.
471	
472	For "normal" VMAs, mlock_vma_pages_range() calls __mlock_vma_pages_range() to
473	fault/allocate the pages and mlock them.  Again, like mlock_fixup(),
474	mlock_vma_pages_range() downgrades the mmap semaphore to read mode before
475	attempting to fault/allocate and mlock the pages and "upgrades" the semaphore
476	back to write mode before returning.
477	
478	The callers of mlock_vma_pages_range() will have already added the memory range
479	to be mlocked to the task's "locked_vm".  To account for filtered VMAs,
480	mlock_vma_pages_range() returns the number of pages NOT mlocked.  All of the
481	callers then subtract a non-negative return value from the task's locked_vm.  A
482	negative return value represent an error - for example, from get_user_pages()
483	attempting to fault in a VMA with PROT_NONE access.  In this case, we leave the
484	memory range accounted as locked_vm, as the protections could be changed later
485	and pages allocated into that region.
486	
487	
488	munmap()/exit()/exec() SYSTEM CALL HANDLING
489	-------------------------------------------
490	
491	When unmapping an mlocked region of memory, whether by an explicit call to
492	munmap() or via an internal unmap from exit() or exec() processing, we must
493	munlock the pages if we're removing the last VM_LOCKED VMA that maps the pages.
494	Before the unevictable/mlock changes, mlocking did not mark the pages in any
495	way, so unmapping them required no processing.
496	
497	To munlock a range of memory under the unevictable/mlock infrastructure, the
498	munmap() handler and task address space call tear down function
499	munlock_vma_pages_all().  The name reflects the observation that one always
500	specifies the entire VMA range when munlock()ing during unmap of a region.
501	Because of the VMA filtering when mlocking() regions, only "normal" VMAs that
502	actually contain mlocked pages will be passed to munlock_vma_pages_all().
503	
504	munlock_vma_pages_all() clears the VM_LOCKED VMA flag and, like mlock_fixup()
505	for the munlock case, calls __munlock_vma_pages_range() to walk the page table
506	for the VMA's memory range and munlock_vma_page() each resident page mapped by
507	the VMA.  This effectively munlocks the page, only if this is the last
508	VM_LOCKED VMA that maps the page.
509	
510	
511	try_to_unmap()
512	--------------
513	
514	Pages can, of course, be mapped into multiple VMAs.  Some of these VMAs may
515	have VM_LOCKED flag set.  It is possible for a page mapped into one or more
516	VM_LOCKED VMAs not to have the PG_mlocked flag set and therefore reside on one
517	of the active or inactive LRU lists.  This could happen if, for example, a task
518	in the process of munlocking the page could not isolate the page from the LRU.
519	As a result, vmscan/shrink_page_list() might encounter such a page as described
520	in section "vmscan's handling of unevictable pages".  To handle this situation,
521	try_to_unmap() checks for VM_LOCKED VMAs while it is walking a page's reverse
522	map.
523	
524	try_to_unmap() is always called, by either vmscan for reclaim or for page
525	migration, with the argument page locked and isolated from the LRU.  Separate
526	functions handle anonymous and mapped file pages, as these types of pages have
527	different reverse map mechanisms.
528	
529	 (*) try_to_unmap_anon()
530	
531	     To unmap anonymous pages, each VMA in the list anchored in the anon_vma
532	     must be visited - at least until a VM_LOCKED VMA is encountered.  If the
533	     page is being unmapped for migration, VM_LOCKED VMAs do not stop the
534	     process because mlocked pages are migratable.  However, for reclaim, if
535	     the page is mapped into a VM_LOCKED VMA, the scan stops.
536	
537	     try_to_unmap_anon() attempts to acquire in read mode the mmap semaphore of
538	     the mm_struct to which the VMA belongs.  If this is successful, it will
539	     mlock the page via mlock_vma_page() - we wouldn't have gotten to
540	     try_to_unmap_anon() if the page were already mlocked - and will return
541	     SWAP_MLOCK, indicating that the page is unevictable.
542	
543	     If the mmap semaphore cannot be acquired, we are not sure whether the page
544	     is really unevictable or not.  In this case, try_to_unmap_anon() will
545	     return SWAP_AGAIN.
546	
547	 (*) try_to_unmap_file() - linear mappings
548	
549	     Unmapping of a mapped file page works the same as for anonymous mappings,
550	     except that the scan visits all VMAs that map the page's index/page offset
551	     in the page's mapping's reverse map priority search tree.  It also visits
552	     each VMA in the page's mapping's non-linear list, if the list is
553	     non-empty.
554	
555	     As for anonymous pages, on encountering a VM_LOCKED VMA for a mapped file
556	     page, try_to_unmap_file() will attempt to acquire the associated
557	     mm_struct's mmap semaphore to mlock the page, returning SWAP_MLOCK if this
558	     is successful, and SWAP_AGAIN, if not.
559	
560	 (*) try_to_unmap_file() - non-linear mappings
561	
562	     If a page's mapping contains a non-empty non-linear mapping VMA list, then
563	     try_to_un{map|lock}() must also visit each VMA in that list to determine
564	     whether the page is mapped in a VM_LOCKED VMA.  Again, the scan must visit
565	     all VMAs in the non-linear list to ensure that the pages is not/should not
566	     be mlocked.
567	
568	     If a VM_LOCKED VMA is found in the list, the scan could terminate.
569	     However, there is no easy way to determine whether the page is actually
570	     mapped in a given VMA - either for unmapping or testing whether the
571	     VM_LOCKED VMA actually pins the page.
572	
573	     try_to_unmap_file() handles non-linear mappings by scanning a certain
574	     number of pages - a "cluster" - in each non-linear VMA associated with the
575	     page's mapping, for each file mapped page that vmscan tries to unmap.  If
576	     this happens to unmap the page we're trying to unmap, try_to_unmap() will
577	     notice this on return (page_mapcount(page) will be 0) and return
578	     SWAP_SUCCESS.  Otherwise, it will return SWAP_AGAIN, causing vmscan to
579	     recirculate this page.  We take advantage of the cluster scan in
580	     try_to_unmap_cluster() as follows:
581	
582		For each non-linear VMA, try_to_unmap_cluster() attempts to acquire the
583		mmap semaphore of the associated mm_struct for read without blocking.
584	
585		If this attempt is successful and the VMA is VM_LOCKED,
586		try_to_unmap_cluster() will retain the mmap semaphore for the scan;
587		otherwise it drops it here.
588	
589		Then, for each page in the cluster, if we're holding the mmap semaphore
590		for a locked VMA, try_to_unmap_cluster() calls mlock_vma_page() to
591		mlock the page.  This call is a no-op if the page is already locked,
592		but will mlock any pages in the non-linear mapping that happen to be
593		unlocked.
594	
595		If one of the pages so mlocked is the page passed in to try_to_unmap(),
596		try_to_unmap_cluster() will return SWAP_MLOCK, rather than the default
597		SWAP_AGAIN.  This will allow vmscan to cull the page, rather than
598		recirculating it on the inactive list.
599	
600		Again, if try_to_unmap_cluster() cannot acquire the VMA's mmap sem, it
601		returns SWAP_AGAIN, indicating that the page is mapped by a VM_LOCKED
602		VMA, but couldn't be mlocked.
603	
604	
605	try_to_munlock() REVERSE MAP SCAN
606	---------------------------------
607	
608	 [!] TODO/FIXME: a better name might be page_mlocked() - analogous to the
609	     page_referenced() reverse map walker.
610	
611	When munlock_vma_page() [see section "munlock()/munlockall() System Call
612	Handling" above] tries to munlock a page, it needs to determine whether or not
613	the page is mapped by any VM_LOCKED VMA without actually attempting to unmap
614	all PTEs from the page.  For this purpose, the unevictable/mlock infrastructure
615	introduced a variant of try_to_unmap() called try_to_munlock().
616	
617	try_to_munlock() calls the same functions as try_to_unmap() for anonymous and
618	mapped file pages with an additional argument specifying unlock versus unmap
619	processing.  Again, these functions walk the respective reverse maps looking
620	for VM_LOCKED VMAs.  When such a VMA is found for anonymous pages and file
621	pages mapped in linear VMAs, as in the try_to_unmap() case, the functions
622	attempt to acquire the associated mmap semaphore, mlock the page via
623	mlock_vma_page() and return SWAP_MLOCK.  This effectively undoes the
624	pre-clearing of the page's PG_mlocked done by munlock_vma_page.
625	
626	If try_to_unmap() is unable to acquire a VM_LOCKED VMA's associated mmap
627	semaphore, it will return SWAP_AGAIN.  This will allow shrink_page_list() to
628	recycle the page on the inactive list and hope that it has better luck with the
629	page next time.
630	
631	For file pages mapped into non-linear VMAs, the try_to_munlock() logic works
632	slightly differently.  On encountering a VM_LOCKED non-linear VMA that might
633	map the page, try_to_munlock() returns SWAP_AGAIN without actually mlocking the
634	page.  munlock_vma_page() will just leave the page unlocked and let vmscan deal
635	with it - the usual fallback position.
636	
637	Note that try_to_munlock()'s reverse map walk must visit every VMA in a page's
638	reverse map to determine that a page is NOT mapped into any VM_LOCKED VMA.
639	However, the scan can terminate when it encounters a VM_LOCKED VMA and can
640	successfully acquire the VMA's mmap semaphore for read and mlock the page.
641	Although try_to_munlock() might be called a great many times when munlocking a
642	large region or tearing down a large address space that has been mlocked via
643	mlockall(), overall this is a fairly rare event.
644	
645	
646	PAGE RECLAIM IN shrink_*_list()
647	-------------------------------
648	
649	shrink_active_list() culls any obviously unevictable pages - i.e.
650	!page_evictable(page) - diverting these to the unevictable list.
651	However, shrink_active_list() only sees unevictable pages that made it onto the
652	active/inactive lru lists.  Note that these pages do not have PageUnevictable
653	set - otherwise they would be on the unevictable list and shrink_active_list
654	would never see them.
655	
656	Some examples of these unevictable pages on the LRU lists are:
657	
658	 (1) ramfs pages that have been placed on the LRU lists when first allocated.
659	
660	 (2) SHM_LOCK'd shared memory pages.  shmctl(SHM_LOCK) does not attempt to
661	     allocate or fault in the pages in the shared memory region.  This happens
662	     when an application accesses the page the first time after SHM_LOCK'ing
663	     the segment.
664	
665	 (3) mlocked pages that could not be isolated from the LRU and moved to the
666	     unevictable list in mlock_vma_page().
667	
668	 (4) Pages mapped into multiple VM_LOCKED VMAs, but try_to_munlock() couldn't
669	     acquire the VMA's mmap semaphore to test the flags and set PageMlocked.
670	     munlock_vma_page() was forced to let the page back on to the normal LRU
671	     list for vmscan to handle.
672	
673	shrink_inactive_list() also diverts any unevictable pages that it finds on the
674	inactive lists to the appropriate zone's unevictable list.
675	
676	shrink_inactive_list() should only see SHM_LOCK'd pages that became SHM_LOCK'd
677	after shrink_active_list() had moved them to the inactive list, or pages mapped
678	into VM_LOCKED VMAs that munlock_vma_page() couldn't isolate from the LRU to
679	recheck via try_to_munlock().  shrink_inactive_list() won't notice the latter,
680	but will pass on to shrink_page_list().
681	
682	shrink_page_list() again culls obviously unevictable pages that it could
683	encounter for similar reason to shrink_inactive_list().  Pages mapped into
684	VM_LOCKED VMAs but without PG_mlocked set will make it all the way to
685	try_to_unmap().  shrink_page_list() will divert them to the unevictable list
686	when try_to_unmap() returns SWAP_MLOCK, as discussed above.
Hide Line Numbers
About Kernel Documentation Linux Kernel Contact Linux Resources Linux Blog

Information is copyright its respective author. All material is available from the Linux Kernel Source distributed under a GPL License. This page is provided as a free service by mjmwired.net.