xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

mm/doc: editorial pass on page migration

Browse Source 
Add Sphinx reference links to HMM and CPUSETS, and numerous small
editorial changes to make the page_migration.rst document more readable.

Signed-off-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20200902225247.15213-1-rcampbell@nvidia.com
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
This commit is contained in:
Ralph Campbell
2020-09-02 15:52:47 -07:00
committed by Jonathan Corbet
parent 17dca05023
commit 50aab9b142
3 changed files with 84 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
Documentation/admin-guide/cgroup-v1/cpusets.rst
View File

@@ -1,3 +1,5 @@
.. _cpusets:
======= =======
CPUSETS CPUSETS
======= =======

2
Documentation/vm/hmm.rst
View File

@@ -1,4 +1,4 @@
.. hmm: .. _hmm:
===================================== =====================================
Heterogeneous Memory Management (HMM) Heterogeneous Memory Management (HMM)

158
Documentation/vm/page_migration.rst
View File

@@ -4,25 +4,28 @@
Page migration Page migration
============== ==============
Page migration allows the moving of the physical location of pages between Page migration allows moving the physical location of pages between
nodes in a numa system while the process is running. This means that the nodes in a NUMA system while the process is running. This means that the
virtual addresses that the process sees do not change. However, the virtual addresses that the process sees do not change. However, the
system rearranges the physical location of those pages. system rearranges the physical location of those pages.
The main intend of page migration is to reduce the latency of memory access Also see :ref:`Heterogeneous Memory Management (HMM) <hmm>`
for migrating pages to or from device private memory.
The main intent of page migration is to reduce the latency of memory accesses
by moving pages near to the processor where the process accessing that memory by moving pages near to the processor where the process accessing that memory
is running. is running.
Page migration allows a process to manually relocate the node on which its Page migration allows a process to manually relocate the node on which its
pages are located through the MF_MOVE and MF_MOVE_ALL options while setting pages are located through the MF_MOVE and MF_MOVE_ALL options while setting
a new memory policy via mbind(). The pages of process can also be relocated a new memory policy via mbind(). The pages of a process can also be relocated
from another process using the sys_migrate_pages() function call. The from another process using the sys_migrate_pages() function call. The
migrate_pages function call takes two sets of nodes and moves pages of a migrate_pages() function call takes two sets of nodes and moves pages of a
process that are located on the from nodes to the destination nodes. process that are located on the from nodes to the destination nodes.
Page migration functions are provided by the numactl package by Andi Kleen Page migration functions are provided by the numactl package by Andi Kleen
(a version later than 0.9.3 is required. Get it from (a version later than 0.9.3 is required. Get it from
ftp://oss.sgi.com/www/projects/libnuma/download/). numactl provides libnuma https://github.com/numactl/numactl.git). numactl provides libnuma
which provides an interface similar to other numa functionality for page which provides an interface similar to other NUMA functionality for page
migration. cat ``/proc/<pid>/numa_maps`` allows an easy review of where the migration. cat ``/proc/<pid>/numa_maps`` allows an easy review of where the
pages of a process are located. See also the numa_maps documentation in the pages of a process are located. See also the numa_maps documentation in the
proc(5) man page. proc(5) man page.
@@ -30,19 +33,19 @@ proc(5) man page.
Manual migration is useful if for example the scheduler has relocated Manual migration is useful if for example the scheduler has relocated
a process to a processor on a distant node. A batch scheduler or an a process to a processor on a distant node. A batch scheduler or an
administrator may detect the situation and move the pages of the process administrator may detect the situation and move the pages of the process
nearer to the new processor. The kernel itself does only provide nearer to the new processor. The kernel itself only provides
manual page migration support. Automatic page migration may be implemented manual page migration support. Automatic page migration may be implemented
through user space processes that move pages. A special function call through user space processes that move pages. A special function call
"move_pages" allows the moving of individual pages within a process. "move_pages" allows the moving of individual pages within a process.
A NUMA profiler may f.e. obtain a log showing frequent off node For example, A NUMA profiler may obtain a log showing frequent off-node
accesses and may use the result to move pages to more advantageous accesses and may use the result to move pages to more advantageous
locations. locations.
Larger installations usually partition the system using cpusets into Larger installations usually partition the system using cpusets into
sections of nodes. Paul Jackson has equipped cpusets with the ability to sections of nodes. Paul Jackson has equipped cpusets with the ability to
move pages when a task is moved to another cpuset (See move pages when a task is moved to another cpuset (See
Documentation/admin-guide/cgroup-v1/cpusets.rst). :ref:`CPUSETS <cpusets>`).
Cpusets allows the automation of process locality. If a task is moved to Cpusets allow the automation of process locality. If a task is moved to
a new cpuset then also all its pages are moved with it so that the a new cpuset then also all its pages are moved with it so that the
performance of the process does not sink dramatically. Also the pages performance of the process does not sink dramatically. Also the pages
of processes in a cpuset are moved if the allowed memory nodes of a of processes in a cpuset are moved if the allowed memory nodes of a
@@ -67,9 +70,9 @@ In kernel use of migrate_pages()
Lists of pages to be migrated are generated by scanning over Lists of pages to be migrated are generated by scanning over
pages and moving them into lists. This is done by pages and moving them into lists. This is done by
calling isolate_lru_page(). calling isolate_lru_page().
Calling isolate_lru_page increases the references to the page Calling isolate_lru_page() increases the references to the page
so that it cannot vanish while the page migration occurs. so that it cannot vanish while the page migration occurs.
It also prevents the swapper or other scans to encounter It also prevents the swapper or other scans from encountering
the page. the page.
2. We need to have a function of type new_page_t that can be 2. We need to have a function of type new_page_t that can be
@@ -91,23 +94,24 @@ is increased so that the page cannot be freed while page migration occurs.
Steps: Steps:
1. Lock the page to be migrated 1. Lock the page to be migrated.
2. Ensure that writeback is complete. 2. Ensure that writeback is complete.
3. Lock the new page that we want to move to. It is locked so that accesses to 3. Lock the new page that we want to move to. It is locked so that accesses to
this (not yet uptodate) page immediately lock while the move is in progress. this (not yet uptodate) page immediately block while the move is in progress.
4. All the page table references to the page are converted to migration 4. All the page table references to the page are converted to migration
entries. This decreases the mapcount of a page. If the resulting entries. This decreases the mapcount of a page. If the resulting
mapcount is not zero then we do not migrate the page. All user space mapcount is not zero then we do not migrate the page. All user space
processes that attempt to access the page will now wait on the page lock. processes that attempt to access the page will now wait on the page lock
or wait for the migration page table entry to be removed.
5. The i_pages lock is taken. This will cause all processes trying 5. The i_pages lock is taken. This will cause all processes trying
to access the page via the mapping to block on the spinlock. to access the page via the mapping to block on the spinlock.
6. The refcount of the page is examined and we back out if references remain 6. The refcount of the page is examined and we back out if references remain.
otherwise we know that we are the only one referencing this page. Otherwise, we know that we are the only one referencing this page.
7. The radix tree is checked and if it does not contain the pointer to this 7. The radix tree is checked and if it does not contain the pointer to this
page then we back out because someone else modified the radix tree. page then we back out because someone else modified the radix tree.
@@ -134,124 +138,124 @@ Steps:
15. Queued up writeback on the new page is triggered. 15. Queued up writeback on the new page is triggered.
16. If migration entries were page then replace them with real ptes. Doing 16. If migration entries were inserted into the page table, then replace them
so will enable access for user space processes not already waiting for with real ptes. Doing so will enable access for user space processes not
the page lock. already waiting for the page lock.
19. The page locks are dropped from the old and new page. 17. The page locks are dropped from the old and new page.
Processes waiting on the page lock will redo their page faults Processes waiting on the page lock will redo their page faults
and will reach the new page. and will reach the new page.
20. The new page is moved to the LRU and can be scanned by the swapper 18. The new page is moved to the LRU and can be scanned by the swapper,
etc again. etc. again.
Non-LRU page migration Non-LRU page migration
====================== ======================
Although original migration aimed for reducing the latency of memory access Although migration originally aimed for reducing the latency of memory accesses
for NUMA, compaction who want to create high-order page is also main customer. for NUMA, compaction also uses migration to create high-order pages.
Current problem of the implementation is that it is designed to migrate only Current problem of the implementation is that it is designed to migrate only
*LRU* pages. However, there are potential non-lru pages which can be migrated *LRU* pages. However, there are potential non-LRU pages which can be migrated
in drivers, for example, zsmalloc, virtio-balloon pages. in drivers, for example, zsmalloc, virtio-balloon pages.
For virtio-balloon pages, some parts of migration code path have been hooked For virtio-balloon pages, some parts of migration code path have been hooked
up and added virtio-balloon specific functions to intercept migration logics. up and added virtio-balloon specific functions to intercept migration logics.
It's too specific to a driver so other drivers who want to make their pages It's too specific to a driver so other drivers who want to make their pages
movable would have to add own specific hooks in migration path. movable would have to add their own specific hooks in the migration path.
To overclome the problem, VM supports non-LRU page migration which provides To overcome the problem, VM supports non-LRU page migration which provides
generic functions for non-LRU movable pages without driver specific hooks generic functions for non-LRU movable pages without driver specific hooks
migration path. in the migration path.
If a driver want to make own pages movable, it should define three functions If a driver wants to make its pages movable, it should define three functions
which are function pointers of struct address_space_operations. which are function pointers of struct address_space_operations.
1. ``bool (*isolate_page) (struct page *page, isolate_mode_t mode);`` 1. ``bool (*isolate_page) (struct page *page, isolate_mode_t mode);``
What VM expects on isolate_page function of driver is to return *true* What VM expects from isolate_page() function of driver is to return *true*
if driver isolates page successfully. On returing true, VM marks the page if driver isolates the page successfully. On returning true, VM marks the page
as PG_isolated so concurrent isolation in several CPUs skip the page as PG_isolated so concurrent isolation in several CPUs skip the page
for isolation. If a driver cannot isolate the page, it should return *false*. for isolation. If a driver cannot isolate the page, it should return *false*.
Once page is successfully isolated, VM uses page.lru fields so driver Once page is successfully isolated, VM uses page.lru fields so driver
shouldn't expect to preserve values in that fields. shouldn't expect to preserve values in those fields.
2. ``int (*migratepage) (struct address_space *mapping,`` 2. ``int (*migratepage) (struct address_space *mapping,``
| ``struct page *newpage, struct page *oldpage, enum migrate_mode);`` | ``struct page *newpage, struct page *oldpage, enum migrate_mode);``
After isolation, VM calls migratepage of driver with isolated page. After isolation, VM calls migratepage() of driver with the isolated page.
The function of migratepage is to move content of the old page to new page The function of migratepage() is to move the contents of the old page to the
new page
and set up fields of struct page newpage. Keep in mind that you should and set up fields of struct page newpage. Keep in mind that you should
indicate to the VM the oldpage is no longer movable via __ClearPageMovable() indicate to the VM the oldpage is no longer movable via __ClearPageMovable()
under page_lock if you migrated the oldpage successfully and returns under page_lock if you migrated the oldpage successfully and returned
MIGRATEPAGE_SUCCESS. If driver cannot migrate the page at the moment, driver MIGRATEPAGE_SUCCESS. If driver cannot migrate the page at the moment, driver
can return -EAGAIN. On -EAGAIN, VM will retry page migration in a short time can return -EAGAIN. On -EAGAIN, VM will retry page migration in a short time
because VM interprets -EAGAIN as "temporal migration failure". On returning because VM interprets -EAGAIN as "temporary migration failure". On returning
any error except -EAGAIN, VM will give up the page migration without retrying any error except -EAGAIN, VM will give up the page migration without
in this time. retrying.
Driver shouldn't touch page.lru field VM using in the functions. Driver shouldn't touch the page.lru field while in the migratepage() function.
3. ``void (*putback_page)(struct page *);`` 3. ``void (*putback_page)(struct page *);``
If migration fails on isolated page, VM should return the isolated page If migration fails on the isolated page, VM should return the isolated page
to the driver so VM calls driver's putback_page with migration failed page. to the driver so VM calls the driver's putback_page() with the isolated page.
In this function, driver should put the isolated page back to the own data In this function, the driver should put the isolated page back into its own data
structure. structure.
4. non-lru movable page flags 4. non-LRU movable page flags
There are two page flags for supporting non-lru movable page. There are two page flags for supporting non-LRU movable page.
* PG_movable * PG_movable
Driver should use the below function to make page movable under page_lock:: Driver should use the function below to make page movable under page_lock::
void __SetPageMovable(struct page *page, struct address_space *mapping) void __SetPageMovable(struct page *page, struct address_space *mapping)
It needs argument of address_space for registering migration It needs argument of address_space for registering migration
family functions which will be called by VM. Exactly speaking, family functions which will be called by VM. Exactly speaking,
PG_movable is not a real flag of struct page. Rather than, VM PG_movable is not a real flag of struct page. Rather, VM
reuses page->mapping's lower bits to represent it. reuses the page->mapping's lower bits to represent it::
::
#define PAGE_MAPPING_MOVABLE 0x2 #define PAGE_MAPPING_MOVABLE 0x2
page->mapping = page->mapping | PAGE_MAPPING_MOVABLE; page->mapping = page->mapping | PAGE_MAPPING_MOVABLE;
so driver shouldn't access page->mapping directly. Instead, driver should so driver shouldn't access page->mapping directly. Instead, driver should
use page_mapping which mask off the low two bits of page->mapping under use page_mapping() which masks off the low two bits of page->mapping under
page lock so it can get right struct address_space. page lock so it can get the right struct address_space.
For testing of non-lru movable page, VM supports __PageMovable function. For testing of non-LRU movable pages, VM supports __PageMovable() function.
However, it doesn't guarantee to identify non-lru movable page because However, it doesn't guarantee to identify non-LRU movable pages because
page->mapping field is unified with other variables in struct page. the page->mapping field is unified with other variables in struct page.
As well, if driver releases the page after isolation by VM, page->mapping If the driver releases the page after isolation by VM, page->mapping
doesn't have stable value although it has PAGE_MAPPING_MOVABLE doesn't have a stable value although it has PAGE_MAPPING_MOVABLE set
(Look at __ClearPageMovable). But __PageMovable is cheap to catch whether (look at __ClearPageMovable). But __PageMovable() is cheap to call whether
page is LRU or non-lru movable once the page has been isolated. Because page is LRU or non-LRU movable once the page has been isolated because LRU
LRU pages never can have PAGE_MAPPING_MOVABLE in page->mapping. It is also pages can never have PAGE_MAPPING_MOVABLE set in page->mapping. It is also
good for just peeking to test non-lru movable pages before more expensive good for just peeking to test non-LRU movable pages before more expensive
checking with lock_page in pfn scanning to select victim. checking with lock_page() in pfn scanning to select a victim.
For guaranteeing non-lru movable page, VM provides PageMovable function. For guaranteeing non-LRU movable page, VM provides PageMovable() function.
Unlike __PageMovable, PageMovable functions validates page->mapping and Unlike __PageMovable(), PageMovable() validates page->mapping and
mapping->a_ops->isolate_page under lock_page. The lock_page prevents sudden mapping->a_ops->isolate_page under lock_page(). The lock_page() prevents
destroying of page->mapping. sudden destroying of page->mapping.
Driver using __SetPageMovable should clear the flag via __ClearMovablePage Drivers using __SetPageMovable() should clear the flag via
under page_lock before the releasing the page. __ClearMovablePage() under page_lock() before the releasing the page.
* PG_isolated * PG_isolated
To prevent concurrent isolation among several CPUs, VM marks isolated page To prevent concurrent isolation among several CPUs, VM marks isolated page
as PG_isolated under lock_page. So if a CPU encounters PG_isolated non-lru as PG_isolated under lock_page(). So if a CPU encounters PG_isolated
movable page, it can skip it. Driver doesn't need to manipulate the flag non-LRU movable page, it can skip it. Driver doesn't need to manipulate the
because VM will set/clear it automatically. Keep in mind that if driver flag because VM will set/clear it automatically. Keep in mind that if the
sees PG_isolated page, it means the page have been isolated by VM so it driver sees a PG_isolated page, it means the page has been isolated by the
shouldn't touch page.lru field. VM so it shouldn't touch the page.lru field.
PG_isolated is alias with PG_reclaim flag so driver shouldn't use the flag The PG_isolated flag is aliased with the PG_reclaim flag so drivers
for own purpose. shouldn't use PG_isolated for its own purposes.
Monitoring Migration Monitoring Migration
===================== =====================
@@ -266,8 +270,8 @@ The following events (counters) can be used to monitor page migration.
512. 512.
2. PGMIGRATE_FAIL: Normal page migration failure. Same counting rules as for 2. PGMIGRATE_FAIL: Normal page migration failure. Same counting rules as for
_SUCCESS, above: this will be increased by the number of subpages, if it was PGMIGRATE_SUCCESS, above: this will be increased by the number of subpages,
a THP. if it was a THP.
3. THP_MIGRATION_SUCCESS: A THP was migrated without being split. 3. THP_MIGRATION_SUCCESS: A THP was migrated without being split.