€•ù²Œsphinx.addnodes”Œdocument”“”)”}”(Œ rawsource”Œ”Œchildren”]”(Œ translations”Œ LanguagesNode”“”)”}”(hhh]”(hŒ pending_xref”“”)”}”(hhh]”Œdocutils.nodes”ŒText”“”ŒChinese (Simplified)”…””}”Œparent”hsbaŒ attributes”}”(Œids”]”Œclasses”]”Œnames”]”Œdupnames”]”Œbackrefs”]”Œ refdomain”Œstd”Œreftype”Œdoc”Œ reftarget”Œ%/translations/zh_CN/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuŒtagname”hhh ubh)”}”(hhh]”hŒChinese (Traditional)”…””}”hh2sbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/zh_TW/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubh)”}”(hhh]”hŒItalian”…””}”hhFsbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/it_IT/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubh)”}”(hhh]”hŒJapanese”…””}”hhZsbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/ja_JP/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubh)”}”(hhh]”hŒKorean”…””}”hhnsbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/ko_KR/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubh)”}”(hhh]”hŒPortuguese (Brazilian)”…””}”hh‚sbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/pt_BR/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubh)”}”(hhh]”hŒSpanish”…””}”hh–sbah}”(h]”h ]”h"]”h$]”h&]”Œ refdomain”h)Œreftype”h+Œ reftarget”Œ%/translations/sp_SP/mm/page_migration”Œmodname”NŒ classname”NŒ refexplicit”ˆuh1hhh ubeh}”(h]”h ]”h"]”h$]”h&]”Œcurrent_language”ŒEnglish”uh1h hhŒ _document”hŒsource”NŒline”NubhŒsection”“”)”}”(hhh]”(hŒtitle”“”)”}”(hŒPage migration”h]”hŒPage migration”…””}”(hh¼h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhh·h²hh³Œ?/var/lib/git/docbuild/linux/Documentation/mm/page_migration.rst”h´KubhŒ paragraph”“”)”}”(hXPage migration allows moving the physical location of pages between 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 system rearranges the physical location of those pages.”h]”hXPage migration allows moving the physical location of pages between 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 system rearranges the physical location of those pages.”…””}”(hhÍh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hŒWAlso see Documentation/mm/hmm.rst for migrating pages to or from device private memory.”h]”hŒWAlso see Documentation/mm/hmm.rst for migrating pages to or from device private memory.”…””}”(hhÛh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K hh·h²hubhÌ)”}”(hŒ§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 is running.”h]”hŒ§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 is running.”…””}”(hhéh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K hh·h²hubhÌ)”}”(hXpPage migration allows a process to manually relocate the node on which its pages are located through the MF_MOVE and MF_MOVE_ALL options while setting a new memory policy via mbind(). The pages of a process can also be relocated from another process using the sys_migrate_pages() function call. The migrate_pages() function call takes two sets of nodes and moves pages of a process that are located on the from nodes to the destination nodes. Page migration functions are provided by the numactl package by Andi Kleen (a version later than 0.9.3 is required. Get it from https://github.com/numactl/numactl.git). numactl provides libnuma which provides an interface similar to other NUMA functionality for page migration. cat ``/proc//numa_maps`` allows an easy review of where the pages of a process are located. See also the numa_maps documentation in the proc(5) man page.”h]”(hX;Page migration allows a process to manually relocate the node on which its pages are located through the MF_MOVE and MF_MOVE_ALL options while setting a new memory policy via mbind(). The pages of a process can also be relocated from another process using the sys_migrate_pages() function call. The migrate_pages() function call takes two sets of nodes and moves pages of a process that are located on the from nodes to the destination nodes. Page migration functions are provided by the numactl package by Andi Kleen (a version later than 0.9.3 is required. Get it from ”…””}”(hh÷h²hh³Nh´NubhŒ reference”“”)”}”(hŒ&https://github.com/numactl/numactl.git”h]”hŒ&https://github.com/numactl/numactl.git”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”juh1hÿhh÷ubhŒu). numactl provides libnuma which provides an interface similar to other NUMA functionality for page migration. cat ”…””}”(hh÷h²hh³Nh´NubhŒliteral”“”)”}”(hŒ``/proc//numa_maps``”h]”hŒ/proc//numa_maps”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhh÷ubhŒ allows an easy review of where the pages of a process are located. See also the numa_maps documentation in the proc(5) man page.”…””}”(hh÷h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hX{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 administrator may detect the situation and move the pages of the process nearer to the new processor. The kernel itself only provides manual page migration support. Automatic page migration may be implemented through user space processes that move pages. A special function call "move_pages" allows the moving of individual pages within a process. 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 locations.”h]”hXManual migration is useful if for example the scheduler has relocated 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 nearer to the new processor. The kernel itself only provides manual page migration support. Automatic page migration may be implemented through user space processes that move pages. A special function call “move_pages†allows the moving of individual pages within a process. 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 locations.”…””}”(hj.h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hXLarger installations usually partition the system using cpusets into sections of nodes. Paul Jackson has equipped cpusets with the ability to move pages when a task is moved to another cpuset (See :ref:`CPUSETS `). Cpusets allow the automation of process locality. If a task is moved to a new cpuset then also all its pages are moved with it so that the performance of the process does not sink dramatically. Also the pages of processes in a cpuset are moved if the allowed memory nodes of a cpuset are changed.”h]”(hŒÅLarger installations usually partition the system using cpusets into sections of nodes. Paul Jackson has equipped cpusets with the ability to move pages when a task is moved to another cpuset (See ”…””}”(hj<h²hh³Nh´Nubh)”}”(hŒ:ref:`CPUSETS `”h]”hŒinline”“”)”}”(hjFh]”hŒCPUSETS”…””}”(hjJh²hh³Nh´Nubah}”(h]”h ]”(Œxref”Œstd”Œstd-ref”eh"]”h$]”h&]”uh1jHhjDubah}”(h]”h ]”h"]”h$]”h&]”Œrefdoc”Œmm/page_migration”Œ refdomain”jUŒreftype”Œref”Œ refexplicit”ˆŒrefwarn”ˆŒ reftarget”Œcpusets”uh1hh³hÊh´K*hj<ubhX+). Cpusets allow the automation of process locality. If a task is moved to a new cpuset then also all its pages are moved with it so that the performance of the process does not sink dramatically. Also the pages of processes in a cpuset are moved if the allowed memory nodes of a cpuset are changed.”…””}”(hj<h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K*hh·h²hubhÌ)”}”(hX\Page migration allows the preservation of the relative location of pages within a group of nodes for all migration techniques which will preserve a particular memory allocation pattern generated even after migrating a process. This is necessary in order to preserve the memory latencies. Processes will run with similar performance after migration.”h]”hX\Page migration allows the preservation of the relative location of pages within a group of nodes for all migration techniques which will preserve a particular memory allocation pattern generated even after migrating a process. This is necessary in order to preserve the memory latencies. Processes will run with similar performance after migration.”…””}”(hjsh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K4hh·h²hubhÌ)”}”(hX Page migration occurs in several steps. First a high level description for those trying to use migrate_pages() from the kernel (for userspace usage see the Andi Kleen's numactl package mentioned above) and then a low level description of how the low level details work.”h]”hXPage migration occurs in several steps. First a high level description for those trying to use migrate_pages() from the kernel (for userspace usage see the Andi Kleen’s numactl package mentioned above) and then a low level description of how the low level details work.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K:hh·h²hubh¶)”}”(hhh]”(h»)”}”(hŒ In kernel use of migrate_pages()”h]”hŒ In kernel use of migrate_pages()”…””}”(hj’h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhjh²hh³hÊh´K@ubhŒenumerated_list”“”)”}”(hhh]”(hŒ list_item”“”)”}”(hXqRemove folios from the LRU. Lists of folios to be migrated are generated by scanning over folios and moving them into lists. This is done by calling folio_isolate_lru(). Calling folio_isolate_lru() increases the references to the folio so that it cannot vanish while the folio migration occurs. It also prevents the swapper or other scans from encountering the folio. ”h]”(hÌ)”}”(hŒRemove folios from the LRU.”h]”hŒRemove folios from the LRU.”…””}”(hj«h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KBhj§ubhÌ)”}”(hXSLists of folios to be migrated are generated by scanning over folios and moving them into lists. This is done by calling folio_isolate_lru(). Calling folio_isolate_lru() increases the references to the folio so that it cannot vanish while the folio migration occurs. It also prevents the swapper or other scans from encountering the folio.”h]”hXSLists of folios to be migrated are generated by scanning over folios and moving them into lists. This is done by calling folio_isolate_lru(). Calling folio_isolate_lru() increases the references to the folio so that it cannot vanish while the folio migration occurs. It also prevents the swapper or other scans from encountering the folio.”…””}”(hj¹h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KDhj§ubeh}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj¢h²hh³hÊh´Nubj¦)”}”(hŒ±We need to have a function of type new_folio_t that can be passed to migrate_pages(). This function should figure out how to allocate the correct new folio given the old folio. ”h]”hÌ)”}”(hŒ°We need to have a function of type new_folio_t that can be passed to migrate_pages(). This function should figure out how to allocate the correct new folio given the old folio.”h]”hŒ°We need to have a function of type new_folio_t that can be passed to migrate_pages(). This function should figure out how to allocate the correct new folio given the old folio.”…””}”(hjÑh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KLhjÍubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj¢h²hh³hÊh´Nubj¦)”}”(hŒ­The migrate_pages() function is called which attempts to do the migration. It will call the function to allocate the new folio for each folio that is considered for moving. ”h]”hÌ)”}”(hŒ¬The migrate_pages() function is called which attempts to do the migration. It will call the function to allocate the new folio for each folio that is considered for moving.”h]”hŒ¬The migrate_pages() function is called which attempts to do the migration. It will call the function to allocate the new folio for each folio that is considered for moving.”…””}”(hjéh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KPhjåubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj¢h²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”Œenumtype”Œarabic”Œprefix”hŒsuffix”Œ.”uh1j hjh²hh³hÊh´KBubeh}”(h]”Œin-kernel-use-of-migrate-pages”ah ]”h"]”Œ in kernel use of migrate_pages()”ah$]”h&]”uh1hµhh·h²hh³hÊh´K@ubh¶)”}”(hhh]”(h»)”}”(hŒHow migrate_pages() works”h]”hŒHow migrate_pages() works”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhjh²hh³hÊh´KUubhÌ)”}”(hX+migrate_pages() does several passes over its list of folios. A folio is moved if all references to a folio are removable at the time. The folio has already been removed from the LRU via folio_isolate_lru() and the refcount is increased so that the folio cannot be freed while folio migration occurs.”h]”hX+migrate_pages() does several passes over its list of folios. A folio is moved if all references to a folio are removable at the time. The folio has already been removed from the LRU via folio_isolate_lru() and the refcount is increased so that the folio cannot be freed while folio migration occurs.”…””}”(hj!h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KWhjh²hubhÌ)”}”(hŒSteps:”h]”hŒSteps:”…””}”(hj/h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K\hjh²hubj¡)”}”(hhh]”(j¦)”}”(hŒLock the page to be migrated. ”h]”hÌ)”}”(hŒLock the page to be migrated.”h]”hŒLock the page to be migrated.”…””}”(hjDh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K^hj@ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ#Ensure that writeback is complete. ”h]”hÌ)”}”(hŒ"Ensure that writeback is complete.”h]”hŒ"Ensure that writeback is complete.”…””}”(hj\h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K`hjXubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒœLock the new page that we want to move to. It is locked so that accesses to this (not yet up-to-date) page immediately block while the move is in progress. ”h]”hÌ)”}”(hŒ›Lock the new page that we want to move to. It is locked so that accesses to this (not yet up-to-date) page immediately block while the move is in progress.”h]”hŒ›Lock the new page that we want to move to. It is locked so that accesses to this (not yet up-to-date) page immediately block while the move is in progress.”…””}”(hjth²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kbhjpubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hXNAll the page table references to the page are converted to migration entries. This decreases the mapcount of a page. If the resulting mapcount is not zero then we do not migrate the page. All user space processes that attempt to access the page will now wait on the page lock or wait for the migration page table entry to be removed. ”h]”hÌ)”}”(hXMAll the page table references to the page are converted to migration entries. This decreases the mapcount of a page. If the resulting mapcount is not zero then we do not migrate the page. All user space processes that attempt to access the page will now wait on the page lock or wait for the migration page table entry to be removed.”h]”hXMAll the page table references to the page are converted to migration entries. This decreases the mapcount of a page. If the resulting mapcount is not zero then we do not migrate the page. All user space processes that attempt to access the page will now wait on the page lock or wait for the migration page table entry to be removed.”…””}”(hjŒh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kehjˆubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ}The i_pages lock is taken. This will cause all processes trying to access the page via the mapping to block on the spinlock. ”h]”hÌ)”}”(hŒ|The i_pages lock is taken. This will cause all processes trying to access the page via the mapping to block on the spinlock.”h]”hŒ|The i_pages lock is taken. This will cause all processes trying to access the page via the mapping to block on the spinlock.”…””}”(hj¤h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kkhj ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒŽThe refcount of the page is examined and we back out if references remain. Otherwise, we know that we are the only one referencing this page. ”h]”hÌ)”}”(hŒThe refcount of the page is examined and we back out if references remain. Otherwise, we know that we are the only one referencing this page.”h]”hŒThe refcount of the page is examined and we back out if references remain. Otherwise, we know that we are the only one referencing this page.”…””}”(hj¼h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Knhj¸ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒThe radix tree is checked and if it does not contain the pointer to this page then we back out because someone else modified the radix tree. ”h]”hÌ)”}”(hŒŒThe radix tree is checked and if it does not contain the pointer to this page then we back out because someone else modified the radix tree.”h]”hŒŒThe radix tree is checked and if it does not contain the pointer to this page then we back out because someone else modified the radix tree.”…””}”(hjÔh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KqhjÐubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒŽThe new page is prepped with some settings from the old page so that accesses to the new page will discover a page with the correct settings. ”h]”hÌ)”}”(hŒThe new page is prepped with some settings from the old page so that accesses to the new page will discover a page with the correct settings.”h]”hŒThe new page is prepped with some settings from the old page so that accesses to the new page will discover a page with the correct settings.”…””}”(hjìh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kthjèubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ4The radix tree is changed to point to the new page. ”h]”hÌ)”}”(hŒ3The radix tree is changed to point to the new page.”h]”hŒ3The radix tree is changed to point to the new page.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kwhjubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒÀThe reference count of the old page is dropped because the address space reference is gone. A reference to the new page is established because the new page is referenced by the address space. ”h]”hÌ)”}”(hŒ¿The reference count of the old page is dropped because the address space reference is gone. A reference to the new page is established because the new page is referenced by the address space.”h]”hŒ¿The reference count of the old page is dropped because the address space reference is gone. A reference to the new page is established because the new page is referenced by the address space.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kyhjubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ§The i_pages lock is dropped. With that lookups in the mapping become possible again. Processes will move from spinning on the lock to sleeping on the locked new page. ”h]”hÌ)”}”(hŒ¦The i_pages lock is dropped. With that lookups in the mapping become possible again. Processes will move from spinning on the lock to sleeping on the locked new page.”h]”hŒ¦The i_pages lock is dropped. With that lookups in the mapping become possible again. Processes will move from spinning on the lock to sleeping on the locked new page.”…””}”(hj4h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K}hj0ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ.The page contents are copied to the new page. ”h]”hÌ)”}”(hŒ-The page contents are copied to the new page.”h]”hŒ-The page contents are copied to the new page.”…””}”(hjLh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KhjHubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ5The remaining page flags are copied to the new page. ”h]”hÌ)”}”(hŒ4The remaining page flags are copied to the new page.”h]”hŒ4The remaining page flags are copied to the new page.”…””}”(hjdh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kƒhj`ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒcThe old page flags are cleared to indicate that the page does not provide any information anymore. ”h]”hÌ)”}”(hŒbThe old page flags are cleared to indicate that the page does not provide any information anymore.”h]”hŒbThe old page flags are cleared to indicate that the page does not provide any information anymore.”…””}”(hj|h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K…hjxubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ2Queued up writeback on the new page is triggered. ”h]”hÌ)”}”(hŒ1Queued up writeback on the new page is triggered.”h]”hŒ1Queued up writeback on the new page is triggered.”…””}”(hj”h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kˆhjubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ¶If migration entries were inserted into the page table, then replace them with real ptes. Doing so will enable access for user space processes not already waiting for the page lock. ”h]”hÌ)”}”(hŒµIf migration entries were inserted into the page table, then replace them with real ptes. Doing so will enable access for user space processes not already waiting for the page lock.”h]”hŒµIf migration entries were inserted into the page table, then replace them with real ptes. Doing so will enable access for user space processes not already waiting for the page lock.”…””}”(hj¬h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KŠhj¨ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒ’The page locks are dropped from the old and new page. Processes waiting on the page lock will redo their page faults and will reach the new page. ”h]”hÌ)”}”(hŒ‘The page locks are dropped from the old and new page. Processes waiting on the page lock will redo their page faults and will reach the new page.”h]”hŒ‘The page locks are dropped from the old and new page. Processes waiting on the page lock will redo their page faults and will reach the new page.”…””}”(hjÄh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KŽhjÀubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubj¦)”}”(hŒPThe new page is moved to the LRU and can be scanned by the swapper, etc. again. ”h]”hÌ)”}”(hŒOThe new page is moved to the LRU and can be scanned by the swapper, etc. again.”h]”hŒOThe new page is moved to the LRU and can be scanned by the swapper, etc. again.”…””}”(hjÜh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K’hjØubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hj=h²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”jjjhjjuh1j hjh²hh³hÊh´K^ubeh}”(h]”Œhow-migrate-pages-works”ah ]”h"]”Œhow migrate_pages() works”ah$]”h&]”uh1hµhh·h²hh³hÊh´KUubh¶)”}”(hhh]”(h»)”}”(hŒmovable_ops page migration”h]”hŒmovable_ops page migration”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhjþh²hh³hÊh´K–ubhÌ)”}”(hŒ•Selected typed, non-folio pages (e.g., pages inflated in a memory balloon, zsmalloc pages) can be migrated using the movable_ops migration framework.”h]”hŒ•Selected typed, non-folio pages (e.g., pages inflated in a memory balloon, zsmalloc pages) can be migrated using the movable_ops migration framework.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K˜hjþh²hubhÌ)”}”(hŒŠThe "struct movable_operations" provide callbacks specific to a page type for isolating, migrating and un-isolating (putback) these pages.”h]”hŒŽThe “struct movable_operations†provide callbacks specific to a page type for isolating, migrating and un-isolating (putback) these pages.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K›hjþh²hubhÌ)”}”(hŒäOnce a page is indicated as having movable_ops, that condition must not change until the page was freed back to the buddy. This includes not changing/clearing the page type and not changing/clearing the PG_movable_ops page flag.”h]”hŒäOnce a page is indicated as having movable_ops, that condition must not change until the page was freed back to the buddy. This includes not changing/clearing the page type and not changing/clearing the PG_movable_ops page flag.”…””}”(hj+h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kžhjþh²hubhÌ)”}”(hŒNArbitrary drivers cannot currently make use of this framework, as it requires:”h]”hŒNArbitrary drivers cannot currently make use of this framework, as it requires:”…””}”(hj9h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K£hjþh²hubj¡)”}”(hhh]”(j¦)”}”(hŒ a page type”h]”hÌ)”}”(hjLh]”hŒ a page type”…””}”(hjNh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K¦hjJubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjGh²hh³hÊh´Nubj¦)”}”(hŒ_indicating them as possibly having movable_ops in page_has_movable_ops() based on the page type”h]”hÌ)”}”(hŒ_indicating them as possibly having movable_ops in page_has_movable_ops() based on the page type”h]”hŒ_indicating them as possibly having movable_ops in page_has_movable_ops() based on the page type”…””}”(hjeh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K§hjaubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjGh²hh³hÊh´Nubj¦)”}”(hŒHreturning the movable_ops from page_movable_ops() based on the page type”h]”hÌ)”}”(hŒHreturning the movable_ops from page_movable_ops() based on the page type”h]”hŒHreturning the movable_ops from page_movable_ops() based on the page type”…””}”(hj}h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K©hjyubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjGh²hh³hÊh´Nubj¦)”}”(hŒYnot reusing the PG_movable_ops and PG_movable_ops_isolated page flags for other purposes ”h]”hÌ)”}”(hŒXnot reusing the PG_movable_ops and PG_movable_ops_isolated page flags for other purposes”h]”hŒXnot reusing the PG_movable_ops and PG_movable_ops_isolated page flags for other purposes”…””}”(hj•h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K«hj‘ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjGh²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”jŒ loweralpha”jŒ(”jŒ)”uh1j hjþh²hh³hÊh´K¦ubhÌ)”}”(hŒ†For example, balloon drivers can make use of this framework through the balloon-compaction infrastructure residing in the core kernel.”h]”hŒ†For example, balloon drivers can make use of this framework through the balloon-compaction infrastructure residing in the core kernel.”…””}”(hj²h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K®hjþh²hubeh}”(h]”Œmovable-ops-page-migration”ah ]”h"]”Œmovable_ops page migration”ah$]”h&]”uh1hµhh·h²hh³hÊh´K–ubh¶)”}”(hhh]”(h»)”}”(hŒMonitoring Migration”h]”hŒMonitoring Migration”…””}”(hjËh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhjÈh²hh³hÊh´K²ubhÌ)”}”(hŒFThe following events (counters) can be used to monitor page migration.”h]”hŒFThe following events (counters) can be used to monitor page migration.”…””}”(hjÙh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K´hjÈh²hubj¡)”}”(hhh]”(j¦)”}”(hXžPGMIGRATE_SUCCESS: Normal page migration success. Each count means that a page was migrated. If the page was a non-THP and non-hugetlb page, then this counter is increased by one. If the page was a THP or hugetlb, then this counter is increased by the number of THP or hugetlb subpages. For example, migration of a single 2MB THP that has 4KB-size base pages (subpages) will cause this counter to increase by 512. ”h]”hÌ)”}”(hXPGMIGRATE_SUCCESS: Normal page migration success. Each count means that a page was migrated. If the page was a non-THP and non-hugetlb page, then this counter is increased by one. If the page was a THP or hugetlb, then this counter is increased by the number of THP or hugetlb subpages. For example, migration of a single 2MB THP that has 4KB-size base pages (subpages) will cause this counter to increase by 512.”h]”hXPGMIGRATE_SUCCESS: Normal page migration success. Each count means that a page was migrated. If the page was a non-THP and non-hugetlb page, then this counter is increased by one. If the page was a THP or hugetlb, then this counter is increased by the number of THP or hugetlb subpages. For example, migration of a single 2MB THP that has 4KB-size base pages (subpages) will cause this counter to increase by 512.”…””}”(hjîh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K¶hjêubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjçh²hh³hÊh´Nubj¦)”}”(hŒ²PGMIGRATE_FAIL: Normal page migration failure. Same counting rules as for PGMIGRATE_SUCCESS, above: this will be increased by the number of subpages, if it was a THP or hugetlb. ”h]”hÌ)”}”(hŒ±PGMIGRATE_FAIL: Normal page migration failure. Same counting rules as for PGMIGRATE_SUCCESS, above: this will be increased by the number of subpages, if it was a THP or hugetlb.”h]”hŒ±PGMIGRATE_FAIL: Normal page migration failure. Same counting rules as for PGMIGRATE_SUCCESS, above: this will be increased by the number of subpages, if it was a THP or hugetlb.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K½hjubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjçh²hh³hÊh´Nubj¦)”}”(hŒ?THP_MIGRATION_SUCCESS: A THP was migrated without being split. ”h]”hÌ)”}”(hŒ>THP_MIGRATION_SUCCESS: A THP was migrated without being split.”h]”hŒ>THP_MIGRATION_SUCCESS: A THP was migrated without being split.”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÁhjubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjçh²hh³hÊh´Nubj¦)”}”(hŒGTHP_MIGRATION_FAIL: A THP could not be migrated nor it could be split. ”h]”hÌ)”}”(hŒFTHP_MIGRATION_FAIL: A THP could not be migrated nor it could be split.”h]”hŒFTHP_MIGRATION_FAIL: A THP could not be migrated nor it could be split.”…””}”(hj6h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÃhj2ubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjçh²hh³hÊh´Nubj¦)”}”(hŒ™THP_MIGRATION_SPLIT: A THP was migrated, but not as such: first, the THP had to be split. After splitting, a migration retry was used for its sub-pages. ”h]”hÌ)”}”(hŒ˜THP_MIGRATION_SPLIT: A THP was migrated, but not as such: first, the THP had to be split. After splitting, a migration retry was used for its sub-pages.”h]”hŒ˜THP_MIGRATION_SPLIT: A THP was migrated, but not as such: first, the THP had to be split. After splitting, a migration retry was used for its sub-pages.”…””}”(hjNh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÅhjJubah}”(h]”h ]”h"]”h$]”h&]”uh1j¥hjçh²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”jjjhjjuh1j hjÈh²hh³hÊh´K¶ubhÌ)”}”(hŒÆTHP_MIGRATION_* events also update the appropriate PGMIGRATE_SUCCESS or PGMIGRATE_FAIL events. For example, a THP migration failure will cause both THP_MIGRATION_FAIL and PGMIGRATE_FAIL to increase.”h]”hŒÆTHP_MIGRATION_* events also update the appropriate PGMIGRATE_SUCCESS or PGMIGRATE_FAIL events. For example, a THP migration failure will cause both THP_MIGRATION_FAIL and PGMIGRATE_FAIL to increase.”…””}”(hjhh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÈhjÈh²hubhÌ)”}”(hŒ:Christoph Lameter, May 8, 2006. Minchan Kim, Mar 28, 2016.”h]”hŒ:Christoph Lameter, May 8, 2006. Minchan Kim, Mar 28, 2016.”…””}”(hjvh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÌhjÈh²hubhŒindex”“”)”}”(hhh]”h}”(h]”h ]”h"]”h$]”h&]”Œentries”]”(Œsingle”Œmovable_operations (C struct)”Œc.movable_operations”hNt”auh1j„hjÈh²hh³Nh´NubhŒdesc”“”)”}”(hhh]”(hŒdesc_signature”“”)”}”(hŒmovable_operations”h]”hŒdesc_signature_line”“”)”}”(hŒstruct movable_operations”h]”(hŒdesc_sig_keyword”“”)”}”(hŒstruct”h]”hŒstruct”…””}”(hj¨h²hh³Nh´Nubah}”(h]”h ]”Œk”ah"]”h$]”h&]”uh1j¦hj¢h²hh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´KubhŒdesc_sig_space”“”)”}”(hŒ ”h]”hŒ ”…””}”(hjºh²hh³Nh´Nubah}”(h]”h ]”Œw”ah"]”h$]”h&]”uh1j¸hj¢h²hh³j·h´KubhŒ desc_name”“”)”}”(hŒmovable_operations”h]”hŒ desc_sig_name”“”)”}”(hjžh]”hŒmovable_operations”…””}”(hjÑh²hh³Nh´Nubah}”(h]”h ]”Œn”ah"]”h$]”h&]”uh1jÏhjËubah}”(h]”h ]”(Œsig-name”Œdescname”eh"]”h$]”h&]”Œ xml:space”Œpreserve”uh1jÉhj¢h²hh³j·h´Kubeh}”(h]”h ]”h"]”h$]”h&]”jçjèŒ add_permalink”ˆuh1j Œsphinx_line_type”Œ declarator”hjœh²hh³j·h´Kubah}”(h]”j“ah ]”(Œsig”Œ sig-object”eh"]”h$]”h&]”Œ is_multiline”ˆŒ _toc_parts”)Œ _toc_name”huh1jšh³j·h´Khj—h²hubhŒ desc_content”“”)”}”(hhh]”hÌ)”}”(hŒDriver page migration”h]”hŒDriver page migration”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khjÿh²hubah}”(h]”h ]”h"]”h$]”h&]”uh1jýhj—h²hh³j·h´Kubeh}”(h]”h ]”(Œc”Œstruct”eh"]”h$]”h&]”Œdomain”jŒobjtype”jŒdesctype”jŒnoindex”‰Œ noindexentry”‰Œnocontentsentry”‰uh1j•h²hhjÈh³Nh´NubhŒ container”“”)”}”(hX **Definition**:: struct movable_operations { bool (*isolate_page)(struct page *, isolate_mode_t); int (*migrate_page)(struct page *dst, struct page *src, enum migrate_mode); void (*putback_page)(struct page *); }; **Members** ``isolate_page`` The VM calls this function to prepare the page to be moved. The page is locked and the driver should not unlock it. The driver should return ``true`` if the page is movable and ``false`` if it is not currently movable. After this function returns, the VM uses the page->lru field, so the driver must preserve any information which is usually stored here. ``migrate_page`` After isolation, the VM calls this function with the isolated **src** page. The driver should copy the contents of the **src** page to the **dst** page and set up the fields of **dst** page. Both pages are locked. If page migration is successful, the driver should return 0. If the driver cannot migrate the page at the moment, it can return -EAGAIN. The VM interprets this as a temporary migration failure and will retry it later. Any other error value is a permanent migration failure and migration will not be retried. The driver shouldn't touch the **src->lru** field while in the migrate_page() function. It may write to **dst->lru**. ``putback_page`` If migration fails on the isolated page, the VM informs the driver that the page is no longer a candidate for migration by calling this function. The driver should put the isolated page back into its own data structure.”h]”(hÌ)”}”(hŒ**Definition**::”h]”(hŒstrong”“”)”}”(hŒ**Definition**”h]”hŒ Definition”…””}”(hj1h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hj+ubhŒ:”…””}”(hj+h²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khj'ubhŒ literal_block”“”)”}”(hŒÐstruct movable_operations { bool (*isolate_page)(struct page *, isolate_mode_t); int (*migrate_page)(struct page *dst, struct page *src, enum migrate_mode); void (*putback_page)(struct page *); };”h]”hŒÐstruct movable_operations { bool (*isolate_page)(struct page *, isolate_mode_t); int (*migrate_page)(struct page *dst, struct page *src, enum migrate_mode); void (*putback_page)(struct page *); };”…””}”hjLsbah}”(h]”h ]”h"]”h$]”h&]”jçjèuh1jJh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khj'ubhÌ)”}”(hŒ **Members**”h]”j0)”}”(hj]h]”hŒMembers”…””}”(hj_h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hj[ubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khj'ubhŒdefinition_list”“”)”}”(hhh]”(hŒdefinition_list_item”“”)”}”(hXw``isolate_page`` The VM calls this function to prepare the page to be moved. The page is locked and the driver should not unlock it. The driver should return ``true`` if the page is movable and ``false`` if it is not currently movable. After this function returns, the VM uses the page->lru field, so the driver must preserve any information which is usually stored here. ”h]”(hŒterm”“”)”}”(hŒ``isolate_page``”h]”j)”}”(hj‚h]”hŒ isolate_page”…””}”(hj„h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhj€ubah}”(h]”h ]”h"]”h$]”h&]”uh1j~h³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´KhjzubhŒ definition”“”)”}”(hhh]”hÌ)”}”(hXeThe VM calls this function to prepare the page to be moved. The page is locked and the driver should not unlock it. The driver should return ``true`` if the page is movable and ``false`` if it is not currently movable. After this function returns, the VM uses the page->lru field, so the driver must preserve any information which is usually stored here.”h]”(hŒThe VM calls this function to prepare the page to be moved. The page is locked and the driver should not unlock it. The driver should return ”…””}”(hjh²hh³Nh´Nubj)”}”(hŒ``true``”h]”hŒtrue”…””}”(hj¥h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhjubhŒ if the page is movable and ”…””}”(hjh²hh³Nh´Nubj)”}”(hŒ ``false``”h]”hŒfalse”…””}”(hj·h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhjubhŒ© if it is not currently movable. After this function returns, the VM uses the page->lru field, so the driver must preserve any information which is usually stored here.”…””}”(hjh²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khjšubah}”(h]”h ]”h"]”h$]”h&]”uh1j˜hjzubeh}”(h]”h ]”h"]”h$]”h&]”uh1jxh³j—h´Khjuubjy)”}”(hX•``migrate_page`` After isolation, the VM calls this function with the isolated **src** page. The driver should copy the contents of the **src** page to the **dst** page and set up the fields of **dst** page. Both pages are locked. If page migration is successful, the driver should return 0. If the driver cannot migrate the page at the moment, it can return -EAGAIN. The VM interprets this as a temporary migration failure and will retry it later. Any other error value is a permanent migration failure and migration will not be retried. The driver shouldn't touch the **src->lru** field while in the migrate_page() function. It may write to **dst->lru**. ”h]”(j)”}”(hŒ``migrate_page``”h]”j)”}”(hjâh]”hŒ migrate_page”…””}”(hjäh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhjàubah}”(h]”h ]”h"]”h$]”h&]”uh1j~h³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´K$hjÜubj™)”}”(hhh]”hÌ)”}”(hXƒAfter isolation, the VM calls this function with the isolated **src** page. The driver should copy the contents of the **src** page to the **dst** page and set up the fields of **dst** page. Both pages are locked. If page migration is successful, the driver should return 0. If the driver cannot migrate the page at the moment, it can return -EAGAIN. The VM interprets this as a temporary migration failure and will retry it later. Any other error value is a permanent migration failure and migration will not be retried. The driver shouldn't touch the **src->lru** field while in the migrate_page() function. It may write to **dst->lru**.”h]”(hŒ>After isolation, the VM calls this function with the isolated ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ**src**”h]”hŒsrc”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhŒ3 page. The driver should copy the contents of the ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ**src**”h]”hŒsrc”…””}”(hjh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhŒ page to the ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ**dst**”h]”hŒdst”…””}”(hj'h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhŒ page and set up the fields of ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ**dst**”h]”hŒdst”…””}”(hj9h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhXu page. Both pages are locked. If page migration is successful, the driver should return 0. If the driver cannot migrate the page at the moment, it can return -EAGAIN. The VM interprets this as a temporary migration failure and will retry it later. Any other error value is a permanent migration failure and migration will not be retried. The driver shouldn’t touch the ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ **src->lru**”h]”hŒsrc->lru”…””}”(hjKh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhŒ> field while in the migrate_page() function. It may write to ”…””}”(hjûh²hh³Nh´Nubj0)”}”(hŒ **dst->lru**”h]”hŒdst->lru”…””}”(hj]h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1j/hjûubhŒ.”…””}”(hjûh²hh³Nh´Nubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´Khjøubah}”(h]”h ]”h"]”h$]”h&]”uh1j˜hjÜubeh}”(h]”h ]”h"]”h$]”h&]”uh1jxh³j÷h´K$hjuubjy)”}”(hŒí``putback_page`` If migration fails on the isolated page, the VM informs the driver that the page is no longer a candidate for migration by calling this function. The driver should put the isolated page back into its own data structure.”h]”(j)”}”(hŒ``putback_page``”h]”j)”}”(hjˆh]”hŒ putback_page”…””}”(hjŠh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1jhj†ubah}”(h]”h ]”h"]”h$]”h&]”uh1j~h³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´K)hj‚ubj™)”}”(hhh]”hÌ)”}”(hŒÜIf migration fails on the isolated page, the VM informs the driver that the page is no longer a candidate for migration by calling this function. The driver should put the isolated page back into its own data structure.”h]”hŒÜIf migration fails on the isolated page, the VM informs the driver that the page is no longer a candidate for migration by calling this function. The driver should put the isolated page back into its own data structure.”…””}”(hj¡h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³ŒZ/var/lib/git/docbuild/linux/Documentation/mm/page_migration:207: ./include/linux/migrate.h”h´K'hjžubah}”(h]”h ]”h"]”h$]”h&]”uh1j˜hj‚ubeh}”(h]”h ]”h"]”h$]”h&]”uh1jxh³jh´K)hjuubeh}”(h]”h ]”h"]”h$]”h&]”uh1jshj'ubeh}”(h]”h ]”Œ kernelindent”ah"]”h$]”h&]”uh1j%hjÈh²hh³Nh´Nubeh}”(h]”Œmonitoring-migration”ah ]”h"]”Œmonitoring migration”ah$]”h&]”uh1hµhh·h²hh³hÊh´K²ubeh}”(h]”Œpage-migration”ah ]”h"]”Œpage migration”ah$]”h&]”uh1hµhhh²hh³hÊh´Kubeh}”(h]”h ]”h"]”h$]”h&]”Œsource”hÊuh1hŒcurrent_source”NŒ current_line”NŒsettings”Œdocutils.frontend”ŒValues”“”)”}”(hºNŒ generator”NŒ datestamp”NŒ source_link”NŒ source_url”NŒ toc_backlinks”Œentry”Œfootnote_backlinks”KŒ sectnum_xform”KŒstrip_comments”NŒstrip_elements_with_classes”NŒ strip_classes”NŒ report_level”KŒ halt_level”KŒexit_status_level”KŒdebug”NŒwarning_stream”NŒ traceback”ˆŒinput_encoding”Œ utf-8-sig”Œinput_encoding_error_handler”Œstrict”Œoutput_encoding”Œutf-8”Œoutput_encoding_error_handler”jüŒerror_encoding”Œutf-8”Œerror_encoding_error_handler”Œbackslashreplace”Œ language_code”Œen”Œrecord_dependencies”NŒconfig”NŒ id_prefix”hŒauto_id_prefix”Œid”Œ dump_settings”NŒdump_internals”NŒdump_transforms”NŒdump_pseudo_xml”NŒexpose_internals”NŒstrict_visitor”NŒ_disable_config”NŒ_source”hÊŒ _destination”NŒ _config_files”]”Œ7/var/lib/git/docbuild/linux/Documentation/docutils.conf”aŒfile_insertion_enabled”ˆŒ raw_enabled”KŒline_length_limit”M'Œpep_references”NŒ pep_base_url”Œhttps://peps.python.org/”Œpep_file_url_template”Œpep-%04d”Œrfc_references”NŒ rfc_base_url”Œ&https://datatracker.ietf.org/doc/html/”Œ tab_width”KŒtrim_footnote_reference_space”‰Œsyntax_highlight”Œlong”Œ smart_quotes”ˆŒsmartquotes_locales”]”Œcharacter_level_inline_markup”‰Œdoctitle_xform”‰Œ docinfo_xform”KŒsectsubtitle_xform”‰Œ image_loading”Œlink”Œembed_stylesheet”‰Œcloak_email_addresses”ˆŒsection_self_link”‰Œenv”NubŒreporter”NŒindirect_targets”]”Œsubstitution_defs”}”Œsubstitution_names”}”Œrefnames”}”Œrefids”}”Œnameids”}”(jÖjÓj j jûjøjÅjÂjÎjËuŒ nametypes”}”(jÖ‰j ‰jû‰jʼnjΉuh}”(jÓh·j jjøjjÂjþjËjÈj“jœuŒ footnote_refs”}”Œ citation_refs”}”Œ autofootnotes”]”Œautofootnote_refs”]”Œsymbol_footnotes”]”Œsymbol_footnote_refs”]”Œ footnotes”]”Œ citations”]”Œautofootnote_start”KŒsymbol_footnote_start”KŒ id_counter”Œ collections”ŒCounter”“”}”…”R”Œparse_messages”]”Œtransform_messages”]”Œ transformer”NŒ include_log”]”Œ decoration”Nh²hub.