Based on kernel version 4.7.2. Page generated on 2016-08-22 22:48 EST.
1 Page migration 2 -------------- 3 4 Page migration allows the moving of the physical location of pages between 5 nodes in a numa system while the process is running. This means that the 6 virtual addresses that the process sees do not change. However, the 7 system rearranges the physical location of those pages. 8 9 The main intend of page migration is to reduce the latency of memory access 10 by moving pages near to the processor where the process accessing that memory 11 is running. 12 13 Page migration allows a process to manually relocate the node on which its 14 pages are located through the MF_MOVE and MF_MOVE_ALL options while setting 15 a new memory policy via mbind(). The pages of process can also be relocated 16 from another process using the sys_migrate_pages() function call. The 17 migrate_pages function call takes two sets of nodes and moves pages of a 18 process that are located on the from nodes to the destination nodes. 19 Page migration functions are provided by the numactl package by Andi Kleen 20 (a version later than 0.9.3 is required. Get it from 21 ftp://oss.sgi.com/www/projects/libnuma/download/). numactl provides libnuma 22 which provides an interface similar to other numa functionality for page 23 migration. cat /proc/<pid>/numa_maps allows an easy review of where the 24 pages of a process are located. See also the numa_maps documentation in the 25 proc(5) man page. 26 27 Manual migration is useful if for example the scheduler has relocated 28 a process to a processor on a distant node. A batch scheduler or an 29 administrator may detect the situation and move the pages of the process 30 nearer to the new processor. The kernel itself does only provide 31 manual page migration support. Automatic page migration may be implemented 32 through user space processes that move pages. A special function call 33 "move_pages" allows the moving of individual pages within a process. 34 A NUMA profiler may f.e. obtain a log showing frequent off node 35 accesses and may use the result to move pages to more advantageous 36 locations. 37 38 Larger installations usually partition the system using cpusets into 39 sections of nodes. Paul Jackson has equipped cpusets with the ability to 40 move pages when a task is moved to another cpuset (See 41 Documentation/cgroups/cpusets.txt). 42 Cpusets allows the automation of process locality. If a task is moved to 43 a new cpuset then also all its pages are moved with it so that the 44 performance of the process does not sink dramatically. Also the pages 45 of processes in a cpuset are moved if the allowed memory nodes of a 46 cpuset are changed. 47 48 Page migration allows the preservation of the relative location of pages 49 within a group of nodes for all migration techniques which will preserve a 50 particular memory allocation pattern generated even after migrating a 51 process. This is necessary in order to preserve the memory latencies. 52 Processes will run with similar performance after migration. 53 54 Page migration occurs in several steps. First a high level 55 description for those trying to use migrate_pages() from the kernel 56 (for userspace usage see the Andi Kleen's numactl package mentioned above) 57 and then a low level description of how the low level details work. 58 59 A. In kernel use of migrate_pages() 60 ----------------------------------- 61 62 1. Remove pages from the LRU. 63 64 Lists of pages to be migrated are generated by scanning over 65 pages and moving them into lists. This is done by 66 calling isolate_lru_page(). 67 Calling isolate_lru_page increases the references to the page 68 so that it cannot vanish while the page migration occurs. 69 It also prevents the swapper or other scans to encounter 70 the page. 71 72 2. We need to have a function of type new_page_t that can be 73 passed to migrate_pages(). This function should figure out 74 how to allocate the correct new page given the old page. 75 76 3. The migrate_pages() function is called which attempts 77 to do the migration. It will call the function to allocate 78 the new page for each page that is considered for 79 moving. 80 81 B. How migrate_pages() works 82 ---------------------------- 83 84 migrate_pages() does several passes over its list of pages. A page is moved 85 if all references to a page are removable at the time. The page has 86 already been removed from the LRU via isolate_lru_page() and the refcount 87 is increased so that the page cannot be freed while page migration occurs. 88 89 Steps: 90 91 1. Lock the page to be migrated 92 93 2. Insure that writeback is complete. 94 95 3. Lock the new page that we want to move to. It is locked so that accesses to 96 this (not yet uptodate) page immediately lock while the move is in progress. 97 98 4. All the page table references to the page are converted to migration 99 entries. This decreases the mapcount of a page. If the resulting 100 mapcount is not zero then we do not migrate the page. All user space 101 processes that attempt to access the page will now wait on the page lock. 102 103 5. The radix tree lock is taken. This will cause all processes trying 104 to access the page via the mapping to block on the radix tree spinlock. 105 106 6. The refcount of the page is examined and we back out if references remain 107 otherwise we know that we are the only one referencing this page. 108 109 7. The radix tree is checked and if it does not contain the pointer to this 110 page then we back out because someone else modified the radix tree. 111 112 8. The new page is prepped with some settings from the old page so that 113 accesses to the new page will discover a page with the correct settings. 114 115 9. The radix tree is changed to point to the new page. 116 117 10. The reference count of the old page is dropped because the radix tree 118 reference is gone. A reference to the new page is established because 119 the new page is referenced to by the radix tree. 120 121 11. The radix tree lock is dropped. With that lookups in the mapping 122 become possible again. Processes will move from spinning on the tree_lock 123 to sleeping on the locked new page. 124 125 12. The page contents are copied to the new page. 126 127 13. The remaining page flags are copied to the new page. 128 129 14. The old page flags are cleared to indicate that the page does 130 not provide any information anymore. 131 132 15. Queued up writeback on the new page is triggered. 133 134 16. If migration entries were page then replace them with real ptes. Doing 135 so will enable access for user space processes not already waiting for 136 the page lock. 137 138 19. The page locks are dropped from the old and new page. 139 Processes waiting on the page lock will redo their page faults 140 and will reach the new page. 141 142 20. The new page is moved to the LRU and can be scanned by the swapper 143 etc again. 144 145 Christoph Lameter, May 8, 2006.