1============== 2Memory Hotplug 3============== 4 5:Created: Jul 28 2007 6:Updated: Add description of notifier of memory hotplug: Oct 11 2007 7 8This document is about memory hotplug including how-to-use and current status. 9Because Memory Hotplug is still under development, contents of this text will 10be changed often. 11 12.. CONTENTS 13 14 1. Introduction 15 1.1 purpose of memory hotplug 16 1.2. Phases of memory hotplug 17 1.3. Unit of Memory online/offline operation 18 2. Kernel Configuration 19 3. sysfs files for memory hotplug 20 4. Physical memory hot-add phase 21 4.1 Hardware(Firmware) Support 22 4.2 Notify memory hot-add event by hand 23 5. Logical Memory hot-add phase 24 5.1. State of memory 25 5.2. How to online memory 26 6. Logical memory remove 27 6.1 Memory offline and ZONE_MOVABLE 28 6.2. How to offline memory 29 7. Physical memory remove 30 8. Memory hotplug event notifier 31 9. Future Work List 32 33 34.. note:: 35 36 (1) x86_64's has special implementation for memory hotplug. 37 This text does not describe it. 38 (2) This text assumes that sysfs is mounted at /sys. 39 40 41Introduction 42============ 43 44purpose of memory hotplug 45------------------------- 46 47Memory Hotplug allows users to increase/decrease the amount of memory. 48Generally, there are two purposes. 49 50(A) For changing the amount of memory. 51 This is to allow a feature like capacity on demand. 52(B) For installing/removing DIMMs or NUMA-nodes physically. 53 This is to exchange DIMMs/NUMA-nodes, reduce power consumption, etc. 54 55(A) is required by highly virtualized environments and (B) is required by 56hardware which supports memory power management. 57 58Linux memory hotplug is designed for both purpose. 59 60 61Phases of memory hotplug 62------------------------ 63 64There are 2 phases in Memory Hotplug: 65 66 1) Physical Memory Hotplug phase 67 2) Logical Memory Hotplug phase. 68 69The First phase is to communicate hardware/firmware and make/erase 70environment for hotplugged memory. Basically, this phase is necessary 71for the purpose (B), but this is good phase for communication between 72highly virtualized environments too. 73 74When memory is hotplugged, the kernel recognizes new memory, makes new memory 75management tables, and makes sysfs files for new memory's operation. 76 77If firmware supports notification of connection of new memory to OS, 78this phase is triggered automatically. ACPI can notify this event. If not, 79"probe" operation by system administration is used instead. 80(see :ref:`memory_hotplug_physical_mem`). 81 82Logical Memory Hotplug phase is to change memory state into 83available/unavailable for users. Amount of memory from user's view is 84changed by this phase. The kernel makes all memory in it as free pages 85when a memory range is available. 86 87In this document, this phase is described as online/offline. 88 89Logical Memory Hotplug phase is triggered by write of sysfs file by system 90administrator. For the hot-add case, it must be executed after Physical Hotplug 91phase by hand. 92(However, if you writes udev's hotplug scripts for memory hotplug, these 93phases can be execute in seamless way.) 94 95 96Unit of Memory online/offline operation 97--------------------------------------- 98 99Memory hotplug uses SPARSEMEM memory model which allows memory to be divided 100into chunks of the same size. These chunks are called "sections". The size of 101a memory section is architecture dependent. For example, power uses 16MiB, ia64 102uses 1GiB. 103 104Memory sections are combined into chunks referred to as "memory blocks". The 105size of a memory block is architecture dependent and represents the logical 106unit upon which memory online/offline operations are to be performed. The 107default size of a memory block is the same as memory section size unless an 108architecture specifies otherwise. (see :ref:`memory_hotplug_sysfs_files`.) 109 110To determine the size (in bytes) of a memory block please read this file: 111 112/sys/devices/system/memory/block_size_bytes 113 114 115Kernel Configuration 116==================== 117 118To use memory hotplug feature, kernel must be compiled with following 119config options. 120 121- For all memory hotplug: 122 - Memory model -> Sparse Memory (CONFIG_SPARSEMEM) 123 - Allow for memory hot-add (CONFIG_MEMORY_HOTPLUG) 124 125- To enable memory removal, the following are also necessary: 126 - Allow for memory hot remove (CONFIG_MEMORY_HOTREMOVE) 127 - Page Migration (CONFIG_MIGRATION) 128 129- For ACPI memory hotplug, the following are also necessary: 130 - Memory hotplug (under ACPI Support menu) (CONFIG_ACPI_HOTPLUG_MEMORY) 131 - This option can be kernel module. 132 133- As a related configuration, if your box has a feature of NUMA-node hotplug 134 via ACPI, then this option is necessary too. 135 136 - ACPI0004,PNP0A05 and PNP0A06 Container Driver (under ACPI Support menu) 137 (CONFIG_ACPI_CONTAINER). 138 139 This option can be kernel module too. 140 141 142.. _memory_hotplug_sysfs_files: 143 144sysfs files for memory hotplug 145============================== 146 147All memory blocks have their device information in sysfs. Each memory block 148is described under /sys/devices/system/memory as: 149 150 /sys/devices/system/memory/memoryXXX 151 (XXX is the memory block id.) 152 153For the memory block covered by the sysfs directory. It is expected that all 154memory sections in this range are present and no memory holes exist in the 155range. Currently there is no way to determine if there is a memory hole, but 156the existence of one should not affect the hotplug capabilities of the memory 157block. 158 159For example, assume 1GiB memory block size. A device for a memory starting at 1600x100000000 is /sys/device/system/memory/memory4:: 161 162 (0x100000000 / 1Gib = 4) 163 164This device covers address range [0x100000000 ... 0x140000000) 165 166Under each memory block, you can see 5 files: 167 168- /sys/devices/system/memory/memoryXXX/phys_index 169- /sys/devices/system/memory/memoryXXX/phys_device 170- /sys/devices/system/memory/memoryXXX/state 171- /sys/devices/system/memory/memoryXXX/removable 172- /sys/devices/system/memory/memoryXXX/valid_zones 173 174=================== ============================================================ 175``phys_index`` read-only and contains memory block id, same as XXX. 176``state`` read-write 177 178 - at read: contains online/offline state of memory. 179 - at write: user can specify "online_kernel", 180 181 "online_movable", "online", "offline" command 182 which will be performed on all sections in the block. 183``phys_device`` read-only: designed to show the name of physical memory 184 device. This is not well implemented now. 185``removable`` read-only: contains an integer value indicating 186 whether the memory block is removable or not 187 removable. A value of 1 indicates that the memory 188 block is removable and a value of 0 indicates that 189 it is not removable. A memory block is removable only if 190 every section in the block is removable. 191``valid_zones`` read-only: designed to show which zones this memory block 192 can be onlined to. 193 194 The first column shows it`s default zone. 195 196 "memory6/valid_zones: Normal Movable" shows this memoryblock 197 can be onlined to ZONE_NORMAL by default and to ZONE_MOVABLE 198 by online_movable. 199 200 "memory7/valid_zones: Movable Normal" shows this memoryblock 201 can be onlined to ZONE_MOVABLE by default and to ZONE_NORMAL 202 by online_kernel. 203=================== ============================================================ 204 205.. note:: 206 207 These directories/files appear after physical memory hotplug phase. 208 209If CONFIG_NUMA is enabled the memoryXXX/ directories can also be accessed 210via symbolic links located in the /sys/devices/system/node/node* directories. 211 212For example: 213/sys/devices/system/node/node0/memory9 -> ../../memory/memory9 214 215A backlink will also be created: 216/sys/devices/system/memory/memory9/node0 -> ../../node/node0 217 218.. _memory_hotplug_physical_mem: 219 220Physical memory hot-add phase 221============================= 222 223Hardware(Firmware) Support 224-------------------------- 225 226On x86_64/ia64 platform, memory hotplug by ACPI is supported. 227 228In general, the firmware (ACPI) which supports memory hotplug defines 229memory class object of _HID "PNP0C80". When a notify is asserted to PNP0C80, 230Linux's ACPI handler does hot-add memory to the system and calls a hotplug udev 231script. This will be done automatically. 232 233But scripts for memory hotplug are not contained in generic udev package(now). 234You may have to write it by yourself or online/offline memory by hand. 235Please see :ref:`memory_hotplug_how_to_online_memory` and 236:ref:`memory_hotplug_how_to_offline_memory`. 237 238If firmware supports NUMA-node hotplug, and defines an object _HID "ACPI0004", 239"PNP0A05", or "PNP0A06", notification is asserted to it, and ACPI handler 240calls hotplug code for all of objects which are defined in it. 241If memory device is found, memory hotplug code will be called. 242 243 244Notify memory hot-add event by hand 245----------------------------------- 246 247On some architectures, the firmware may not notify the kernel of a memory 248hotplug event. Therefore, the memory "probe" interface is supported to 249explicitly notify the kernel. This interface depends on 250CONFIG_ARCH_MEMORY_PROBE and can be configured on powerpc, sh, and x86 251if hotplug is supported, although for x86 this should be handled by ACPI 252notification. 253 254Probe interface is located at 255/sys/devices/system/memory/probe 256 257You can tell the physical address of new memory to the kernel by:: 258 259 % echo start_address_of_new_memory > /sys/devices/system/memory/probe 260 261Then, [start_address_of_new_memory, start_address_of_new_memory + 262memory_block_size] memory range is hot-added. In this case, hotplug script is 263not called (in current implementation). You'll have to online memory by 264yourself. Please see :ref:`memory_hotplug_how_to_online_memory`. 265 266 267Logical Memory hot-add phase 268============================ 269 270State of memory 271--------------- 272 273To see (online/offline) state of a memory block, read 'state' file:: 274 275 % cat /sys/device/system/memory/memoryXXX/state 276 277 278- If the memory block is online, you'll read "online". 279- If the memory block is offline, you'll read "offline". 280 281 282.. _memory_hotplug_how_to_online_memory: 283 284How to online memory 285-------------------- 286 287When the memory is hot-added, the kernel decides whether or not to "online" 288it according to the policy which can be read from "auto_online_blocks" file:: 289 290 % cat /sys/devices/system/memory/auto_online_blocks 291 292The default depends on the CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 293option. If it is disabled the default is "offline" which means the newly added 294memory is not in a ready-to-use state and you have to "online" the newly added 295memory blocks manually. Automatic onlining can be requested by writing "online" 296to "auto_online_blocks" file:: 297 298 % echo online > /sys/devices/system/memory/auto_online_blocks 299 300This sets a global policy and impacts all memory blocks that will subsequently 301be hotplugged. Currently offline blocks keep their state. It is possible, under 302certain circumstances, that some memory blocks will be added but will fail to 303online. User space tools can check their "state" files 304(/sys/devices/system/memory/memoryXXX/state) and try to online them manually. 305 306If the automatic onlining wasn't requested, failed, or some memory block was 307offlined it is possible to change the individual block's state by writing to the 308"state" file:: 309 310 % echo online > /sys/devices/system/memory/memoryXXX/state 311 312This onlining will not change the ZONE type of the target memory block, 313If the memory block doesn't belong to any zone an appropriate kernel zone 314(usually ZONE_NORMAL) will be used unless movable_node kernel command line 315option is specified when ZONE_MOVABLE will be used. 316 317You can explicitly request to associate it with ZONE_MOVABLE by:: 318 319 % echo online_movable > /sys/devices/system/memory/memoryXXX/state 320 321.. note:: current limit: this memory block must be adjacent to ZONE_MOVABLE 322 323Or you can explicitly request a kernel zone (usually ZONE_NORMAL) by:: 324 325 % echo online_kernel > /sys/devices/system/memory/memoryXXX/state 326 327.. note:: current limit: this memory block must be adjacent to ZONE_NORMAL 328 329An explicit zone onlining can fail (e.g. when the range is already within 330and existing and incompatible zone already). 331 332After this, memory block XXX's state will be 'online' and the amount of 333available memory will be increased. 334 335This may be changed in future. 336 337 338 339Logical memory remove 340===================== 341 342Memory offline and ZONE_MOVABLE 343------------------------------- 344 345Memory offlining is more complicated than memory online. Because memory offline 346has to make the whole memory block be unused, memory offline can fail if 347the memory block includes memory which cannot be freed. 348 349In general, memory offline can use 2 techniques. 350 351(1) reclaim and free all memory in the memory block. 352(2) migrate all pages in the memory block. 353 354In the current implementation, Linux's memory offline uses method (2), freeing 355all pages in the memory block by page migration. But not all pages are 356migratable. Under current Linux, migratable pages are anonymous pages and 357page caches. For offlining a memory block by migration, the kernel has to 358guarantee that the memory block contains only migratable pages. 359 360Now, a boot option for making a memory block which consists of migratable pages 361is supported. By specifying "kernelcore=" or "movablecore=" boot option, you can 362create ZONE_MOVABLE...a zone which is just used for movable pages. 363(See also Documentation/admin-guide/kernel-parameters.rst) 364 365Assume the system has "TOTAL" amount of memory at boot time, this boot option 366creates ZONE_MOVABLE as following. 367 3681) When kernelcore=YYYY boot option is used, 369 Size of memory not for movable pages (not for offline) is YYYY. 370 Size of memory for movable pages (for offline) is TOTAL-YYYY. 371 3722) When movablecore=ZZZZ boot option is used, 373 Size of memory not for movable pages (not for offline) is TOTAL - ZZZZ. 374 Size of memory for movable pages (for offline) is ZZZZ. 375 376.. note:: 377 378 Unfortunately, there is no information to show which memory block belongs 379 to ZONE_MOVABLE. This is TBD. 380 381.. _memory_hotplug_how_to_offline_memory: 382 383How to offline memory 384--------------------- 385 386You can offline a memory block by using the same sysfs interface that was used 387in memory onlining:: 388 389 % echo offline > /sys/devices/system/memory/memoryXXX/state 390 391If offline succeeds, the state of the memory block is changed to be "offline". 392If it fails, some error core (like -EBUSY) will be returned by the kernel. 393Even if a memory block does not belong to ZONE_MOVABLE, you can try to offline 394it. If it doesn't contain 'unmovable' memory, you'll get success. 395 396A memory block under ZONE_MOVABLE is considered to be able to be offlined 397easily. But under some busy state, it may return -EBUSY. Even if a memory 398block cannot be offlined due to -EBUSY, you can retry offlining it and may be 399able to offline it (or not). (For example, a page is referred to by some kernel 400internal call and released soon.) 401 402Consideration: 403 Memory hotplug's design direction is to make the possibility of memory 404 offlining higher and to guarantee unplugging memory under any situation. But 405 it needs more work. Returning -EBUSY under some situation may be good because 406 the user can decide to retry more or not by himself. Currently, memory 407 offlining code does some amount of retry with 120 seconds timeout. 408 409Physical memory remove 410====================== 411 412Need more implementation yet.... 413 - Notification completion of remove works by OS to firmware. 414 - Guard from remove if not yet. 415 416Memory hotplug event notifier 417============================= 418 419Hotplugging events are sent to a notification queue. 420 421There are six types of notification defined in include/linux/memory.h: 422 423MEM_GOING_ONLINE 424 Generated before new memory becomes available in order to be able to 425 prepare subsystems to handle memory. The page allocator is still unable 426 to allocate from the new memory. 427 428MEM_CANCEL_ONLINE 429 Generated if MEMORY_GOING_ONLINE fails. 430 431MEM_ONLINE 432 Generated when memory has successfully brought online. The callback may 433 allocate pages from the new memory. 434 435MEM_GOING_OFFLINE 436 Generated to begin the process of offlining memory. Allocations are no 437 longer possible from the memory but some of the memory to be offlined 438 is still in use. The callback can be used to free memory known to a 439 subsystem from the indicated memory block. 440 441MEM_CANCEL_OFFLINE 442 Generated if MEMORY_GOING_OFFLINE fails. Memory is available again from 443 the memory block that we attempted to offline. 444 445MEM_OFFLINE 446 Generated after offlining memory is complete. 447 448A callback routine can be registered by calling:: 449 450 hotplug_memory_notifier(callback_func, priority) 451 452Callback functions with higher values of priority are called before callback 453functions with lower values. 454 455A callback function must have the following prototype:: 456 457 int callback_func( 458 struct notifier_block *self, unsigned long action, void *arg); 459 460The first argument of the callback function (self) is a pointer to the block 461of the notifier chain that points to the callback function itself. 462The second argument (action) is one of the event types described above. 463The third argument (arg) passes a pointer of struct memory_notify:: 464 465 struct memory_notify { 466 unsigned long start_pfn; 467 unsigned long nr_pages; 468 int status_change_nid_normal; 469 int status_change_nid_high; 470 int status_change_nid; 471 } 472 473- start_pfn is start_pfn of online/offline memory. 474- nr_pages is # of pages of online/offline memory. 475- status_change_nid_normal is set node id when N_NORMAL_MEMORY of nodemask 476 is (will be) set/clear, if this is -1, then nodemask status is not changed. 477- status_change_nid_high is set node id when N_HIGH_MEMORY of nodemask 478 is (will be) set/clear, if this is -1, then nodemask status is not changed. 479- status_change_nid is set node id when N_MEMORY of nodemask is (will be) 480 set/clear. It means a new(memoryless) node gets new memory by online and a 481 node loses all memory. If this is -1, then nodemask status is not changed. 482 483 If status_changed_nid* >= 0, callback should create/discard structures for the 484 node if necessary. 485 486The callback routine shall return one of the values 487NOTIFY_DONE, NOTIFY_OK, NOTIFY_BAD, NOTIFY_STOP 488defined in include/linux/notifier.h 489 490NOTIFY_DONE and NOTIFY_OK have no effect on the further processing. 491 492NOTIFY_BAD is used as response to the MEM_GOING_ONLINE, MEM_GOING_OFFLINE, 493MEM_ONLINE, or MEM_OFFLINE action to cancel hotplugging. It stops 494further processing of the notification queue. 495 496NOTIFY_STOP stops further processing of the notification queue. 497 498Future Work 499=========== 500 501 - allowing memory hot-add to ZONE_MOVABLE. maybe we need some switch like 502 sysctl or new control file. 503 - showing memory block and physical device relationship. 504 - test and make it better memory offlining. 505 - support HugeTLB page migration and offlining. 506 - memmap removing at memory offline. 507 - physical remove memory. 508
