linux/Documentation/memory-hotplug.txt
<<
>>
Prefs
   1==============
   2Memory Hotplug
   3==============
   4
   5Created:                                        Jul 28 2007
   6Add 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
  121. Introduction
  13  1.1 purpose of memory hotplug
  14  1.2. Phases of memory hotplug
  15  1.3. Unit of Memory online/offline operation
  162. Kernel Configuration
  173. sysfs files for memory hotplug
  184. Physical memory hot-add phase
  19  4.1 Hardware(Firmware) Support
  20  4.2 Notify memory hot-add event by hand
  215. Logical Memory hot-add phase
  22  5.1. State of memory
  23  5.2. How to online memory
  246. Logical memory remove
  25  6.1 Memory offline and ZONE_MOVABLE
  26  6.2. How to offline memory
  277. Physical memory remove
  288. Memory hotplug event notifier
  299. Future Work List
  30
  31Note(1): x86_64's has special implementation for memory hotplug.
  32         This text does not describe it.
  33Note(2): This text assumes that sysfs is mounted at /sys.
  34
  35
  36---------------
  371. Introduction
  38---------------
  39
  401.1 purpose of memory hotplug
  41------------
  42Memory Hotplug allows users to increase/decrease the amount of memory.
  43Generally, there are two purposes.
  44
  45(A) For changing the amount of memory.
  46    This is to allow a feature like capacity on demand.
  47(B) For installing/removing DIMMs or NUMA-nodes physically.
  48    This is to exchange DIMMs/NUMA-nodes, reduce power consumption, etc.
  49
  50(A) is required by highly virtualized environments and (B) is required by
  51hardware which supports memory power management.
  52
  53Linux memory hotplug is designed for both purpose.
  54
  55
  561.2. Phases of memory hotplug
  57---------------
  58There are 2 phases in Memory Hotplug.
  59  1) Physical Memory Hotplug phase
  60  2) Logical Memory Hotplug phase.
  61
  62The First phase is to communicate hardware/firmware and make/erase
  63environment for hotplugged memory. Basically, this phase is necessary
  64for the purpose (B), but this is good phase for communication between
  65highly virtualized environments too.
  66
  67When memory is hotplugged, the kernel recognizes new memory, makes new memory
  68management tables, and makes sysfs files for new memory's operation.
  69
  70If firmware supports notification of connection of new memory to OS,
  71this phase is triggered automatically. ACPI can notify this event. If not,
  72"probe" operation by system administration is used instead.
  73(see Section 4.).
  74
  75Logical Memory Hotplug phase is to change memory state into
  76available/unavailable for users. Amount of memory from user's view is
  77changed by this phase. The kernel makes all memory in it as free pages
  78when a memory range is available.
  79
  80In this document, this phase is described as online/offline.
  81
  82Logical Memory Hotplug phase is triggered by write of sysfs file by system
  83administrator. For the hot-add case, it must be executed after Physical Hotplug
  84phase by hand.
  85(However, if you writes udev's hotplug scripts for memory hotplug, these
  86 phases can be execute in seamless way.)
  87
  88
  891.3. Unit of Memory online/offline operation
  90------------
  91Memory hotplug uses SPARSEMEM memory model. SPARSEMEM divides the whole memory
  92into chunks of the same size. The chunk is called a "section". The size of
  93a section is architecture dependent. For example, power uses 16MiB, ia64 uses
  941GiB. The unit of online/offline operation is "one section". (see Section 3.)
  95
  96To determine the size of sections, please read this file:
  97
  98/sys/devices/system/memory/block_size_bytes
  99
 100This file shows the size of sections in byte.
 101
 102-----------------------
 1032. Kernel Configuration
 104-----------------------
 105To use memory hotplug feature, kernel must be compiled with following
 106config options.
 107
 108- For all memory hotplug
 109    Memory model -> Sparse Memory  (CONFIG_SPARSEMEM)
 110    Allow for memory hot-add       (CONFIG_MEMORY_HOTPLUG)
 111
 112- To enable memory removal, the followings are also necessary
 113    Allow for memory hot remove    (CONFIG_MEMORY_HOTREMOVE)
 114    Page Migration                 (CONFIG_MIGRATION)
 115
 116- For ACPI memory hotplug, the followings are also necessary
 117    Memory hotplug (under ACPI Support menu) (CONFIG_ACPI_HOTPLUG_MEMORY)
 118    This option can be kernel module.
 119
 120- As a related configuration, if your box has a feature of NUMA-node hotplug
 121  via ACPI, then this option is necessary too.
 122    ACPI0004,PNP0A05 and PNP0A06 Container Driver (under ACPI Support menu)
 123    (CONFIG_ACPI_CONTAINER).
 124    This option can be kernel module too.
 125
 126--------------------------------
 1274 sysfs files for memory hotplug
 128--------------------------------
 129All sections have their device information under /sys/devices/system/memory as
 130
 131/sys/devices/system/memory/memoryXXX
 132(XXX is section id.)
 133
 134Now, XXX is defined as start_address_of_section / section_size.
 135
 136For example, assume 1GiB section size. A device for a memory starting at
 1370x100000000 is /sys/device/system/memory/memory4
 138(0x100000000 / 1Gib = 4)
 139This device covers address range [0x100000000 ... 0x140000000)
 140
 141Under each section, you can see 4 files.
 142
 143/sys/devices/system/memory/memoryXXX/phys_index
 144/sys/devices/system/memory/memoryXXX/phys_device
 145/sys/devices/system/memory/memoryXXX/state
 146/sys/devices/system/memory/memoryXXX/removable
 147
 148'phys_index' : read-only and contains section id, same as XXX.
 149'state'      : read-write
 150               at read:  contains online/offline state of memory.
 151               at write: user can specify "online", "offline" command
 152'phys_device': read-only: designed to show the name of physical memory device.
 153               This is not well implemented now.
 154'removable'  : read-only: contains an integer value indicating
 155               whether the memory section is removable or not
 156               removable.  A value of 1 indicates that the memory
 157               section is removable and a value of 0 indicates that
 158               it is not removable.
 159
 160NOTE:
 161  These directories/files appear after physical memory hotplug phase.
 162
 163If CONFIG_NUMA is enabled the memoryXXX/ directories can also be accessed
 164via symbolic links located in the /sys/devices/system/node/node* directories.
 165
 166For example:
 167/sys/devices/system/node/node0/memory9 -> ../../memory/memory9
 168
 169A backlink will also be created:
 170/sys/devices/system/memory/memory9/node0 -> ../../node/node0
 171
 172--------------------------------
 1734. Physical memory hot-add phase
 174--------------------------------
 175
 1764.1 Hardware(Firmware) Support
 177------------
 178On x86_64/ia64 platform, memory hotplug by ACPI is supported.
 179
 180In general, the firmware (ACPI) which supports memory hotplug defines
 181memory class object of _HID "PNP0C80". When a notify is asserted to PNP0C80,
 182Linux's ACPI handler does hot-add memory to the system and calls a hotplug udev
 183script. This will be done automatically.
 184
 185But scripts for memory hotplug are not contained in generic udev package(now).
 186You may have to write it by yourself or online/offline memory by hand.
 187Please see "How to online memory", "How to offline memory" in this text.
 188
 189If firmware supports NUMA-node hotplug, and defines an object _HID "ACPI0004",
 190"PNP0A05", or "PNP0A06", notification is asserted to it, and ACPI handler
 191calls hotplug code for all of objects which are defined in it.
 192If memory device is found, memory hotplug code will be called.
 193
 194
 1954.2 Notify memory hot-add event by hand
 196------------
 197In some environments, especially virtualized environment, firmware will not
 198notify memory hotplug event to the kernel. For such environment, "probe"
 199interface is supported. This interface depends on CONFIG_ARCH_MEMORY_PROBE.
 200
 201Now, CONFIG_ARCH_MEMORY_PROBE is supported only by powerpc but it does not
 202contain highly architecture codes. Please add config if you need "probe"
 203interface.
 204
 205Probe interface is located at
 206/sys/devices/system/memory/probe
 207
 208You can tell the physical address of new memory to the kernel by
 209
 210% echo start_address_of_new_memory > /sys/devices/system/memory/probe
 211
 212Then, [start_address_of_new_memory, start_address_of_new_memory + section_size)
 213memory range is hot-added. In this case, hotplug script is not called (in
 214current implementation). You'll have to online memory by yourself.
 215Please see "How to online memory" in this text.
 216
 217
 218
 219------------------------------
 2205. Logical Memory hot-add phase
 221------------------------------
 222
 2235.1. State of memory
 224------------
 225To see (online/offline) state of memory section, read 'state' file.
 226
 227% cat /sys/device/system/memory/memoryXXX/state
 228
 229
 230If the memory section is online, you'll read "online".
 231If the memory section is offline, you'll read "offline".
 232
 233
 2345.2. How to online memory
 235------------
 236Even if the memory is hot-added, it is not at ready-to-use state.
 237For using newly added memory, you have to "online" the memory section.
 238
 239For onlining, you have to write "online" to the section's state file as:
 240
 241% echo online > /sys/devices/system/memory/memoryXXX/state
 242
 243After this, section memoryXXX's state will be 'online' and the amount of
 244available memory will be increased.
 245
 246Currently, newly added memory is added as ZONE_NORMAL (for powerpc, ZONE_DMA).
 247This may be changed in future.
 248
 249
 250
 251------------------------
 2526. Logical memory remove
 253------------------------
 254
 2556.1 Memory offline and ZONE_MOVABLE
 256------------
 257Memory offlining is more complicated than memory online. Because memory offline
 258has to make the whole memory section be unused, memory offline can fail if
 259the section includes memory which cannot be freed.
 260
 261In general, memory offline can use 2 techniques.
 262
 263(1) reclaim and free all memory in the section.
 264(2) migrate all pages in the section.
 265
 266In the current implementation, Linux's memory offline uses method (2), freeing
 267all  pages in the section by page migration. But not all pages are
 268migratable. Under current Linux, migratable pages are anonymous pages and
 269page caches. For offlining a section by migration, the kernel has to guarantee
 270that the section contains only migratable pages.
 271
 272Now, a boot option for making a section which consists of migratable pages is
 273supported. By specifying "kernelcore=" or "movablecore=" boot option, you can
 274create ZONE_MOVABLE...a zone which is just used for movable pages.
 275(See also Documentation/kernel-parameters.txt)
 276
 277Assume the system has "TOTAL" amount of memory at boot time, this boot option
 278creates ZONE_MOVABLE as following.
 279
 2801) When kernelcore=YYYY boot option is used,
 281  Size of memory not for movable pages (not for offline) is YYYY.
 282  Size of memory for movable pages (for offline) is TOTAL-YYYY.
 283
 2842) When movablecore=ZZZZ boot option is used,
 285  Size of memory not for movable pages (not for offline) is TOTAL - ZZZZ.
 286  Size of memory for movable pages (for offline) is ZZZZ.
 287
 288
 289Note) Unfortunately, there is no information to show which section belongs
 290to ZONE_MOVABLE. This is TBD.
 291
 292
 2936.2. How to offline memory
 294------------
 295You can offline a section by using the same sysfs interface that was used in
 296memory onlining.
 297
 298% echo offline > /sys/devices/system/memory/memoryXXX/state
 299
 300If offline succeeds, the state of the memory section is changed to be "offline".
 301If it fails, some error core (like -EBUSY) will be returned by the kernel.
 302Even if a section does not belong to ZONE_MOVABLE, you can try to offline it.
 303If it doesn't contain 'unmovable' memory, you'll get success.
 304
 305A section under ZONE_MOVABLE is considered to be able to be offlined easily.
 306But under some busy state, it may return -EBUSY. Even if a memory section
 307cannot be offlined due to -EBUSY, you can retry offlining it and may be able to
 308offline it (or not).
 309(For example, a page is referred to by some kernel internal call and released
 310 soon.)
 311
 312Consideration:
 313Memory hotplug's design direction is to make the possibility of memory offlining
 314higher and to guarantee unplugging memory under any situation. But it needs
 315more work. Returning -EBUSY under some situation may be good because the user
 316can decide to retry more or not by himself. Currently, memory offlining code
 317does some amount of retry with 120 seconds timeout.
 318
 319-------------------------
 3207. Physical memory remove
 321-------------------------
 322Need more implementation yet....
 323 - Notification completion of remove works by OS to firmware.
 324 - Guard from remove if not yet.
 325
 326--------------------------------
 3278. Memory hotplug event notifier
 328--------------------------------
 329Memory hotplug has event notifer. There are 6 types of notification.
 330
 331MEMORY_GOING_ONLINE
 332  Generated before new memory becomes available in order to be able to
 333  prepare subsystems to handle memory. The page allocator is still unable
 334  to allocate from the new memory.
 335
 336MEMORY_CANCEL_ONLINE
 337  Generated if MEMORY_GOING_ONLINE fails.
 338
 339MEMORY_ONLINE
 340  Generated when memory has successfully brought online. The callback may
 341  allocate pages from the new memory.
 342
 343MEMORY_GOING_OFFLINE
 344  Generated to begin the process of offlining memory. Allocations are no
 345  longer possible from the memory but some of the memory to be offlined
 346  is still in use. The callback can be used to free memory known to a
 347  subsystem from the indicated memory section.
 348
 349MEMORY_CANCEL_OFFLINE
 350  Generated if MEMORY_GOING_OFFLINE fails. Memory is available again from
 351  the section that we attempted to offline.
 352
 353MEMORY_OFFLINE
 354  Generated after offlining memory is complete.
 355
 356A callback routine can be registered by
 357  hotplug_memory_notifier(callback_func, priority)
 358
 359The second argument of callback function (action) is event types of above.
 360The third argument is passed by pointer of struct memory_notify.
 361
 362struct memory_notify {
 363       unsigned long start_pfn;
 364       unsigned long nr_pages;
 365       int status_change_nid;
 366}
 367
 368start_pfn is start_pfn of online/offline memory.
 369nr_pages is # of pages of online/offline memory.
 370status_change_nid is set node id when N_HIGH_MEMORY of nodemask is (will be)
 371set/clear. It means a new(memoryless) node gets new memory by online and a
 372node loses all memory. If this is -1, then nodemask status is not changed.
 373If status_changed_nid >= 0, callback should create/discard structures for the
 374node if necessary.
 375
 376--------------
 3779. Future Work
 378--------------
 379  - allowing memory hot-add to ZONE_MOVABLE. maybe we need some switch like
 380    sysctl or new control file.
 381  - showing memory section and physical device relationship.
 382  - showing memory section is under ZONE_MOVABLE or not
 383  - test and make it better memory offlining.
 384  - support HugeTLB page migration and offlining.
 385  - memmap removing at memory offline.
 386  - physical remove memory.
 387
 388
lxr.linux.no kindly hosted by Redpill Linpro AS, provider of Linux consulting and operations services since 1995.