1============================ 2Transparent Hugepage Support 3============================ 4 5Objective 6========= 7 8Performance critical computing applications dealing with large memory 9working sets are already running on top of libhugetlbfs and in turn 10hugetlbfs. Transparent HugePage Support (THP) is an alternative mean of 11using huge pages for the backing of virtual memory with huge pages 12that supports the automatic promotion and demotion of page sizes and 13without the shortcomings of hugetlbfs. 14 15Currently THP only works for anonymous memory mappings and tmpfs/shmem. 16But in the future it can expand to other filesystems. 17 18.. note:: 19 in the examples below we presume that the basic page size is 4K and 20 the huge page size is 2M, although the actual numbers may vary 21 depending on the CPU architecture. 22 23The reason applications are running faster is because of two 24factors. The first factor is almost completely irrelevant and it's not 25of significant interest because it'll also have the downside of 26requiring larger clear-page copy-page in page faults which is a 27potentially negative effect. The first factor consists in taking a 28single page fault for each 2M virtual region touched by userland (so 29reducing the enter/exit kernel frequency by a 512 times factor). This 30only matters the first time the memory is accessed for the lifetime of 31a memory mapping. The second long lasting and much more important 32factor will affect all subsequent accesses to the memory for the whole 33runtime of the application. The second factor consist of two 34components: 35 361) the TLB miss will run faster (especially with virtualization using 37 nested pagetables but almost always also on bare metal without 38 virtualization) 39 402) a single TLB entry will be mapping a much larger amount of virtual 41 memory in turn reducing the number of TLB misses. With 42 virtualization and nested pagetables the TLB can be mapped of 43 larger size only if both KVM and the Linux guest are using 44 hugepages but a significant speedup already happens if only one of 45 the two is using hugepages just because of the fact the TLB miss is 46 going to run faster. 47 48Modern kernels support "multi-size THP" (mTHP), which introduces the 49ability to allocate memory in blocks that are bigger than a base page 50but smaller than traditional PMD-size (as described above), in 51increments of a power-of-2 number of pages. mTHP can back anonymous 52memory (for example 16K, 32K, 64K, etc). These THPs continue to be 53PTE-mapped, but in many cases can still provide similar benefits to 54those outlined above: Page faults are significantly reduced (by a 55factor of e.g. 4, 8, 16, etc), but latency spikes are much less 56prominent because the size of each page isn't as huge as the PMD-sized 57variant and there is less memory to clear in each page fault. Some 58architectures also employ TLB compression mechanisms to squeeze more 59entries in when a set of PTEs are virtually and physically contiguous 60and approporiately aligned. In this case, TLB misses will occur less 61often. 62 63THP can be enabled system wide or restricted to certain tasks or even 64memory ranges inside task's address space. Unless THP is completely 65disabled, there is ``khugepaged`` daemon that scans memory and 66collapses sequences of basic pages into PMD-sized huge pages. 67 68The THP behaviour is controlled via :ref:`sysfs <thp_sysfs>` 69interface and using madvise(2) and prctl(2) system calls. 70 71Transparent Hugepage Support maximizes the usefulness of free memory 72if compared to the reservation approach of hugetlbfs by allowing all 73unused memory to be used as cache or other movable (or even unmovable 74entities). It doesn't require reservation to prevent hugepage 75allocation failures to be noticeable from userland. It allows paging 76and all other advanced VM features to be available on the 77hugepages. It requires no modifications for applications to take 78advantage of it. 79 80Applications however can be further optimized to take advantage of 81this feature, like for example they've been optimized before to avoid 82a flood of mmap system calls for every malloc(4k). Optimizing userland 83is by far not mandatory and khugepaged already can take care of long 84lived page allocations even for hugepage unaware applications that 85deals with large amounts of memory. 86 87In certain cases when hugepages are enabled system wide, application 88may end up allocating more memory resources. An application may mmap a 89large region but only touch 1 byte of it, in that case a 2M page might 90be allocated instead of a 4k page for no good. This is why it's 91possible to disable hugepages system-wide and to only have them inside 92MADV_HUGEPAGE madvise regions. 93 94Embedded systems should enable hugepages only inside madvise regions 95to eliminate any risk of wasting any precious byte of memory and to 96only run faster. 97 98Applications that gets a lot of benefit from hugepages and that don't 99risk to lose memory by using hugepages, should use 100madvise(MADV_HUGEPAGE) on their critical mmapped regions. 101 102.. _thp_sysfs: 103 104sysfs 105===== 106 107Global THP controls 108------------------- 109 110Transparent Hugepage Support for anonymous memory can be disabled 111(mostly for debugging purposes) or only enabled inside MADV_HUGEPAGE 112regions (to avoid the risk of consuming more memory resources) or enabled 113system wide. This can be achieved per-supported-THP-size with one of:: 114 115 echo always >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled 116 echo madvise >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled 117 echo never >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled 118 119where <size> is the hugepage size being addressed, the available sizes 120for which vary by system. 121 122.. note:: Setting "never" in all sysfs THP controls does **not** disable 123 Transparent Huge Pages globally. This is because ``madvise(..., 124 MADV_COLLAPSE)`` ignores these settings and collapses ranges to 125 PMD-sized huge pages unconditionally. 126 127For example:: 128 129 echo always >/sys/kernel/mm/transparent_hugepage/hugepages-2048kB/enabled 130 131Alternatively it is possible to specify that a given hugepage size 132will inherit the top-level "enabled" value:: 133 134 echo inherit >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled 135 136For example:: 137 138 echo inherit >/sys/kernel/mm/transparent_hugepage/hugepages-2048kB/enabled 139 140The top-level setting (for use with "inherit") can be set by issuing 141one of the following commands:: 142 143 echo always >/sys/kernel/mm/transparent_hugepage/enabled 144 echo madvise >/sys/kernel/mm/transparent_hugepage/enabled 145 echo never >/sys/kernel/mm/transparent_hugepage/enabled 146 147By default, PMD-sized hugepages have enabled="inherit" and all other 148hugepage sizes have enabled="never". If enabling multiple hugepage 149sizes, the kernel will select the most appropriate enabled size for a 150given allocation. 151 152It's also possible to limit defrag efforts in the VM to generate 153anonymous hugepages in case they're not immediately free to madvise 154regions or to never try to defrag memory and simply fallback to regular 155pages unless hugepages are immediately available. Clearly if we spend CPU 156time to defrag memory, we would expect to gain even more by the fact we 157use hugepages later instead of regular pages. This isn't always 158guaranteed, but it may be more likely in case the allocation is for a 159MADV_HUGEPAGE region. 160 161:: 162 163 echo always >/sys/kernel/mm/transparent_hugepage/defrag 164 echo defer >/sys/kernel/mm/transparent_hugepage/defrag 165 echo defer+madvise >/sys/kernel/mm/transparent_hugepage/defrag 166 echo madvise >/sys/kernel/mm/transparent_hugepage/defrag 167 echo never >/sys/kernel/mm/transparent_hugepage/defrag 168 169always 170 means that an application requesting THP will stall on 171 allocation failure and directly reclaim pages and compact 172 memory in an effort to allocate a THP immediately. This may be 173 desirable for virtual machines that benefit heavily from THP 174 use and are willing to delay the VM start to utilise them. 175 176defer 177 means that an application will wake kswapd in the background 178 to reclaim pages and wake kcompactd to compact memory so that 179 THP is available in the near future. It's the responsibility 180 of khugepaged to then install the THP pages later. 181 182defer+madvise 183 will enter direct reclaim and compaction like ``always``, but 184 only for regions that have used madvise(MADV_HUGEPAGE); all 185 other regions will wake kswapd in the background to reclaim 186 pages and wake kcompactd to compact memory so that THP is 187 available in the near future. 188 189madvise 190 will enter direct reclaim like ``always`` but only for regions 191 that are have used madvise(MADV_HUGEPAGE). This is the default 192 behaviour. 193 194never 195 should be self-explanatory. Note that ``madvise(..., 196 MADV_COLLAPSE)`` can still cause transparent huge pages to be 197 obtained even if this mode is specified everywhere. 198 199By default kernel tries to use huge, PMD-mappable zero page on read 200page fault to anonymous mapping. It's possible to disable huge zero 201page by writing 0 or enable it back by writing 1:: 202 203 echo 0 >/sys/kernel/mm/transparent_hugepage/use_zero_page 204 echo 1 >/sys/kernel/mm/transparent_hugepage/use_zero_page 205 206Some userspace (such as a test program, or an optimized memory 207allocation library) may want to know the size (in bytes) of a 208PMD-mappable transparent hugepage:: 209 210 cat /sys/kernel/mm/transparent_hugepage/hpage_pmd_size 211 212All THPs at fault and collapse time will be added to _deferred_list, 213and will therefore be split under memory presure if they are considered 214"underused". A THP is underused if the number of zero-filled pages in 215the THP is above max_ptes_none (see below). It is possible to disable 216this behaviour by writing 0 to shrink_underused, and enable it by writing 2171 to it:: 218 219 echo 0 > /sys/kernel/mm/transparent_hugepage/shrink_underused 220 echo 1 > /sys/kernel/mm/transparent_hugepage/shrink_underused 221 222khugepaged will be automatically started when PMD-sized THP is enabled 223(either of the per-size anon control or the top-level control are set 224to "always" or "madvise"), and it'll be automatically shutdown when 225PMD-sized THP is disabled (when both the per-size anon control and the 226top-level control are "never") 227 228process THP controls 229-------------------- 230 231A process can control its own THP behaviour using the ``PR_SET_THP_DISABLE`` 232and ``PR_GET_THP_DISABLE`` pair of prctl(2) calls. The THP behaviour set using 233``PR_SET_THP_DISABLE`` is inherited across fork(2) and execve(2). These calls 234support the following arguments:: 235 236 prctl(PR_SET_THP_DISABLE, 1, 0, 0, 0): 237 This will disable THPs completely for the process, irrespective 238 of global THP controls or madvise(..., MADV_COLLAPSE) being used. 239 240 prctl(PR_SET_THP_DISABLE, 1, PR_THP_DISABLE_EXCEPT_ADVISED, 0, 0): 241 This will disable THPs for the process except when the usage of THPs is 242 advised. Consequently, THPs will only be used when: 243 - Global THP controls are set to "always" or "madvise" and 244 madvise(..., MADV_HUGEPAGE) or madvise(..., MADV_COLLAPSE) is used. 245 - Global THP controls are set to "never" and madvise(..., MADV_COLLAPSE) 246 is used. This is the same behavior as if THPs would not be disabled on 247 a process level. 248 Note that MADV_COLLAPSE is currently always rejected if 249 madvise(..., MADV_NOHUGEPAGE) is set on an area. 250 251 prctl(PR_SET_THP_DISABLE, 0, 0, 0, 0): 252 This will re-enable THPs for the process, as if they were never disabled. 253 Whether THPs will actually be used depends on global THP controls and 254 madvise() calls. 255 256 prctl(PR_GET_THP_DISABLE, 0, 0, 0, 0): 257 This returns a value whose bits indicate how THP-disable is configured: 258 Bits 259 1 0 Value Description 260 |0|0| 0 No THP-disable behaviour specified. 261 |0|1| 1 THP is entirely disabled for this process. 262 |1|1| 3 THP-except-advised mode is set for this process. 263 264Khugepaged controls 265------------------- 266 267.. note:: 268 khugepaged currently only searches for opportunities to collapse to 269 PMD-sized THP and no attempt is made to collapse to other THP 270 sizes. 271 272khugepaged runs usually at low frequency so while one may not want to 273invoke defrag algorithms synchronously during the page faults, it 274should be worth invoking defrag at least in khugepaged. However it's 275also possible to disable defrag in khugepaged by writing 0 or enable 276defrag in khugepaged by writing 1:: 277 278 echo 0 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag 279 echo 1 >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag 280 281You can also control how many pages khugepaged should scan at each 282pass:: 283 284 /sys/kernel/mm/transparent_hugepage/khugepaged/pages_to_scan 285 286and how many milliseconds to wait in khugepaged between each pass (you 287can set this to 0 to run khugepaged at 100% utilization of one core):: 288 289 /sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs 290 291and how many milliseconds to wait in khugepaged if there's an hugepage 292allocation failure to throttle the next allocation attempt:: 293 294 /sys/kernel/mm/transparent_hugepage/khugepaged/alloc_sleep_millisecs 295 296The khugepaged progress can be seen in the number of pages collapsed (note 297that this counter may not be an exact count of the number of pages 298collapsed, since "collapsed" could mean multiple things: (1) A PTE mapping 299being replaced by a PMD mapping, or (2) All 4K physical pages replaced by 300one 2M hugepage. Each may happen independently, or together, depending on 301the type of memory and the failures that occur. As such, this value should 302be interpreted roughly as a sign of progress, and counters in /proc/vmstat 303consulted for more accurate accounting):: 304 305 /sys/kernel/mm/transparent_hugepage/khugepaged/pages_collapsed 306 307for each pass:: 308 309 /sys/kernel/mm/transparent_hugepage/khugepaged/full_scans 310 311``max_ptes_none`` specifies how many extra small pages (that are 312not already mapped) can be allocated when collapsing a group 313of small pages into one large page:: 314 315 /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none 316 317A higher value leads to use additional memory for programs. 318A lower value leads to gain less thp performance. Value of 319max_ptes_none can waste cpu time very little, you can 320ignore it. 321 322``max_ptes_swap`` specifies how many pages can be brought in from 323swap when collapsing a group of pages into a transparent huge page:: 324 325 /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_swap 326 327A higher value can cause excessive swap IO and waste 328memory. A lower value can prevent THPs from being 329collapsed, resulting fewer pages being collapsed into 330THPs, and lower memory access performance. 331 332``max_ptes_shared`` specifies how many pages can be shared across multiple 333processes. khugepaged might treat pages of THPs as shared if any page of 334that THP is shared. Exceeding the number would block the collapse:: 335 336 /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_shared 337 338A higher value may increase memory footprint for some workloads. 339 340Boot parameters 341=============== 342 343You can change the sysfs boot time default for the top-level "enabled" 344control by passing the parameter ``transparent_hugepage=always`` or 345``transparent_hugepage=madvise`` or ``transparent_hugepage=never`` to the 346kernel command line. 347 348Alternatively, each supported anonymous THP size can be controlled by 349passing ``thp_anon=<size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state>``, 350where ``<size>`` is the THP size (must be a power of 2 of PAGE_SIZE and 351supported anonymous THP) and ``<state>`` is one of ``always``, ``madvise``, 352``never`` or ``inherit``. 353 354For example, the following will set 16K, 32K, 64K THP to ``always``, 355set 128K, 512K to ``inherit``, set 256K to ``madvise`` and 1M, 2M 356to ``never``:: 357 358 thp_anon=16K-64K:always;128K,512K:inherit;256K:madvise;1M-2M:never 359 360``thp_anon=`` may be specified multiple times to configure all THP sizes as 361required. If ``thp_anon=`` is specified at least once, any anon THP sizes 362not explicitly configured on the command line are implicitly set to 363``never``. 364 365``transparent_hugepage`` setting only affects the global toggle. If 366``thp_anon`` is not specified, PMD_ORDER THP will default to ``inherit``. 367However, if a valid ``thp_anon`` setting is provided by the user, the 368PMD_ORDER THP policy will be overridden. If the policy for PMD_ORDER 369is not defined within a valid ``thp_anon``, its policy will default to 370``never``. 371 372Similarly to ``transparent_hugepage``, you can control the hugepage 373allocation policy for the internal shmem mount by using the kernel parameter 374``transparent_hugepage_shmem=<policy>``, where ``<policy>`` is one of the 375seven valid policies for shmem (``always``, ``within_size``, ``advise``, 376``never``, ``deny``, and ``force``). 377 378Similarly to ``transparent_hugepage_shmem``, you can control the default 379hugepage allocation policy for the tmpfs mount by using the kernel parameter 380``transparent_hugepage_tmpfs=<policy>``, where ``<policy>`` is one of the 381four valid policies for tmpfs (``always``, ``within_size``, ``advise``, 382``never``). The tmpfs mount default policy is ``never``. 383 384Additionally, Kconfig options are available to set the default hugepage 385policies for shmem (``CONFIG_TRANSPARENT_HUGEPAGE_SHMEM_HUGE_*``) and tmpfs 386(``CONFIG_TRANSPARENT_HUGEPAGE_TMPFS_HUGE_*``) at build time. Refer to the 387Kconfig help for more details. 388 389In the same manner as ``thp_anon`` controls each supported anonymous THP 390size, ``thp_shmem`` controls each supported shmem THP size. ``thp_shmem`` 391has the same format as ``thp_anon``, but also supports the policy 392``within_size``. 393 394``thp_shmem=`` may be specified multiple times to configure all THP sizes 395as required. If ``thp_shmem=`` is specified at least once, any shmem THP 396sizes not explicitly configured on the command line are implicitly set to 397``never``. 398 399``transparent_hugepage_shmem`` setting only affects the global toggle. If 400``thp_shmem`` is not specified, PMD_ORDER hugepage will default to 401``inherit``. However, if a valid ``thp_shmem`` setting is provided by the 402user, the PMD_ORDER hugepage policy will be overridden. If the policy for 403PMD_ORDER is not defined within a valid ``thp_shmem``, its policy will 404default to ``never``. 405 406Hugepages in tmpfs/shmem 407======================== 408 409Traditionally, tmpfs only supported a single huge page size ("PMD"). Today, 410it also supports smaller sizes just like anonymous memory, often referred 411to as "multi-size THP" (mTHP). Huge pages of any size are commonly 412represented in the kernel as "large folios". 413 414While there is fine control over the huge page sizes to use for the internal 415shmem mount (see below), ordinary tmpfs mounts will make use of all available 416huge page sizes without any control over the exact sizes, behaving more like 417other file systems. 418 419tmpfs mounts 420------------ 421 422The THP allocation policy for tmpfs mounts can be adjusted using the mount 423option: ``huge=``. It can have following values: 424 425always 426 Attempt to allocate huge pages every time we need a new page; 427 Always try PMD-sized huge pages first, and fall back to smaller-sized 428 huge pages if the PMD-sized huge page allocation fails; 429 430never 431 Do not allocate huge pages. Note that ``madvise(..., MADV_COLLAPSE)`` 432 can still cause transparent huge pages to be obtained even if this mode 433 is specified everywhere; 434 435within_size 436 Only allocate huge page if it will be fully within i_size; 437 Always try PMD-sized huge pages first, and fall back to smaller-sized 438 huge pages if the PMD-sized huge page allocation fails; 439 Also respect madvise() hints; 440 441advise 442 Only allocate huge pages if requested with madvise(); 443 444Remember, that the kernel may use huge pages of all available sizes, and 445that no fine control as for the internal tmpfs mount is available. 446 447The default policy in the past was ``never``, but it can now be adjusted 448using the kernel parameter ``transparent_hugepage_tmpfs=<policy>``. 449 450``mount -o remount,huge= /mountpoint`` works fine after mount: remounting 451``huge=never`` will not attempt to break up huge pages at all, just stop more 452from being allocated. 453 454In addition to policies listed above, the sysfs knob 455/sys/kernel/mm/transparent_hugepage/shmem_enabled will affect the 456allocation policy of tmpfs mounts, when set to the following values: 457 458deny 459 For use in emergencies, to force the huge option off from 460 all mounts; 461force 462 Force the huge option on for all - very useful for testing; 463 464shmem / internal tmpfs 465---------------------- 466The mount internal tmpfs mount is used for SysV SHM, memfds, shared anonymous 467mmaps (of /dev/zero or MAP_ANONYMOUS), GPU drivers' DRM objects, Ashmem. 468 469To control the THP allocation policy for this internal tmpfs mount, the 470sysfs knob /sys/kernel/mm/transparent_hugepage/shmem_enabled and the knobs 471per THP size in 472'/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/shmem_enabled' 473can be used. 474 475The global knob has the same semantics as the ``huge=`` mount options 476for tmpfs mounts, except that the different huge page sizes can be controlled 477individually, and will only use the setting of the global knob when the 478per-size knob is set to 'inherit'. 479 480The options 'force' and 'deny' are dropped for the individual sizes, which 481are rather testing artifacts from the old ages. 482 483always 484 Attempt to allocate <size> huge pages every time we need a new page; 485 486inherit 487 Inherit the top-level "shmem_enabled" value. By default, PMD-sized hugepages 488 have enabled="inherit" and all other hugepage sizes have enabled="never"; 489 490never 491 Do not allocate <size> huge pages. Note that ``madvise(..., 492 MADV_COLLAPSE)`` can still cause transparent huge pages to be obtained 493 even if this mode is specified everywhere; 494 495within_size 496 Only allocate <size> huge page if it will be fully within i_size. 497 Also respect madvise() hints; 498 499advise 500 Only allocate <size> huge pages if requested with madvise(); 501 502Need of application restart 503=========================== 504 505The transparent_hugepage/enabled and 506transparent_hugepage/hugepages-<size>kB/enabled values and tmpfs mount 507option only affect future behavior. So to make them effective you need 508to restart any application that could have been using hugepages. This 509also applies to the regions registered in khugepaged. 510 511Monitoring usage 512================ 513 514The number of PMD-sized anonymous transparent huge pages currently used by the 515system is available by reading the AnonHugePages field in ``/proc/meminfo``. 516To identify what applications are using PMD-sized anonymous transparent huge 517pages, it is necessary to read ``/proc/PID/smaps`` and count the AnonHugePages 518fields for each mapping. (Note that AnonHugePages only applies to traditional 519PMD-sized THP for historical reasons and should have been called 520AnonHugePmdMapped). 521 522The number of file transparent huge pages mapped to userspace is available 523by reading ShmemPmdMapped and ShmemHugePages fields in ``/proc/meminfo``. 524To identify what applications are mapping file transparent huge pages, it 525is necessary to read ``/proc/PID/smaps`` and count the FilePmdMapped fields 526for each mapping. 527 528Note that reading the smaps file is expensive and reading it 529frequently will incur overhead. 530 531There are a number of counters in ``/proc/vmstat`` that may be used to 532monitor how successfully the system is providing huge pages for use. 533 534thp_fault_alloc 535 is incremented every time a huge page is successfully 536 allocated and charged to handle a page fault. 537 538thp_collapse_alloc 539 is incremented by khugepaged when it has found 540 a range of pages to collapse into one huge page and has 541 successfully allocated a new huge page to store the data. 542 543thp_fault_fallback 544 is incremented if a page fault fails to allocate or charge 545 a huge page and instead falls back to using small pages. 546 547thp_fault_fallback_charge 548 is incremented if a page fault fails to charge a huge page and 549 instead falls back to using small pages even though the 550 allocation was successful. 551 552thp_collapse_alloc_failed 553 is incremented if khugepaged found a range 554 of pages that should be collapsed into one huge page but failed 555 the allocation. 556 557thp_file_alloc 558 is incremented every time a shmem huge page is successfully 559 allocated (Note that despite being named after "file", the counter 560 measures only shmem). 561 562thp_file_fallback 563 is incremented if a shmem huge page is attempted to be allocated 564 but fails and instead falls back to using small pages. (Note that 565 despite being named after "file", the counter measures only shmem). 566 567thp_file_fallback_charge 568 is incremented if a shmem huge page cannot be charged and instead 569 falls back to using small pages even though the allocation was 570 successful. (Note that despite being named after "file", the 571 counter measures only shmem). 572 573thp_file_mapped 574 is incremented every time a file or shmem huge page is mapped into 575 user address space. 576 577thp_split_page 578 is incremented every time a huge page is split into base 579 pages. This can happen for a variety of reasons but a common 580 reason is that a huge page is old and is being reclaimed. 581 This action implies splitting all PMD the page mapped with. 582 583thp_split_page_failed 584 is incremented if kernel fails to split huge 585 page. This can happen if the page was pinned by somebody. 586 587thp_deferred_split_page 588 is incremented when a huge page is put onto split 589 queue. This happens when a huge page is partially unmapped and 590 splitting it would free up some memory. Pages on split queue are 591 going to be split under memory pressure. 592 593thp_underused_split_page 594 is incremented when a huge page on the split queue was split 595 because it was underused. A THP is underused if the number of 596 zero pages in the THP is above a certain threshold 597 (/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none). 598 599thp_split_pmd 600 is incremented every time a PMD split into table of PTEs. 601 This can happen, for instance, when application calls mprotect() or 602 munmap() on part of huge page. It doesn't split huge page, only 603 page table entry. 604 605thp_zero_page_alloc 606 is incremented every time a huge zero page used for thp is 607 successfully allocated. Note, it doesn't count every map of 608 the huge zero page, only its allocation. 609 610thp_zero_page_alloc_failed 611 is incremented if kernel fails to allocate 612 huge zero page and falls back to using small pages. 613 614thp_swpout 615 is incremented every time a huge page is swapout in one 616 piece without splitting. 617 618thp_swpout_fallback 619 is incremented if a huge page has to be split before swapout. 620 Usually because failed to allocate some continuous swap space 621 for the huge page. 622 623In /sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/stats, There are 624also individual counters for each huge page size, which can be utilized to 625monitor the system's effectiveness in providing huge pages for usage. Each 626counter has its own corresponding file. 627 628anon_fault_alloc 629 is incremented every time a huge page is successfully 630 allocated and charged to handle a page fault. 631 632anon_fault_fallback 633 is incremented if a page fault fails to allocate or charge 634 a huge page and instead falls back to using huge pages with 635 lower orders or small pages. 636 637anon_fault_fallback_charge 638 is incremented if a page fault fails to charge a huge page and 639 instead falls back to using huge pages with lower orders or 640 small pages even though the allocation was successful. 641 642zswpout 643 is incremented every time a huge page is swapped out to zswap in one 644 piece without splitting. 645 646swpin 647 is incremented every time a huge page is swapped in from a non-zswap 648 swap device in one piece. 649 650swpin_fallback 651 is incremented if swapin fails to allocate or charge a huge page 652 and instead falls back to using huge pages with lower orders or 653 small pages. 654 655swpin_fallback_charge 656 is incremented if swapin fails to charge a huge page and instead 657 falls back to using huge pages with lower orders or small pages 658 even though the allocation was successful. 659 660swpout 661 is incremented every time a huge page is swapped out to a non-zswap 662 swap device in one piece without splitting. 663 664swpout_fallback 665 is incremented if a huge page has to be split before swapout. 666 Usually because failed to allocate some continuous swap space 667 for the huge page. 668 669shmem_alloc 670 is incremented every time a shmem huge page is successfully 671 allocated. 672 673shmem_fallback 674 is incremented if a shmem huge page is attempted to be allocated 675 but fails and instead falls back to using small pages. 676 677shmem_fallback_charge 678 is incremented if a shmem huge page cannot be charged and instead 679 falls back to using small pages even though the allocation was 680 successful. 681 682split 683 is incremented every time a huge page is successfully split into 684 smaller orders. This can happen for a variety of reasons but a 685 common reason is that a huge page is old and is being reclaimed. 686 687split_failed 688 is incremented if kernel fails to split huge 689 page. This can happen if the page was pinned by somebody. 690 691split_deferred 692 is incremented when a huge page is put onto split queue. 693 This happens when a huge page is partially unmapped and splitting 694 it would free up some memory. Pages on split queue are going to 695 be split under memory pressure, if splitting is possible. 696 697nr_anon 698 the number of anonymous THP we have in the whole system. These THPs 699 might be currently entirely mapped or have partially unmapped/unused 700 subpages. 701 702nr_anon_partially_mapped 703 the number of anonymous THP which are likely partially mapped, possibly 704 wasting memory, and have been queued for deferred memory reclamation. 705 Note that in corner some cases (e.g., failed migration), we might detect 706 an anonymous THP as "partially mapped" and count it here, even though it 707 is not actually partially mapped anymore. 708 709As the system ages, allocating huge pages may be expensive as the 710system uses memory compaction to copy data around memory to free a 711huge page for use. There are some counters in ``/proc/vmstat`` to help 712monitor this overhead. 713 714compact_stall 715 is incremented every time a process stalls to run 716 memory compaction so that a huge page is free for use. 717 718compact_success 719 is incremented if the system compacted memory and 720 freed a huge page for use. 721 722compact_fail 723 is incremented if the system tries to compact memory 724 but failed. 725 726It is possible to establish how long the stalls were using the function 727tracer to record how long was spent in __alloc_pages() and 728using the mm_page_alloc tracepoint to identify which allocations were 729for huge pages. 730 731Optimizing the applications 732=========================== 733 734To be guaranteed that the kernel will map a THP immediately in any 735memory region, the mmap region has to be hugepage naturally 736aligned. posix_memalign() can provide that guarantee. 737 738Hugetlbfs 739========= 740 741You can use hugetlbfs on a kernel that has transparent hugepage 742support enabled just fine as always. No difference can be noted in 743hugetlbfs other than there will be less overall fragmentation. All 744usual features belonging to hugetlbfs are preserved and 745unaffected. libhugetlbfs will also work fine as usual. 746