]>
Commit | Line | Data |
---|---|---|
ec8f24b7 | 1 | # SPDX-License-Identifier: GPL-2.0-only |
59e0b520 CH |
2 | |
3 | menu "Memory Management options" | |
4 | ||
7b42f104 JW |
5 | # |
6 | # For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can | |
7 | # add proper SWAP support to them, in which case this can be remove. | |
8 | # | |
9 | config ARCH_NO_SWAP | |
10 | bool | |
11 | ||
b3fbd58f JW |
12 | config ZPOOL |
13 | bool | |
14 | ||
519bcb79 | 15 | menuconfig SWAP |
7b42f104 JW |
16 | bool "Support for paging of anonymous memory (swap)" |
17 | depends on MMU && BLOCK && !ARCH_NO_SWAP | |
18 | default y | |
19 | help | |
20 | This option allows you to choose whether you want to have support | |
21 | for so called swap devices or swap files in your kernel that are | |
22 | used to provide more virtual memory than the actual RAM present | |
23 | in your computer. If unsure say Y. | |
24 | ||
519bcb79 | 25 | config ZSWAP |
fcab9b44 | 26 | bool "Compressed cache for swap pages" |
b3fbd58f | 27 | depends on SWAP |
b3fbd58f | 28 | select CRYPTO |
519bcb79 JW |
29 | select ZPOOL |
30 | help | |
31 | A lightweight compressed cache for swap pages. It takes | |
32 | pages that are in the process of being swapped out and attempts to | |
33 | compress them into a dynamically allocated RAM-based memory pool. | |
34 | This can result in a significant I/O reduction on swap device and, | |
1a44131d | 35 | in the case where decompressing from RAM is faster than swap device |
519bcb79 JW |
36 | reads, can also improve workload performance. |
37 | ||
b3fbd58f JW |
38 | config ZSWAP_DEFAULT_ON |
39 | bool "Enable the compressed cache for swap pages by default" | |
40 | depends on ZSWAP | |
41 | help | |
42 | If selected, the compressed cache for swap pages will be enabled | |
43 | at boot, otherwise it will be disabled. | |
44 | ||
45 | The selection made here can be overridden by using the kernel | |
46 | command line 'zswap.enabled=' option. | |
47 | ||
b5ba474f NP |
48 | config ZSWAP_SHRINKER_DEFAULT_ON |
49 | bool "Shrink the zswap pool on memory pressure" | |
50 | depends on ZSWAP | |
51 | default n | |
52 | help | |
53 | If selected, the zswap shrinker will be enabled, and the pages | |
54 | stored in the zswap pool will become available for reclaim (i.e | |
55 | written back to the backing swap device) on memory pressure. | |
56 | ||
57 | This means that zswap writeback could happen even if the pool is | |
58 | not yet full, or the cgroup zswap limit has not been reached, | |
59 | reducing the chance that cold pages will reside in the zswap pool | |
60 | and consume memory indefinitely. | |
61 | ||
519bcb79 | 62 | choice |
b3fbd58f | 63 | prompt "Default compressor" |
519bcb79 JW |
64 | depends on ZSWAP |
65 | default ZSWAP_COMPRESSOR_DEFAULT_LZO | |
66 | help | |
67 | Selects the default compression algorithm for the compressed cache | |
68 | for swap pages. | |
69 | ||
70 | For an overview what kind of performance can be expected from | |
71 | a particular compression algorithm please refer to the benchmarks | |
72 | available at the following LWN page: | |
73 | https://lwn.net/Articles/751795/ | |
74 | ||
75 | If in doubt, select 'LZO'. | |
76 | ||
77 | The selection made here can be overridden by using the kernel | |
78 | command line 'zswap.compressor=' option. | |
79 | ||
80 | config ZSWAP_COMPRESSOR_DEFAULT_DEFLATE | |
81 | bool "Deflate" | |
82 | select CRYPTO_DEFLATE | |
83 | help | |
84 | Use the Deflate algorithm as the default compression algorithm. | |
85 | ||
86 | config ZSWAP_COMPRESSOR_DEFAULT_LZO | |
87 | bool "LZO" | |
88 | select CRYPTO_LZO | |
89 | help | |
90 | Use the LZO algorithm as the default compression algorithm. | |
91 | ||
92 | config ZSWAP_COMPRESSOR_DEFAULT_842 | |
93 | bool "842" | |
94 | select CRYPTO_842 | |
95 | help | |
96 | Use the 842 algorithm as the default compression algorithm. | |
97 | ||
98 | config ZSWAP_COMPRESSOR_DEFAULT_LZ4 | |
99 | bool "LZ4" | |
100 | select CRYPTO_LZ4 | |
101 | help | |
102 | Use the LZ4 algorithm as the default compression algorithm. | |
103 | ||
104 | config ZSWAP_COMPRESSOR_DEFAULT_LZ4HC | |
105 | bool "LZ4HC" | |
106 | select CRYPTO_LZ4HC | |
107 | help | |
108 | Use the LZ4HC algorithm as the default compression algorithm. | |
109 | ||
110 | config ZSWAP_COMPRESSOR_DEFAULT_ZSTD | |
111 | bool "zstd" | |
112 | select CRYPTO_ZSTD | |
113 | help | |
114 | Use the zstd algorithm as the default compression algorithm. | |
115 | endchoice | |
116 | ||
117 | config ZSWAP_COMPRESSOR_DEFAULT | |
118 | string | |
119 | depends on ZSWAP | |
120 | default "deflate" if ZSWAP_COMPRESSOR_DEFAULT_DEFLATE | |
121 | default "lzo" if ZSWAP_COMPRESSOR_DEFAULT_LZO | |
122 | default "842" if ZSWAP_COMPRESSOR_DEFAULT_842 | |
123 | default "lz4" if ZSWAP_COMPRESSOR_DEFAULT_LZ4 | |
124 | default "lz4hc" if ZSWAP_COMPRESSOR_DEFAULT_LZ4HC | |
125 | default "zstd" if ZSWAP_COMPRESSOR_DEFAULT_ZSTD | |
126 | default "" | |
127 | ||
128 | choice | |
b3fbd58f | 129 | prompt "Default allocator" |
519bcb79 | 130 | depends on ZSWAP |
64d4d49c | 131 | default ZSWAP_ZPOOL_DEFAULT_ZSMALLOC if MMU |
519bcb79 JW |
132 | default ZSWAP_ZPOOL_DEFAULT_ZBUD |
133 | help | |
134 | Selects the default allocator for the compressed cache for | |
135 | swap pages. | |
136 | The default is 'zbud' for compatibility, however please do | |
137 | read the description of each of the allocators below before | |
138 | making a right choice. | |
139 | ||
140 | The selection made here can be overridden by using the kernel | |
141 | command line 'zswap.zpool=' option. | |
142 | ||
143 | config ZSWAP_ZPOOL_DEFAULT_ZBUD | |
144 | bool "zbud" | |
145 | select ZBUD | |
146 | help | |
147 | Use the zbud allocator as the default allocator. | |
148 | ||
149 | config ZSWAP_ZPOOL_DEFAULT_Z3FOLD | |
150 | bool "z3fold" | |
151 | select Z3FOLD | |
152 | help | |
153 | Use the z3fold allocator as the default allocator. | |
154 | ||
155 | config ZSWAP_ZPOOL_DEFAULT_ZSMALLOC | |
156 | bool "zsmalloc" | |
157 | select ZSMALLOC | |
158 | help | |
159 | Use the zsmalloc allocator as the default allocator. | |
160 | endchoice | |
161 | ||
162 | config ZSWAP_ZPOOL_DEFAULT | |
163 | string | |
164 | depends on ZSWAP | |
165 | default "zbud" if ZSWAP_ZPOOL_DEFAULT_ZBUD | |
166 | default "z3fold" if ZSWAP_ZPOOL_DEFAULT_Z3FOLD | |
167 | default "zsmalloc" if ZSWAP_ZPOOL_DEFAULT_ZSMALLOC | |
168 | default "" | |
169 | ||
519bcb79 | 170 | config ZBUD |
b3fbd58f JW |
171 | tristate "2:1 compression allocator (zbud)" |
172 | depends on ZSWAP | |
519bcb79 JW |
173 | help |
174 | A special purpose allocator for storing compressed pages. | |
175 | It is designed to store up to two compressed pages per physical | |
176 | page. While this design limits storage density, it has simple and | |
177 | deterministic reclaim properties that make it preferable to a higher | |
178 | density approach when reclaim will be used. | |
179 | ||
180 | config Z3FOLD | |
b3fbd58f JW |
181 | tristate "3:1 compression allocator (z3fold)" |
182 | depends on ZSWAP | |
519bcb79 JW |
183 | help |
184 | A special purpose allocator for storing compressed pages. | |
185 | It is designed to store up to three compressed pages per physical | |
186 | page. It is a ZBUD derivative so the simplicity and determinism are | |
187 | still there. | |
188 | ||
189 | config ZSMALLOC | |
b3fbd58f JW |
190 | tristate |
191 | prompt "N:1 compression allocator (zsmalloc)" if ZSWAP | |
519bcb79 JW |
192 | depends on MMU |
193 | help | |
194 | zsmalloc is a slab-based memory allocator designed to store | |
b3fbd58f JW |
195 | pages of various compression levels efficiently. It achieves |
196 | the highest storage density with the least amount of fragmentation. | |
519bcb79 JW |
197 | |
198 | config ZSMALLOC_STAT | |
199 | bool "Export zsmalloc statistics" | |
200 | depends on ZSMALLOC | |
201 | select DEBUG_FS | |
202 | help | |
203 | This option enables code in the zsmalloc to collect various | |
204 | statistics about what's happening in zsmalloc and exports that | |
205 | information to userspace via debugfs. | |
206 | If unsure, say N. | |
207 | ||
4ff93b29 SS |
208 | config ZSMALLOC_CHAIN_SIZE |
209 | int "Maximum number of physical pages per-zspage" | |
b46402fa | 210 | default 8 |
4ff93b29 SS |
211 | range 4 16 |
212 | depends on ZSMALLOC | |
213 | help | |
214 | This option sets the upper limit on the number of physical pages | |
215 | that a zmalloc page (zspage) can consist of. The optimal zspage | |
216 | chain size is calculated for each size class during the | |
217 | initialization of the pool. | |
218 | ||
219 | Changing this option can alter the characteristics of size classes, | |
220 | such as the number of pages per zspage and the number of objects | |
221 | per zspage. This can also result in different configurations of | |
222 | the pool, as zsmalloc merges size classes with similar | |
223 | characteristics. | |
224 | ||
225 | For more information, see zsmalloc documentation. | |
226 | ||
2a19be61 | 227 | menu "Slab allocator options" |
7b42f104 JW |
228 | |
229 | config SLUB | |
2a19be61 | 230 | def_bool y |
eb07c4f3 | 231 | |
e240e53a | 232 | config SLUB_TINY |
2a19be61 VB |
233 | bool "Configure for minimal memory footprint" |
234 | depends on EXPERT | |
e240e53a VB |
235 | select SLAB_MERGE_DEFAULT |
236 | help | |
2a19be61 | 237 | Configures the slab allocator in a way to achieve minimal memory |
e240e53a VB |
238 | footprint, sacrificing scalability, debugging and other features. |
239 | This is intended only for the smallest system that had used the | |
240 | SLOB allocator and is not recommended for systems with more than | |
241 | 16MB RAM. | |
242 | ||
243 | If unsure, say N. | |
244 | ||
7b42f104 JW |
245 | config SLAB_MERGE_DEFAULT |
246 | bool "Allow slab caches to be merged" | |
247 | default y | |
7b42f104 JW |
248 | help |
249 | For reduced kernel memory fragmentation, slab caches can be | |
250 | merged when they share the same size and other characteristics. | |
251 | This carries a risk of kernel heap overflows being able to | |
252 | overwrite objects from merged caches (and more easily control | |
253 | cache layout), which makes such heap attacks easier to exploit | |
254 | by attackers. By keeping caches unmerged, these kinds of exploits | |
255 | can usually only damage objects in the same cache. To disable | |
256 | merging at runtime, "slab_nomerge" can be passed on the kernel | |
257 | command line. | |
258 | ||
259 | config SLAB_FREELIST_RANDOM | |
260 | bool "Randomize slab freelist" | |
2a19be61 | 261 | depends on !SLUB_TINY |
7b42f104 JW |
262 | help |
263 | Randomizes the freelist order used on creating new pages. This | |
264 | security feature reduces the predictability of the kernel slab | |
265 | allocator against heap overflows. | |
266 | ||
267 | config SLAB_FREELIST_HARDENED | |
268 | bool "Harden slab freelist metadata" | |
2a19be61 | 269 | depends on !SLUB_TINY |
7b42f104 JW |
270 | help |
271 | Many kernel heap attacks try to target slab cache metadata and | |
272 | other infrastructure. This options makes minor performance | |
273 | sacrifices to harden the kernel slab allocator against common | |
2a19be61 | 274 | freelist exploit methods. |
7b42f104 | 275 | |
0710d012 VB |
276 | config SLUB_STATS |
277 | default n | |
2a19be61 VB |
278 | bool "Enable performance statistics" |
279 | depends on SYSFS && !SLUB_TINY | |
0710d012 | 280 | help |
2a19be61 | 281 | The statistics are useful to debug slab allocation behavior in |
0710d012 VB |
282 | order find ways to optimize the allocator. This should never be |
283 | enabled for production use since keeping statistics slows down | |
284 | the allocator by a few percentage points. The slabinfo command | |
285 | supports the determination of the most active slabs to figure | |
286 | out which slabs are relevant to a particular load. | |
287 | Try running: slabinfo -DA | |
288 | ||
519bcb79 JW |
289 | config SLUB_CPU_PARTIAL |
290 | default y | |
2a19be61 VB |
291 | depends on SMP && !SLUB_TINY |
292 | bool "Enable per cpu partial caches" | |
519bcb79 JW |
293 | help |
294 | Per cpu partial caches accelerate objects allocation and freeing | |
295 | that is local to a processor at the price of more indeterminism | |
296 | in the latency of the free. On overflow these caches will be cleared | |
297 | which requires the taking of locks that may cause latency spikes. | |
298 | Typically one would choose no for a realtime system. | |
299 | ||
3c615294 GR |
300 | config RANDOM_KMALLOC_CACHES |
301 | default n | |
2a19be61 | 302 | depends on !SLUB_TINY |
3c615294 GR |
303 | bool "Randomize slab caches for normal kmalloc" |
304 | help | |
305 | A hardening feature that creates multiple copies of slab caches for | |
306 | normal kmalloc allocation and makes kmalloc randomly pick one based | |
307 | on code address, which makes the attackers more difficult to spray | |
308 | vulnerable memory objects on the heap for the purpose of exploiting | |
309 | memory vulnerabilities. | |
310 | ||
311 | Currently the number of copies is set to 16, a reasonably large value | |
312 | that effectively diverges the memory objects allocated for different | |
313 | subsystems or modules into different caches, at the expense of a | |
314 | limited degree of memory and CPU overhead that relates to hardware and | |
315 | system workload. | |
316 | ||
2a19be61 | 317 | endmenu # Slab allocator options |
519bcb79 | 318 | |
7b42f104 JW |
319 | config SHUFFLE_PAGE_ALLOCATOR |
320 | bool "Page allocator randomization" | |
321 | default SLAB_FREELIST_RANDOM && ACPI_NUMA | |
322 | help | |
323 | Randomization of the page allocator improves the average | |
324 | utilization of a direct-mapped memory-side-cache. See section | |
325 | 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI | |
326 | 6.2a specification for an example of how a platform advertises | |
327 | the presence of a memory-side-cache. There are also incidental | |
328 | security benefits as it reduces the predictability of page | |
329 | allocations to compliment SLAB_FREELIST_RANDOM, but the | |
5e0a760b | 330 | default granularity of shuffling on the MAX_PAGE_ORDER i.e, 10th |
23baf831 KS |
331 | order of pages is selected based on cache utilization benefits |
332 | on x86. | |
7b42f104 JW |
333 | |
334 | While the randomization improves cache utilization it may | |
335 | negatively impact workloads on platforms without a cache. For | |
b413f9cd MC |
336 | this reason, by default, the randomization is not enabled even |
337 | if SHUFFLE_PAGE_ALLOCATOR=y. The randomization may be force enabled | |
338 | with the 'page_alloc.shuffle' kernel command line parameter. | |
7b42f104 JW |
339 | |
340 | Say Y if unsure. | |
341 | ||
0710d012 VB |
342 | config COMPAT_BRK |
343 | bool "Disable heap randomization" | |
344 | default y | |
345 | help | |
346 | Randomizing heap placement makes heap exploits harder, but it | |
347 | also breaks ancient binaries (including anything libc5 based). | |
348 | This option changes the bootup default to heap randomization | |
349 | disabled, and can be overridden at runtime by setting | |
350 | /proc/sys/kernel/randomize_va_space to 2. | |
351 | ||
352 | On non-ancient distros (post-2000 ones) N is usually a safe choice. | |
353 | ||
354 | config MMAP_ALLOW_UNINITIALIZED | |
355 | bool "Allow mmapped anonymous memory to be uninitialized" | |
356 | depends on EXPERT && !MMU | |
357 | default n | |
358 | help | |
359 | Normally, and according to the Linux spec, anonymous memory obtained | |
360 | from mmap() has its contents cleared before it is passed to | |
361 | userspace. Enabling this config option allows you to request that | |
362 | mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus | |
363 | providing a huge performance boost. If this option is not enabled, | |
364 | then the flag will be ignored. | |
365 | ||
366 | This is taken advantage of by uClibc's malloc(), and also by | |
367 | ELF-FDPIC binfmt's brk and stack allocator. | |
368 | ||
369 | Because of the obvious security issues, this option should only be | |
370 | enabled on embedded devices where you control what is run in | |
371 | userspace. Since that isn't generally a problem on no-MMU systems, | |
372 | it is normally safe to say Y here. | |
373 | ||
374 | See Documentation/admin-guide/mm/nommu-mmap.rst for more information. | |
375 | ||
e1785e85 DH |
376 | config SELECT_MEMORY_MODEL |
377 | def_bool y | |
a8826eeb | 378 | depends on ARCH_SELECT_MEMORY_MODEL |
e1785e85 | 379 | |
3a9da765 DH |
380 | choice |
381 | prompt "Memory model" | |
e1785e85 | 382 | depends on SELECT_MEMORY_MODEL |
d41dee36 | 383 | default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT |
e1785e85 | 384 | default FLATMEM_MANUAL |
d66d109d MR |
385 | help |
386 | This option allows you to change some of the ways that | |
387 | Linux manages its memory internally. Most users will | |
388 | only have one option here selected by the architecture | |
389 | configuration. This is normal. | |
3a9da765 | 390 | |
e1785e85 | 391 | config FLATMEM_MANUAL |
3a9da765 | 392 | bool "Flat Memory" |
bb1c50d3 | 393 | depends on !ARCH_SPARSEMEM_ENABLE || ARCH_FLATMEM_ENABLE |
3a9da765 | 394 | help |
d66d109d MR |
395 | This option is best suited for non-NUMA systems with |
396 | flat address space. The FLATMEM is the most efficient | |
397 | system in terms of performance and resource consumption | |
398 | and it is the best option for smaller systems. | |
399 | ||
400 | For systems that have holes in their physical address | |
401 | spaces and for features like NUMA and memory hotplug, | |
dd33d29a | 402 | choose "Sparse Memory". |
d41dee36 AW |
403 | |
404 | If unsure, choose this option (Flat Memory) over any other. | |
3a9da765 | 405 | |
d41dee36 AW |
406 | config SPARSEMEM_MANUAL |
407 | bool "Sparse Memory" | |
408 | depends on ARCH_SPARSEMEM_ENABLE | |
409 | help | |
410 | This will be the only option for some systems, including | |
d66d109d | 411 | memory hot-plug systems. This is normal. |
d41dee36 | 412 | |
d66d109d MR |
413 | This option provides efficient support for systems with |
414 | holes is their physical address space and allows memory | |
415 | hot-plug and hot-remove. | |
d41dee36 | 416 | |
d66d109d | 417 | If unsure, choose "Flat Memory" over this option. |
d41dee36 | 418 | |
3a9da765 DH |
419 | endchoice |
420 | ||
d41dee36 AW |
421 | config SPARSEMEM |
422 | def_bool y | |
1a83e175 | 423 | depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL |
d41dee36 | 424 | |
e1785e85 DH |
425 | config FLATMEM |
426 | def_bool y | |
bb1c50d3 | 427 | depends on !SPARSEMEM || FLATMEM_MANUAL |
d41dee36 | 428 | |
3e347261 BP |
429 | # |
430 | # SPARSEMEM_EXTREME (which is the default) does some bootmem | |
c89ab04f | 431 | # allocations when sparse_init() is called. If this cannot |
3e347261 BP |
432 | # be done on your architecture, select this option. However, |
433 | # statically allocating the mem_section[] array can potentially | |
434 | # consume vast quantities of .bss, so be careful. | |
435 | # | |
436 | # This option will also potentially produce smaller runtime code | |
437 | # with gcc 3.4 and later. | |
438 | # | |
439 | config SPARSEMEM_STATIC | |
9ba16087 | 440 | bool |
3e347261 | 441 | |
802f192e | 442 | # |
44c09201 | 443 | # Architecture platforms which require a two level mem_section in SPARSEMEM |
802f192e BP |
444 | # must select this option. This is usually for architecture platforms with |
445 | # an extremely sparse physical address space. | |
446 | # | |
3e347261 BP |
447 | config SPARSEMEM_EXTREME |
448 | def_bool y | |
449 | depends on SPARSEMEM && !SPARSEMEM_STATIC | |
4c21e2f2 | 450 | |
29c71111 | 451 | config SPARSEMEM_VMEMMAP_ENABLE |
9ba16087 | 452 | bool |
29c71111 AW |
453 | |
454 | config SPARSEMEM_VMEMMAP | |
a5ee6daa GL |
455 | bool "Sparse Memory virtual memmap" |
456 | depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE | |
457 | default y | |
458 | help | |
19fa40a0 KK |
459 | SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise |
460 | pfn_to_page and page_to_pfn operations. This is the most | |
461 | efficient option when sufficient kernel resources are available. | |
0b376f1e AK |
462 | # |
463 | # Select this config option from the architecture Kconfig, if it is preferred | |
464 | # to enable the feature of HugeTLB/dev_dax vmemmap optimization. | |
465 | # | |
0b6f1582 AK |
466 | config ARCH_WANT_OPTIMIZE_DAX_VMEMMAP |
467 | bool | |
468 | ||
469 | config ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP | |
0b376f1e | 470 | bool |
29c71111 | 471 | |
70210ed9 | 472 | config HAVE_MEMBLOCK_PHYS_MAP |
6341e62b | 473 | bool |
70210ed9 | 474 | |
25176ad0 | 475 | config HAVE_GUP_FAST |
050a9adc | 476 | depends on MMU |
6341e62b | 477 | bool |
2667f50e | 478 | |
52219aea DH |
479 | # Don't discard allocated memory used to track "memory" and "reserved" memblocks |
480 | # after early boot, so it can still be used to test for validity of memory. | |
481 | # Also, memblocks are updated with memory hot(un)plug. | |
350e88ba | 482 | config ARCH_KEEP_MEMBLOCK |
6341e62b | 483 | bool |
c378ddd5 | 484 | |
1e5d8e1e DW |
485 | # Keep arch NUMA mapping infrastructure post-init. |
486 | config NUMA_KEEP_MEMINFO | |
487 | bool | |
488 | ||
ee6f509c | 489 | config MEMORY_ISOLATION |
6341e62b | 490 | bool |
ee6f509c | 491 | |
a9e7b8d4 DH |
492 | # IORESOURCE_SYSTEM_RAM regions in the kernel resource tree that are marked |
493 | # IORESOURCE_EXCLUSIVE cannot be mapped to user space, for example, via | |
494 | # /dev/mem. | |
495 | config EXCLUSIVE_SYSTEM_RAM | |
496 | def_bool y | |
497 | depends on !DEVMEM || STRICT_DEVMEM | |
498 | ||
46723bfa YI |
499 | # |
500 | # Only be set on architectures that have completely implemented memory hotplug | |
501 | # feature. If you are not sure, don't touch it. | |
502 | # | |
503 | config HAVE_BOOTMEM_INFO_NODE | |
504 | def_bool n | |
505 | ||
91024b3c AK |
506 | config ARCH_ENABLE_MEMORY_HOTPLUG |
507 | bool | |
508 | ||
519bcb79 JW |
509 | config ARCH_ENABLE_MEMORY_HOTREMOVE |
510 | bool | |
511 | ||
3947be19 | 512 | # eventually, we can have this option just 'select SPARSEMEM' |
519bcb79 JW |
513 | menuconfig MEMORY_HOTPLUG |
514 | bool "Memory hotplug" | |
b30c5927 | 515 | select MEMORY_ISOLATION |
71b6f2dd | 516 | depends on SPARSEMEM |
40b31360 | 517 | depends on ARCH_ENABLE_MEMORY_HOTPLUG |
7ec58a2b | 518 | depends on 64BIT |
1e5d8e1e | 519 | select NUMA_KEEP_MEMINFO if NUMA |
3947be19 | 520 | |
519bcb79 JW |
521 | if MEMORY_HOTPLUG |
522 | ||
8604d9e5 | 523 | config MEMORY_HOTPLUG_DEFAULT_ONLINE |
19fa40a0 KK |
524 | bool "Online the newly added memory blocks by default" |
525 | depends on MEMORY_HOTPLUG | |
526 | help | |
8604d9e5 VK |
527 | This option sets the default policy setting for memory hotplug |
528 | onlining policy (/sys/devices/system/memory/auto_online_blocks) which | |
529 | determines what happens to newly added memory regions. Policy setting | |
530 | can always be changed at runtime. | |
cb1aaebe | 531 | See Documentation/admin-guide/mm/memory-hotplug.rst for more information. |
8604d9e5 VK |
532 | |
533 | Say Y here if you want all hot-plugged memory blocks to appear in | |
534 | 'online' state by default. | |
535 | Say N here if you want the default policy to keep all hot-plugged | |
536 | memory blocks in 'offline' state. | |
537 | ||
0c0e6195 KH |
538 | config MEMORY_HOTREMOVE |
539 | bool "Allow for memory hot remove" | |
f7e3334a | 540 | select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64) |
0c0e6195 KH |
541 | depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE |
542 | depends on MIGRATION | |
543 | ||
a08a2ae3 OS |
544 | config MHP_MEMMAP_ON_MEMORY |
545 | def_bool y | |
546 | depends on MEMORY_HOTPLUG && SPARSEMEM_VMEMMAP | |
547 | depends on ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE | |
548 | ||
519bcb79 JW |
549 | endif # MEMORY_HOTPLUG |
550 | ||
04d5ea46 AK |
551 | config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE |
552 | bool | |
553 | ||
4c21e2f2 HD |
554 | # Heavily threaded applications may benefit from splitting the mm-wide |
555 | # page_table_lock, so that faults on different parts of the user address | |
556 | # space can be handled with less contention: split it at this NR_CPUS. | |
557 | # Default to 4 for wider testing, though 8 might be more appropriate. | |
558 | # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. | |
7b6ac9df | 559 | # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. |
60bccaa6 WD |
560 | # SPARC32 allocates multiple pte tables within a single page, and therefore |
561 | # a per-page lock leads to problems when multiple tables need to be locked | |
562 | # at the same time (e.g. copy_page_range()). | |
a70caa8b | 563 | # DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page. |
4c21e2f2 HD |
564 | # |
565 | config SPLIT_PTLOCK_CPUS | |
566 | int | |
9164550e | 567 | default "999999" if !MMU |
a70caa8b HD |
568 | default "999999" if ARM && !CPU_CACHE_VIPT |
569 | default "999999" if PARISC && !PA20 | |
60bccaa6 | 570 | default "999999" if SPARC32 |
4c21e2f2 | 571 | default "4" |
7cbe34cf | 572 | |
e009bb30 | 573 | config ARCH_ENABLE_SPLIT_PMD_PTLOCK |
6341e62b | 574 | bool |
e009bb30 | 575 | |
09316c09 KK |
576 | # |
577 | # support for memory balloon | |
578 | config MEMORY_BALLOON | |
6341e62b | 579 | bool |
09316c09 | 580 | |
18468d93 RA |
581 | # |
582 | # support for memory balloon compaction | |
583 | config BALLOON_COMPACTION | |
584 | bool "Allow for balloon memory compaction/migration" | |
cd14b018 | 585 | default y |
09316c09 | 586 | depends on COMPACTION && MEMORY_BALLOON |
18468d93 RA |
587 | help |
588 | Memory fragmentation introduced by ballooning might reduce | |
589 | significantly the number of 2MB contiguous memory blocks that can be | |
590 | used within a guest, thus imposing performance penalties associated | |
591 | with the reduced number of transparent huge pages that could be used | |
592 | by the guest workload. Allowing the compaction & migration for memory | |
593 | pages enlisted as being part of memory balloon devices avoids the | |
594 | scenario aforementioned and helps improving memory defragmentation. | |
595 | ||
e9e96b39 MG |
596 | # |
597 | # support for memory compaction | |
598 | config COMPACTION | |
599 | bool "Allow for memory compaction" | |
cd14b018 | 600 | default y |
e9e96b39 | 601 | select MIGRATION |
33a93877 | 602 | depends on MMU |
e9e96b39 | 603 | help |
19fa40a0 KK |
604 | Compaction is the only memory management component to form |
605 | high order (larger physically contiguous) memory blocks | |
606 | reliably. The page allocator relies on compaction heavily and | |
607 | the lack of the feature can lead to unexpected OOM killer | |
608 | invocations for high order memory requests. You shouldn't | |
609 | disable this option unless there really is a strong reason for | |
610 | it and then we would be really interested to hear about that at | |
611 | [email protected]. | |
e9e96b39 | 612 | |
c7e0b3d0 TG |
613 | config COMPACT_UNEVICTABLE_DEFAULT |
614 | int | |
615 | depends on COMPACTION | |
616 | default 0 if PREEMPT_RT | |
617 | default 1 | |
618 | ||
36e66c55 AD |
619 | # |
620 | # support for free page reporting | |
621 | config PAGE_REPORTING | |
622 | bool "Free page reporting" | |
36e66c55 AD |
623 | help |
624 | Free page reporting allows for the incremental acquisition of | |
625 | free pages from the buddy allocator for the purpose of reporting | |
626 | those pages to another entity, such as a hypervisor, so that the | |
627 | memory can be freed within the host for other uses. | |
628 | ||
7cbe34cf CL |
629 | # |
630 | # support for page migration | |
631 | # | |
632 | config MIGRATION | |
b20a3503 | 633 | bool "Page migration" |
cd14b018 | 634 | default y |
de32a817 | 635 | depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU |
b20a3503 CL |
636 | help |
637 | Allows the migration of the physical location of pages of processes | |
e9e96b39 MG |
638 | while the virtual addresses are not changed. This is useful in |
639 | two situations. The first is on NUMA systems to put pages nearer | |
640 | to the processors accessing. The second is when allocating huge | |
641 | pages as migration can relocate pages to satisfy a huge page | |
642 | allocation instead of reclaiming. | |
6550e07f | 643 | |
76cbbead | 644 | config DEVICE_MIGRATION |
d90a25f8 | 645 | def_bool MIGRATION && ZONE_DEVICE |
76cbbead | 646 | |
c177c81e | 647 | config ARCH_ENABLE_HUGEPAGE_MIGRATION |
6341e62b | 648 | bool |
c177c81e | 649 | |
9c670ea3 NH |
650 | config ARCH_ENABLE_THP_MIGRATION |
651 | bool | |
652 | ||
4bfb68a0 AK |
653 | config HUGETLB_PAGE_SIZE_VARIABLE |
654 | def_bool n | |
655 | help | |
656 | Allows the pageblock_order value to be dynamic instead of just standard | |
657 | HUGETLB_PAGE_ORDER when there are multiple HugeTLB page sizes available | |
658 | on a platform. | |
659 | ||
5e0a760b KS |
660 | Note that the pageblock_order cannot exceed MAX_PAGE_ORDER and will be |
661 | clamped down to MAX_PAGE_ORDER. | |
b3d40a2b | 662 | |
8df995f6 | 663 | config CONTIG_ALLOC |
19fa40a0 | 664 | def_bool (MEMORY_ISOLATION && COMPACTION) || CMA |
8df995f6 | 665 | |
52166607 YH |
666 | config PCP_BATCH_SCALE_MAX |
667 | int "Maximum scale factor of PCP (Per-CPU pageset) batch allocate/free" | |
668 | default 5 | |
669 | range 0 6 | |
670 | help | |
671 | In page allocator, PCP (Per-CPU pageset) is refilled and drained in | |
672 | batches. The batch number is scaled automatically to improve page | |
673 | allocation/free throughput. But too large scale factor may hurt | |
674 | latency. This option sets the upper limit of scale factor to limit | |
675 | the maximum latency. | |
676 | ||
600715dc | 677 | config PHYS_ADDR_T_64BIT |
d4a451d5 | 678 | def_bool 64BIT |
600715dc | 679 | |
2a7326b5 | 680 | config BOUNCE |
9ca24e2e VM |
681 | bool "Enable bounce buffers" |
682 | default y | |
ce288e05 | 683 | depends on BLOCK && MMU && HIGHMEM |
9ca24e2e | 684 | help |
ce288e05 CH |
685 | Enable bounce buffers for devices that cannot access the full range of |
686 | memory available to the CPU. Enabled by default when HIGHMEM is | |
687 | selected, but you may say n to override this. | |
2a7326b5 | 688 | |
cddb8a5c AA |
689 | config MMU_NOTIFIER |
690 | bool | |
99cb252f | 691 | select INTERVAL_TREE |
fc4d5c29 | 692 | |
f8af4da3 HD |
693 | config KSM |
694 | bool "Enable KSM for page merging" | |
695 | depends on MMU | |
59e1a2f4 | 696 | select XXHASH |
f8af4da3 HD |
697 | help |
698 | Enable Kernel Samepage Merging: KSM periodically scans those areas | |
699 | of an application's address space that an app has advised may be | |
700 | mergeable. When it finds pages of identical content, it replaces | |
d0f209f6 | 701 | the many instances by a single page with that content, so |
f8af4da3 HD |
702 | saving memory until one or another app needs to modify the content. |
703 | Recommended for use with KVM, or with other duplicative applications. | |
ee65728e | 704 | See Documentation/mm/ksm.rst for more information: KSM is inactive |
c73602ad HD |
705 | until a program has madvised that an area is MADV_MERGEABLE, and |
706 | root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set). | |
f8af4da3 | 707 | |
e0a94c2a | 708 | config DEFAULT_MMAP_MIN_ADDR |
19fa40a0 | 709 | int "Low address space to protect from user allocation" |
6e141546 | 710 | depends on MMU |
19fa40a0 KK |
711 | default 4096 |
712 | help | |
e0a94c2a CL |
713 | This is the portion of low virtual memory which should be protected |
714 | from userspace allocation. Keeping a user from writing to low pages | |
715 | can help reduce the impact of kernel NULL pointer bugs. | |
716 | ||
e99fb98d | 717 | For most ppc64 and x86 users with lots of address space |
e0a94c2a CL |
718 | a value of 65536 is reasonable and should cause no problems. |
719 | On arm and other archs it should not be higher than 32768. | |
788084ab EP |
720 | Programs which use vm86 functionality or have some need to map |
721 | this low address space will need CAP_SYS_RAWIO or disable this | |
722 | protection by setting the value to 0. | |
e0a94c2a CL |
723 | |
724 | This value can be changed after boot using the | |
725 | /proc/sys/vm/mmap_min_addr tunable. | |
726 | ||
d949f36f LT |
727 | config ARCH_SUPPORTS_MEMORY_FAILURE |
728 | bool | |
e0a94c2a | 729 | |
6a46079c AK |
730 | config MEMORY_FAILURE |
731 | depends on MMU | |
d949f36f | 732 | depends on ARCH_SUPPORTS_MEMORY_FAILURE |
6a46079c | 733 | bool "Enable recovery from hardware memory errors" |
ee6f509c | 734 | select MEMORY_ISOLATION |
97f0b134 | 735 | select RAS |
6a46079c AK |
736 | help |
737 | Enables code to recover from some memory failures on systems | |
738 | with MCA recovery. This allows a system to continue running | |
739 | even when some of its memory has uncorrected errors. This requires | |
740 | special hardware support and typically ECC memory. | |
741 | ||
cae681fc | 742 | config HWPOISON_INJECT |
413f9efb | 743 | tristate "HWPoison pages injector" |
27df5068 | 744 | depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS |
478c5ffc | 745 | select PROC_PAGE_MONITOR |
cae681fc | 746 | |
fc4d5c29 DH |
747 | config NOMMU_INITIAL_TRIM_EXCESS |
748 | int "Turn on mmap() excess space trimming before booting" | |
749 | depends on !MMU | |
750 | default 1 | |
751 | help | |
752 | The NOMMU mmap() frequently needs to allocate large contiguous chunks | |
753 | of memory on which to store mappings, but it can only ask the system | |
754 | allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently | |
755 | more than it requires. To deal with this, mmap() is able to trim off | |
756 | the excess and return it to the allocator. | |
757 | ||
758 | If trimming is enabled, the excess is trimmed off and returned to the | |
759 | system allocator, which can cause extra fragmentation, particularly | |
760 | if there are a lot of transient processes. | |
761 | ||
762 | If trimming is disabled, the excess is kept, but not used, which for | |
763 | long-term mappings means that the space is wasted. | |
764 | ||
765 | Trimming can be dynamically controlled through a sysctl option | |
766 | (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of | |
767 | excess pages there must be before trimming should occur, or zero if | |
768 | no trimming is to occur. | |
769 | ||
770 | This option specifies the initial value of this option. The default | |
771 | of 1 says that all excess pages should be trimmed. | |
772 | ||
dd19d293 | 773 | See Documentation/admin-guide/mm/nommu-mmap.rst for more information. |
bbddff05 | 774 | |
519bcb79 JW |
775 | config ARCH_WANT_GENERAL_HUGETLB |
776 | bool | |
777 | ||
778 | config ARCH_WANTS_THP_SWAP | |
779 | def_bool n | |
780 | ||
781 | menuconfig TRANSPARENT_HUGEPAGE | |
13ece886 | 782 | bool "Transparent Hugepage Support" |
554b0f3c | 783 | depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT |
5d689240 | 784 | select COMPACTION |
3a08cd52 | 785 | select XARRAY_MULTI |
4c76d9d1 AA |
786 | help |
787 | Transparent Hugepages allows the kernel to use huge pages and | |
788 | huge tlb transparently to the applications whenever possible. | |
789 | This feature can improve computing performance to certain | |
790 | applications by speeding up page faults during memory | |
791 | allocation, by reducing the number of tlb misses and by speeding | |
792 | up the pagetable walking. | |
793 | ||
794 | If memory constrained on embedded, you may want to say N. | |
795 | ||
519bcb79 JW |
796 | if TRANSPARENT_HUGEPAGE |
797 | ||
13ece886 AA |
798 | choice |
799 | prompt "Transparent Hugepage Support sysfs defaults" | |
800 | depends on TRANSPARENT_HUGEPAGE | |
801 | default TRANSPARENT_HUGEPAGE_ALWAYS | |
802 | help | |
803 | Selects the sysfs defaults for Transparent Hugepage Support. | |
804 | ||
805 | config TRANSPARENT_HUGEPAGE_ALWAYS | |
806 | bool "always" | |
807 | help | |
808 | Enabling Transparent Hugepage always, can increase the | |
809 | memory footprint of applications without a guaranteed | |
810 | benefit but it will work automatically for all applications. | |
811 | ||
812 | config TRANSPARENT_HUGEPAGE_MADVISE | |
813 | bool "madvise" | |
814 | help | |
815 | Enabling Transparent Hugepage madvise, will only provide a | |
816 | performance improvement benefit to the applications using | |
817 | madvise(MADV_HUGEPAGE) but it won't risk to increase the | |
818 | memory footprint of applications without a guaranteed | |
819 | benefit. | |
683ec99f DM |
820 | |
821 | config TRANSPARENT_HUGEPAGE_NEVER | |
822 | bool "never" | |
823 | help | |
824 | Disable Transparent Hugepage by default. It can still be | |
825 | enabled at runtime via sysfs. | |
13ece886 AA |
826 | endchoice |
827 | ||
38d8b4e6 YH |
828 | config THP_SWAP |
829 | def_bool y | |
dad6a5eb | 830 | depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP && 64BIT |
38d8b4e6 YH |
831 | help |
832 | Swap transparent huge pages in one piece, without splitting. | |
14fef284 YH |
833 | XXX: For now, swap cluster backing transparent huge page |
834 | will be split after swapout. | |
38d8b4e6 YH |
835 | |
836 | For selection by architectures with reasonable THP sizes. | |
837 | ||
519bcb79 JW |
838 | config READ_ONLY_THP_FOR_FS |
839 | bool "Read-only THP for filesystems (EXPERIMENTAL)" | |
840 | depends on TRANSPARENT_HUGEPAGE && SHMEM | |
841 | ||
842 | help | |
843 | Allow khugepaged to put read-only file-backed pages in THP. | |
844 | ||
845 | This is marked experimental because it is a new feature. Write | |
846 | support of file THPs will be developed in the next few release | |
847 | cycles. | |
848 | ||
849 | endif # TRANSPARENT_HUGEPAGE | |
850 | ||
ac3830c3 PX |
851 | # |
852 | # The architecture supports pgtable leaves that is larger than PAGE_SIZE | |
853 | # | |
854 | config PGTABLE_HAS_HUGE_LEAVES | |
855 | def_bool TRANSPARENT_HUGEPAGE || HUGETLB_PAGE | |
856 | ||
bbddff05 TH |
857 | # |
858 | # UP and nommu archs use km based percpu allocator | |
859 | # | |
860 | config NEED_PER_CPU_KM | |
3583521a | 861 | depends on !SMP || !MMU |
bbddff05 TH |
862 | bool |
863 | default y | |
077b1f83 | 864 | |
7ecd19cf KW |
865 | config NEED_PER_CPU_EMBED_FIRST_CHUNK |
866 | bool | |
867 | ||
868 | config NEED_PER_CPU_PAGE_FIRST_CHUNK | |
869 | bool | |
870 | ||
871 | config USE_PERCPU_NUMA_NODE_ID | |
872 | bool | |
873 | ||
874 | config HAVE_SETUP_PER_CPU_AREA | |
875 | bool | |
876 | ||
f825c736 AK |
877 | config CMA |
878 | bool "Contiguous Memory Allocator" | |
aca52c39 | 879 | depends on MMU |
f825c736 AK |
880 | select MIGRATION |
881 | select MEMORY_ISOLATION | |
882 | help | |
883 | This enables the Contiguous Memory Allocator which allows other | |
884 | subsystems to allocate big physically-contiguous blocks of memory. | |
885 | CMA reserves a region of memory and allows only movable pages to | |
886 | be allocated from it. This way, the kernel can use the memory for | |
887 | pagecache and when a subsystem requests for contiguous area, the | |
888 | allocated pages are migrated away to serve the contiguous request. | |
889 | ||
890 | If unsure, say "n". | |
891 | ||
28b24c1f SL |
892 | config CMA_DEBUGFS |
893 | bool "CMA debugfs interface" | |
894 | depends on CMA && DEBUG_FS | |
895 | help | |
896 | Turns on the DebugFS interface for CMA. | |
897 | ||
43ca106f MK |
898 | config CMA_SYSFS |
899 | bool "CMA information through sysfs interface" | |
900 | depends on CMA && SYSFS | |
901 | help | |
902 | This option exposes some sysfs attributes to get information | |
903 | from CMA. | |
904 | ||
a254129e JK |
905 | config CMA_AREAS |
906 | int "Maximum count of the CMA areas" | |
907 | depends on CMA | |
73307523 AK |
908 | default 20 if NUMA |
909 | default 8 | |
a254129e JK |
910 | help |
911 | CMA allows to create CMA areas for particular purpose, mainly, | |
912 | used as device private area. This parameter sets the maximum | |
913 | number of CMA area in the system. | |
914 | ||
73307523 | 915 | If unsure, leave the default value "8" in UMA and "20" in NUMA. |
a254129e | 916 | |
af8d417a DS |
917 | config MEM_SOFT_DIRTY |
918 | bool "Track memory changes" | |
919 | depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS | |
920 | select PROC_PAGE_MONITOR | |
4e2e2770 | 921 | help |
af8d417a DS |
922 | This option enables memory changes tracking by introducing a |
923 | soft-dirty bit on pte-s. This bit it set when someone writes | |
924 | into a page just as regular dirty bit, but unlike the latter | |
925 | it can be cleared by hands. | |
926 | ||
1ad1335d | 927 | See Documentation/admin-guide/mm/soft-dirty.rst for more details. |
4e2e2770 | 928 | |
9e5c33d7 MS |
929 | config GENERIC_EARLY_IOREMAP |
930 | bool | |
042d27ac | 931 | |
22ee3ea5 HD |
932 | config STACK_MAX_DEFAULT_SIZE_MB |
933 | int "Default maximum user stack size for 32-bit processes (MB)" | |
934 | default 100 | |
042d27ac HD |
935 | range 8 2048 |
936 | depends on STACK_GROWSUP && (!64BIT || COMPAT) | |
937 | help | |
938 | This is the maximum stack size in Megabytes in the VM layout of 32-bit | |
939 | user processes when the stack grows upwards (currently only on parisc | |
22ee3ea5 | 940 | arch) when the RLIMIT_STACK hard limit is unlimited. |
042d27ac | 941 | |
22ee3ea5 | 942 | A sane initial value is 100 MB. |
3a80a7fa | 943 | |
3a80a7fa | 944 | config DEFERRED_STRUCT_PAGE_INIT |
1ce22103 | 945 | bool "Defer initialisation of struct pages to kthreads" |
d39f8fb4 | 946 | depends on SPARSEMEM |
ab1e8d89 | 947 | depends on !NEED_PER_CPU_KM |
889c695d | 948 | depends on 64BIT |
e4443149 | 949 | select PADATA |
3a80a7fa MG |
950 | help |
951 | Ordinarily all struct pages are initialised during early boot in a | |
952 | single thread. On very large machines this can take a considerable | |
953 | amount of time. If this option is set, large machines will bring up | |
e4443149 DJ |
954 | a subset of memmap at boot and then initialise the rest in parallel. |
955 | This has a potential performance impact on tasks running early in the | |
1ce22103 VB |
956 | lifetime of the system until these kthreads finish the |
957 | initialisation. | |
033fbae9 | 958 | |
1c676e0d SP |
959 | config PAGE_IDLE_FLAG |
960 | bool | |
961 | select PAGE_EXTENSION if !64BIT | |
962 | help | |
963 | This adds PG_idle and PG_young flags to 'struct page'. PTE Accessed | |
964 | bit writers can set the state of the bit in the flags so that PTE | |
965 | Accessed bit readers may avoid disturbance. | |
966 | ||
33c3fc71 VD |
967 | config IDLE_PAGE_TRACKING |
968 | bool "Enable idle page tracking" | |
969 | depends on SYSFS && MMU | |
1c676e0d | 970 | select PAGE_IDLE_FLAG |
33c3fc71 VD |
971 | help |
972 | This feature allows to estimate the amount of user pages that have | |
973 | not been touched during a given period of time. This information can | |
974 | be useful to tune memory cgroup limits and/or for job placement | |
975 | within a compute cluster. | |
976 | ||
1ad1335d MR |
977 | See Documentation/admin-guide/mm/idle_page_tracking.rst for |
978 | more details. | |
33c3fc71 | 979 | |
8690bbcf MD |
980 | # Architectures which implement cpu_dcache_is_aliasing() to query |
981 | # whether the data caches are aliased (VIVT or VIPT with dcache | |
982 | # aliasing) need to select this. | |
983 | config ARCH_HAS_CPU_CACHE_ALIASING | |
984 | bool | |
985 | ||
c2280be8 AK |
986 | config ARCH_HAS_CACHE_LINE_SIZE |
987 | bool | |
988 | ||
2792d84e KC |
989 | config ARCH_HAS_CURRENT_STACK_POINTER |
990 | bool | |
991 | help | |
992 | In support of HARDENED_USERCOPY performing stack variable lifetime | |
993 | checking, an architecture-agnostic way to find the stack pointer | |
994 | is needed. Once an architecture defines an unsigned long global | |
995 | register alias named "current_stack_pointer", this config can be | |
996 | selected. | |
997 | ||
17596731 | 998 | config ARCH_HAS_PTE_DEVMAP |
65f7d049 OH |
999 | bool |
1000 | ||
63703f37 KW |
1001 | config ARCH_HAS_ZONE_DMA_SET |
1002 | bool | |
1003 | ||
1004 | config ZONE_DMA | |
1005 | bool "Support DMA zone" if ARCH_HAS_ZONE_DMA_SET | |
1006 | default y if ARM64 || X86 | |
1007 | ||
1008 | config ZONE_DMA32 | |
1009 | bool "Support DMA32 zone" if ARCH_HAS_ZONE_DMA_SET | |
1010 | depends on !X86_32 | |
1011 | default y if ARM64 | |
1012 | ||
033fbae9 | 1013 | config ZONE_DEVICE |
5042db43 | 1014 | bool "Device memory (pmem, HMM, etc...) hotplug support" |
033fbae9 DW |
1015 | depends on MEMORY_HOTPLUG |
1016 | depends on MEMORY_HOTREMOVE | |
99490f16 | 1017 | depends on SPARSEMEM_VMEMMAP |
17596731 | 1018 | depends on ARCH_HAS_PTE_DEVMAP |
3a08cd52 | 1019 | select XARRAY_MULTI |
033fbae9 DW |
1020 | |
1021 | help | |
1022 | Device memory hotplug support allows for establishing pmem, | |
1023 | or other device driver discovered memory regions, in the | |
1024 | memmap. This allows pfn_to_page() lookups of otherwise | |
1025 | "device-physical" addresses which is needed for using a DAX | |
1026 | mapping in an O_DIRECT operation, among other things. | |
1027 | ||
1028 | If FS_DAX is enabled, then say Y. | |
06a660ad | 1029 | |
9c240a7b CH |
1030 | # |
1031 | # Helpers to mirror range of the CPU page tables of a process into device page | |
1032 | # tables. | |
1033 | # | |
c0b12405 | 1034 | config HMM_MIRROR |
9c240a7b | 1035 | bool |
f442c283 | 1036 | depends on MMU |
c0b12405 | 1037 | |
14b80582 DW |
1038 | config GET_FREE_REGION |
1039 | depends on SPARSEMEM | |
1040 | bool | |
1041 | ||
5042db43 JG |
1042 | config DEVICE_PRIVATE |
1043 | bool "Unaddressable device memory (GPU memory, ...)" | |
7328d9cc | 1044 | depends on ZONE_DEVICE |
14b80582 | 1045 | select GET_FREE_REGION |
5042db43 JG |
1046 | |
1047 | help | |
1048 | Allows creation of struct pages to represent unaddressable device | |
1049 | memory; i.e., memory that is only accessible from the device (or | |
1050 | group of devices). You likely also want to select HMM_MIRROR. | |
1051 | ||
3e9a9e25 CH |
1052 | config VMAP_PFN |
1053 | bool | |
1054 | ||
63c17fb8 DH |
1055 | config ARCH_USES_HIGH_VMA_FLAGS |
1056 | bool | |
66d37570 DH |
1057 | config ARCH_HAS_PKEYS |
1058 | bool | |
30a5b536 | 1059 | |
b0284cd2 CM |
1060 | config ARCH_USES_PG_ARCH_X |
1061 | bool | |
1062 | help | |
1063 | Enable the definition of PG_arch_x page flags with x > 1. Only | |
1064 | suitable for 64-bit architectures with CONFIG_FLATMEM or | |
1065 | CONFIG_SPARSEMEM_VMEMMAP enabled, otherwise there may not be | |
1066 | enough room for additional bits in page->flags. | |
1067 | ||
0710d012 VB |
1068 | config VM_EVENT_COUNTERS |
1069 | default y | |
1070 | bool "Enable VM event counters for /proc/vmstat" if EXPERT | |
1071 | help | |
1072 | VM event counters are needed for event counts to be shown. | |
1073 | This option allows the disabling of the VM event counters | |
1074 | on EXPERT systems. /proc/vmstat will only show page counts | |
1075 | if VM event counters are disabled. | |
1076 | ||
30a5b536 DZ |
1077 | config PERCPU_STATS |
1078 | bool "Collect percpu memory statistics" | |
30a5b536 DZ |
1079 | help |
1080 | This feature collects and exposes statistics via debugfs. The | |
1081 | information includes global and per chunk statistics, which can | |
1082 | be used to help understand percpu memory usage. | |
64c349f4 | 1083 | |
9c84f229 JH |
1084 | config GUP_TEST |
1085 | bool "Enable infrastructure for get_user_pages()-related unit tests" | |
d0de8241 | 1086 | depends on DEBUG_FS |
64c349f4 | 1087 | help |
9c84f229 JH |
1088 | Provides /sys/kernel/debug/gup_test, which in turn provides a way |
1089 | to make ioctl calls that can launch kernel-based unit tests for | |
1090 | the get_user_pages*() and pin_user_pages*() family of API calls. | |
64c349f4 | 1091 | |
9c84f229 JH |
1092 | These tests include benchmark testing of the _fast variants of |
1093 | get_user_pages*() and pin_user_pages*(), as well as smoke tests of | |
1094 | the non-_fast variants. | |
1095 | ||
f4f9bda4 JH |
1096 | There is also a sub-test that allows running dump_page() on any |
1097 | of up to eight pages (selected by command line args) within the | |
1098 | range of user-space addresses. These pages are either pinned via | |
1099 | pin_user_pages*(), or pinned via get_user_pages*(), as specified | |
1100 | by other command line arguments. | |
1101 | ||
baa489fa | 1102 | See tools/testing/selftests/mm/gup_test.c |
3010a5ea | 1103 | |
d0de8241 BS |
1104 | comment "GUP_TEST needs to have DEBUG_FS enabled" |
1105 | depends on !GUP_TEST && !DEBUG_FS | |
3010a5ea | 1106 | |
6ca297d4 | 1107 | config GUP_GET_PXX_LOW_HIGH |
39656e83 CH |
1108 | bool |
1109 | ||
def85743 KB |
1110 | config DMAPOOL_TEST |
1111 | tristate "Enable a module to run time tests on dma_pool" | |
1112 | depends on HAS_DMA | |
1113 | help | |
1114 | Provides a test module that will allocate and free many blocks of | |
1115 | various sizes and report how long it takes. This is intended to | |
1116 | provide a consistent way to measure how changes to the | |
1117 | dma_pool_alloc/free routines affect performance. | |
1118 | ||
3010a5ea LD |
1119 | config ARCH_HAS_PTE_SPECIAL |
1120 | bool | |
59e0b520 | 1121 | |
cbd34da7 CH |
1122 | # |
1123 | # Some architectures require a special hugepage directory format that is | |
1124 | # required to support multiple hugepage sizes. For example a4fe3ce76 | |
1125 | # "powerpc/mm: Allow more flexible layouts for hugepage pagetables" | |
1126 | # introduced it on powerpc. This allows for a more flexible hugepage | |
1127 | # pagetable layouts. | |
1128 | # | |
1129 | config ARCH_HAS_HUGEPD | |
1130 | bool | |
1131 | ||
c5acad84 TH |
1132 | config MAPPING_DIRTY_HELPERS |
1133 | bool | |
1134 | ||
298fa1ad TG |
1135 | config KMAP_LOCAL |
1136 | bool | |
1137 | ||
825c43f5 AB |
1138 | config KMAP_LOCAL_NON_LINEAR_PTE_ARRAY |
1139 | bool | |
1140 | ||
1fbaf8fc CH |
1141 | # struct io_mapping based helper. Selected by drivers that need them |
1142 | config IO_MAPPING | |
1143 | bool | |
1507f512 | 1144 | |
626e98cb TW |
1145 | config MEMFD_CREATE |
1146 | bool "Enable memfd_create() system call" if EXPERT | |
1147 | ||
1507f512 | 1148 | config SECRETMEM |
74947724 LB |
1149 | default y |
1150 | bool "Enable memfd_secret() system call" if EXPERT | |
1151 | depends on ARCH_HAS_SET_DIRECT_MAP | |
1152 | help | |
1153 | Enable the memfd_secret() system call with the ability to create | |
1154 | memory areas visible only in the context of the owning process and | |
1155 | not mapped to other processes and other kernel page tables. | |
1507f512 | 1156 | |
9a10064f CC |
1157 | config ANON_VMA_NAME |
1158 | bool "Anonymous VMA name support" | |
1159 | depends on PROC_FS && ADVISE_SYSCALLS && MMU | |
1160 | ||
1161 | help | |
1162 | Allow naming anonymous virtual memory areas. | |
1163 | ||
1164 | This feature allows assigning names to virtual memory areas. Assigned | |
1165 | names can be later retrieved from /proc/pid/maps and /proc/pid/smaps | |
1166 | and help identifying individual anonymous memory areas. | |
1167 | Assigning a name to anonymous virtual memory area might prevent that | |
1168 | area from being merged with adjacent virtual memory areas due to the | |
1169 | difference in their name. | |
1170 | ||
430529b5 PX |
1171 | config HAVE_ARCH_USERFAULTFD_WP |
1172 | bool | |
1173 | help | |
1174 | Arch has userfaultfd write protection support | |
1175 | ||
1176 | config HAVE_ARCH_USERFAULTFD_MINOR | |
1177 | bool | |
1178 | help | |
1179 | Arch has userfaultfd minor fault support | |
1180 | ||
97219cc3 PX |
1181 | menuconfig USERFAULTFD |
1182 | bool "Enable userfaultfd() system call" | |
1183 | depends on MMU | |
1184 | help | |
1185 | Enable the userfaultfd() system call that allows to intercept and | |
1186 | handle page faults in userland. | |
1187 | ||
1188 | if USERFAULTFD | |
1db9dbc2 | 1189 | config PTE_MARKER_UFFD_WP |
81e0f15f PX |
1190 | bool "Userfaultfd write protection support for shmem/hugetlbfs" |
1191 | default y | |
1192 | depends on HAVE_ARCH_USERFAULTFD_WP | |
1db9dbc2 PX |
1193 | |
1194 | help | |
1195 | Allows to create marker PTEs for userfaultfd write protection | |
1196 | purposes. It is required to enable userfaultfd write protection on | |
1197 | file-backed memory types like shmem and hugetlbfs. | |
97219cc3 | 1198 | endif # USERFAULTFD |
1db9dbc2 | 1199 | |
ac35a490 | 1200 | # multi-gen LRU { |
ec1c86b2 YZ |
1201 | config LRU_GEN |
1202 | bool "Multi-Gen LRU" | |
1203 | depends on MMU | |
1204 | # make sure folio->flags has enough spare bits | |
1205 | depends on 64BIT || !SPARSEMEM || SPARSEMEM_VMEMMAP | |
1206 | help | |
07017acb YZ |
1207 | A high performance LRU implementation to overcommit memory. See |
1208 | Documentation/admin-guide/mm/multigen_lru.rst for details. | |
ec1c86b2 | 1209 | |
354ed597 YZ |
1210 | config LRU_GEN_ENABLED |
1211 | bool "Enable by default" | |
1212 | depends on LRU_GEN | |
1213 | help | |
1214 | This option enables the multi-gen LRU by default. | |
1215 | ||
ac35a490 YZ |
1216 | config LRU_GEN_STATS |
1217 | bool "Full stats for debugging" | |
1218 | depends on LRU_GEN | |
1219 | help | |
1220 | Do not enable this option unless you plan to look at historical stats | |
1221 | from evicted generations for debugging purpose. | |
1222 | ||
1223 | This option has a per-memcg and per-node memory overhead. | |
61dd3f24 KH |
1224 | |
1225 | config LRU_GEN_WALKS_MMU | |
1226 | def_bool y | |
1227 | depends on LRU_GEN && ARCH_HAS_HW_PTE_YOUNG | |
ac35a490 YZ |
1228 | # } |
1229 | ||
0b6cc04f SB |
1230 | config ARCH_SUPPORTS_PER_VMA_LOCK |
1231 | def_bool n | |
1232 | ||
1233 | config PER_VMA_LOCK | |
1234 | def_bool y | |
1235 | depends on ARCH_SUPPORTS_PER_VMA_LOCK && MMU && SMP | |
1236 | help | |
1237 | Allow per-vma locking during page fault handling. | |
1238 | ||
1239 | This feature allows locking each virtual memory area separately when | |
1240 | handling page faults instead of taking mmap_lock. | |
1241 | ||
c2508ec5 LT |
1242 | config LOCK_MM_AND_FIND_VMA |
1243 | bool | |
1244 | depends on !STACK_GROWSUP | |
1245 | ||
8f23f5db JG |
1246 | config IOMMU_MM_DATA |
1247 | bool | |
1248 | ||
12af2b83 MRI |
1249 | config EXECMEM |
1250 | bool | |
1251 | ||
2224d848 SP |
1252 | source "mm/damon/Kconfig" |
1253 | ||
59e0b520 | 1254 | endmenu |