| Kconfig (e240e53ae0abb0896e0f399bdfef41c69cec3123) | Kconfig (b0284cd29a957e62d60c2886fd663be93c56f9c0) |
|---|---|
| 1# SPDX-License-Identifier: GPL-2.0-only 2 3menu "Memory Management options" 4 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# --- 216 unchanged lines hidden (view full) --- 225 depends on !PREEMPT_RT 226 help 227 SLOB replaces the stock allocator with a drastically simpler 228 allocator. SLOB is generally more space efficient but 229 does not perform as well on large systems. 230 231endchoice 232 | 1# SPDX-License-Identifier: GPL-2.0-only 2 3menu "Memory Management options" 4 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# --- 216 unchanged lines hidden (view full) --- 225 depends on !PREEMPT_RT 226 help 227 SLOB replaces the stock allocator with a drastically simpler 228 allocator. SLOB is generally more space efficient but 229 does not perform as well on large systems. 230 231endchoice 232 |
| 233config SLUB_TINY 234 bool "Configure SLUB for minimal memory footprint" 235 depends on SLUB && EXPERT 236 select SLAB_MERGE_DEFAULT 237 help 238 Configures the SLUB allocator in a way to achieve minimal memory 239 footprint, sacrificing scalability, debugging and other features. 240 This is intended only for the smallest system that had used the 241 SLOB allocator and is not recommended for systems with more than 242 16MB RAM. 243 244 If unsure, say N. 245 | |
| 246config SLAB_MERGE_DEFAULT 247 bool "Allow slab caches to be merged" 248 default y 249 depends on SLAB || SLUB 250 help 251 For reduced kernel memory fragmentation, slab caches can be 252 merged when they share the same size and other characteristics. 253 This carries a risk of kernel heap overflows being able to 254 overwrite objects from merged caches (and more easily control 255 cache layout), which makes such heap attacks easier to exploit 256 by attackers. By keeping caches unmerged, these kinds of exploits 257 can usually only damage objects in the same cache. To disable 258 merging at runtime, "slab_nomerge" can be passed on the kernel 259 command line. 260 261config SLAB_FREELIST_RANDOM 262 bool "Randomize slab freelist" | 233config SLAB_MERGE_DEFAULT 234 bool "Allow slab caches to be merged" 235 default y 236 depends on SLAB || SLUB 237 help 238 For reduced kernel memory fragmentation, slab caches can be 239 merged when they share the same size and other characteristics. 240 This carries a risk of kernel heap overflows being able to 241 overwrite objects from merged caches (and more easily control 242 cache layout), which makes such heap attacks easier to exploit 243 by attackers. By keeping caches unmerged, these kinds of exploits 244 can usually only damage objects in the same cache. To disable 245 merging at runtime, "slab_nomerge" can be passed on the kernel 246 command line. 247 248config SLAB_FREELIST_RANDOM 249 bool "Randomize slab freelist" |
| 263 depends on SLAB || (SLUB && !SLUB_TINY) | 250 depends on SLAB || SLUB |
| 264 help 265 Randomizes the freelist order used on creating new pages. This 266 security feature reduces the predictability of the kernel slab 267 allocator against heap overflows. 268 269config SLAB_FREELIST_HARDENED 270 bool "Harden slab freelist metadata" | 251 help 252 Randomizes the freelist order used on creating new pages. This 253 security feature reduces the predictability of the kernel slab 254 allocator against heap overflows. 255 256config SLAB_FREELIST_HARDENED 257 bool "Harden slab freelist metadata" |
| 271 depends on SLAB || (SLUB && !SLUB_TINY) | 258 depends on SLAB || SLUB |
| 272 help 273 Many kernel heap attacks try to target slab cache metadata and 274 other infrastructure. This options makes minor performance 275 sacrifices to harden the kernel slab allocator against common 276 freelist exploit methods. Some slab implementations have more 277 sanity-checking than others. This option is most effective with 278 CONFIG_SLUB. 279 280config SLUB_STATS 281 default n 282 bool "Enable SLUB performance statistics" | 259 help 260 Many kernel heap attacks try to target slab cache metadata and 261 other infrastructure. This options makes minor performance 262 sacrifices to harden the kernel slab allocator against common 263 freelist exploit methods. Some slab implementations have more 264 sanity-checking than others. This option is most effective with 265 CONFIG_SLUB. 266 267config SLUB_STATS 268 default n 269 bool "Enable SLUB performance statistics" |
| 283 depends on SLUB && SYSFS && !SLUB_TINY | 270 depends on SLUB && SYSFS |
| 284 help 285 SLUB statistics are useful to debug SLUBs allocation behavior in 286 order find ways to optimize the allocator. This should never be 287 enabled for production use since keeping statistics slows down 288 the allocator by a few percentage points. The slabinfo command 289 supports the determination of the most active slabs to figure 290 out which slabs are relevant to a particular load. 291 Try running: slabinfo -DA 292 293config SLUB_CPU_PARTIAL 294 default y | 271 help 272 SLUB statistics are useful to debug SLUBs allocation behavior in 273 order find ways to optimize the allocator. This should never be 274 enabled for production use since keeping statistics slows down 275 the allocator by a few percentage points. The slabinfo command 276 supports the determination of the most active slabs to figure 277 out which slabs are relevant to a particular load. 278 Try running: slabinfo -DA 279 280config SLUB_CPU_PARTIAL 281 default y |
| 295 depends on SLUB && SMP && !SLUB_TINY | 282 depends on SLUB && SMP |
| 296 bool "SLUB per cpu partial cache" 297 help 298 Per cpu partial caches accelerate objects allocation and freeing 299 that is local to a processor at the price of more indeterminism 300 in the latency of the free. On overflow these caches will be cleared 301 which requires the taking of locks that may cause latency spikes. 302 Typically one would choose no for a realtime system. 303 --- 709 unchanged lines hidden (view full) --- 1013config VMAP_PFN 1014 bool 1015 1016config ARCH_USES_HIGH_VMA_FLAGS 1017 bool 1018config ARCH_HAS_PKEYS 1019 bool 1020 | 283 bool "SLUB per cpu partial cache" 284 help 285 Per cpu partial caches accelerate objects allocation and freeing 286 that is local to a processor at the price of more indeterminism 287 in the latency of the free. On overflow these caches will be cleared 288 which requires the taking of locks that may cause latency spikes. 289 Typically one would choose no for a realtime system. 290 --- 709 unchanged lines hidden (view full) --- 1000config VMAP_PFN 1001 bool 1002 1003config ARCH_USES_HIGH_VMA_FLAGS 1004 bool 1005config ARCH_HAS_PKEYS 1006 bool 1007 |
| 1008config ARCH_USES_PG_ARCH_X 1009 bool 1010 help 1011 Enable the definition of PG_arch_x page flags with x > 1. Only 1012 suitable for 64-bit architectures with CONFIG_FLATMEM or 1013 CONFIG_SPARSEMEM_VMEMMAP enabled, otherwise there may not be 1014 enough room for additional bits in page->flags. 1015 |
|
| 1021config VM_EVENT_COUNTERS 1022 default y 1023 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1024 help 1025 VM event counters are needed for event counts to be shown. 1026 This option allows the disabling of the VM event counters 1027 on EXPERT systems. /proc/vmstat will only show page counts 1028 if VM event counters are disabled. --- 140 unchanged lines hidden --- | 1016config VM_EVENT_COUNTERS 1017 default y 1018 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1019 help 1020 VM event counters are needed for event counts to be shown. 1021 This option allows the disabling of the VM event counters 1022 on EXPERT systems. /proc/vmstat will only show page counts 1023 if VM event counters are disabled. --- 140 unchanged lines hidden --- |