Lines Matching full:memory
3 menu "Memory Management options"
13 bool "Support for paging of anonymous memory (swap)"
19 used to provide more virtual memory than the actual RAM present
30 compress them into a dynamically allocated RAM-based memory pool.
46 bool "Shrink the zswap pool on memory pressure"
52 written back to the backing swap device) on memory pressure.
57 and consume memory indefinitely.
177 bool "Configure for minimal memory footprint"
181 Configures the slab allocator in a way to achieve minimal memory
193 For reduced kernel memory fragmentation, slab caches can be
231 memory fragmentation, though in practice it's only a handful
258 vulnerable memory objects on the heap for the purpose of exploiting
259 memory vulnerabilities.
262 that effectively diverges the memory objects allocated for different
264 limited degree of memory and CPU overhead that relates to hardware and
274 utilization of a direct-mapped memory-side-cache. See section
275 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI
277 the presence of a memory-side-cache. There are also incidental
305 bool "Allow mmapped anonymous memory to be uninitialized"
309 Normally, and according to the Linux spec, anonymous memory obtained
331 prompt "Memory model"
337 Linux manages its memory internally. Most users will
342 bool "Flat Memory"
351 spaces and for features like NUMA and memory hotplug,
352 choose "Sparse Memory".
354 If unsure, choose this option (Flat Memory) over any other.
357 bool "Sparse Memory"
361 memory hot-plug systems. This is normal.
364 holes is their physical address space and allows memory
367 If unsure, choose "Flat Memory" over this option.
434 # Enable memblock support for scratch memory which is needed for kexec handover
438 # Don't discard allocated memory used to track "memory" and "reserved" memblocks
439 # after early boot, so it can still be used to test for validity of memory.
440 # Also, memblocks are updated with memory hot(un)plug.
459 # Only be set on architectures that have completely implemented memory hotplug
473 bool "Memory hotplug"
483 prompt "Memory Hotplug Default Online Type"
486 Default memory type for hotplugged memory.
488 This option sets the default policy setting for memory hotplug
489 onlining policy (/sys/devices/system/memory/auto_online_blocks) which
490 determines what happens to newly added memory regions. Policy setting
497 Select online_kernel to generally allow kernel usage of this memory.
498 Select online_movable to generally disallow kernel usage of this memory.
502 See Documentation/admin-guide/mm/memory-hotplug.rst for more information.
507 Hotplugged memory will not be onlined by default.
509 handle onlining of hotplug memory policy.
515 hotplugged memory into the zone it thinks is reasonable.
516 This memory may be utilized for kernel data.
522 hotplugged memory into a zone capable of being used for kernel
529 hotplug memory into ZONE_MOVABLE. This memory will generally
533 ZONE_NORMAL memory is available to describe hotplug memory,
534 otherwise hotplug memory may fail to online. For example,
535 sufficient kernel-capable memory (ZONE_NORMAL) must be
541 bool "Allow for memory hot remove"
584 # support for memory balloon
589 # support for memory balloon page migration
591 bool "Allow for balloon memory migration"
595 Allow for migration of pages inflated in a memory balloon such that
596 they can be allocated from memory areas only available for movable
598 migrated for memory defragmentation purposes by memory compaction.
601 # support for memory compaction
603 bool "Allow for memory compaction"
608 Compaction is the only memory management component to form
609 high order (larger physically contiguous) memory blocks
612 invocations for high order memory requests. You shouldn't
631 memory can be freed within the host for other uses.
697 saving memory until one or another app needs to modify the content.
708 This is the portion of low virtual memory which should be protected
728 bool "Enable recovery from hardware memory errors"
731 Enables code to recover from some memory failures on systems
733 even when some of its memory has uncorrected errors. This requires
734 special hardware support and typically ECC memory.
747 of memory on which to store mappings, but it can only ask the system
782 from access to large folios for zeroing memory.
785 once and never freed. One full huge page's worth of memory shall
788 Say Y if your system has lots of memory. Say N if you are
789 memory constrained.
804 applications by speeding up page faults during memory
808 If memory constrained on embedded, you may want to say N.
823 memory footprint of applications without a guaranteed
832 memory footprint of applications without a guaranteed
867 increase the memory footprint of applications without a
912 increase the memory footprint of applications without a
1030 bool "Contiguous Memory Allocator"
1035 This enables the Contiguous Memory Allocator which allows other
1036 subsystems to allocate big physically-contiguous blocks of memory.
1037 CMA reserves a region of memory and allows only movable pages to
1038 be allocated from it. This way, the kernel can use the memory for
1105 bool "Track memory changes"
1109 This option enables memory changes tracking by introducing a
1162 be useful to tune memory cgroup limits and/or for job placement
1199 bool "Device memory (pmem, HMM, etc...) hotplug support"
1206 Device memory hotplug support allows for establishing pmem,
1207 or other device driver discovered memory regions, in the
1212 Enabling this option will reduce the entropy of x86 KASLR memory
1216 (5 level page table) and physical memory present on the system.
1230 bool "Unaddressable device memory (GPU memory, ...)"
1236 memory; i.e., memory that is only accessible from the device (or
1262 bool "Collect percpu memory statistics"
1266 be used to help understand percpu memory usage.
1324 memory areas visible only in the context of the owning process and
1332 Allow naming anonymous virtual memory areas.
1334 This feature allows assigning names to virtual memory areas. Assigned
1336 and help identifying individual anonymous memory areas.
1337 Assigning a name to anonymous virtual memory area might prevent that
1338 area from being merged with adjacent virtual memory areas due to the
1367 file-backed memory types like shmem and hugetlbfs.
1377 A high performance LRU implementation to overcommit memory. See
1393 This option has a per-memcg and per-node memory overhead.
1409 This feature allows locking each virtual memory area separately when