linux-6.12.1/mm/Kconfig

1 # SPDX-License-Identifier: GPL-2.0-only
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.
15 menuconfig SWAP  config
16 	bool "Support for paging of anonymous memory (swap)"
21 	  for so called swap devices or swap files in your kernel that are
26 	bool "Compressed cache for swap pages"
27 	depends on SWAP
31 	  A lightweight compressed cache for swap pages.  It takes
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,
35 	  in the case where decompressing from RAM is faster than swap device
39 	bool "Enable the compressed cache for swap pages by default"
42 	  If selected, the compressed cache for swap pages will be enabled
55 	  written back to the backing swap device) on memory pressure.
68 	  for swap pages.
135 	  swap pages.
189 	  linux-mm@kvack.org and the zswap maintainers.
207 	  zsmalloc is a slab-based memory allocator designed to store
222 	int "Maximum number of physical pages per-zspage"
295 	  specifically-sized allocations with user-controlled contents
299 	  user-controlled allocations. This may very slightly increase
301 	  of extra pages since the bulk of user-controlled allocations
302 	  are relatively long-lived.
317 	  Try running: slabinfo -DA
354 	  utilization of a direct-mapped memory-side-cache. See section
357 	  the presence of a memory-side-cache. There are also incidental
382 	  On non-ancient distros (post-2000 ones) N is usually a safe choice.
397 	  ELF-FDPIC binfmt's brk and stack allocator.
401 	  userspace.  Since that isn't generally a problem on no-MMU systems,
404 	  See Documentation/admin-guide/mm/nommu-mmap.rst for more information.
425 	  This option is best suited for non-NUMA systems with
441 	  memory hot-plug systems.  This is normal.
445 	  hot-plug and hot-remove.
511 # Also, memblocks are updated with memory hot(un)plug.
515 # Keep arch NUMA mapping infrastructure post-init.
561 	  See Documentation/admin-guide/mm/memory-hotplug.rst for more information.
563 	  Say Y here if you want all hot-plugged memory blocks to appear in
565 	  Say N here if you want the default policy to keep all hot-plugged
569 	bool "Allow for memory hot remove"
584 # Heavily threaded applications may benefit from splitting the mm-wide
588 # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
589 # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
591 # a per-page lock leads to problems when multiple tables need to be locked
646 	  linux-mm@kvack.org.
702 	int "Maximum scale factor of PCP (Per-CPU pageset) batch allocate/free"
706 	  In page allocator, PCP (Per-CPU pageset) is refilled and drained in
789 	  allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
798 	  long-term mappings means that the space is wasted.
808 	  See Documentation/admin-guide/mm/nommu-mmap.rst for more information.
865 	depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP && 64BIT
867 	  Swap transparent huge pages in one piece, without splitting.
868 	  XXX: For now, swap cluster backing transparent huge page
874 	bool "Read-only THP for filesystems (EXPERIMENTAL)"
878 	  Allow khugepaged to put read-only file-backed pages in THP.
932 	  subsystems to allocate big physically-contiguous blocks of memory.
971 	  soft-dirty bit on pte-s. This bit it set when someone writes
975 	  See Documentation/admin-guide/mm/soft-dirty.rst for more details.
981 	int "Default maximum user stack size for 32-bit processes (MB)"
986 	  This is the maximum stack size in Megabytes in the VM layout of 32-bit
1026 	  See Documentation/admin-guide/mm/idle_page_tracking.rst for
1042 	  checking, an architecture-agnostic way to find the stack pointer
1074 	  "device-physical" addresses which is needed for using a DAX
1130 	bool "Enable infrastructure for get_user_pages()-related unit tests"
1134 	  to make ioctl calls that can launch kernel-based unit tests for
1139 	  the non-_fast variants.
1141 	  There is also a sub-test that allows running dump_page() on any
1143 	  range of user-space addresses. These pages are either pinned via
1232 	  file-backed memory types like shmem and hugetlbfs.
1235 # multi-gen LRU {
1237 	bool "Multi-Gen LRU"
1239 	# make sure folio->flags has enough spare bits
1243 	  Documentation/admin-guide/mm/multigen_lru.rst for details.
1249 	  This option enables the multi-gen LRU by default.
1258 	  This option has a per-memcg and per-node memory overhead.
1272 	  Allow per-vma locking during page fault handling.