| /linux-4.1.27/Documentation/vm/ |
| D | numa | 3 What is NUMA?
8 From the hardware perspective, a NUMA system is a computer platform that
17 may not be populated on any given cell. The cells of the NUMA system are
19 point-to-point link are common types of NUMA system interconnects. Both of
20 these types of interconnects can be aggregated to create NUMA platforms with
23 For Linux, the NUMA platforms of interest are primarily what is known as Cache
24 Coherent NUMA or ccNUMA systems. With ccNUMA systems, all memory is visible
32 bandwidths than accesses to memory on other, remote cells. NUMA platforms
35 Platform vendors don't build NUMA systems just to make software developers'
42 This leads to the Linux software view of a NUMA system:
[all …]
|
| D | 00-INDEX | 20 - information about NUMA specific code in the Linux vm.
26 - description of page migration in NUMA systems.
|
| D | hugetlbpage.txt | 90 On a NUMA platform, the kernel will attempt to distribute the huge page pool
91 over all the set of allowed nodes specified by the NUMA memory policy of the
102 some nodes in a NUMA system, it will attempt to make up the difference by
111 distribution of huge pages in a NUMA system, use:
171 the /sysfs interface using the nr_hugepages_mempolicy attribute, the NUMA
173 NUMA memory policy of the task that modifies the nr_hugepages_mempolicy
237 NUMA node with memory in:
|
| D | ksm.txt | 63 reside in the memory area of same NUMA node. That brings
65 nodes, at significant NUMA distances, are likely to benefit
|
| D | numa_memory_policy.txt | 5 allocate memory in a NUMA system or in an emulated NUMA system. Linux has
15 programming interface that a NUMA-aware application can take advantage of. When
88 inheritable across exec(). Thus, only NUMA-aware applications may
311 and NUMA nodes. "Usage" here means one of the following:
358 on different NUMA nodes. This extra overhead can be avoided by always
|
| D | page_migration | 34 A NUMA profiler may f.e. obtain a log showing frequent off node
|
| D | unevictable-lru.txt | 59 To illustrate this with an example, a non-NUMA x86_64 platform with 128GB of
|
| /linux-4.1.27/kernel/ |
| D | Kconfig.hz | 11 beneficial for servers and NUMA systems that do not need to have
22 100 Hz is a typical choice for servers, SMP and NUMA systems
31 on SMP and NUMA systems. If you are going to be using NTSC video
39 on SMP and NUMA systems and exactly dividing by both PAL and
|
| /linux-4.1.27/arch/mips/sgi-ip27/ |
| D | Kconfig | 30 NUMA. If you need to understand it, read the source code.
38 nodes in a NUMA cluster. This trades memory for speed.
45 across multiple nodes in a NUMA cluster. This trades memory for
|
| /linux-4.1.27/arch/frv/include/asm/ |
| D | topology.h | 6 #error NUMA not supported yet
|
| /linux-4.1.27/arch/ia64/ |
| D | Kconfig | 169 select NUMA
216 select NUMA
227 select NUMA
415 for architectures which are either NUMA (Non-Uniform Memory Access)
431 config NUMA config
432 bool "NUMA support"
437 Say Y to compile the kernel to support NUMA (Non-Uniform Memory
469 def_bool NUMA && SPARSEMEM
473 depends on NUMA
477 depends on NUMA
[all …]
|
| /linux-4.1.27/arch/metag/mm/ |
| D | Kconfig | 41 config NUMA config
42 bool "Non Uniform Memory Access (NUMA) Support"
87 depends on !NUMA
|
| /linux-4.1.27/Documentation/x86/x86_64/ |
| D | fake-numa-for-cpusets | 6 you can create fake NUMA nodes that represent contiguous chunks of memory and
16 For the purposes of this introduction, we'll assume a very primitive NUMA
58 case (i.e. running the same 'dd' command without assigning it to a fake NUMA
|
| D | boot-options.txt | 139 NUMA
141 numa=off Only set up a single NUMA node spanning all memory.
143 numa=noacpi Don't parse the SRAT table for NUMA setup
|
| /linux-4.1.27/Documentation/scheduler/ |
| D | sched-domains.txt | 55 node. Each group being a single physical CPU. Then with NUMA, the parent
57 cpumask of a node. Or, you could do multi-level NUMA or Opteron, for example,
58 might have just one domain covering its one NUMA level.
66 retain the traditional SMT->SMP->NUMA topology (or some subset of that). This
|
| /linux-4.1.27/kernel/sched/ |
| D | features.h | 82 SCHED_FEAT(NUMA, false)
|
| D | sched.h | 996 #define numabalancing_enabled sched_feat_numa(NUMA)
|
| D | fair.c | 5515 if (!sched_feat(NUMA) || !sched_feat(NUMA_RESIST_LOWER)) in migrate_degrades_locality()
|
| D | core.c | 6515 SD_INIT_NAME(NUMA) in sched_init_numa()
|
| /linux-4.1.27/kernel/irq/ |
| D | Kconfig | 98 ( Sparse irqs can also be beneficial on NUMA boxes, as they spread
99 out the interrupt descriptors in a more NUMA-friendly way. )
|
| /linux-4.1.27/arch/x86/realmode/rm/ |
| D | trampoline_32.S | 32 wbinvd # Needed for NUMA-Q should be harmless for others
|
| /linux-4.1.27/Documentation/filesystems/ |
| D | tmpfs.txt | 81 tmpfs has a mount option to set the NUMA memory allocation policy for
105 NUMA memory allocation policies have optional flags that can be used in
118 running kernel does not support NUMA; and will fail if its nodelist
121 NUMA capability (perhaps a safe recovery kernel), or with fewer nodes
|
| D | proc.txt | 217 Ngid NUMA group ID (0 if none)
536 "policy" reports the NUMA memory policy set for the mapping (see vm/numa_memory_policy.txt);
941 N<node>=nr (Only on NUMA kernels)
|
| /linux-4.1.27/mm/ |
| D | ksm.c | 45 #define NUMA(x) (x) macro
48 #define NUMA(x) (0) macro
492 return ksm_merge_across_nodes ? 0 : NUMA(pfn_to_nid(kpfn)); in get_kpfn_nid()
513 root_stable_tree + NUMA(stable_node->nid)); in remove_node_from_stable_tree()
649 root_unstable_tree + NUMA(rmap_item->nid)); in remove_rmap_item_from_tree()
1203 NUMA(stable_node->nid)) { in stable_tree_search()
1427 get_kpfn_nid(stable_node->kpfn) != NUMA(stable_node->nid)) { in cmp_and_merge_page()
1429 root_stable_tree + NUMA(stable_node->nid)); in cmp_and_merge_page()
|
| D | Kconfig | 21 Some users of more advanced features like NUMA and
43 Many NUMA configurations will have this as the only option.
81 # Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
87 depends on DISCONTIGMEM || NUMA
157 depends on NUMA
269 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
273 two situations. The first is on NUMA systems to put pages nearer
|
| D | memory.c | 75 #warning Unfortunate NUMA and NUMA Balancing config, growing page-frame for last_cpupid.
|
| /linux-4.1.27/tools/perf/Documentation/ |
| D | perf-bench.txt | 56 NUMA scheduling and MM benchmarks.
203 Suite for evaluating NUMA workloads.
|
| /linux-4.1.27/arch/sh/mm/ |
| D | Kconfig | 111 config NUMA config
112 bool "Non Uniform Memory Access (NUMA) Support"
131 depends on !NUMA
|
| /linux-4.1.27/arch/tile/ |
| D | Kconfig | 257 # We do not currently support disabling NUMA.
258 config NUMA config
259 bool # "NUMA Memory Allocation and Scheduler Support"
263 NUMA memory allocation is required for TILE processors
|
| /linux-4.1.27/Documentation/ABI/testing/ |
| D | sysfs-kernel-mm-ksm | 49 Description: Control merging pages across different NUMA nodes.
|
| D | sysfs-devices-system-cpu | 59 Description: Discover NUMA node a CPU belongs to
62 to the corresponding NUMA node directory.
65 in NUMA node 2:
|
| D | sysfs-devices-memory | 77 points to the corresponding NUMA node directory.
|
| D | sysfs-bus-pci | 289 This file contains the NUMA node to which the PCI device is
|
| D | sysfs-kernel-slab | 116 are active and their NUMA locality.
|
| /linux-4.1.27/arch/x86/ |
| D | Kconfig | 417 depends on NUMA
442 depends on NUMA
839 bool "Enable Maximum number of SMP Processors and NUMA Nodes"
843 Enable maximum number of CPUS and NUMA Nodes for this architecture.
1304 # Common NUMA Features
1305 config NUMA config in Processor type and features""choice609114cf0104
1311 Enable NUMA (Non Uniform Memory Access) support.
1315 NUMA awareness to the kernel.
1318 (or later), AMD Opteron, or EM64T NUMA.
1321 kernel on a 64-bit NUMA platform.
[all …]
|
| /linux-4.1.27/Documentation/sysctl/ |
| D | vm.txt | 442 This is available only on NUMA kernels.
447 This insures that the slab growth stays under control even in NUMA
460 This is available only on NUMA kernels.
524 This sysctl is only for NUMA.
529 In non-NUMA case, a zonelist for GFP_KERNEL is ordered as following.
534 In NUMA case, you can think of following 2 types of order.
535 Assume 2 node NUMA and below is zonelist of Node(0)'s GFP_KERNEL
798 such that each partition fits within a NUMA node and that accessing remote
|
| D | kernel.txt | 444 Enables/disables automatic page fault based NUMA memory
448 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
457 guarantee. If the target workload is already bound to NUMA nodes then this
459 feature is too high then the rate the kernel samples for NUMA hinting
469 Automatic NUMA balancing scans tasks address space and unmaps pages to
|
| /linux-4.1.27/Documentation/cgroups/ |
| D | cpusets.txt | 74 non-uniform access times (NUMA) presents additional challenges for
92 * NUMA systems running large HPC applications with demanding
326 except perhaps as modified by the task's NUMA mempolicy or cpuset
333 or slab caches to ignore the task's NUMA mempolicy and be spread
334 instead. Tasks using mbind() or set_mempolicy() calls to set NUMA
337 is turned off, then the currently specified NUMA mempolicy once again
560 3 : search cpus in a node [= system wide on non-NUMA system]
561 4 : search nodes in a chunk of node [on NUMA system]
562 5 : search system wide [on NUMA system]
610 was MPOL_BIND bound to the new cpuset (even though its NUMA placement,
|
| D | memcg_test.txt | 138 For NUMA, migration is an another special case. To do easy test, cpuset
|
| /linux-4.1.27/Documentation/ |
| D | memory-hotplug.txt | 47 (B) For installing/removing DIMMs or NUMA-nodes physically.
48 This is to exchange DIMMs/NUMA-nodes, reduce power consumption, etc.
125 - As a related configuration, if your box has a feature of NUMA-node hotplug
214 If firmware supports NUMA-node hotplug, and defines an object _HID "ACPI0004",
|
| D | kernel-parameters.txt | 91 NUMA NUMA support is enabled.
1226 hashdist= [KNL,NUMA] Large hashes allocated during boot
1227 are distributed across NUMA nodes. Defaults on
1228 for 64-bit NUMA, off otherwise.
2311 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2536 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2539 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3501 pernode one pool for each NUMA node (equivalent
3502 to global on non-NUMA machines)
3626 topology_updates= [KNL, PPC, NUMA]
[all …]
|
| D | cpu-hotplug.txt | 27 support NUMA hardware which are hot pluggable as well, where physical
|
| D | memory-barriers.txt | 2178 Under certain circumstances (especially involving NUMA), I/O accesses within
|
| /linux-4.1.27/Documentation/devicetree/bindings/arm/ |
| D | fw-cfg.txt | 31 virtual NUMA topology, and so on.
|
| /linux-4.1.27/arch/powerpc/ |
| D | Kconfig | 461 config NUMA config
462 bool "NUMA support"
474 depends on NUMA
478 depends on NUMA
486 depends on (PPC64 && !NUMA) || PPC32
506 # Some NUMA nodes have memory ranges that span
|
| /linux-4.1.27/arch/m32r/ |
| D | Kconfig | 316 # Common NUMA Features
317 config NUMA config
|
| /linux-4.1.27/arch/sparc/ |
| D | Kconfig | 283 config NUMA config
284 bool "NUMA support"
292 # Some NUMA nodes have memory ranges that span
|
| /linux-4.1.27/arch/alpha/ |
| D | Kconfig | 584 for architectures which are either NUMA (Non-Uniform Memory Access)
590 config NUMA config
591 bool "NUMA Support (EXPERIMENTAL)"
594 Say Y to compile the kernel to support NUMA (Non-Uniform Memory
|
| /linux-4.1.27/init/ |
| D | Kconfig | 632 strong NUMA behavior.
900 # For architectures that want to enable the support for NUMA-affine scheduler
912 # For architectures that (ab)use NUMA to represent different memory regions
919 bool "Memory placement aware NUMA scheduler"
922 depends on SMP && NUMA && MIGRATION
924 This option adds support for automatic NUMA aware memory/task placement.
931 bool "Automatically enable NUMA aware memory/task placement"
935 If set, automatic NUMA balancing will be enabled if running on a NUMA
982 This is primarily useful on large SMP or NUMA systems.
|
| /linux-4.1.27/Documentation/networking/ |
| D | ixgbe.txt | 190 0 - n: where n is the number of NUMA nodes (i.e. 0 - 3) currently online in
194 The Node parameter will allow you to pick which NUMA node you want to have
|
| D | scaling.txt | 56 share a particular memory level (L1, L2, NUMA node, etc.).
156 CPU. If NUMA locality is not an issue, this could also be all CPUs in
|
| /linux-4.1.27/Documentation/arm64/ |
| D | acpi_object_usage.txt | 210 Optional in general, but required for NUMA systems.
223 To support NUMA, this table is required.
592 -- Section 17: NUMA support (prototypes have been submitted for
|
| /linux-4.1.27/drivers/acpi/ |
| D | Kconfig | 228 bool "NUMA support"
229 depends on NUMA
|
| /linux-4.1.27/Documentation/virtual/kvm/ |
| D | timekeeping.txt | 370 3.3) TSC and multi-socket / NUMA
385 It is recommended not to trust the TSCs to remain synchronized on NUMA or
445 if so, the TSCs in multi-sockets or NUMA systems may still run independently
|
| /linux-4.1.27/arch/mn10300/ |
| D | Kconfig | 38 config NUMA config
|
| /linux-4.1.27/arch/mips/ |
| D | Kconfig | 2396 depends on !NUMA && !CPU_LOONGSON2
2403 for architectures which are either NUMA (Non-Uniform Memory Access)
2411 config NUMA config
2412 bool "NUMA Support"
2415 Say Y to compile the kernel to support NUMA (Non-Uniform Memory
|
| /linux-4.1.27/Documentation/RCU/ |
| D | RTFP.txt | 496 Describes the parallel RCU infrastructure. Includes NUMA aspect
516 SMP/NUMA computer systems.
818 Performance of dcache RCU on kernbench for 16x NUMA-Q and 1x,
|
| /linux-4.1.27/ |
| D | CREDITS | 1884 D: network, x86, NUMA, various hacks
2085 D: NUMA support, Slab allocators, Page migration
|