| /linux/Documentation/admin-guide/mm/ |
| H A D | memory-hotplug.rst | 2 Memory Hot(Un)Plug
5 This document describes generic Linux support for memory hot(un)plug with
13 Memory hot(un)plug allows for increasing and decreasing the size of physical
14 memory available to a machine at runtime. In the simplest case, it consists of
18 Memory hot(un)plug is used for various purposes:
20 - The physical memory available to a machine can be adjusted at runtime, up- or
21 downgrading the memory capacity. This dynamic memory resizing, sometimes
26 example is replacing failing memory modules.
28 - Reducing energy consumption either by physically unplugging memory modules or
29 by logically unplugging (parts of) memory modules from Linux.
[all …]
|
| H A D | concepts.rst | 5 The memory management in Linux is a complex system that evolved over the
7 systems from MMU-less microcontrollers to supercomputers. The memory
16 Virtual Memory Primer
19 The physical memory in a computer system is a limited resource and
20 even for systems that support memory hotplug there is a hard limit on
21 the amount of memory that can be installed. The physical memory is not
27 All this makes dealing directly with physical memory quite complex and
28 to avoid this complexity a concept of virtual memory was developed.
30 The virtual memory abstracts the details of physical memory from the
32 physical memory (demand paging) and provides a mechanism for the
[all …]
|
| H A D | numaperf.rst | 2 NUMA Memory Performance
8 Some platforms may have multiple types of memory attached to a compute
9 node. These disparate memory ranges may share some characteristics, such
13 A system supports such heterogeneous memory by grouping each memory type
15 characteristics. Some memory may share the same node as a CPU, and others
16 are provided as memory only nodes. While memory only nodes do not provide
19 nodes with local memory and a memory only node for each of compute node::
30 A "memory initiator" is a node containing one or more devices such as
31 CPUs or separate memory I/O devices that can initiate memory requests.
32 A "memory target" is a node containing one or more physical address
[all …]
|
| /linux/tools/testing/selftests/memory-hotplug/ |
| H A D | mem-on-off-test.sh | 25 if ! ls $SYSFS/devices/system/memory/memory* > /dev/null 2>&1; then
26 echo $msg memory hotplug is not supported >&2
30 if ! grep -q 1 $SYSFS/devices/system/memory/memory*/removable; then
31 echo $msg no hot-pluggable memory >&2
37 # list all hot-pluggable memory
43 for memory in $SYSFS/devices/system/memory/memory*; do
44 if grep -q 1 $memory/removable &&
45 grep -q $state $memory/state; then
46 echo ${memory##/*/memory}
63 grep -q online $SYSFS/devices/system/memory/memory$1/state
[all …]
|
| /linux/Documentation/admin-guide/cgroup-v1/ |
| H A D | memory.rst | 2 Memory Resource Controller
12 The Memory Resource Controller has generically been referred to as the
13 memory controller in this document. Do not confuse memory controller
14 used here with the memory controller that is used in hardware.
17 When we mention a cgroup (cgroupfs's directory) with memory controller,
18 we call it "memory cgroup". When you see git-log and source code, you'll
22 Benefits and Purpose of the memory controller
25 The memory controller isolates the memory behaviour of a group of tasks
27 uses of the memory controller. The memory controller can be used to
30 Memory-hungry applications can be isolated and limited to a smaller
[all …]
|
| /linux/Documentation/ABI/testing/ |
| H A D | sysfs-devices-memory | 1 What: /sys/devices/system/memory
5 The /sys/devices/system/memory contains a snapshot of the
6 internal state of the kernel memory blocks. Files could be
9 Users: hotplug memory add/remove tools
12 What: /sys/devices/system/memory/memoryX/removable
16 The file /sys/devices/system/memory/memoryX/removable is a
17 legacy interface used to indicated whether a memory block is
19 "1" if and only if the kernel supports memory offlining.
20 Users: hotplug memory remove tools
24 What: /sys/devices/system/memory/memoryX/phys_device
[all …]
|
| H A D | sysfs-edac-memory-repair | 7 pertains to the memory media repair features control, such as
8 PPR (Post Package Repair), memory sparing etc, where <dev-name>
10 device driver for the memory repair features.
12 Post Package Repair is a maintenance operation requests the memory
13 device to perform a repair operation on its media. It is a memory
14 self-healing feature that fixes a failing memory location by
16 CXL memory device with DRAM components that support PPR features may
28 decoders have been configured), memory devices (e.g. CXL)
30 physical address map. As such, the memory to repair must be
40 (RO) Memory repair type. For eg. post package repair,
[all …]
|
| /linux/Documentation/mm/ |
| H A D | memory-model.rst | 4 Physical Memory Model
7 Physical memory in a system may be addressed in different ways. The
8 simplest case is when the physical memory starts at address 0 and
13 different memory banks are attached to different CPUs.
15 Linux abstracts this diversity using one of the two memory models:
17 memory models it supports, what the default memory model is and
20 All the memory models track the status of physical page frames using
23 Regardless of the selected memory model, there exists one-to-one
27 Each memory model defines :c:func:`pfn_to_page` and :c:func:`page_to_pfn`
34 The simplest memory model is FLATMEM. This model is suitable for
[all …]
|
| H A D | hmm.rst | 2 Heterogeneous Memory Management (HMM)
5 Provide infrastructure and helpers to integrate non-conventional memory (device
6 memory like GPU on board memory) into regular kernel path, with the cornerstone
7 of this being specialized struct page for such memory (see sections 5 to 7 of
10 HMM also provides optional helpers for SVM (Share Virtual Memory), i.e.,
18 related to using device specific memory allocators. In the second section, I
22 fifth section deals with how device memory is represented inside the kernel.
28 Problems of using a device specific memory allocator
31 Devices with a large amount of on board memory (several gigabytes) like GPUs
32 have historically managed their memory through dedicated driver specific APIs.
[all …]
|
| H A D | numa.rst | 12 or more CPUs, local memory, and/or IO buses. For brevity and to
26 Coherent NUMA or ccNUMA systems. With ccNUMA systems, all memory is visible
30 Memory access time and effective memory bandwidth varies depending on how far
31 away the cell containing the CPU or IO bus making the memory access is from the
32 cell containing the target memory. For example, access to memory by CPUs
34 bandwidths than accesses to memory on other, remote cells. NUMA platforms
39 memory bandwidth. However, to achieve scalable memory bandwidth, system and
40 application software must arrange for a large majority of the memory references
41 [cache misses] to be to "local" memory--memory on the same cell, if any--or
42 to the closest cell with memory.
[all …]
|
| /linux/Documentation/devicetree/bindings/memory-controllers/fsl/ |
| H A D | fsl,ddr.yaml | 4 $id: http://devicetree.org/schemas/memory-controllers/fsl/fsl,ddr.yaml#
7 title: Freescale DDR memory controller
15 pattern: "^memory-controller@[0-9a-f]+$"
21 - fsl,qoriq-memory-controller-v4.4
22 - fsl,qoriq-memory-controller-v4.5
23 - fsl,qoriq-memory-controller-v4.7
24 - fsl,qoriq-memory-controller-v5.0
25 - const: fsl,qoriq-memory-controller
27 - fsl,bsc9132-memory-controller
28 - fsl,mpc8536-memory-controller
[all …]
|
| /linux/Documentation/edac/ |
| H A D | memory_repair.rst | 4 EDAC Memory Repair Control
20 Some memory devices support repair operations to address issues in their
21 memory media. Post Package Repair (PPR) and memory sparing are examples of
27 Post Package Repair is a maintenance operation which requests the memory
28 device to perform repair operation on its media. It is a memory self-healing
29 feature that fixes a failing memory location by replacing it with a spare row
32 For example, a CXL memory device with DRAM components that support PPR
42 The data may not be retained and memory requests may not be correctly
46 For example, for CXL memory devices, see CXL spec rev 3.1 [1]_ sections
50 Memory Sparing
[all …]
|
| /linux/drivers/cxl/ |
| H A D | Kconfig | 16 memory targets, the CXL.io protocol is equivalent to PCI Express.
26 The CXL specification defines a "CXL memory device" sub-class in the
27 PCI "memory controller" base class of devices. Device's identified by
29 memory to be mapped into the system address map (Host-managed Device
30 Memory (HDM)).
32 Say 'y/m' to enable a driver that will attach to CXL memory expander
33 devices enumerated by the memory device class code for configuration
40 bool "RAW Command Interface for Memory Devices"
53 potential impact to memory currently in use by the kernel.
66 Enable support for host managed device memory (HDM) resources
[all …]
|
| /linux/Documentation/arch/arm64/ |
| H A D | kdump.rst | 2 crashkernel memory reservation on arm64
9 reserved memory is needed to pre-load the kdump kernel and boot such
12 That reserved memory for kdump is adapted to be able to minimally
19 Through the kernel parameters below, memory can be reserved accordingly
21 large chunk of memomy can be found. The low memory reservation needs to
22 be considered if the crashkernel is reserved from the high memory area.
28 Low memory and high memory
31 For kdump reservations, low memory is the memory area under a specific
34 vmcore dumping can be ignored. On arm64, the low memory upper bound is
37 whole system RAM is low memory. Outside of the low memory described
[all …]
|
| /linux/Documentation/core-api/ |
| H A D | memory-hotplug.rst | 4 Memory hotplug
7 Memory hotplug event notifier
12 Memory notifier
15 There are six types of notification defined in ``include/linux/memory.h``:
18 Generated before new memory becomes available in order to be able to
19 prepare subsystems to handle memory. The page allocator is still unable
20 to allocate from the new memory.
26 Generated when memory has successfully brought online. The callback may
27 allocate pages from the new memory.
30 Generated to begin the process of offlining memory. Allocations are no
[all …]
|
| /linux/tools/testing/selftests/cgroup/ |
| H A D | test_memcontrol.c | 107 * the memory controller.
115 /* Create two nested cgroups with the memory controller enabled */ in test_memcg_subtree_control()
124 if (cg_write(parent, "cgroup.subtree_control", "+memory")) in test_memcg_subtree_control()
130 if (cg_read_strstr(child, "cgroup.controllers", "memory")) in test_memcg_subtree_control()
133 /* Create two nested cgroups without enabling memory controller */ in test_memcg_subtree_control()
148 if (!cg_read_strstr(child2, "cgroup.controllers", "memory")) in test_memcg_subtree_control()
187 current = cg_read_long(cgroup, "memory.current"); in alloc_anon_50M_check()
194 anon = cg_read_key_long(cgroup, "memory.stat", "anon "); in alloc_anon_50M_check()
221 current = cg_read_long(cgroup, "memory.current"); in alloc_pagecache_50M_check()
225 file = cg_read_key_long(cgroup, "memory.stat", "file "); in alloc_pagecache_50M_check()
[all …]
|
| /linux/tools/perf/pmu-events/arch/arm64/arm/neoverse-v3/ |
| H A D | memory.json | 4 …memory accesses issued by the CPU load store unit, where those accesses are issued due to load or …
8 …memory errors (ECC or parity) in protected CPUs RAMs. On the core, this event counts errors in the…
16 …memory accesses issued by the CPU due to load operations. The event counts any memory load access,…
20 …memory accesses issued by the CPU due to store operations. The event counts any memory store acces…
24 …"PublicDescription": "Counts the number of memory read and write accesses in a cycle that incurred…
28 …"PublicDescription": "Counts the number of memory read accesses in a cycle that incurred additiona…
32 …"PublicDescription": "Counts the number of memory write access in a cycle that incurred additional…
36 …cription": "Counts the number of memory read and write accesses counted by MEM_ACCESS that are tag…
40 …"PublicDescription": "Counts the number of memory read accesses in a cycle that are tag checked by…
44 …"PublicDescription": "Counts the number of memory write accesses in a cycle that is tag checked by…
[all …]
|
| /linux/drivers/base/ |
| H A D | memory.c | 3 * Memory subsystem support
9 * a SPARSEMEM-memory-model system's physical memory in /sysfs.
19 #include <linux/memory.h>
30 #define MEMORY_CLASS_NAME "memory"
66 * Memory blocks are cached in a local radix tree to avoid
73 * Memory groups, indexed by memory group id (mgid).
106 * memory_block_advise_max_size() - advise memory hotplug on the max suggested
157 /* Show the memory block ID, relative to the memory block size */
241 * they describe (they remain until the memory is unplugged), doing in memory_block_online()
242 * their initialization and accounting at memory onlining/offlining in memory_block_online()
[all …]
|
| /linux/Documentation/driver-api/cxl/platform/ |
| H A D | bios-and-efi.rst | 19 * BIOS/EFI create the system memory map (EFI Memory Map, E820, etc)
24 static memory map configuration. More detail on these tables can be found
29 on physical memory region size and alignment, memory holes, HDM interleave,
39 When this is enabled, this bit tells linux to defer management of a memory
40 region to a driver (in this case, the CXL driver). Otherwise, the memory is
41 treated as "normal memory", and is exposed to the page allocator during
60 Memory Attribute` field. This may be called something else on your platform.
62 :code:`uefisettings get "CXL Memory Attribute"` ::
67 name: "CXL Memory Attribute",
72 Physical Memory Map
[all …]
|
| /linux/tools/perf/pmu-events/arch/arm64/arm/neoverse-n3/ |
| H A D | memory.json | 4 …memory accesses issued by the CPU load store unit, where those accesses are issued due to load or …
12 …memory accesses issued by the CPU due to load operations. The event counts any memory load access,…
16 …memory accesses issued by the CPU due to store operations. The event counts any memory store acces…
20 …"PublicDescription": "Counts the number of memory read and write accesses in a cycle that incurred…
24 …"PublicDescription": "Counts the number of memory read accesses in a cycle that incurred additiona…
28 …"PublicDescription": "Counts the number of memory write access in a cycle that incurred additional…
32 …cription": "Counts the number of memory read and write accesses counted by MEM_ACCESS that are tag…
36 …"PublicDescription": "Counts the number of memory read accesses in a cycle that are tag checked by…
40 …"PublicDescription": "Counts the number of memory write accesses in a cycle that is tag checked by…
48 … "PublicDescription": "Counts the number of outstanding loads or memory read accesses per cycle."
[all …]
|
| /linux/Documentation/driver-api/cxl/linux/ |
| H A D | memory-hotplug.rst | 4 Memory Hotplug
6 The final phase of surfacing CXL memory to the kernel page allocator is for
7 the `DAX` driver to surface a `Driver Managed` memory region via the
8 memory-hotplug component.
13 2) Hotplug Memory Block size
14 3) Memory Map Resource location
15 4) Driver-Managed Memory Designation
19 The default-online behavior of hotplug memory is dictated by the following,
24 - :code:`/sys/devices/system/memory/auto_online_blocks` value
26 These dictate whether hotplugged memory blocks arrive in one of three states:
[all …]
|
| /linux/drivers/gpu/drm/nouveau/nvkm/core/ |
| H A D | memory.c | 24 #include <core/memory.h>
30 nvkm_memory_tags_put(struct nvkm_memory *memory, struct nvkm_device *device, in nvkm_memory_tags_put() argument
39 kfree(memory->tags); in nvkm_memory_tags_put()
40 memory->tags = NULL; in nvkm_memory_tags_put()
48 nvkm_memory_tags_get(struct nvkm_memory *memory, struct nvkm_device *device, in nvkm_memory_tags_get() argument
56 if ((tags = memory->tags)) { in nvkm_memory_tags_get()
57 /* If comptags exist for the memory, but a different amount in nvkm_memory_tags_get()
84 * As memory can be mapped in multiple places, we still in nvkm_memory_tags_get()
94 *ptags = memory->tags = tags; in nvkm_memory_tags_get()
101 struct nvkm_memory *memory) in nvkm_memory_ctor() argument
[all …]
|
| /linux/Documentation/admin-guide/mm/damon/ |
| H A D | reclaim.rst | 8 be used for proactive and lightweight reclamation under light memory pressure.
10 to be selectively used for different level of memory pressure and requirements.
15 On general memory over-committed systems, proactively reclaiming cold pages
16 helps saving memory and reducing latency spikes that incurred by the direct
20 Free Pages Reporting [3]_ based memory over-commit virtualization systems are
22 memory to host, and the host reallocates the reported memory to other guests.
23 As a result, the memory of the systems are fully utilized. However, the
24 guests could be not so memory-frugal, mainly because some kernel subsystems and
25 user-space applications are designed to use as much memory as available. Then,
26 guests could report only small amount of memory as free to host, results in
[all …]
|
| /linux/Documentation/arch/powerpc/ |
| H A D | firmware-assisted-dump.rst | 14 - Fadump uses the same firmware interfaces and memory reservation model
16 - Unlike phyp dump, FADump exports the memory dump through /proc/vmcore
21 - Unlike phyp dump, FADump allows user to release all the memory reserved
35 - Once the dump is copied out, the memory that held the dump
44 - The first kernel registers the sections of memory with the
46 These registered sections of memory are reserved by the first
50 low memory regions (boot memory) from source to destination area.
54 The term 'boot memory' means size of the low memory chunk
56 booted with restricted memory. By default, the boot memory
58 Alternatively, user can also specify boot memory size
[all …]
|
| /linux/Documentation/driver-api/cxl/devices/ |
| H A D | device-types.rst | 7 The type of CXL device (Memory, Accelerator, etc) dictates many configuration steps. This section
23 other than memory (CXL.mem) or cache (CXL.cache) operations.
31 The mechanism by which a device may coherently access and cache host memory.
37 The mechanism by which the CPU may coherently access and cache device memory.
53 * Does NOT have host-managed device memory (HDM)
56 directly operate on host-memory (DMA) to store incoming packets. These
57 devices largely rely on CPU-attached memory.
65 * Optionally implements coherent cache and Host-Managed Device Memory
66 * Is typically an accelerator device with high bandwidth memory.
69 of host-managed device memory, which allows the device to operate on a
[all …]
|