Commit | Line | Data |
---|---|---|
ec8f24b7 | 1 | # SPDX-License-Identifier: GPL-2.0-only |
18da2c9e | 2 | menuconfig LIBNVDIMM |
b94d5230 DW |
3 | tristate "NVDIMM (Non-Volatile Memory Device) Support" |
4 | depends on PHYS_ADDR_T_64BIT | |
29b9aa0a | 5 | depends on HAS_IOMEM |
b94d5230 | 6 | depends on BLK_DEV |
33dd7075 | 7 | select MEMREGION |
b94d5230 DW |
8 | help |
9 | Generic support for non-volatile memory devices including | |
10 | ACPI-6-NFIT defined resources. On platforms that define an | |
11 | NFIT, or otherwise can discover NVDIMM resources, a libnvdimm | |
12 | bus is registered to advertise PMEM (persistent memory) | |
f8669f1d | 13 | namespaces (/dev/pmemX). A PMEM namespace refers to a |
5212e11f | 14 | memory resource that may span multiple DIMMs and support DAX |
f8669f1d | 15 | (see CONFIG_DAX). |
18da2c9e DW |
16 | |
17 | if LIBNVDIMM | |
18 | ||
19 | config BLK_DEV_PMEM | |
20 | tristate "PMEM: Persistent memory block device support" | |
21 | default LIBNVDIMM | |
afd586f0 | 22 | select DAX |
5212e11f | 23 | select ND_BTT if BTT |
e1455744 | 24 | select ND_PFN if NVDIMM_PFN |
18da2c9e DW |
25 | help |
26 | Memory ranges for PMEM are described by either an NFIT | |
27 | (NVDIMM Firmware Interface Table, see CONFIG_NFIT_ACPI), a | |
28 | non-standard OEM-specific E820 memory type (type-12, see | |
29 | CONFIG_X86_PMEM_LEGACY), or it is manually specified by the | |
30 | 'memmap=nn[KMG]!ss[KMG]' kernel command line (see | |
8c27ceff | 31 | Documentation/admin-guide/kernel-parameters.rst). This driver converts |
18da2c9e DW |
32 | these persistent memory ranges into block devices that are |
33 | capable of DAX (direct-access) file system mappings. See | |
ae4a0502 | 34 | Documentation/driver-api/nvdimm/nvdimm.rst for more details. |
18da2c9e DW |
35 | |
36 | Say Y if you want to use an NVDIMM | |
37 | ||
e1455744 DW |
38 | config ND_CLAIM |
39 | bool | |
40 | ||
5212e11f VV |
41 | config ND_BTT |
42 | tristate | |
43 | ||
8c2f7e86 | 44 | config BTT |
5212e11f VV |
45 | bool "BTT: Block Translation Table (atomic sector updates)" |
46 | default y if LIBNVDIMM | |
e1455744 | 47 | select ND_CLAIM |
5212e11f VV |
48 | help |
49 | The Block Translation Table (BTT) provides atomic sector | |
50 | update semantics for persistent memory devices, so that | |
51 | applications that rely on sector writes not being torn (a | |
52 | guarantee that typical disks provide) can continue to do so. | |
53 | The BTT manifests itself as an alternate personality for an | |
f8669f1d DW |
54 | NVDIMM namespace, i.e. a namespace can be in raw mode pmemX, |
55 | or 'sectored' mode. | |
5212e11f VV |
56 | |
57 | Select Y if unsure | |
8c2f7e86 | 58 | |
e1455744 DW |
59 | config ND_PFN |
60 | tristate | |
61 | ||
62 | config NVDIMM_PFN | |
63 | bool "PFN: Map persistent (device) memory" | |
64 | default LIBNVDIMM | |
32ab0a3f | 65 | depends on ZONE_DEVICE |
e1455744 DW |
66 | select ND_CLAIM |
67 | help | |
68 | Map persistent memory, i.e. advertise it to the memory | |
69 | management sub-system. By default persistent memory does | |
70 | not support direct I/O, RDMA, or any other usage that | |
71 | requires a 'struct page' to mediate an I/O request. This | |
72 | driver allocates and initializes the infrastructure needed | |
73 | to support those use cases. | |
74 | ||
75 | Select Y if unsure | |
76 | ||
cd03412a | 77 | config NVDIMM_DAX |
867dfe34 | 78 | bool "NVDIMM DAX: Raw access to persistent memory" |
cd03412a DW |
79 | default LIBNVDIMM |
80 | depends on NVDIMM_PFN | |
81 | help | |
82 | Support raw device dax access to a persistent memory | |
83 | namespace. For environments that want to hard partition | |
83fc61a5 | 84 | persistent memory, this capability provides a mechanism to |
cd03412a DW |
85 | sub-divide a namespace into character devices that can only be |
86 | accessed via DAX (mmap(2)). | |
87 | ||
88 | Select Y if unsure | |
89 | ||
71719760 | 90 | config OF_PMEM |
f22acf82 | 91 | tristate "Device-tree support for persistent memory regions" |
71719760 OH |
92 | depends on OF |
93 | default LIBNVDIMM | |
94 | help | |
95 | Allows regions of persistent memory to be described in the | |
96 | device-tree. | |
97 | ||
98 | Select Y if unsure. | |
99 | ||
4c6926a2 DJ |
100 | config NVDIMM_KEYS |
101 | def_bool y | |
102 | depends on ENCRYPTED_KEYS | |
103 | depends on (LIBNVDIMM=ENCRYPTED_KEYS) || LIBNVDIMM=m | |
104 | ||
c91d7136 DW |
105 | config NVDIMM_KMSAN |
106 | bool | |
107 | depends on KMSAN | |
108 | help | |
109 | KMSAN, and other memory debug facilities, increase the size of | |
110 | 'struct page' to contain extra metadata. This collides with | |
111 | the NVDIMM capability to store a potentially | |
112 | larger-than-"System RAM" size 'struct page' array in a | |
113 | reservation of persistent memory rather than limited / | |
114 | precious DRAM. However, that reservation needs to persist for | |
115 | the life of the given NVDIMM namespace. If you are using KMSAN | |
116 | to debug an issue unrelated to NVDIMMs or DAX then say N to this | |
117 | option. Otherwise, say Y but understand that any namespaces | |
118 | (with the page array stored pmem) created with this build of | |
119 | the kernel will permanently reserve and strand excess | |
120 | capacity compared to the CONFIG_KMSAN=n case. | |
121 | ||
122 | Select N if unsure. | |
123 | ||
62974fc3 DW |
124 | config NVDIMM_TEST_BUILD |
125 | tristate "Build the unit test core" | |
126 | depends on m | |
127 | depends on COMPILE_TEST && X86_64 | |
128 | default m if COMPILE_TEST | |
129 | help | |
130 | Build the core of the unit test infrastructure. The result of | |
131 | this build is non-functional for unit test execution, but it | |
132 | otherwise helps catch build errors induced by changes to the | |
133 | core devm_memremap_pages() implementation and other | |
134 | infrastructure. | |
135 | ||
15a83487 DJ |
136 | config NVDIMM_SECURITY_TEST |
137 | bool "Enable NVDIMM security unit tests" | |
138 | depends on NVDIMM_KEYS | |
139 | help | |
140 | The NVDIMM and CXL subsystems support unit testing of their device | |
141 | security state machines. The NVDIMM_SECURITY_TEST option disables CPU | |
142 | cache maintenance operations around events like secure erase and | |
143 | overwrite. Also, when enabled, the NVDIMM subsystem core helps the unit | |
144 | test implement a mock state machine. | |
145 | ||
146 | Select N if unsure. | |
147 | ||
18da2c9e | 148 | endif |