1# SPDX-License-Identifier: GPL-2.0-only 2menuconfig LIBNVDIMM 3 tristate "NVDIMM (Non-Volatile Memory Device) Support" 4 depends on PHYS_ADDR_T_64BIT 5 depends on HAS_IOMEM 6 depends on BLK_DEV 7 select MEMREGION 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) 13 namespaces (/dev/pmemX) and BLK (sliding mmio window(s)) 14 namespaces (/dev/ndblkX.Y). A PMEM namespace refers to a 15 memory resource that may span multiple DIMMs and support DAX 16 (see CONFIG_DAX). A BLK namespace refers to an NVDIMM control 17 region which exposes an mmio register set for windowed access 18 mode to non-volatile memory. 19 20if LIBNVDIMM 21 22config BLK_DEV_PMEM 23 tristate "PMEM: Persistent memory block device support" 24 default LIBNVDIMM 25 select DAX_DRIVER 26 select ND_BTT if BTT 27 select ND_PFN if NVDIMM_PFN 28 help 29 Memory ranges for PMEM are described by either an NFIT 30 (NVDIMM Firmware Interface Table, see CONFIG_NFIT_ACPI), a 31 non-standard OEM-specific E820 memory type (type-12, see 32 CONFIG_X86_PMEM_LEGACY), or it is manually specified by the 33 'memmap=nn[KMG]!ss[KMG]' kernel command line (see 34 Documentation/admin-guide/kernel-parameters.rst). This driver converts 35 these persistent memory ranges into block devices that are 36 capable of DAX (direct-access) file system mappings. See 37 Documentation/driver-api/nvdimm/nvdimm.rst for more details. 38 39 Say Y if you want to use an NVDIMM 40 41config ND_BLK 42 tristate "BLK: Block data window (aperture) device support" 43 default LIBNVDIMM 44 select ND_BTT if BTT 45 help 46 Support NVDIMMs, or other devices, that implement a BLK-mode 47 access capability. BLK-mode access uses memory-mapped-i/o 48 apertures to access persistent media. 49 50 Say Y if your platform firmware emits an ACPI.NFIT table 51 (CONFIG_ACPI_NFIT), or otherwise exposes BLK-mode 52 capabilities. 53 54config ND_CLAIM 55 bool 56 57config ND_BTT 58 tristate 59 60config BTT 61 bool "BTT: Block Translation Table (atomic sector updates)" 62 default y if LIBNVDIMM 63 select ND_CLAIM 64 help 65 The Block Translation Table (BTT) provides atomic sector 66 update semantics for persistent memory devices, so that 67 applications that rely on sector writes not being torn (a 68 guarantee that typical disks provide) can continue to do so. 69 The BTT manifests itself as an alternate personality for an 70 NVDIMM namespace, i.e. a namespace can be in raw mode (pmemX, 71 ndblkX.Y, etc...), or 'sectored' mode, (pmemXs, ndblkX.Ys, 72 etc...). 73 74 Select Y if unsure 75 76config ND_PFN 77 tristate 78 79config NVDIMM_PFN 80 bool "PFN: Map persistent (device) memory" 81 default LIBNVDIMM 82 depends on ZONE_DEVICE 83 select ND_CLAIM 84 help 85 Map persistent memory, i.e. advertise it to the memory 86 management sub-system. By default persistent memory does 87 not support direct I/O, RDMA, or any other usage that 88 requires a 'struct page' to mediate an I/O request. This 89 driver allocates and initializes the infrastructure needed 90 to support those use cases. 91 92 Select Y if unsure 93 94config NVDIMM_DAX 95 bool "NVDIMM DAX: Raw access to persistent memory" 96 default LIBNVDIMM 97 depends on NVDIMM_PFN 98 help 99 Support raw device dax access to a persistent memory 100 namespace. For environments that want to hard partition 101 persistent memory, this capability provides a mechanism to 102 sub-divide a namespace into character devices that can only be 103 accessed via DAX (mmap(2)). 104 105 Select Y if unsure 106 107config OF_PMEM 108 tristate "Device-tree support for persistent memory regions" 109 depends on OF 110 default LIBNVDIMM 111 help 112 Allows regions of persistent memory to be described in the 113 device-tree. 114 115 Select Y if unsure. 116 117config NVDIMM_KEYS 118 def_bool y 119 depends on ENCRYPTED_KEYS 120 depends on (LIBNVDIMM=ENCRYPTED_KEYS) || LIBNVDIMM=m 121 122config NVDIMM_TEST_BUILD 123 tristate "Build the unit test core" 124 depends on m 125 depends on COMPILE_TEST && X86_64 126 default m if COMPILE_TEST 127 help 128 Build the core of the unit test infrastructure. The result of 129 this build is non-functional for unit test execution, but it 130 otherwise helps catch build errors induced by changes to the 131 core devm_memremap_pages() implementation and other 132 infrastructure. 133 134endif 135