1 .\" 2 .\" CDDL HEADER START 3 .\" 4 .\" The contents of this file are subject to the terms of the 5 .\" Common Development and Distribution License (the "License"). 6 .\" You may not use this file except in compliance with the License. 7 .\" 8 .\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE 9 .\" or http://www.opensolaris.org/os/licensing. 10 .\" See the License for the specific language governing permissions 11 .\" and limitations under the License. 12 .\" 13 .\" When distributing Covered Code, include this CDDL HEADER in each 14 .\" file and include the License file at usr/src/OPENSOLARIS.LICENSE. 15 .\" If applicable, add the following below this CDDL HEADER, with the 16 .\" fields enclosed by brackets "[]" replaced with your own identifying 17 .\" information: Portions Copyright [yyyy] [name of copyright owner] 18 .\" 19 .\" CDDL HEADER END 20 .\" 21 .\" 22 .\" Copyright 2010 Sun Microsystems, Inc. All rights reserved. 23 .\" Use is subject to license terms. 24 .\" 25 .\" Copyright (c) 2012, 2015 by Delphix. All rights reserved. 26 .\" Copyright (c) 2012, Joyent, Inc. All rights reserved. 27 .\" 28 .\" The text of this is derived from section 1 of the big theory statement in 29 .\" usr/src/uts/common/os/vmem.c, the traditional location of this text. They 30 .\" should largely be updated in tandem. 31 .Dd Jan 18, 2017 32 .Dt VMEM 9 33 .Os 34 .Sh NAME 35 .Nm vmem 36 .Nd virtual memory allocator 37 .Sh DESCRIPTION 38 .Ss Overview 39 An address space is divided into a number of logically distinct pieces, or 40 .Em arenas : 41 text, data, heap, stack, and so on. 42 Within these 43 arenas we often subdivide further; for example, we use heap addresses 44 not only for the kernel heap 45 .Po 46 .Fn kmem_alloc 47 space 48 .Pc , 49 but also for DVMA, 50 .Fn bp_mapin , 51 .Pa /dev/kmem , 52 and even some device mappings. 53 .Pp 54 The kernel address space, therefore, is most accurately described as 55 a tree of arenas in which each node of the tree 56 .Em imports 57 some subset of its parent. 58 The virtual memory allocator manages these arenas 59 and supports their natural hierarchical structure. 60 .Ss Arenas 61 An arena is nothing more than a set of integers. These integers most 62 commonly represent virtual addresses, but in fact they can represent 63 anything at all. For example, we could use an arena containing the 64 integers minpid through maxpid to allocate process IDs. For uses of this 65 nature, prefer 66 .Xr id_space 9F 67 instead. 68 .Pp 69 .Fn vmem_create 70 and 71 .Fn vmem_destroy 72 create and destroy vmem arenas. In order to differentiate between arenas used 73 for addresses and arenas used for identifiers, the 74 .Dv VMC_IDENTIFIER 75 flag is passed to 76 .Fn vmem_create . 77 This prevents identifier exhaustion from being diagnosed as general memory 78 failure. 79 .Ss Spans 80 We represent the integers in an arena as a collection of 81 .Em spans , 82 or contiguous ranges of integers. For example, the kernel heap consists of 83 just one span: 84 .Li "[kernelheap, ekernelheap)" . 85 Spans can be added to an arena in two ways: explicitly, by 86 .Fn vmem_add ; 87 or implicitly, by importing, as described in 88 .Sx Imported Memory 89 below. 90 .Ss Segments 91 Spans are subdivided into 92 .Em segments , 93 each of which is either allocated or free. A segment, like a span, is a 94 contiguous range of integers. Each allocated segment 95 .Li "[addr, addr + size)" 96 represents exactly one 97 .Li "vmem_alloc(size)" 98 that returned 99 .Sy addr . 100 Free segments represent the space between allocated segments. If two free 101 segments are adjacent, we coalesce them into one larger segment; that is, if 102 segments 103 .Li "[a, b)" 104 and 105 .Li "[b, c)" 106 are both free, we merge them into a single segment 107 .Li "[a, c)" . 108 The segments within a span are linked together in increasing\-address 109 order so we can easily determine whether coalescing is possible. 110 .Pp 111 Segments never cross span boundaries. When all segments within an imported 112 span become free, we return the span to its source. 113 .Ss Imported Memory 114 As mentioned in the overview, some arenas are logical subsets of 115 other arenas. For example, 116 .Sy kmem_va_arena 117 (a virtual address cache 118 that satisfies most 119 .Fn kmem_slab_create 120 requests) is just a subset of 121 .Sy heap_arena 122 (the kernel heap) that provides caching for the most common slab sizes. When 123 .Sy kmem_va_arena 124 runs out of virtual memory, it 125 .Em imports 126 more from the heap; we say that 127 .Sy heap_arena 128 is the 129 .Em "vmem source" 130 for 131 .Sy kmem_va_arena. 132 .Fn vmem_create 133 allows you to specify any existing vmem arena as the source for your new 134 arena. Topologically, since every arena is a child of at most one source, the 135 set of all arenas forms a collection of trees. 136 .Ss Constrained Allocations 137 Some vmem clients are quite picky about the kind of address they want. 138 For example, the DVMA code may need an address that is at a particular 139 phase with respect to some alignment (to get good cache coloring), or 140 that lies within certain limits (the addressable range of a device), 141 or that doesn't cross some boundary (a DMA counter restriction) \(em 142 or all of the above. 143 .Fn vmem_xalloc 144 allows the client to specify any or all of these constraints. 145 .Ss The Vmem Quantum 146 Every arena has a notion of 147 .Sq quantum , 148 specified at 149 .Fn vmem_create 150 time, that defines the arena's minimum unit of currency. Most commonly the 151 quantum is either 1 or 152 .Dv PAGESIZE , 153 but any power of 2 is legal. All vmem allocations are guaranteed to be 154 quantum\-aligned. 155 .Ss Relationship to the Kernel Memory Allocator 156 Every kmem cache has a vmem arena as its slab supplier. The kernel memory 157 allocator uses 158 .Fn vmem_alloc 159 and 160 .Fn vmem_free 161 to create and destroy slabs. 162 .Sh SEE ALSO 163 .Xr id_space 9F , 164 .Xr vmem_add 9F , 165 .Xr vmem_alloc 9F , 166 .Xr vmem_contains 9F , 167 .Xr vmem_create 9F , 168 .Xr vmem_walk 9F 169 .Pp 170 .Rs 171 .%A Jeff Bonwick 172 .%A Jonathan Adams 173 .%T Magazines and vmem: Extending the Slab Allocator to Many CPUs and Arbitrary Resources. 174 .%J Proceedings of the 2001 Usenix Conference 175 .%U http://www.usenix.org/event/usenix01/bonwick.html 176 .Re