Subversion Repositories HelenOS-doc

Rev

Rev 15 | Rev 26 | Go to most recent revision | Show entire file | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed

Rev 15 Rev 17
Line 21... Line 21...
21
  </section>
21
  </section>
22
 
22
 
23
  <section>
23
  <section>
24
    <title>Physical memory management</title>
24
    <title>Physical memory management</title>
25
 
25
 
-
 
26
    <section id="buddy_allocator">
-
 
27
      <title>Buddy allocator</title>
-
 
28
 
-
 
29
      <section>
-
 
30
        <title>Overview</title>
-
 
31
 
-
 
32
        <para>In buddy allocator, memory is broken down into power-of-two
-
 
33
        sized naturally aligned blocks. These blocks are organized in an array
-
 
34
        of lists in which list with index i contains all unallocated blocks of
-
 
35
        the size <mathphrase>2<superscript>i</superscript></mathphrase>. The
-
 
36
        index i is called the order of block. Should there be two adjacent
-
 
37
        equally sized blocks in list <mathphrase>i</mathphrase> (i.e.
-
 
38
        buddies), the buddy allocator would coalesce them and put the
-
 
39
        resulting block in list <mathphrase>i + 1</mathphrase>, provided that
-
 
40
        the resulting block would be naturally aligned. Similarily, when the
-
 
41
        allocator is asked to allocate a block of size
-
 
42
        <mathphrase>2<superscript>i</superscript></mathphrase>, it first tries
-
 
43
        to satisfy the request from list with index i. If the request cannot
-
 
44
        be satisfied (i.e. the list i is empty), the buddy allocator will try
-
 
45
        to allocate and split larger block from list with index i + 1. Both of
-
 
46
        these algorithms are recursive. The recursion ends either when there
-
 
47
        are no blocks to coalesce in the former case or when there are no
-
 
48
        blocks that can be split in the latter case.</para>
-
 
49
 
-
 
50
        <graphic fileref="images/buddy_alloc.eps" format="EPS" />
-
 
51
 
-
 
52
        <para>This approach greatly reduces external fragmentation of memory
-
 
53
        and helps in allocating bigger continuous blocks of memory aligned to
-
 
54
        their size. On the other hand, the buddy allocator suffers increased
-
 
55
        internal fragmentation of memory and is not suitable for general
-
 
56
        kernel allocations. This purpose is better addressed by the <link
-
 
57
        linkend="slab">slab allocator</link>.</para>
-
 
58
      </section>
-
 
59
 
-
 
60
      <section>
-
 
61
        <title>Implementation</title>
-
 
62
 
-
 
63
        <para>The buddy allocator is, in fact, an abstract framework wich can
-
 
64
        be easily specialized to serve one particular task. It knows nothing
-
 
65
        about the nature of memory it helps to allocate. In order to beat the
-
 
66
        lack of this knowledge, the buddy allocator exports an interface that
-
 
67
        each of its clients is required to implement. When supplied an
-
 
68
        implementation of this interface, the buddy allocator can use
-
 
69
        specialized external functions to find buddy for a block, split and
-
 
70
        coalesce blocks, manipulate block order and mark blocks busy or
-
 
71
        available. For precize documentation of this interface, refer to <link
-
 
72
        linkend="???">HelenOS Generic Kernel Reference Manual</link>.</para>
-
 
73
 
-
 
74
        <formalpara>
-
 
75
          <title>Data organization</title>
-
 
76
 
-
 
77
          <para>Each entity allocable by the buddy allocator is required to
-
 
78
          contain space for storing block order number and a link variable
-
 
79
          used to interconnect blocks within the same order.</para>
-
 
80
 
-
 
81
          <para>Whatever entities are allocated by the buddy allocator, the
-
 
82
          first entity within a block is used to represent the entire block.
-
 
83
          The first entity keeps the order of the whole block. Other entities
-
 
84
          within the block are assigned the magic value
-
 
85
          <constant>BUDDY_INNER_BLOCK</constant>. This is especially important
-
 
86
          for effective identification of buddies in one-dimensional array
-
 
87
      because the entity that represents a potential buddy cannot be associated
-
 
88
      with <constant>BUDDY_INNER_BLOCK</constant> (i.e. if it is associated
-
 
89
          with <constant>BUDDY_INNER_BLOCK</constant> then it is not a
-
 
90
          buddy).</para>
-
 
91
        </formalpara>
-
 
92
      </section>
-
 
93
    </section>
-
 
94
 
26
    <section id="zones_and_frames">
95
    <section id="zones_and_frames">
27
      <title>Zones and frames</title>
96
      <title>Zones and frames</title>
28
 
97
 
29
      <para>Physical memory is divided into zones. Each zone represents
98
      <para>Physical memory is divided into zones. Each zone represents
30
      continuous area of physical memory frames. Allocation of frames is
99
      continuous area of physical memory frames. Allocation of frames is
31
      handled by the <link linkend="buddy_allocator">buddy allocator</link>
100
      handled by the <link linkend="frame_allocator">frame allocator</link>
32
      associated with the zone. Zone also contains information about free and
101
      associated with the zone. Zone also contains information about free and
33
      occupied frames and its base addresss in the memory. Some of the
102
      occupied frames and its base addresss in the memory. Some of the
34
      architectures (Mips, PPC) have only one zone, that covers whole physical
103
      architectures (mips32, ppc32) have only one zone, that covers whole
35
      memory. Other architectures (IA32) have multiple zones.</para>
104
      physical memory. Other architectures (ia32) have multiple zones.</para>
36
    </section>
105
    </section>
37
 
106
 
38
    <section id="buddy_allocator">
107
    <section id="frame_allocator">
39
      <title>Buddy allocator</title>
108
      <title>Frame allocator</title>
40
 
109
 
41
      <section>
110
      <section>
42
        <title>Overview</title>
111
        <title>Overview</title>
43
 
112
 
44
        <para>Physical memory allocation inside one <link
113
        <para>Physical memory allocation inside one <link
45
        linkend="zones_and_frames">memory zone</link> is being handled by
114
        linkend="zones_and_frames">memory zone</link> is being handled by an
46
        buddy allocation system. This approach greatly reduces possibility of
-
 
47
        outer memory fragmentation and helps in allocating bigger continious
115
        instance of <link linkend="buddy_allocator">buddy allocator</link>
48
        blocks of physical memory aligned to their size. Problem of inner
-
 
49
        memory fragmentation is being solved by <link linkend="slab">SLAB
-
 
50
        allocation system.</link></para>
116
        tailored to allocate blocks of physical memory frames.</para>
51
 
117
 
52
        <graphic fileref="images/mm1.png" />
118
        <graphic fileref="images/mm1.png" />
53
 
-
 
54
        <para>Frames are grouped into bigger blocks and blocks of the size
-
 
55
        <mathphrase>2<superscript>i</superscript></mathphrase> are stored in
-
 
56
        the list indexed with <varname>i</varname> (so called order index). If
-
 
57
        list contains 2 ajacent blocks (of a same size of cause) they can be
-
 
58
        merged into the bigger one and moved into the list with higher order
-
 
59
        index, thus making possible allocation of a bigger block.</para>
-
 
60
      </section>
119
      </section>
61
 
120
 
62
      <section>
121
      <section>
63
        <title>Implementation</title>
122
        <title>Implementation</title>
64
 
123
 
Line 101... Line 160...
101
          <mathphrase>2<superscript>i+1</superscript></mathphrase> block).
160
          <mathphrase>2<superscript>i+1</superscript></mathphrase> block).
102
          Technically, buddy is a odd/even block for even/odd block
161
          Technically, buddy is a odd/even block for even/odd block
103
          respectively. Plus we can put an extra requirement, that resulting
162
          respectively. Plus we can put an extra requirement, that resulting
104
          block must be aligned to its size. This requirement guarantees
163
          block must be aligned to its size. This requirement guarantees
105
          natural block alignment for the blocks coming out the allocation
164
          natural block alignment for the blocks coming out the allocation
106
          system.
165
          system.</para>
107
      </para>
-
 
108
     
166
 
109
      <para>
167
          <para>Using direct pointer arithmetics,
-
 
168
          <varname>frame_t::ref_count</varname> and
110
      Using direct pointer arithmetics, <varname>frame_t::ref_count</varname> and <varname>frame_t::buddy_order</varname> variables,
169
          <varname>frame_t::buddy_order</varname> variables, finding buddy is
111
      finding buddy is done at constant time.
170
          done at constant time.</para>
112
      </para>
-
 
113
        </formalpara>
171
        </formalpara>
114
      </section>
172
      </section>
115
    </section>
173
    </section>
116
 
174
 
117
    <section id="slab">
175
    <section id="slab">