Subversion Repositories HelenOS-doc

Rev

Rev 53 | Go to most recent revision | Details | Last modification | View Log | RSS feed

Rev Author Line No. Line
52 jermar 1
<?xml version="1.0" encoding="UTF-8"?>
2
 
3
<chapter id="ds"><?dbhtml filename="ds.html"?>
4
    <title>Data structures</title>
5
 
6
    <para>
7
    There is lots of data that either flows through various HelenOS subsystems
8
    or is stored directly by them. Each subsystem uses its own data structures
9
    to represent the data. These data structures need to be kept
10
    somewhere. In order to work efficiently, HelenOS, and especially its
11
    kernel, deploys several house keeping data types that are designed
12
    to faciliate managing other data structures. They serve like generic
13
    containers.
14
    </para>
15
 
16
    <section>
17
      <title>Lists</title>
18
      <para>
19
      HelenOS uses circular doubly linked lists to bind related data
20
      together. Lists are composed of an independent head and links that are always
21
      part of the object that is to be put into the list. Adding items to a list
22
      thus doesn't require any further memory allocations. Head and each link then
23
      contains forward and backward pointer. An empty list is composed of a sole
24
      head whose both pointers reference the head itself. The expense of two times
25
      bigger memory consumption as compared to memory consumption of singly linked
26
      lists is justified by constant insertion and removal times at random positions
27
      within the list.
28
      </para>
29
 
30
      <para>
31
      Lists are frequently used to implement FIFO behaviour (e.g. scheduler run queues
32
      or synchronization wait queues). Contrary to the FIFO type, which is also supported
33
      by HelenOS, they don't take up any unused space and are more general. On the
34
      other hand, they are slower than in-array FIFOs and can be hardly used to implement
35
      buffers.
36
      </para>
37
 
38
    </section>
39
 
40
    <section>
41
      <title>FIFO queues</title>
42
      <para>
43
      FIFO queues are implemented as either statically or dynamically allocated
44
      arrays<footnote><para>Depending on the array size.</para></footnote> of some generic type
45
      with two indices. The first index points to the head of the FIFO queue and the other
46
      points to the tail thereof. There can be as many items in the FIFO as is the number
47
      of elements in the array and no more. The indices are taken modulo size of the queue
48
      because as a consequence of insertions and deletions, the tail can have numericaly
49
      lower index than the head.
50
      </para>
51
 
52
      <para>
53
      FIFO queues are used, for example, in ASID management code to store inactive ASIDs or
54
      in userspace keyboard driver to buffer read characters.
55
      </para>
56
    </section>
57
 
58
    <section>
59
      <title>Hash tables</title>
60
      <para>
61
      The kernel, as well as userspace, makes use of the hash table implementation with
62
      separate chaining. The implementation is very generic in that it forces the user
63
      to supply methods for computing the hash index, comparing items with several keys
64
      and the item removal callback function. Besides these virtual operations,
65
      the hash table is composed of a dynamically allocated array of list heads that
66
      represent each chain, number of chains and the maximal number of keys.
67
      </para>
68
    </section>
69
 
70
    <section>
71
      <title>B+trees</title>
72
      <para>
73
 
74
      </para>
75
    </section>
76
 
77
</chapter>