]>
Commit | Line | Data |
---|---|---|
7c673cae FG |
1 | <html> |
2 | ||
3 | <head> | |
4 | <meta name="GENERATOR" content="Microsoft FrontPage 5.0"> | |
5 | <meta name="ProgId" content="FrontPage.Editor.Document"> | |
6 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> | |
7 | <title>Endian Mini-Review</title> | |
8 | </head> | |
9 | <link href="styles.css" rel="stylesheet"> | |
10 | <body> | |
11 | ||
12 | <table border="0" cellpadding="5" cellspacing="0" style="border-collapse: collapse" bordercolor="#111111" width="100%"> | |
13 | <tr> | |
14 | <td width="339"> | |
15 | <a href="../../../index.html"> | |
16 | <img src="../../../boost.png" alt="Boost logo" align="middle" border="0" width="277" height="86"></a></td> | |
17 | <td align="middle" width="1253"> | |
18 | <b> | |
19 | <font size="6">Endian </font></b><font size="6"><b>Mini-Review</b></font></td> | |
20 | </tr> | |
21 | </table> | |
22 | ||
23 | <table border="0" cellpadding="5" cellspacing="0" style="border-collapse: collapse" | |
24 | bordercolor="#111111" bgcolor="#D7EEFF" width="100%"> | |
25 | <tr> | |
26 | <td><b> | |
27 | <a href="index.html">Endian Home</a> | |
28 | <a href="conversion.html">Conversion Functions</a> | |
29 | <a href="arithmetic.html">Arithmetic Types</a> | |
30 | <a href="buffers.html">Buffer Types</a> | |
31 | <a href="choosing_approach.html">Choosing Approach</a></b></td> | |
32 | </tr> | |
33 | </table> | |
34 | <p></p> | |
35 | ||
36 | <p>The results of the Boost.Endian formal review included a list of issues to be | |
37 | resolved before a mini-review.</p> | |
38 | ||
39 | <p>The issues are shown in <b>bold</b> below, with the resolution indicated.</p> | |
40 | ||
41 | <p><b>1. Common use case scenarios should be developed.</b></p> | |
42 | <blockquote> | |
43 | ||
44 | <p>Done. The documentation have been refactored. A page is now devoted to | |
45 | <a href="choosing_approach.html">Choosing the Approach</a> to endianness. See | |
46 | <a href="choosing_approach.html#Use-cases">Use cases</a> for use case | |
47 | scenarios.</p> | |
48 | </blockquote> | |
49 | <p><b>2. Example programs should be developed for the common use case scenarios.</b></p> | |
50 | <blockquote> | |
51 | <p>Done. See <a href="choosing_approach.html">Choosing the Approach</a>. | |
52 | Example code has been added throughout.</p> | |
53 | </blockquote> | |
54 | <p><b>3. Documentation should illuminate the differences between endian | |
55 | integer/float type and endian conversion approaches to the common use | |
56 | case scenarios, and provide guidelines for choosing the most appropriate | |
57 | approach in user's applications.</b></p> | |
58 | <blockquote> | |
59 | <p>Done. See <a href="choosing_approach.html">Choosing the Approach</a>.</p> | |
60 | </blockquote> | |
61 | <p><b>4 .Conversion functions supplying results via return should be provided.</b></p> | |
62 | <blockquote> | |
63 | <p>Done. See <a href="conversion.html">ConversionFunctions</a>.</p> | |
64 | </blockquote> | |
65 | <p><b>5. Platform specific performance enhancements such as use of compiler intrinsics or relaxed alignment requirements should be supported.</b></p> | |
66 | <blockquote> | |
67 | <p>Done. Compiler (Clang, GCC, VisualC++, etc.) intrinsics and built-in | |
68 | functions are used in the implementation where appropriate, as requested. See | |
69 | <a href="index.html#Intrinsic">Built-in support for Intrinsics</a>. See | |
70 | <a href="index.html#Timings">Timings for Example 2</a> to gauge the impact of | |
71 | intrinsics.</p> | |
72 | </blockquote> | |
73 | <p><b>6. Endian integer (and floating) types should be implemented via the | |
74 | conversion functions. If that can't be done efficiently, consideration | |
75 | should be given to expanding the conversion function signatures to | |
76 | resolve the inefficiencies.</b></p> | |
77 | <blockquote> | |
78 | <p>Done. For the endian types, the implementation uses the endian conversion | |
79 | functions, and thus the intrinsics, as requested.</p> | |
80 | </blockquote> | |
81 | <p><b>7. Benchmarks that measure performance should be provided. It should be | |
82 | possible to compare platform specific performance enhancements against | |
83 | portable base implementations, and to compare endian integer approaches | |
84 | against endian conversion approaches for the common use case scenarios.</b></p> | |
85 | <blockquote> | |
86 | <p>Done. See <a href="index.html#Timings">Timings for Example 2</a>. The <code>endian/test</code> directory | |
87 | also contains several additional benchmark and speed test programs.</p> | |
88 | </blockquote> | |
89 | <p><b>8. Float (32-bits) and double (64-bits) should be supported. IEEE 754 is | |
90 | the primary use case.</b></p> | |
91 | <blockquote> | |
92 | <p>Done. The <a href="buffers.html">endian buffer types</a>, | |
93 | <a href="arithmetic.html">endian arithmetic types</a> and | |
94 | <a href="conversion.html">endian conversion functions</a> now support 32-bit (<code>float)</code> | |
95 | and 64-bit <code>(double)</code> floating point, as requested.</p> | |
96 | </blockquote> | |
97 | <p><b>9. Support for user defined types (UDTs) is desirable, and should be | |
98 | provided where there would be no conflict with the other concerns.</b></p> | |
99 | <blockquote> | |
100 | <p>Done. See <a href="conversion.html#Customization-points">Customization | |
101 | points for user-defined types (UDTs)</a>.</p> | |
102 | </blockquote> | |
103 | <p><b>10. There is some concern that endian integer/float arithmetic operations | |
104 | might used inadvertently or inappropriately. The impact of adding an endian_buffer | |
105 | class without arithmetic operations should be investigated.</b></p> | |
106 | <blockquote> | |
107 | <p>Done. The endian types have been decomposed into class template <code> | |
108 | <a href="buffers.html">endian_buffer</a></code> and class template <code> | |
109 | <a href="arithmetic.html">endian_arithmetic</a></code>. Class | |
110 | <code>endian_buffer</code> is a public base class for <code>endian_arithmetic</code>, | |
111 | and can also be used by users as a stand-alone class.</p> | |
112 | </blockquote> | |
113 | <p><b>11. Stream insertion and extraction of the endian integer/float types should | |
114 | be documented and included in the test coverage.</b></p> | |
115 | <blockquote> | |
116 | <p>Done. See <a href="buffers.html#Stream-inserter">Stream inserter</a> and | |
117 | <a href="buffers.html#Stream-extractor">Stream extractor</a>.</p> | |
118 | </blockquote> | |
119 | <p><b>12. Binary I/O support that was investigated during development of the Endian | |
120 | library should be put up for mini-review for inclusion in the Boost I/O | |
121 | library.</b></p> | |
122 | ||
123 | <blockquote> | |
124 | <p>Not done yet. Will be handled as a separate min-review soon after the | |
125 | Endian mini-review.</p> | |
126 | ||
127 | </blockquote> | |
128 | <p><b>13. Other requested changes.</b></p> | |
129 | ||
130 | <blockquote> | |
131 | <ul> | |
132 | <li>In addition to the named-endianness conversion functions, functions that perform | |
133 | compile-time (via template) and run-time (via function argument) dispatch | |
134 | are now provided.</li> | |
135 | <li><code>order::native</code> is now a synonym for <code>order::big</code> | |
136 | or <code>order::little</code> according to the endianness of the platform. This reduces the number of template specializations required.</li> | |
137 | <li>Headers have been reorganized to make them easier to read, | |
138 | with a synopsis at the front and implementation following.</li> | |
139 | </ul> | |
140 | </blockquote> | |
141 | <hr> | |
142 | <p>Last revised: | |
143 | <!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B, %Y" startspan -->19 January, 2015<!--webbot bot="Timestamp" endspan i-checksum="38903" --></p> | |
144 | <p>© Copyright Beman Dawes, 2014</p> | |
145 | <p>Distributed under the Boost Software License, Version 1.0. See | |
146 | <a href="http://www.boost.org/LICENSE_1_0.txt">www.boost.org/ LICENSE_1_0.txt</a></p> | |
147 | ||
148 | </body> | |
149 | ||
150 | </html> |