Rev 660 | Go to most recent revision | Details | Compare with Previous | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
647 | cejka | 1 | /* |
2 | * Copyright (C) 2005 Josef Cejka |
||
3 | * All rights reserved. |
||
4 | * |
||
5 | * Redistribution and use in source and binary forms, with or without |
||
6 | * modification, are permitted provided that the following conditions |
||
7 | * are met: |
||
8 | * |
||
9 | * - Redistributions of source code must retain the above copyright |
||
10 | * notice, this list of conditions and the following disclaimer. |
||
11 | * - Redistributions in binary form must reproduce the above copyright |
||
12 | * notice, this list of conditions and the following disclaimer in the |
||
13 | * documentation and/or other materials provided with the distribution. |
||
14 | * - The name of the author may not be used to endorse or promote products |
||
15 | * derived from this software without specific prior written permission. |
||
16 | * |
||
17 | * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR |
||
18 | * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES |
||
19 | * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. |
||
20 | * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, |
||
21 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT |
||
22 | * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
||
23 | * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
||
24 | * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
||
25 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF |
||
26 | * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
||
27 | */ |
||
28 | |||
29 | #include<sftypes.h> |
||
30 | #include<arithmetic.h> |
||
660 | cejka | 31 | #include<comparison.h> |
647 | cejka | 32 | |
33 | /** Add two Float32 numbers with same signs |
||
34 | */ |
||
35 | float32 addFloat32(float32 a, float32 b) |
||
36 | { |
||
37 | int expdiff; |
||
38 | __u32 exp1,exp2,mant1,mant2; |
||
39 | |||
40 | expdiff=a.parts.exp - b.parts.exp; |
||
41 | if (expdiff<0) { |
||
42 | if (isFloat32NaN(b)) { |
||
43 | //TODO: fix SigNaN |
||
44 | if (isFloat32SigNaN(b)) { |
||
45 | }; |
||
46 | return b; |
||
47 | }; |
||
48 | |||
49 | if (b.parts.exp==0xFF) { |
||
50 | return b; |
||
51 | } |
||
52 | |||
53 | mant1=b.parts.mantisa; |
||
54 | exp1=b.parts.exp; |
||
55 | mant2=a.parts.mantisa; |
||
56 | exp2=a.parts.exp; |
||
57 | expdiff*=-1; |
||
58 | } else { |
||
59 | if (isFloat32NaN(a)) { |
||
60 | //TODO: fix SigNaN |
||
61 | if ((isFloat32SigNaN(a))||(isFloat32SigNaN(b))) { |
||
62 | }; |
||
63 | return a; |
||
64 | }; |
||
65 | |||
66 | if (a.parts.exp==0xFF) { |
||
67 | return a; |
||
68 | } |
||
69 | |||
70 | mant1=a.parts.mantisa; |
||
71 | exp1=a.parts.exp; |
||
72 | mant2=b.parts.mantisa; |
||
73 | exp2=b.parts.exp; |
||
74 | }; |
||
75 | |||
76 | if (exp1==0) { |
||
77 | //both are denormalized |
||
78 | mant1+=mant2; |
||
79 | if (mant1&0xF00000) { |
||
80 | a.parts.exp=1; |
||
81 | }; |
||
82 | a.parts.mantisa=mant1; |
||
83 | return a; |
||
84 | }; |
||
85 | |||
86 | // create some space for rounding |
||
87 | mant1<<=6; |
||
88 | mant2<<=6; |
||
89 | |||
90 | mant1|=0x20000000; //add hidden bit |
||
91 | |||
92 | |||
93 | if (exp2==0) { |
||
94 | --expdiff; |
||
95 | } else { |
||
96 | mant2|=0x20000000; //hidden bit |
||
97 | }; |
||
98 | |||
99 | if (expdiff>24) { |
||
100 | goto done; |
||
101 | }; |
||
102 | |||
103 | mant2>>=expdiff; |
||
104 | mant1+=mant2; |
||
105 | done: |
||
106 | if (mant1&0x40000000) { |
||
107 | ++exp1; |
||
108 | mant1>>=1; |
||
109 | }; |
||
110 | |||
111 | //rounding - if first bit after mantisa is set then round up |
||
112 | mant1+=0x20; |
||
113 | |||
661 | cejka | 114 | if (mant1&0x40000000) { |
115 | ++exp1; |
||
116 | mant1>>=1; |
||
117 | }; |
||
118 | |||
647 | cejka | 119 | a.parts.exp=exp1; |
661 | cejka | 120 | a.parts.mantisa = ((mant1&(~0x20000000))>>6); /*Clear hidden bit and shift */ |
647 | cejka | 121 | return a; |
122 | }; |
||
123 | |||
661 | cejka | 124 | /** Subtract two float32 numbers with same signs |
647 | cejka | 125 | */ |
126 | float32 subFloat32(float32 a, float32 b) |
||
127 | { |
||
661 | cejka | 128 | int expdiff; |
129 | __u32 exp1,exp2,mant1,mant2; |
||
130 | float32 result; |
||
131 | |||
132 | result.f = 0; |
||
647 | cejka | 133 | |
661 | cejka | 134 | expdiff=a.parts.exp - b.parts.exp; |
135 | if ((expdiff<0)||((expdiff==0)&&(a.parts.mantisa<b.parts.mantisa))) { |
||
136 | if (isFloat32NaN(b)) { |
||
137 | //TODO: fix SigNaN |
||
138 | if (isFloat32SigNaN(b)) { |
||
139 | }; |
||
140 | return b; |
||
141 | }; |
||
142 | |||
143 | if (b.parts.exp==0xFF) { |
||
144 | b.parts.sign=!b.parts.sign; /* num -(+-inf) = -+inf */ |
||
145 | return b; |
||
146 | } |
||
147 | |||
148 | result.parts.sign = !a.parts.sign; |
||
149 | |||
150 | mant1=b.parts.mantisa; |
||
151 | exp1=b.parts.exp; |
||
152 | mant2=a.parts.mantisa; |
||
153 | exp2=a.parts.exp; |
||
154 | expdiff*=-1; |
||
155 | } else { |
||
156 | if (isFloat32NaN(a)) { |
||
157 | //TODO: fix SigNaN |
||
158 | if ((isFloat32SigNaN(a))||(isFloat32SigNaN(b))) { |
||
159 | }; |
||
160 | return a; |
||
161 | }; |
||
162 | |||
163 | if (a.parts.exp==0xFF) { |
||
164 | if (b.parts.exp==0xFF) { |
||
165 | /* inf - inf => nan */ |
||
166 | //TODO: fix exception |
||
167 | result.binary = FLOAT32_NAN; |
||
168 | return result; |
||
169 | }; |
||
170 | return a; |
||
171 | } |
||
172 | |||
173 | result.parts.sign = a.parts.sign; |
||
174 | |||
175 | mant1=a.parts.mantisa; |
||
176 | exp1=a.parts.exp; |
||
177 | mant2=b.parts.mantisa; |
||
178 | exp2=b.parts.exp; |
||
179 | |||
180 | |||
181 | |||
182 | }; |
||
647 | cejka | 183 | |
661 | cejka | 184 | if (exp1==0) { |
185 | //both are denormalized |
||
186 | result.parts.mantisa=mant1-mant2; |
||
187 | if (result.parts.mantisa>mant1) { |
||
188 | //TODO: underflow exception |
||
189 | return result; |
||
190 | }; |
||
191 | result.parts.exp=0; |
||
192 | return result; |
||
193 | }; |
||
194 | |||
195 | // create some space for rounding |
||
196 | mant1<<=6; |
||
197 | mant2<<=6; |
||
198 | |||
199 | mant1|=0x20000000; //add hidden bit |
||
200 | |||
201 | |||
202 | if (exp2==0) { |
||
203 | --expdiff; |
||
204 | } else { |
||
205 | mant2|=0x20000000; //hidden bit |
||
206 | }; |
||
207 | |||
208 | if (expdiff>24) { |
||
209 | goto done; |
||
210 | }; |
||
211 | |||
212 | mant1 = mant1-(mant2>>expdiff); |
||
213 | done: |
||
214 | |||
215 | //TODO: find first nonzero digit and shift result and detect possibly underflow |
||
216 | while ((exp1>0)&&(!(mant1&0x20000000))) { |
||
217 | exp1--; |
||
218 | mant1 <<= 1; |
||
219 | if(mant1 == 0) { |
||
220 | /* Realy is it an underflow? ... */ |
||
221 | //TODO: fix underflow |
||
222 | }; |
||
223 | }; |
||
224 | |||
225 | //rounding - if first bit after mantisa is set then round up |
||
226 | mant1 += 0x20; |
||
227 | |||
228 | if (mant1&0x40000000) { |
||
229 | ++exp1; |
||
230 | mant1>>=1; |
||
231 | }; |
||
232 | |||
233 | result.parts.mantisa = ((mant1&(~0x20000000))>>6); /*Clear hidden bit and shift */ |
||
234 | result.parts.exp = exp1; |
||
235 | |||
236 | return result; |
||
647 | cejka | 237 | }; |
238 |