1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
|
-- CXG2014.A
--
-- Grant of Unlimited Rights
--
-- Under contracts F33600-87-D-0337, F33600-84-D-0280, MDA903-79-C-0687,
-- F08630-91-C-0015, and DCA100-97-D-0025, the U.S. Government obtained
-- unlimited rights in the software and documentation contained herein.
-- Unlimited rights are defined in DFAR 252.227-7013(a)(19). By making
-- this public release, the Government intends to confer upon all
-- recipients unlimited rights equal to those held by the Government.
-- These rights include rights to use, duplicate, release or disclose the
-- released technical data and computer software in whole or in part, in
-- any manner and for any purpose whatsoever, and to have or permit others
-- to do so.
--
-- DISCLAIMER
--
-- ALL MATERIALS OR INFORMATION HEREIN RELEASED, MADE AVAILABLE OR
-- DISCLOSED ARE AS IS. THE GOVERNMENT MAKES NO EXPRESS OR IMPLIED
-- WARRANTY AS TO ANY MATTER WHATSOEVER, INCLUDING THE CONDITIONS OF THE
-- SOFTWARE, DOCUMENTATION OR OTHER INFORMATION RELEASED, MADE AVAILABLE
-- OR DISCLOSED, OR THE OWNERSHIP, MERCHANTABILITY, OR FITNESS FOR A
-- PARTICULAR PURPOSE OF SAID MATERIAL.
--*
--
-- OBJECTIVE:
-- Check that the SINH and COSH functions return
-- results that are within the error bound allowed.
--
-- TEST DESCRIPTION:
-- This test consists of a generic package that is
-- instantiated to check both Float and a long float type.
-- The test for each floating point type is divided into
-- several parts:
-- Special value checks where the result is a known constant.
-- Checks that use an identity for determining the result.
-- Exception checks.
--
-- SPECIAL REQUIREMENTS
-- The Strict Mode for the numerical accuracy must be
-- selected. The method by which this mode is selected
-- is implementation dependent.
--
-- APPLICABILITY CRITERIA:
-- This test applies only to implementations supporting the
-- Numerics Annex.
-- This test only applies to the Strict Mode for numerical
-- accuracy.
--
--
-- CHANGE HISTORY:
-- 15 Mar 96 SAIC Initial release for 2.1
-- 03 Jun 98 EDS In line 80, change 1000 to 1024, making it a model
-- number. Add Taylor Series terms in line 281.
-- 15 Feb 99 RLB Repaired Subtraction_Error_Test to avoid precision
-- problems.
--!
--
-- References:
--
-- Software Manual for the Elementary Functions
-- William J. Cody, Jr. and William Waite
-- Prentice-Hall, 1980
--
-- CRC Standard Mathematical Tables
-- 23rd Edition
--
-- Implementation and Testing of Function Software
-- W. J. Cody
-- Problems and Methodologies in Mathematical Software Production
-- editors P. C. Messina and A. Murli
-- Lecture Notes in Computer Science Volume 142
-- Springer Verlag, 1982
--
with System;
with Report;
with Ada.Numerics.Generic_Elementary_Functions;
procedure CXG2014 is
Verbose : constant Boolean := False;
Max_Samples : constant := 1024;
E : constant := Ada.Numerics.E;
Cosh1 : constant := (E + 1.0 / E) / 2.0; -- cosh(1.0)
generic
type Real is digits <>;
package Generic_Check is
procedure Do_Test;
end Generic_Check;
package body Generic_Check is
package Elementary_Functions is new
Ada.Numerics.Generic_Elementary_Functions (Real);
function Sinh (X : Real) return Real renames
Elementary_Functions.Sinh;
function Cosh (X : Real) return Real renames
Elementary_Functions.Cosh;
function Log (X : Real) return Real renames
Elementary_Functions.Log;
-- flag used to terminate some tests early
Accuracy_Error_Reported : Boolean := False;
procedure Check (Actual, Expected : Real;
Test_Name : String;
MRE : Real) is
Max_Error : Real;
Rel_Error : Real;
Abs_Error : Real;
begin
-- In the case where the expected result is very small or 0
-- we compute the maximum error as a multiple of Model_Small instead
-- of Model_Epsilon and Expected.
Rel_Error := MRE * abs Expected * Real'Model_Epsilon;
Abs_Error := MRE * Real'Model_Small;
if Rel_Error > Abs_Error then
Max_Error := Rel_Error;
else
Max_Error := Abs_Error;
end if;
if abs (Actual - Expected) > Max_Error then
Accuracy_Error_Reported := True;
Report.Failed (Test_Name &
" actual: " & Real'Image (Actual) &
" expected: " & Real'Image (Expected) &
" difference: " & Real'Image (Actual - Expected) &
" max err:" & Real'Image (Max_Error) );
elsif Verbose then
if Actual = Expected then
Report.Comment (Test_Name & " exact result");
else
Report.Comment (Test_Name & " passed");
end if;
end if;
end Check;
procedure Special_Value_Test is
-- In the following tests the expected result is accurate
-- to the machine precision so the minimum guaranteed error
-- bound can be used.
Minimum_Error : constant := 8.0;
begin
Check (Sinh (1.0),
(E - 1.0 / E) / 2.0,
"sinh(1)",
Minimum_Error);
Check (Cosh (1.0),
Cosh1,
"cosh(1)",
Minimum_Error);
Check (Sinh (2.0),
(E * E - (1.0 / (E * E))) / 2.0,
"sinh(2)",
Minimum_Error);
Check (Cosh (2.0),
(E * E + (1.0 / (E * E))) / 2.0,
"cosh(2)",
Minimum_Error);
Check (Sinh (-1.0),
(1.0 / E - E) / 2.0,
"sinh(-1)",
Minimum_Error);
exception
when Constraint_Error =>
Report.Failed ("Constraint_Error raised in special value test");
when others =>
Report.Failed ("exception in special value test");
end Special_Value_Test;
procedure Exact_Result_Test is
No_Error : constant := 0.0;
begin
-- A.5.1(38);6.0
Check (Sinh (0.0), 0.0, "sinh(0)", No_Error);
Check (Cosh (0.0), 1.0, "cosh(0)", No_Error);
exception
when Constraint_Error =>
Report.Failed ("Constraint_Error raised in Exact_Result Test");
when others =>
Report.Failed ("exception in Exact_Result Test");
end Exact_Result_Test;
procedure Identity_1_Test is
-- For the Sinh test use the identity
-- 2 * Sinh(x) * Cosh(1) = Sinh(x+1) + Sinh (x-1)
-- which is transformed to
-- Sinh(x) = ((Sinh(x+1) + Sinh(x-1)) * C
-- where C = 1/(2*Cosh(1))
--
-- For the Cosh test use the identity
-- 2 * Cosh(x) * Cosh(1) = Cosh(x+1) + Cosh(x-1)
-- which is transformed to
-- Cosh(x) = C * (Cosh(x+1) + Cosh(x-1))
-- where C is the same as above
--
-- see Cody pg 230-231 for details on the error analysis.
-- The net result is a relative error bound of 16 * Model_Epsilon.
A : constant := 3.0;
-- large upper bound but not so large as to cause Cosh(B)
-- to overflow
B : constant Real := Log(Real'Safe_Last) - 2.0;
X_Minus_1, X, X_Plus_1 : Real;
Actual1, Actual2 : Real;
C : constant := 1.0 / (2.0 * Cosh1);
begin
Accuracy_Error_Reported := False; -- reset
for I in 1..Max_Samples loop
-- make sure there is no error in x-1, x, and x+1
X_Plus_1 := (B - A) * Real (I) / Real (Max_Samples) + A;
X_Plus_1 := Real'Machine (X_Plus_1);
X := Real'Machine (X_Plus_1 - 1.0);
X_Minus_1 := Real'Machine (X - 1.0);
-- Sinh(x) = ((Sinh(x+1) + Sinh(x-1)) * C
Actual1 := Sinh(X);
Actual2 := C * (Sinh(X_Plus_1) + Sinh(X_Minus_1));
Check (Actual1, Actual2,
"Identity_1_Test " & Integer'Image (I) & ": sinh(" &
Real'Image (X) & ") ",
16.0);
-- Cosh(x) = C * (Cosh(x+1) + Cosh(x-1))
Actual1 := Cosh (X);
Actual2 := C * (Cosh(X_Plus_1) + Cosh (X_Minus_1));
Check (Actual1, Actual2,
"Identity_1_Test " & Integer'Image (I) & ": cosh(" &
Real'Image (X) & ") ",
16.0);
if Accuracy_Error_Reported then
-- only report the first error in this test in order to keep
-- lots of failures from producing a huge error log
return;
end if;
end loop;
exception
when Constraint_Error =>
Report.Failed
("Constraint_Error raised in Identity_1_Test" &
" for X=" & Real'Image (X));
when others =>
Report.Failed ("exception in Identity_1_Test" &
" for X=" & Real'Image (X));
end Identity_1_Test;
procedure Subtraction_Error_Test is
-- This test detects the error resulting from subtraction if
-- the obvious algorithm was used for computing sinh. That is,
-- it it is computed as (e**x - e**-x)/2.
-- We check the result by using a Taylor series expansion that
-- will produce a result accurate to the machine precision for
-- the range under test.
--
-- The maximum relative error bound for this test is
-- 8 for the sinh operation and 7 for the Taylor series
-- for a total of 15 * Model_Epsilon
A : constant := 0.0;
B : constant := 0.5;
X : Real;
X_Squared : Real;
Actual, Expected : Real;
begin
if Real'digits > 15 then
return; -- The approximation below is not accurate beyond
-- 15 digits. Adding more terms makes the error
-- larger, so it makes the test worse for more normal
-- values. Thus, we skip this subtest for larger than
-- 15 digits.
end if;
Accuracy_Error_Reported := False; -- reset
for I in 1..Max_Samples loop
X := (B - A) * Real (I) / Real (Max_Samples) + A;
X_Squared := X * X;
Actual := Sinh(X);
-- The Taylor series regrouped a bit
Expected :=
X * (1.0 + (X_Squared / 6.0) *
(1.0 + (X_Squared/20.0) *
(1.0 + (X_Squared/42.0) *
(1.0 + (X_Squared/72.0) *
(1.0 + (X_Squared/110.0) *
(1.0 + (X_Squared/156.0)
))))));
Check (Actual, Expected,
"Subtraction_Error_Test " & Integer'Image (I) & ": sinh(" &
Real'Image (X) & ") ",
15.0);
if Accuracy_Error_Reported then
-- only report the first error in this test in order to keep
-- lots of failures from producing a huge error log
return;
end if;
end loop;
exception
when Constraint_Error =>
Report.Failed
("Constraint_Error raised in Subtraction_Error_Test");
when others =>
Report.Failed ("exception in Subtraction_Error_Test");
end Subtraction_Error_Test;
procedure Exception_Test is
X1, X2 : Real := 0.0;
begin
-- this part of the test is only applicable if 'Machine_Overflows
-- is true.
if Real'Machine_Overflows then
begin
X1 := Sinh (Real'Safe_Last / 2.0);
Report.Failed ("no exception for sinh overflow");
exception
when Constraint_Error => null;
when others =>
Report.Failed ("wrong exception sinh overflow");
end;
begin
X2 := Cosh (Real'Safe_Last / 2.0);
Report.Failed ("no exception for cosh overflow");
exception
when Constraint_Error => null;
when others =>
Report.Failed ("wrong exception cosh overflow");
end;
end if;
-- optimizer thwarting
if Report.Ident_Bool (False) then
Report.Comment (Real'Image (X1 + X2));
end if;
end Exception_Test;
procedure Do_Test is
begin
Special_Value_Test;
Exact_Result_Test;
Identity_1_Test;
Subtraction_Error_Test;
Exception_Test;
end Do_Test;
end Generic_Check;
-----------------------------------------------------------------------
-----------------------------------------------------------------------
package Float_Check is new Generic_Check (Float);
-- check the floating point type with the most digits
type A_Long_Float is digits System.Max_Digits;
package A_Long_Float_Check is new Generic_Check (A_Long_Float);
-----------------------------------------------------------------------
-----------------------------------------------------------------------
begin
Report.Test ("CXG2014",
"Check the accuracy of the SINH and COSH functions");
if Verbose then
Report.Comment ("checking Standard.Float");
end if;
Float_Check.Do_Test;
if Verbose then
Report.Comment ("checking a digits" &
Integer'Image (System.Max_Digits) &
" floating point type");
end if;
A_Long_Float_Check.Do_Test;
Report.Result;
end CXG2014;
|