3 -- Grant of Unlimited Rights
5 -- Under contracts F33600-87-D-0337, F33600-84-D-0280, MDA903-79-C-0687,
6 -- F08630-91-C-0015, and DCA100-97-D-0025, the U.S. Government obtained
7 -- unlimited rights in the software and documentation contained herein.
8 -- Unlimited rights are defined in DFAR 252.227-7013(a)(19). By making
9 -- this public release, the Government intends to confer upon all
10 -- recipients unlimited rights equal to those held by the Government.
11 -- These rights include rights to use, duplicate, release or disclose the
12 -- released technical data and computer software in whole or in part, in
13 -- any manner and for any purpose whatsoever, and to have or permit others
18 -- ALL MATERIALS OR INFORMATION HEREIN RELEASED, MADE AVAILABLE OR
19 -- DISCLOSED ARE AS IS. THE GOVERNMENT MAKES NO EXPRESS OR IMPLIED
20 -- WARRANTY AS TO ANY MATTER WHATSOEVER, INCLUDING THE CONDITIONS OF THE
21 -- SOFTWARE, DOCUMENTATION OR OTHER INFORMATION RELEASED, MADE AVAILABLE
22 -- OR DISCLOSED, OR THE OWNERSHIP, MERCHANTABILITY, OR FITNESS FOR A
23 -- PARTICULAR PURPOSE OF SAID MATERIAL.
27 -- Check that the ARCSIN and ARCCOS functions return
28 -- results that are within the error bound allowed.
31 -- This test consists of a generic package that is
32 -- instantiated to check both Float and a long float type.
33 -- The test for each floating point type is divided into
35 -- Special value checks where the result is a known constant.
36 -- Checks in a specific range where a Taylor series can be
37 -- used to compute an accurate result for comparison.
39 -- The Taylor series tests are a direct translation of the
40 -- FORTRAN code found in the reference.
42 -- SPECIAL REQUIREMENTS
43 -- The Strict Mode for the numerical accuracy must be
44 -- selected. The method by which this mode is selected
45 -- is implementation dependent.
47 -- APPLICABILITY CRITERIA:
48 -- This test applies only to implementations supporting the
50 -- This test only applies to the Strict Mode for numerical
55 -- 18 Mar 96 SAIC Initial release for 2.1
56 -- 24 Apr 96 SAIC Fixed error bounds.
57 -- 17 Aug 96 SAIC Added reference information and improved
58 -- checking for machines with more than 23
59 -- digits of precision.
60 -- 03 Feb 97 PWB.CTA Removed checks with explicit Cycle => 2.0*Pi
61 -- 22 Dec 99 RLB Added model range checking to "exact" results,
62 -- in order to avoid too strictly requiring a specific
63 -- result, and too weakly checking results.
66 -- According to Ken Dritz, author of the Numerics Annex of the RM,
67 -- one should never specify the cycle 2.0*Pi for the trigonometric
68 -- functions. In particular, if the machine number for the first
69 -- argument is not an exact multiple of the machine number for the
70 -- explicit cycle, then the specified exact results cannot be
71 -- reasonably expected. The affected checks in this test have been
72 -- marked as comments, with the additional notation "pwb-math".
79 -- Software Manual for the Elementary Functions
80 -- William J. Cody, Jr. and William Waite
81 -- Prentice-Hall, 1980
83 -- CRC Standard Mathematical Tables
86 -- Implementation and Testing of Function Software
88 -- Problems and Methodologies in Mathematical Software Production
89 -- editors P. C. Messina and A. Murli
90 -- Lecture Notes in Computer Science Volume 142
91 -- Springer Verlag, 1982
93 -- CELEFUNT: A Portable Test Package for Complex Elementary Functions
94 -- ACM Collected Algorithms number 714
98 with Ada
.Numerics
.Generic_Elementary_Functions
;
100 Verbose
: constant Boolean := False;
101 Max_Samples
: constant := 1000;
104 -- CRC Standard Mathematical Tables; 23rd Edition; pg 738
106 1.41421_35623_73095_04880_16887_24209_69807_85696_71875_37695
;
108 1.73205_08075_68877_29352_74463_41505_87236_69428_05253_81039
;
110 Pi
: constant := Ada
.Numerics
.Pi
;
112 -- relative error bound from G.2.4(7);6.0
113 Minimum_Error
: constant := 4.0;
116 type Real
is digits <>;
117 Half_PI_Low
: in Real
; -- The machine number closest to, but not greater
119 Half_PI_High
: in Real
;-- The machine number closest to, but not less
121 PI_Low
: in Real
; -- The machine number closest to, but not greater
123 PI_High
: in Real
; -- The machine number closest to, but not less
125 package Generic_Check
is
129 package body Generic_Check
is
130 package Elementary_Functions
is new
131 Ada
.Numerics
.Generic_Elementary_Functions
(Real
);
133 function Arcsin
(X
: Real
) return Real
renames
134 Elementary_Functions
.Arcsin
;
135 function Arcsin
(X
, Cycle
: Real
) return Real
renames
136 Elementary_Functions
.Arcsin
;
137 function Arccos
(X
: Real
) return Real
renames
138 Elementary_Functions
.ArcCos
;
139 function Arccos
(X
, Cycle
: Real
) return Real
renames
140 Elementary_Functions
.ArcCos
;
142 -- needed for support
143 function Log
(X
, Base
: Real
) return Real
renames
144 Elementary_Functions
.Log
;
146 -- flag used to terminate some tests early
147 Accuracy_Error_Reported
: Boolean := False;
149 -- The following value is a lower bound on the accuracy
150 -- required. It is normally 0.0 so that the lower bound
151 -- is computed from Model_Epsilon. However, for tests
152 -- where the expected result is only known to a certain
153 -- amount of precision this bound takes on a non-zero
154 -- value to account for that level of precision.
155 Error_Low_Bound
: Real
:= 0.0;
158 procedure Check
(Actual
, Expected
: Real
;
165 -- In the case where the expected result is very small or 0
166 -- we compute the maximum error as a multiple of Model_Epsilon instead
167 -- of Model_Epsilon and Expected.
168 Rel_Error
:= MRE
* abs Expected
* Real
'Model_Epsilon;
169 Abs_Error
:= MRE
* Real
'Model_Epsilon;
170 if Rel_Error
> Abs_Error
then
171 Max_Error
:= Rel_Error
;
173 Max_Error
:= Abs_Error
;
176 -- take into account the low bound on the error
177 if Max_Error
< Error_Low_Bound
then
178 Max_Error
:= Error_Low_Bound
;
181 if abs (Actual
- Expected
) > Max_Error
then
182 Accuracy_Error_Reported
:= True;
183 Report
.Failed
(Test_Name
&
184 " actual: " & Real
'Image (Actual
) &
185 " expected: " & Real
'Image (Expected
) &
186 " difference: " & Real
'Image (Actual
- Expected
) &
187 " max err:" & Real
'Image (Max_Error
) );
189 if Actual
= Expected
then
190 Report
.Comment
(Test_Name
& " exact result");
192 Report
.Comment
(Test_Name
& " passed");
198 procedure Special_Value_Test
is
199 -- In the following tests the expected result is accurate
200 -- to the machine precision so the minimum guaranteed error
201 -- bound can be used.
211 type Test_Data_Type
is array (Positive range <>) of Data_Point
;
213 -- the values in the following tables only involve static
214 -- expressions so no loss of precision occurs. However,
215 -- rounding can be an issue with expressions involving Pi
216 -- and square roots. The error bound specified in the
217 -- table takes the sqrt error into account but not the
218 -- error due to Pi. The Pi error is added in in the
219 -- radians test below.
221 Arcsin_Test_Data
: constant Test_Data_Type
:= (
222 -- degrees radians sine error_bound test #
223 --( 0.0, 0.0, 0.0, 0.0 ), -- 1 - In Exact_Result_Test.
224 ( 30.0, Pi
/6.0, 0.5, 4.0 ), -- 2
225 ( 60.0, Pi
/3.0, Sqrt3
/2.0, 5.0 ), -- 3
226 --( 90.0, Pi/2.0, 1.0, 4.0 ), -- 4 - In Exact_Result_Test.
227 --(-90.0, -Pi/2.0, -1.0, 4.0 ), -- 5 - In Exact_Result_Test.
228 (-60.0, -Pi
/3.0, -Sqrt3
/2.0, 5.0 ), -- 6
229 (-30.0, -Pi
/6.0, -0.5, 4.0 ), -- 7
230 ( 45.0, Pi
/4.0, Sqrt2
/2.0, 5.0 ), -- 8
231 (-45.0, -Pi
/4.0, -Sqrt2
/2.0, 5.0 ) ); -- 9
233 Arccos_Test_Data
: constant Test_Data_Type
:= (
234 -- degrees radians cosine error_bound test #
235 --( 0.0, 0.0, 1.0, 0.0 ), -- 1 - In Exact_Result_Test.
236 ( 30.0, Pi
/6.0, Sqrt3
/2.0, 5.0 ), -- 2
237 ( 60.0, Pi
/3.0, 0.5, 4.0 ), -- 3
238 --( 90.0, Pi/2.0, 0.0, 4.0 ), -- 4 - In Exact_Result_Test.
239 (120.0, 2.0*Pi
/3.0, -0.5, 4.0 ), -- 5
240 (150.0, 5.0*Pi
/6.0, -Sqrt3
/2.0, 5.0 ), -- 6
241 --(180.0, Pi, -1.0, 4.0 ), -- 7 - In Exact_Result_Test.
242 ( 45.0, Pi
/4.0, Sqrt2
/2.0, 5.0 ), -- 8
243 (135.0, 3.0*Pi
/4.0, -Sqrt2
/2.0, 5.0 ) ); -- 9
248 for I
in Arcsin_Test_Data
'Range loop
250 -- note exact result requirements A.5.1(38);6.0 and
252 if Arcsin_Test_Data
(I
).Error_Bound
= 0.0 then
256 Cycle_Error
:= Arcsin_Test_Data
(I
).Error_Bound
;
257 -- allow for rounding error in the specification of Pi
258 Radian_Error
:= Cycle_Error
+ 1.0;
261 Check
(Arcsin
(Arcsin_Test_Data
(I
).Argument
),
262 Arcsin_Test_Data
(I
).Radians
,
263 "test" & Integer'Image (I
) &
265 Real
'Image (Arcsin_Test_Data
(I
).Argument
) &
268 --pwb-math Check (Arcsin (Arcsin_Test_Data (I).Argument, 2.0 * Pi),
269 --pwb-math Arcsin_Test_Data (I).Radians,
270 --pwb-math "test" & Integer'Image (I) &
271 --pwb-math " arcsin(" &
272 --pwb-math Real'Image (Arcsin_Test_Data (I).Argument) &
274 --pwb-math Cycle_Error);
275 Check
(Arcsin
(Arcsin_Test_Data
(I
).Argument
, 360.0),
276 Arcsin_Test_Data
(I
).Degrees
,
277 "test" & Integer'Image (I
) &
279 Real
'Image (Arcsin_Test_Data
(I
).Argument
) &
285 for I
in Arccos_Test_Data
'Range loop
287 -- note exact result requirements A.5.1(39);6.0 and
289 if Arccos_Test_Data
(I
).Error_Bound
= 0.0 then
293 Cycle_Error
:= Arccos_Test_Data
(I
).Error_Bound
;
294 -- allow for rounding error in the specification of Pi
295 Radian_Error
:= Cycle_Error
+ 1.0;
298 Check
(Arccos
(Arccos_Test_Data
(I
).Argument
),
299 Arccos_Test_Data
(I
).Radians
,
300 "test" & Integer'Image (I
) &
302 Real
'Image (Arccos_Test_Data
(I
).Argument
) &
305 --pwb-math Check (Arccos (Arccos_Test_Data (I).Argument, 2.0 * Pi),
306 --pwb-math Arccos_Test_Data (I).Radians,
307 --pwb-math "test" & Integer'Image (I) &
308 --pwb-math " arccos(" &
309 --pwb-math Real'Image (Arccos_Test_Data (I).Argument) &
311 --pwb-math Cycle_Error);
312 Check
(Arccos
(Arccos_Test_Data
(I
).Argument
, 360.0),
313 Arccos_Test_Data
(I
).Degrees
,
314 "test" & Integer'Image (I
) &
316 Real
'Image (Arccos_Test_Data
(I
).Argument
) &
322 when Constraint_Error
=>
323 Report
.Failed
("Constraint_Error raised in special value test");
325 Report
.Failed
("exception in special value test");
326 end Special_Value_Test
;
329 procedure Check_Exact
(Actual
, Expected_Low
, Expected_High
: Real
;
330 Test_Name
: String) is
331 -- If the expected result is not a model number, then Expected_Low is
332 -- the first machine number less than the (exact) expected
333 -- result, and Expected_High is the first machine number greater than
334 -- the (exact) expected result. If the expected result is a model
335 -- number, Expected_Low = Expected_High = the result.
336 Model_Expected_Low
: Real
:= Expected_Low
;
337 Model_Expected_High
: Real
:= Expected_High
;
339 -- Calculate the first model number nearest to, but below (or equal)
340 -- to the expected result:
341 while Real
'Model (Model_Expected_Low
) /= Model_Expected_Low
loop
342 -- Try the next machine number lower:
343 Model_Expected_Low
:= Real
'Adjacent(Model_Expected_Low
, 0.0);
345 -- Calculate the first model number nearest to, but above (or equal)
346 -- to the expected result:
347 while Real
'Model (Model_Expected_High
) /= Model_Expected_High
loop
348 -- Try the next machine number higher:
349 Model_Expected_High
:= Real
'Adjacent(Model_Expected_High
, 100.0);
352 if Actual
< Model_Expected_Low
or Actual
> Model_Expected_High
then
353 Accuracy_Error_Reported
:= True;
354 if Actual
< Model_Expected_Low
then
355 Report
.Failed
(Test_Name
&
356 " actual: " & Real
'Image (Actual
) &
357 " expected low: " & Real
'Image (Model_Expected_Low
) &
358 " expected high: " & Real
'Image (Model_Expected_High
) &
359 " difference: " & Real
'Image (Actual
- Expected_Low
));
361 Report
.Failed
(Test_Name
&
362 " actual: " & Real
'Image (Actual
) &
363 " expected low: " & Real
'Image (Model_Expected_Low
) &
364 " expected high: " & Real
'Image (Model_Expected_High
) &
365 " difference: " & Real
'Image (Expected_High
- Actual
));
368 Report
.Comment
(Test_Name
& " passed");
373 procedure Exact_Result_Test
is
376 Check_Exact
(Arcsin
(0.0), 0.0, 0.0, "arcsin(0)");
377 Check_Exact
(Arcsin
(0.0, 45.0), 0.0, 0.0, "arcsin(0,45)");
380 Check_Exact
(Arccos
(1.0), 0.0, 0.0, "arccos(1)");
381 Check_Exact
(Arccos
(1.0, 75.0), 0.0, 0.0, "arccos(1,75)");
384 Check_Exact
(Arcsin
(1.0), Half_PI_Low
, Half_PI_High
, "arcsin(1)");
385 Check_Exact
(Arcsin
(1.0, 360.0), 90.0, 90.0, "arcsin(1,360)");
387 Check_Exact
(Arcsin
(-1.0), -Half_PI_High
, -Half_PI_Low
, "arcsin(-1)");
388 Check_Exact
(Arcsin
(-1.0, 360.0), -90.0, -90.0, "arcsin(-1,360)");
390 Check_Exact
(Arccos
(0.0), Half_PI_Low
, Half_PI_High
, "arccos(0)");
391 Check_Exact
(Arccos
(0.0, 360.0), 90.0, 90.0, "arccos(0,360)");
393 Check_Exact
(Arccos
(-1.0), PI_Low
, PI_High
, "arccos(-1)");
394 Check_Exact
(Arccos
(-1.0, 360.0), 180.0, 180.0, "arccos(-1,360)");
397 when Constraint_Error
=>
398 Report
.Failed
("Constraint_Error raised in Exact_Result Test");
400 Report
.Failed
("Exception in Exact_Result Test");
401 end Exact_Result_Test
;
404 procedure Arcsin_Taylor_Series_Test
is
405 -- the following range is chosen so that the Taylor series
406 -- used will produce a result accurate to machine precision.
408 -- The following formula is used for the Taylor series:
409 -- TS(x) = x { 1 + (xsq/2) [ (1/3) + (3/4)xsq { (1/5) +
410 -- (5/6)xsq [ (1/7) + (7/8)xsq/9 ] } ] }
413 A
: constant := -0.125;
414 B
: constant := 0.125;
417 Actual
, Sum
, Xm
: Real
;
418 -- terms in Taylor series
419 K
: constant Integer := Integer (
421 Real
(Real
'Machine_Radix) ** Real
'Machine_Mantissa,
424 Accuracy_Error_Reported
:= False; -- reset
425 for I
in 1..Max_Samples
loop
426 -- make sure there is no error in x-1, x, and x+1
427 X
:= (B
- A
) * Real
(I
) / Real
(Max_Samples
) + A
;
432 Xm
:= Real
(K
+ K
+ 1);
434 Sum
:= Y_Sq
* (Sum
+ 1.0/Xm
);
436 Sum
:= Sum
* (Xm
/(Xm
+ 1.0));
440 Sum
:= (Y
- Actual
) + Sum
;
441 if not Real
'Machine_Rounds then
442 Actual
:= Actual
+ (Sum
+ Sum
);
445 Check
(Actual
, Arcsin
(X
),
446 "Taylor Series test" & Integer'Image (I
) & ": arcsin(" &
447 Real
'Image (X
) & ") ",
450 if Accuracy_Error_Reported
then
451 -- only report the first error in this test in order to keep
452 -- lots of failures from producing a huge error log
459 when Constraint_Error
=>
461 ("Constraint_Error raised in Arcsin_Taylor_Series_Test" &
462 " for X=" & Real
'Image (X
));
464 Report
.Failed
("exception in Arcsin_Taylor_Series_Test" &
465 " for X=" & Real
'Image (X
));
466 end Arcsin_Taylor_Series_Test
;
470 procedure Arccos_Taylor_Series_Test
is
471 -- the following range is chosen so that the Taylor series
472 -- used will produce a result accurate to machine precision.
474 -- The following formula is used for the Taylor series:
475 -- TS(x) = x { 1 + (xsq/2) [ (1/3) + (3/4)xsq { (1/5) +
476 -- (5/6)xsq [ (1/7) + (7/8)xsq/9 ] } ] }
477 -- arccos(x) = pi/2 - TS(x)
478 A
: constant := -0.125;
479 B
: constant := 0.125;
483 Actual
, Sum
, Xm
, S
: Real
;
484 -- terms in Taylor series
485 K
: constant Integer := Integer (
487 Real
(Real
'Machine_Radix) ** Real
'Machine_Mantissa,
490 if Real
'Digits > 23 then
491 -- constants in this section only accurate to 23 digits
492 Error_Low_Bound
:= 0.00000_00000_00000_00000_001
;
493 Report
.Comment
("arctan accuracy checked to 23 digits");
496 -- C1 + C2 equals Pi/2 accurate to 23 digits
497 if Real
'Machine_Radix = 10 then
499 C2
:= 7.9632679489661923132E-4;
502 C2
:= 4.8382679489661923132E-4;
505 Accuracy_Error_Reported
:= False; -- reset
506 for I
in 1..Max_Samples
loop
507 -- make sure there is no error in x-1, x, and x+1
508 X
:= (B
- A
) * Real
(I
) / Real
(Max_Samples
) + A
;
513 Xm
:= Real
(K
+ K
+ 1);
515 Sum
:= Y_Sq
* (Sum
+ 1.0/Xm
);
517 Sum
:= Sum
* (Xm
/(Xm
+ 1.0));
521 -- at this point we have arcsin(x).
522 -- We compute arccos(x) = pi/2 - arcsin(x).
523 -- The following code segment is translated directly from
524 -- the CELEFUNT FORTRAN implementation
527 Sum
:= ((C1
- S
) + C2
) - Sum
;
529 Sum
:= ((S
- Actual
) + Sum
) - Y
;
532 Sum
:= (S
- Actual
) + Sum
;
534 if not Real
'Machine_Rounds then
535 Actual
:= Actual
+ (Sum
+ Sum
);
538 Check
(Actual
, Arccos
(X
),
539 "Taylor Series test" & Integer'Image (I
) & ": arccos(" &
540 Real
'Image (X
) & ") ",
543 -- only report the first error in this test in order to keep
544 -- lots of failures from producing a huge error log
545 exit when Accuracy_Error_Reported
;
547 Error_Low_Bound
:= 0.0; -- reset
549 when Constraint_Error
=>
551 ("Constraint_Error raised in Arccos_Taylor_Series_Test" &
552 " for X=" & Real
'Image (X
));
554 Report
.Failed
("exception in Arccos_Taylor_Series_Test" &
555 " for X=" & Real
'Image (X
));
556 end Arccos_Taylor_Series_Test
;
560 procedure Identity_Test
is
561 -- test the identity arcsin(-x) = -arcsin(x)
562 -- range chosen to be most of the valid range of the argument.
563 A
: constant := -0.999;
564 B
: constant := 0.999;
567 Accuracy_Error_Reported
:= False; -- reset
568 for I
in 1..Max_Samples
loop
569 -- make sure there is no error in x-1, x, and x+1
570 X
:= (B
- A
) * Real
(I
) / Real
(Max_Samples
) + A
;
572 Check
(Arcsin
(-X
), -Arcsin
(X
),
573 "Identity test" & Integer'Image (I
) & ": arcsin(" &
574 Real
'Image (X
) & ") ",
575 8.0); -- 2 arcsin evaluations => twice the error bound
577 if Accuracy_Error_Reported
then
578 -- only report the first error in this test in order to keep
579 -- lots of failures from producing a huge error log
586 procedure Exception_Test
is
587 X1
, X2
: Real
:= 0.0;
591 Report
.Failed
("no exception for Arcsin (1.1)");
593 when Constraint_Error
=>
594 Report
.Failed
("Constraint_Error instead of " &
595 "Argument_Error for Arcsin (1.1)");
596 when Ada
.Numerics
.Argument_Error
=>
597 null; -- expected result
599 Report
.Failed
("wrong exception for Arcsin(1.1)");
604 Report
.Failed
("no exception for Arccos (-1.1)");
606 when Constraint_Error
=>
607 Report
.Failed
("Constraint_Error instead of " &
608 "Argument_Error for Arccos (-1.1)");
609 when Ada
.Numerics
.Argument_Error
=>
610 null; -- expected result
612 Report
.Failed
("wrong exception for Arccos(-1.1)");
616 -- optimizer thwarting
617 if Report
.Ident_Bool
(False) then
618 Report
.Comment
(Real
'Image (X1
+ X2
));
627 Arcsin_Taylor_Series_Test
;
628 Arccos_Taylor_Series_Test
;
634 -----------------------------------------------------------------------
635 -----------------------------------------------------------------------
636 -- These expressions must be truly static, which is why we have to do them
637 -- outside of the generic, and we use the named numbers. Note that we know
638 -- that PI is not a machine number (it is irrational), and it should be
639 -- represented to more digits than supported by the target machine.
640 Float_Half_PI_Low
: constant := Float'Adjacent(PI
/2.0, 0.0);
641 Float_Half_PI_High
: constant := Float'Adjacent(PI
/2.0, 10.0);
642 Float_PI_Low
: constant := Float'Adjacent(PI
, 0.0);
643 Float_PI_High
: constant := Float'Adjacent(PI
, 10.0);
644 package Float_Check
is new Generic_Check
(Float,
645 Half_PI_Low
=> Float_Half_PI_Low
,
646 Half_PI_High
=> Float_Half_PI_High
,
647 PI_Low
=> Float_PI_Low
,
648 PI_High
=> Float_PI_High
);
650 -- check the floating point type with the most digits
651 type A_Long_Float
is digits System
.Max_Digits
;
652 A_Long_Float_Half_PI_Low
: constant := A_Long_Float
'Adjacent(PI
/2.0, 0.0);
653 A_Long_Float_Half_PI_High
: constant := A_Long_Float
'Adjacent(PI
/2.0, 10.0);
654 A_Long_Float_PI_Low
: constant := A_Long_Float
'Adjacent(PI
, 0.0);
655 A_Long_Float_PI_High
: constant := A_Long_Float
'Adjacent(PI
, 10.0);
656 package A_Long_Float_Check
is new Generic_Check
(A_Long_Float
,
657 Half_PI_Low
=> A_Long_Float_Half_PI_Low
,
658 Half_PI_High
=> A_Long_Float_Half_PI_High
,
659 PI_Low
=> A_Long_Float_PI_Low
,
660 PI_High
=> A_Long_Float_PI_High
);
662 -----------------------------------------------------------------------
663 -----------------------------------------------------------------------
667 Report
.Test
("CXG2015",
668 "Check the accuracy of the ARCSIN and ARCCOS functions");
671 Report
.Comment
("checking Standard.Float");
677 Report
.Comment
("checking a digits" &
678 Integer'Image (System
.Max_Digits
) &
679 " floating point type");
682 A_Long_Float_Check
.Do_Test
;