cxg2015.a
来自「linux下编程用 编译软件」· A 代码 · 共 687 行 · 第 1/2 页
A
687 行
-- CXG2015.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 ARCSIN and ARCCOS 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 in a specific range where a Taylor series can be-- used to compute an accurate result for comparison.-- Exception checks.-- The Taylor series tests are a direct translation of the-- FORTRAN code found in the reference.---- 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:-- 18 Mar 96 SAIC Initial release for 2.1-- 24 Apr 96 SAIC Fixed error bounds.-- 17 Aug 96 SAIC Added reference information and improved-- checking for machines with more than 23-- digits of precision.-- 03 Feb 97 PWB.CTA Removed checks with explicit Cycle => 2.0*Pi-- 22 Dec 99 RLB Added model range checking to "exact" results,-- in order to avoid too strictly requiring a specific-- result, and too weakly checking results.---- CHANGE NOTE:-- According to Ken Dritz, author of the Numerics Annex of the RM,-- one should never specify the cycle 2.0*Pi for the trigonometric-- functions. In particular, if the machine number for the first-- argument is not an exact multiple of the machine number for the-- explicit cycle, then the specified exact results cannot be-- reasonably expected. The affected checks in this test have been-- marked as comments, with the additional notation "pwb-math".-- Phil Brashear--!---- 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---- CELEFUNT: A Portable Test Package for Complex Elementary Functions-- ACM Collected Algorithms number 714with System;with Report;with Ada.Numerics.Generic_Elementary_Functions;procedure CXG2015 is Verbose : constant Boolean := False; Max_Samples : constant := 1000; -- CRC Standard Mathematical Tables; 23rd Edition; pg 738 Sqrt2 : constant := 1.41421_35623_73095_04880_16887_24209_69807_85696_71875_37695; Sqrt3 : constant := 1.73205_08075_68877_29352_74463_41505_87236_69428_05253_81039; Pi : constant := Ada.Numerics.Pi; -- relative error bound from G.2.4(7);6.0 Minimum_Error : constant := 4.0; generic type Real is digits <>; Half_PI_Low : in Real; -- The machine number closest to, but not greater -- than PI/2.0. Half_PI_High : in Real;-- The machine number closest to, but not less -- than PI/2.0. PI_Low : in Real; -- The machine number closest to, but not greater -- than PI. PI_High : in Real; -- The machine number closest to, but not less -- than PI. 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 Arcsin (X : Real) return Real renames Elementary_Functions.Arcsin; function Arcsin (X, Cycle : Real) return Real renames Elementary_Functions.Arcsin; function Arccos (X : Real) return Real renames Elementary_Functions.ArcCos; function Arccos (X, Cycle : Real) return Real renames Elementary_Functions.ArcCos; -- needed for support function Log (X, Base : Real) return Real renames Elementary_Functions.Log; -- flag used to terminate some tests early Accuracy_Error_Reported : Boolean := False; -- The following value is a lower bound on the accuracy -- required. It is normally 0.0 so that the lower bound -- is computed from Model_Epsilon. However, for tests -- where the expected result is only known to a certain -- amount of precision this bound takes on a non-zero -- value to account for that level of precision. Error_Low_Bound : Real := 0.0; 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_Epsilon instead -- of Model_Epsilon and Expected. Rel_Error := MRE * abs Expected * Real'Model_Epsilon; Abs_Error := MRE * Real'Model_Epsilon; if Rel_Error > Abs_Error then Max_Error := Rel_Error; else Max_Error := Abs_Error; end if; -- take into account the low bound on the error if Max_Error < Error_Low_Bound then Max_Error := Error_Low_Bound; 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. type Data_Point is record Degrees, Radians, Argument, Error_Bound : Real; end record; type Test_Data_Type is array (Positive range <>) of Data_Point; -- the values in the following tables only involve static -- expressions so no loss of precision occurs. However, -- rounding can be an issue with expressions involving Pi -- and square roots. The error bound specified in the -- table takes the sqrt error into account but not the -- error due to Pi. The Pi error is added in in the -- radians test below. Arcsin_Test_Data : constant Test_Data_Type := ( -- degrees radians sine error_bound test # --( 0.0, 0.0, 0.0, 0.0 ), -- 1 - In Exact_Result_Test. ( 30.0, Pi/6.0, 0.5, 4.0 ), -- 2 ( 60.0, Pi/3.0, Sqrt3/2.0, 5.0 ), -- 3 --( 90.0, Pi/2.0, 1.0, 4.0 ), -- 4 - In Exact_Result_Test. --(-90.0, -Pi/2.0, -1.0, 4.0 ), -- 5 - In Exact_Result_Test. (-60.0, -Pi/3.0, -Sqrt3/2.0, 5.0 ), -- 6 (-30.0, -Pi/6.0, -0.5, 4.0 ), -- 7 ( 45.0, Pi/4.0, Sqrt2/2.0, 5.0 ), -- 8 (-45.0, -Pi/4.0, -Sqrt2/2.0, 5.0 ) ); -- 9 Arccos_Test_Data : constant Test_Data_Type := ( -- degrees radians cosine error_bound test # --( 0.0, 0.0, 1.0, 0.0 ), -- 1 - In Exact_Result_Test. ( 30.0, Pi/6.0, Sqrt3/2.0, 5.0 ), -- 2 ( 60.0, Pi/3.0, 0.5, 4.0 ), -- 3 --( 90.0, Pi/2.0, 0.0, 4.0 ), -- 4 - In Exact_Result_Test. (120.0, 2.0*Pi/3.0, -0.5, 4.0 ), -- 5 (150.0, 5.0*Pi/6.0, -Sqrt3/2.0, 5.0 ), -- 6 --(180.0, Pi, -1.0, 4.0 ), -- 7 - In Exact_Result_Test. ( 45.0, Pi/4.0, Sqrt2/2.0, 5.0 ), -- 8 (135.0, 3.0*Pi/4.0, -Sqrt2/2.0, 5.0 ) ); -- 9 Cycle_Error, Radian_Error : Real; begin for I in Arcsin_Test_Data'Range loop -- note exact result requirements A.5.1(38);6.0 and -- G.2.4(12);6.0 if Arcsin_Test_Data (I).Error_Bound = 0.0 then Cycle_Error := 0.0; Radian_Error := 0.0; else Cycle_Error := Arcsin_Test_Data (I).Error_Bound; -- allow for rounding error in the specification of Pi Radian_Error := Cycle_Error + 1.0; end if; Check (Arcsin (Arcsin_Test_Data (I).Argument), Arcsin_Test_Data (I).Radians, "test" & Integer'Image (I) & " arcsin(" & Real'Image (Arcsin_Test_Data (I).Argument) & ")", Radian_Error);--pwb-math Check (Arcsin (Arcsin_Test_Data (I).Argument, 2.0 * Pi),--pwb-math Arcsin_Test_Data (I).Radians,--pwb-math "test" & Integer'Image (I) &--pwb-math " arcsin(" &--pwb-math Real'Image (Arcsin_Test_Data (I).Argument) &--pwb-math ", 2pi)",--pwb-math Cycle_Error); Check (Arcsin (Arcsin_Test_Data (I).Argument, 360.0), Arcsin_Test_Data (I).Degrees, "test" & Integer'Image (I) & " arcsin(" & Real'Image (Arcsin_Test_Data (I).Argument) & ", 360)", Cycle_Error); end loop; for I in Arccos_Test_Data'Range loop -- note exact result requirements A.5.1(39);6.0 and -- G.2.4(12);6.0 if Arccos_Test_Data (I).Error_Bound = 0.0 then Cycle_Error := 0.0; Radian_Error := 0.0; else Cycle_Error := Arccos_Test_Data (I).Error_Bound; -- allow for rounding error in the specification of Pi Radian_Error := Cycle_Error + 1.0; end if; Check (Arccos (Arccos_Test_Data (I).Argument), Arccos_Test_Data (I).Radians, "test" & Integer'Image (I) & " arccos(" & Real'Image (Arccos_Test_Data (I).Argument) & ")", Radian_Error);--pwb-math Check (Arccos (Arccos_Test_Data (I).Argument, 2.0 * Pi),--pwb-math Arccos_Test_Data (I).Radians,--pwb-math "test" & Integer'Image (I) &--pwb-math " arccos(" &--pwb-math Real'Image (Arccos_Test_Data (I).Argument) &--pwb-math ", 2pi)",--pwb-math Cycle_Error); Check (Arccos (Arccos_Test_Data (I).Argument, 360.0), Arccos_Test_Data (I).Degrees, "test" & Integer'Image (I) & " arccos(" & Real'Image (Arccos_Test_Data (I).Argument) & ", 360)", Cycle_Error); end loop; 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 Check_Exact (Actual, Expected_Low, Expected_High : Real; Test_Name : String) is -- If the expected result is not a model number, then Expected_Low is -- the first machine number less than the (exact) expected -- result, and Expected_High is the first machine number greater than -- the (exact) expected result. If the expected result is a model -- number, Expected_Low = Expected_High = the result. Model_Expected_Low : Real := Expected_Low; Model_Expected_High : Real := Expected_High; begin -- Calculate the first model number nearest to, but below (or equal) -- to the expected result: while Real'Model (Model_Expected_Low) /= Model_Expected_Low loop -- Try the next machine number lower: Model_Expected_Low := Real'Adjacent(Model_Expected_Low, 0.0); end loop;
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?