cfgexpand: Expand comment on when non-var clobbers can show up
[official-gcc.git] / gcc / ada / libgnat / s-fore_f.adb
blob91d45c0b40e4f2c4d32f7429057cbb4f9c05cca0
1 ------------------------------------------------------------------------------
2 -- --
3 -- GNAT RUN-TIME COMPONENTS --
4 -- --
5 -- S Y S T E M . F O R E _ F --
6 -- --
7 -- B o d y --
8 -- --
9 -- Copyright (C) 2020-2024, Free Software Foundation, Inc. --
10 -- --
11 -- GNAT is free software; you can redistribute it and/or modify it under --
12 -- terms of the GNU General Public License as published by the Free Soft- --
13 -- ware Foundation; either version 3, or (at your option) any later ver- --
14 -- sion. GNAT is distributed in the hope that it will be useful, but WITH- --
15 -- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY --
16 -- or FITNESS FOR A PARTICULAR PURPOSE. --
17 -- --
18 -- As a special exception under Section 7 of GPL version 3, you are granted --
19 -- additional permissions described in the GCC Runtime Library Exception, --
20 -- version 3.1, as published by the Free Software Foundation. --
21 -- --
22 -- You should have received a copy of the GNU General Public License and --
23 -- a copy of the GCC Runtime Library Exception along with this program; --
24 -- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see --
25 -- <http://www.gnu.org/licenses/>. --
26 -- --
27 -- GNAT was originally developed by the GNAT team at New York University. --
28 -- Extensive contributions were provided by Ada Core Technologies Inc. --
29 -- --
30 ------------------------------------------------------------------------------
32 package body System.Fore_F is
34 Maxdigs : constant Natural := Int'Width - 2;
35 -- Maximum number of decimal digits that can be represented in an Int.
36 -- The "-2" accounts for the sign and one extra digit, since we need the
37 -- maximum number of 9's that can be represented, e.g. for the 64-bit case,
38 -- Integer_64'Width is 20 since the maximum value is approximately 9.2E+18
39 -- and has 19 digits, but the maximum number of 9's that can be represented
40 -- in Integer_64 is only 18.
42 -- The first prerequisite of the implementation is that the scaled divide
43 -- does not overflow, which means that the absolute value of the bounds of
44 -- the subtype must be smaller than 10**Maxdigs * 2**(Int'Size - 1).
45 -- Otherwise Constraint_Error is raised by the scaled divide operation.
47 -- The second prerequisite is that the computation of the operands does not
48 -- overflow, which means that, if the small is larger than 1, it is either
49 -- an integer or its numerator and denominator must be both smaller than
50 -- the power 10**(Maxdigs - 1).
52 ----------------
53 -- Fore_Fixed --
54 ----------------
56 function Fore_Fixed (Lo, Hi, Num, Den : Int; Scale : Integer) return Natural
58 pragma Assert (Num < 0 and then Den < 0);
59 -- Accept only negative numbers to allow -2**(Int'Size - 1)
61 function Negative_Abs (Val : Int) return Int is
62 (if Val <= 0 then Val else -Val);
63 -- Return the opposite of the absolute value of Val
65 T : Int := Int'Min (Negative_Abs (Lo), Negative_Abs (Hi));
66 F : Natural;
68 Q, R : Int;
70 begin
71 -- Initial value of 2 allows for sign and mandatory single digit
73 F := 2;
75 -- The easy case is when Num is not larger than Den in magnitude,
76 -- i.e. if S = Num / Den, then S <= 1, in which case we can just
77 -- compute the product Q = T * S.
79 if Num >= Den then
80 Scaled_Divide (T, Num, Den, Q, R, Round => False);
81 T := Q;
83 -- Otherwise S > 1 and thus Scale <= 0, compute Q and R such that
85 -- T * Num = Q * (Den * 10**(-D)) + R
87 -- with
89 -- D = Integer'Max (-Maxdigs, Scale - 1)
91 -- then reason on Q if it is non-zero or else on R / Den.
93 -- This works only if Den * 10**(-D) does not overflow, which is true
94 -- if Den = 1. Suppose that Num corresponds to the maximum value of -D,
95 -- i.e. Maxdigs and 10**(-D) = 10**Maxdigs. If you change Den into 10,
96 -- then S becomes 10 times smaller and, therefore, Scale is incremented
97 -- by 1, which means that -D is decremented by 1 provided that Scale was
98 -- initially not smaller than 1 - Maxdigs, so the multiplication still
99 -- does not overflow. But you need to reach 10 to trigger this effect,
100 -- which means that a leeway of 10 is required, so let's restrict this
101 -- to a Num for which 10**(-D) <= 10**(Maxdigs - 1). To sum up, if S is
102 -- the ratio of two integers with
104 -- 1 < Den < Num <= B
106 -- where B is a fixed limit, then the multiplication does not overflow.
107 -- B can be taken as the largest integer Small such that D = 1 - Maxdigs
108 -- i.e. such that Scale = 2 - Maxdigs, which is 10**(Maxdigs - 1) - 1.
110 else
111 declare
112 D : constant Integer := Integer'Max (-Maxdigs, Scale - 1);
114 begin
115 Scaled_Divide (T, Num, Den * 10**(-D), Q, R, Round => False);
117 if Q /= 0 then
118 T := Q;
119 F := F - D;
120 else
121 T := R / Den;
122 end if;
123 end;
124 end if;
126 -- Loop to increase Fore as needed to include full range of values
128 while T <= -10 or else T >= 10 loop
129 T := T / 10;
130 F := F + 1;
131 end loop;
133 return F;
134 end Fore_Fixed;
136 end System.Fore_F;