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
|
const builtin = @import("builtin");
const std = @import("std");
const maxInt = std.math.maxInt;
const FLT_MANT_DIG = 24;
inline fn floatXisf(comptime T: type, arg: T) f32 {
@setRuntimeSafety(builtin.is_test);
const bits = @typeInfo(T).Int.bits;
const Z = std.meta.Int(.unsigned, bits);
const S = std.meta.Int(.unsigned, bits - @clz(Z, @as(Z, bits) - 1));
if (arg == 0) {
return @as(f32, 0.0);
}
var ai = arg;
const N: u32 = bits;
const si = ai >> @intCast(S, (N - 1));
ai = ((ai ^ si) -% si);
var a = @bitCast(Z, ai);
const sd = @bitCast(i32, N - @clz(Z, a)); // number of significant digits
var e: i32 = sd - 1; // exponent
if (sd > FLT_MANT_DIG) {
// start: 0000000000000000000001xxxxxxxxxxxxxxxxxxxxxxPQxxxxxxxxxxxxxxxxxx
// finish: 000000000000000000000000000000000000001xxxxxxxxxxxxxxxxxxxxxxPQR
// 12345678901234567890123456
// 1 = msb 1 bit
// P = bit FLT_MANT_DIG-1 bits to the right of 1
// Q = bit FLT_MANT_DIG bits to the right of 1
// R = "or" of all bits to the right of Q
switch (sd) {
FLT_MANT_DIG + 1 => {
a <<= 1;
},
FLT_MANT_DIG + 2 => {},
else => {
const shift1_amt = @intCast(i32, sd - (FLT_MANT_DIG + 2));
const shift1_amt_u7 = @intCast(S, shift1_amt);
const shift2_amt = @intCast(i32, N + (FLT_MANT_DIG + 2)) - sd;
const shift2_amt_u7 = @intCast(S, shift2_amt);
a = (a >> shift1_amt_u7) | @boolToInt((a & (@intCast(Z, maxInt(Z)) >> shift2_amt_u7)) != 0);
},
}
// finish
a |= @boolToInt((a & 4) != 0); // Or P into R
a += 1; // round - this step may add a significant bit
a >>= 2; // dump Q and R
// a is now rounded to FLT_MANT_DIG or FLT_MANT_DIG+1 bits
if ((a & (@as(Z, 1) << FLT_MANT_DIG)) != 0) {
a >>= 1;
e += 1;
}
// a is now rounded to FLT_MANT_DIG bits
} else {
a <<= @intCast(S, FLT_MANT_DIG - sd);
// a is now rounded to FLT_MANT_DIG bits
}
const s = @bitCast(Z, arg) >> (@typeInfo(T).Int.bits - 32);
const r = (@intCast(u32, s) & 0x80000000) | // sign
(@intCast(u32, (e + 127)) << 23) | // exponent
(@truncate(u32, a) & 0x007fffff); // mantissa-high
return @bitCast(f32, r);
}
pub fn __floatdisf(arg: i64) callconv(.C) f32 {
return floatXisf(i64, arg);
}
pub fn __floattisf(arg: i128) callconv(.C) f32 {
return floatXisf(i128, arg);
}
pub fn __aeabi_l2f(arg: i64) callconv(.AAPCS) f32 {
return floatXisf(i64, arg);
}
test {
_ = @import("floattisf_test.zig");
}
test {
_ = @import("floattisf_test.zig");
}
|
