zig/lib/compiler_rt/int_from_float.zig
2025-03-08 11:22:46 -08:00

107 lines
4 KiB
Zig

const std = @import("std");
const Int = std.meta.Int;
const math = std.math;
const Log2Int = math.Log2Int;
pub inline fn intFromFloat(comptime I: type, a: anytype) I {
const F = @TypeOf(a);
const float_bits = @typeInfo(F).float.bits;
const int_bits = @typeInfo(I).int.bits;
const rep_t = Int(.unsigned, float_bits);
const sig_bits = math.floatMantissaBits(F);
const exp_bits = math.floatExponentBits(F);
const fractional_bits = math.floatFractionalBits(F);
const implicit_bit = if (F != f80) (@as(rep_t, 1) << sig_bits) else 0;
const max_exp = (1 << (exp_bits - 1));
const exp_bias = max_exp - 1;
const sig_mask = (@as(rep_t, 1) << sig_bits) - 1;
// Break a into sign, exponent, significand
const a_rep: rep_t = @bitCast(a);
const negative = (a_rep >> (float_bits - 1)) != 0;
const exponent = @as(i32, @intCast((a_rep << 1) >> (sig_bits + 1))) - exp_bias;
const significand: rep_t = (a_rep & sig_mask) | implicit_bit;
// If the exponent is negative, the result rounds to zero.
if (exponent < 0) return 0;
// If the value is too large for the integer type, saturate.
switch (@typeInfo(I).int.signedness) {
.unsigned => {
if (negative) return 0;
if (@as(c_uint, @intCast(exponent)) >= @min(int_bits, max_exp)) return math.maxInt(I);
},
.signed => if (@as(c_uint, @intCast(exponent)) >= @min(int_bits - 1, max_exp)) {
return if (negative) math.minInt(I) else math.maxInt(I);
},
}
// If 0 <= exponent < sig_bits, right shift to get the result.
// Otherwise, shift left.
var result: I = undefined;
if (exponent < fractional_bits) {
result = @intCast(significand >> @intCast(fractional_bits - exponent));
} else {
result = @as(I, @intCast(significand)) << @intCast(exponent - fractional_bits);
}
if ((@typeInfo(I).int.signedness == .signed) and negative)
return ~result +% 1;
return result;
}
pub inline fn bigIntFromFloat(comptime signedness: std.builtin.Signedness, result: []u32, a: anytype) void {
switch (result.len) {
0 => return,
inline 1...4 => |limbs_len| {
result[0..limbs_len].* = @bitCast(@as(
@Type(.{ .int = .{ .signedness = signedness, .bits = 32 * limbs_len } }),
@intFromFloat(a),
));
return;
},
else => {},
}
// sign implicit fraction
const significand_bits = 1 + math.floatFractionalBits(@TypeOf(a));
const I = @Type(comptime .{ .int = .{
.signedness = signedness,
.bits = @as(u16, @intFromBool(signedness == .signed)) + significand_bits,
} });
const parts = math.frexp(a);
const significand_bits_adjusted_to_handle_smin = @as(i32, significand_bits) +
@intFromBool(signedness == .signed and parts.exponent == 32 * result.len);
const exponent = @max(parts.exponent - significand_bits_adjusted_to_handle_smin, 0);
const int: I = @intFromFloat(switch (exponent) {
0 => a,
else => math.ldexp(parts.significand, significand_bits_adjusted_to_handle_smin),
});
switch (signedness) {
.signed => {
const endian = @import("builtin").cpu.arch.endian();
const exponent_limb = switch (endian) {
.little => exponent / 32,
.big => result.len - 1 - exponent / 32,
};
const sign_bits: u32 = if (int < 0) math.maxInt(u32) else 0;
@memset(result[0..exponent_limb], switch (endian) {
.little => 0,
.big => sign_bits,
});
result[exponent_limb] = sign_bits << @truncate(exponent);
@memset(result[exponent_limb + 1 ..], switch (endian) {
.little => sign_bits,
.big => 0,
});
},
.unsigned => @memset(result, 0),
}
std.mem.writePackedIntNative(I, std.mem.sliceAsBytes(result), exponent, int);
}
test {
_ = @import("int_from_float_test.zig");
}