//! a raised to integer power of b //! ported from https://github.com/llvm-mirror/compiler-rt/blob/release_80/lib/builtins/powisf2.c //! Multiplication order (left-to-right or right-to-left) does not matter for //! error propagation and this method is optimized for performance, not accuracy. const builtin = @import("builtin"); const common = @import("common.zig"); const std = @import("std"); pub const panic = common.panic; comptime { @export(&__powihf2, .{ .name = "__powihf2", .linkage = common.linkage, .visibility = common.visibility }); @export(&__powisf2, .{ .name = "__powisf2", .linkage = common.linkage, .visibility = common.visibility }); @export(&__powidf2, .{ .name = "__powidf2", .linkage = common.linkage, .visibility = common.visibility }); if (common.want_ppc_abi) @export(&__powitf2, .{ .name = "__powikf2", .linkage = common.linkage, .visibility = common.visibility }); @export(&__powitf2, .{ .name = "__powitf2", .linkage = common.linkage, .visibility = common.visibility }); @export(&__powixf2, .{ .name = "__powixf2", .linkage = common.linkage, .visibility = common.visibility }); } inline fn powiXf2(comptime FT: type, a: FT, b: i32) FT { var x_a: FT = a; var x_b: i32 = b; const is_recip: bool = b < 0; var r: FT = 1.0; while (true) { if (@as(u32, @bitCast(x_b)) & @as(u32, 1) != 0) { r *= x_a; } x_b = @divTrunc(x_b, @as(i32, 2)); if (x_b == 0) break; x_a *= x_a; // Multiplication of x_a propagates the error } return if (is_recip) 1 / r else r; } pub fn __powihf2(a: f16, b: i32) callconv(.c) f16 { return powiXf2(f16, a, b); } pub fn __powisf2(a: f32, b: i32) callconv(.c) f32 { return powiXf2(f32, a, b); } pub fn __powidf2(a: f64, b: i32) callconv(.c) f64 { return powiXf2(f64, a, b); } pub fn __powitf2(a: f128, b: i32) callconv(.c) f128 { return powiXf2(f128, a, b); } pub fn __powixf2(a: f80, b: i32) callconv(.c) f80 { return powiXf2(f80, a, b); } test { _ = @import("powiXf2_test.zig"); }