diff options
Diffstat (limited to 'math/aarch64/sve/atan2.c')
| -rw-r--r-- | math/aarch64/sve/atan2.c | 118 |
1 files changed, 118 insertions, 0 deletions
diff --git a/math/aarch64/sve/atan2.c b/math/aarch64/sve/atan2.c new file mode 100644 index 000000000000..1e1d00678b1d --- /dev/null +++ b/math/aarch64/sve/atan2.c @@ -0,0 +1,118 @@ +/* + * Double-precision vector atan2(x) function. + * + * Copyright (c) 2021-2024, Arm Limited. + * SPDX-License-Identifier: MIT OR Apache-2.0 WITH LLVM-exception + */ + +#include "sv_math.h" +#include "test_sig.h" +#include "test_defs.h" +#include "sv_poly_f64.h" + +static const struct data +{ + float64_t poly[20]; + float64_t pi_over_2; +} data = { + /* Coefficients of polynomial P such that atan(x)~x+x*P(x^2) on + [2**-1022, 1.0]. */ + .poly = { -0x1.5555555555555p-2, 0x1.99999999996c1p-3, -0x1.2492492478f88p-3, + 0x1.c71c71bc3951cp-4, -0x1.745d160a7e368p-4, 0x1.3b139b6a88ba1p-4, + -0x1.11100ee084227p-4, 0x1.e1d0f9696f63bp-5, -0x1.aebfe7b418581p-5, + 0x1.842dbe9b0d916p-5, -0x1.5d30140ae5e99p-5, 0x1.338e31eb2fbbcp-5, + -0x1.00e6eece7de8p-5, 0x1.860897b29e5efp-6, -0x1.0051381722a59p-6, + 0x1.14e9dc19a4a4ep-7, -0x1.d0062b42fe3bfp-9, 0x1.17739e210171ap-10, + -0x1.ab24da7be7402p-13, 0x1.358851160a528p-16, }, + .pi_over_2 = 0x1.921fb54442d18p+0, +}; + +/* Special cases i.e. 0, infinity, nan (fall back to scalar calls). */ +static svfloat64_t NOINLINE +special_case (svfloat64_t y, svfloat64_t x, svfloat64_t ret, + const svbool_t cmp) +{ + return sv_call2_f64 (atan2, y, x, ret, cmp); +} + +/* Returns a predicate indicating true if the input is the bit representation + of 0, infinity or nan. */ +static inline svbool_t +zeroinfnan (svuint64_t i, const svbool_t pg) +{ + return svcmpge (pg, svsub_x (pg, svlsl_x (pg, i, 1), 1), + sv_u64 (2 * asuint64 (INFINITY) - 1)); +} + +/* Fast implementation of SVE atan2. Errors are greatest when y and + x are reasonably close together. The greatest observed error is 2.28 ULP: + _ZGVsMxvv_atan2 (-0x1.5915b1498e82fp+732, 0x1.54d11ef838826p+732) + got -0x1.954f42f1fa841p-1 want -0x1.954f42f1fa843p-1. */ +svfloat64_t SV_NAME_D2 (atan2) (svfloat64_t y, svfloat64_t x, + const svbool_t pg) +{ + const struct data *data_ptr = ptr_barrier (&data); + + svuint64_t ix = svreinterpret_u64 (x); + svuint64_t iy = svreinterpret_u64 (y); + + svbool_t cmp_x = zeroinfnan (ix, pg); + svbool_t cmp_y = zeroinfnan (iy, pg); + svbool_t cmp_xy = svorr_z (pg, cmp_x, cmp_y); + + svfloat64_t ax = svabs_x (pg, x); + svfloat64_t ay = svabs_x (pg, y); + svuint64_t iax = svreinterpret_u64 (ax); + svuint64_t iay = svreinterpret_u64 (ay); + + svuint64_t sign_x = sveor_x (pg, ix, iax); + svuint64_t sign_y = sveor_x (pg, iy, iay); + svuint64_t sign_xy = sveor_x (pg, sign_x, sign_y); + + svbool_t pred_aygtax = svcmpgt (pg, ay, ax); + + /* Set up z for call to atan. */ + svfloat64_t n = svsel (pred_aygtax, svneg_x (pg, ax), ay); + svfloat64_t d = svsel (pred_aygtax, ay, ax); + svfloat64_t z = svdiv_x (pg, n, d); + + /* Work out the correct shift. */ + svfloat64_t shift = svreinterpret_f64 (svlsr_x (pg, sign_x, 1)); + shift = svsel (pred_aygtax, sv_f64 (1.0), shift); + shift = svreinterpret_f64 (svorr_x (pg, sign_x, svreinterpret_u64 (shift))); + shift = svmul_x (pg, shift, data_ptr->pi_over_2); + + /* Use split Estrin scheme for P(z^2) with deg(P)=19. */ + svfloat64_t z2 = svmul_x (pg, z, z); + svfloat64_t x2 = svmul_x (pg, z2, z2); + svfloat64_t x4 = svmul_x (pg, x2, x2); + svfloat64_t x8 = svmul_x (pg, x4, x4); + + svfloat64_t ret = svmla_x ( + pg, sv_estrin_7_f64_x (pg, z2, x2, x4, data_ptr->poly), + sv_estrin_11_f64_x (pg, z2, x2, x4, x8, data_ptr->poly + 8), x8); + + /* y = shift + z + z^3 * P(z^2). */ + svfloat64_t z3 = svmul_x (pg, z2, z); + ret = svmla_x (pg, z, z3, ret); + + ret = svadd_m (pg, ret, shift); + + /* Account for the sign of x and y. */ + if (unlikely (svptest_any (pg, cmp_xy))) + return special_case ( + y, x, + svreinterpret_f64 (sveor_x (pg, svreinterpret_u64 (ret), sign_xy)), + cmp_xy); + return svreinterpret_f64 (sveor_x (pg, svreinterpret_u64 (ret), sign_xy)); +} + +/* Arity of 2 means no mathbench entry emitted. See test/mathbench_funcs.h. */ +TEST_SIG (SV, D, 2, atan2) +TEST_ULP (SV_NAME_D2 (atan2), 1.78) +TEST_DISABLE_FENV (SV_NAME_D2 (atan2)) +TEST_INTERVAL (SV_NAME_D2 (atan2), 0.0, 1.0, 40000) +TEST_INTERVAL (SV_NAME_D2 (atan2), 1.0, 100.0, 40000) +TEST_INTERVAL (SV_NAME_D2 (atan2), 100, inf, 40000) +TEST_INTERVAL (SV_NAME_D2 (atan2), -0, -inf, 40000) +CLOSE_SVE_ATTR |