COMPMID-3870: Create ActivationLayer SVE/SVE2
Adds support for ActivationLayer for SVE and SVE2.
Datatypes supported:
*FP32
*FP16
*QASYMM8
*QASYMM8_SIGNED
*QSYMM16
Change-Id: Ia3583891795cda4ca2f9fa27c440731a5c27710d
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4566
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/core/NEON/SVEMath.inl b/src/core/NEON/SVEMath.inl
new file mode 100644
index 0000000..5ebfeaa
--- /dev/null
+++ b/src/core/NEON/SVEMath.inl
@@ -0,0 +1,266 @@
+/*
+ * Copyright (c) 2020 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <cmath>
+#include <limits>
+
+#if defined(__ARM_FEATURE_SVE)
+
+namespace arm_compute
+{
+inline svfloat32_t svtaylor_poly_f32_z(svbool_t pg, svfloat32_t x, const std::array<svfloat32_t, 8> &coeffs)
+{
+ const auto A = svmla_f32_z(pg, coeffs[0], coeffs[4], x);
+ const auto B = svmla_f32_z(pg, coeffs[2], coeffs[6], x);
+ const auto C = svmla_f32_z(pg, coeffs[1], coeffs[5], x);
+ const auto D = svmla_f32_z(pg, coeffs[3], coeffs[7], x);
+ const auto x2 = svmul_f32_z(pg, x, x);
+ const auto x4 = svmul_f32_z(pg, x2, x2);
+ const auto res = svmla_f32_z(pg, svmla_f32_z(pg, A, B, x2), svmla_f32_z(pg, C, D, x2), x4);
+ return res;
+}
+
+inline svfloat16_t svtaylor_poly_f16_z(svbool_t pg, svfloat16_t x, const std::array<svfloat16_t, 8> &coeffs)
+{
+ const auto A = svmla_f16_z(pg, coeffs[0], coeffs[4], x);
+ const auto B = svmla_f16_z(pg, coeffs[2], coeffs[6], x);
+ const auto C = svmla_f16_z(pg, coeffs[1], coeffs[5], x);
+ const auto D = svmla_f16_z(pg, coeffs[3], coeffs[7], x);
+ const auto x2 = svmul_f16_z(pg, x, x);
+ const auto x4 = svmul_f16_z(pg, x2, x2);
+ const auto res = svmla_f16_z(pg, svmla_f16_z(pg, A, B, x2), svmla_f16_z(pg, C, D, x2), x4);
+ return res;
+}
+
+inline svfloat16_t svinv_f16_z(svbool_t pg, svfloat16_t x)
+{
+ auto recip = svrecpe_f16(x);
+ recip = svmul_f16_z(pg, svrecps_f16(x, recip), recip);
+ recip = svmul_f16_z(pg, svrecps_f16(x, recip), recip);
+ return recip;
+}
+
+inline svfloat32_t svinv_f32_z(svbool_t pg, svfloat32_t x)
+{
+ auto recip = svrecpe_f32(x);
+ recip = svmul_f32_z(pg, svrecps_f32(x, recip), recip);
+ recip = svmul_f32_z(pg, svrecps_f32(x, recip), recip);
+ return recip;
+}
+
+inline svfloat32_t svexp_f32_z(svbool_t pg, svfloat32_t x)
+{
+ const auto CONST_LN2 = svdup_n_f32(0.6931471805f); // ln(2)
+ const auto CONST_INV_LN2 = svdup_n_f32(1.4426950408f); // 1/ln(2)
+ const auto CONST_INF = svdup_n_f32(std::numeric_limits<float>::infinity());
+ const auto CONST_MAX_INPUT = svdup_n_f32(88.7f);
+ const auto CONST_0 = svdup_n_f32(0.f);
+ const auto CONST_NEGATIVE_126 = svdup_n_s32(-126);
+
+ /** Exponent polynomial coefficients */
+ const std::array<svfloat32_t, 8> exp_tab =
+ {
+ {
+ svdup_n_f32(1.f),
+ svdup_n_f32(0.0416598916054f),
+ svdup_n_f32(0.500000596046f),
+ svdup_n_f32(0.0014122662833f),
+ svdup_n_f32(1.00000011921f),
+ svdup_n_f32(0.00833693705499f),
+ svdup_n_f32(0.166665703058f),
+ svdup_n_f32(0.000195780929062f),
+ }
+ };
+
+ // Perform range reduction [-log(2),log(2)]
+ auto m = svcvt_s32_f32_z(pg, svmul_f32_z(pg, x, CONST_INV_LN2));
+ auto val = svmls_f32_z(pg, x, svcvt_f32_s32_z(pg, m), CONST_LN2);
+
+ // Polynomial Approximation
+ auto poly = svtaylor_poly_f32_z(pg, val, exp_tab);
+
+ // Reconstruct
+ poly = svreinterpret_f32_s32(svqadd_s32(svreinterpret_s32_f32(poly), svlsl_n_s32_z(pg, m, 23)));
+
+ // Handle underflow
+ svbool_t ltpg = svcmplt_s32(pg, m, CONST_NEGATIVE_126);
+ poly = svsel_f32(ltpg, CONST_0, poly);
+
+ // Handle overflow
+ svbool_t gtpg = svcmpgt_f32(pg, x, CONST_MAX_INPUT);
+ poly = svsel_f32(gtpg, CONST_INF, poly);
+
+ return poly;
+}
+
+inline svfloat16_t svexp_f16_z(svbool_t pg, svfloat16_t x)
+{
+ const auto CONST_LN2 = svdup_n_f16(0.6931471805f); // ln(2)
+ const auto CONST_INV_LN2 = svdup_n_f16(1.4426950408f); // 1/ln(2)
+ const auto CONST_INF = svdup_n_f16(std::numeric_limits<float16_t>::infinity());
+ const auto CONST_MAX_INPUT = svdup_n_f16(88.7f);
+ const auto CONST_0 = svdup_n_f16(0.f);
+ const auto CONST_NEGATIVE_126 = svdup_n_s16(-126);
+
+ /** Exponent polynomial coefficients */
+ const std::array<svfloat16_t, 8> exp_tab =
+ {
+ {
+ svdup_n_f16(1.f),
+ svdup_n_f16(0.0416598916054f),
+ svdup_n_f16(0.500000596046f),
+ svdup_n_f16(0.0014122662833f),
+ svdup_n_f16(1.00000011921f),
+ svdup_n_f16(0.00833693705499f),
+ svdup_n_f16(0.166665703058f),
+ svdup_n_f16(0.000195780929062f),
+ }
+ };
+
+ // Perform range reduction [-log(2),log(2)]
+ auto m = svcvt_s16_f16_z(pg, svmul_f16_z(pg, x, CONST_INV_LN2));
+ auto val = svmls_f16_z(pg, x, svcvt_f16_s16_z(pg, m), CONST_LN2);
+
+ // Polynomial Approximation
+ auto poly = svtaylor_poly_f16_z(pg, val, exp_tab);
+
+ // Reconstruct
+ poly = svreinterpret_f16_s16(svqadd_s16(svreinterpret_s16_f16(poly), svlsl_n_s16_z(pg, m, 11)));
+
+ // Handle underflow
+ svbool_t ltpg = svcmplt_s16(pg, m, CONST_NEGATIVE_126);
+ poly = svsel_f16(ltpg, CONST_0, poly);
+
+ // Handle overflow
+ svbool_t gtpg = svcmpgt_f16(pg, x, CONST_MAX_INPUT);
+ poly = svsel_f16(gtpg, CONST_INF, poly);
+
+ return poly;
+}
+
+inline svfloat32_t svtanh_f32_z(svbool_t pg, svfloat32_t val)
+{
+ const svfloat32_t CONST_1 = svdup_n_f32(1.f);
+ const svfloat32_t CONST_2 = svdup_n_f32(2.f);
+ const svfloat32_t CONST_MIN_TANH = svdup_n_f32(-10.f);
+ const svfloat32_t CONST_MAX_TANH = svdup_n_f32(10.f);
+
+ svfloat32_t x = svmin_f32_z(pg, svmax_f32_z(pg, val, CONST_MIN_TANH), CONST_MAX_TANH);
+ svfloat32_t exp2x = svexp_f32_z(pg, svmul_f32_z(pg, CONST_2, x));
+ svfloat32_t num = svsub_f32_z(pg, exp2x, CONST_1);
+ svfloat32_t den = svadd_f32_z(pg, exp2x, CONST_1);
+ svfloat32_t tanh = svdiv_f32_z(pg, num, den);
+ return tanh;
+}
+
+inline svfloat16_t svtanh_f16_z(svbool_t pg, svfloat16_t val)
+{
+ const svfloat16_t CONST_1 = svdup_n_f16(1.f);
+ const svfloat16_t CONST_2 = svdup_n_f16(2.f);
+ const svfloat16_t CONST_MIN_TANH = svdup_n_f16(-10.f);
+ const svfloat16_t CONST_MAX_TANH = svdup_n_f16(10.f);
+
+ const svfloat16_t x = svmin_f16_z(pg, svmax_f16_z(pg, val, CONST_MIN_TANH), CONST_MAX_TANH);
+ const svfloat16_t exp2x = svexp_f16_z(pg, svmul_f16_z(pg, CONST_2, x));
+ const svfloat16_t num = svsub_f16_z(pg, exp2x, CONST_1);
+ const svfloat16_t den = svadd_f16_z(pg, exp2x, CONST_1);
+ const svfloat16_t tanh = svdiv_f16_z(pg, num, den);
+ return tanh;
+}
+
+inline svfloat32_t svlog_f32_z(svbool_t pg, svfloat32_t x)
+{
+#if defined(__ARM_FEATURE_SVE2)
+ return svcvt_f32_s32_z(pg, svlogb_f32_z(pg, x));
+#else /* !defined(__ARM_FEATURE_SVE2) */
+ /** Logarithm polynomial coefficients */
+ const std::array<svfloat32_t, 8> log_tab =
+ {
+ {
+ svdup_n_f32(-2.29561495781f),
+ svdup_n_f32(-2.47071170807f),
+ svdup_n_f32(-5.68692588806f),
+ svdup_n_f32(-0.165253549814f),
+ svdup_n_f32(5.17591238022f),
+ svdup_n_f32(0.844007015228f),
+ svdup_n_f32(4.58445882797f),
+ svdup_n_f32(0.0141278216615f),
+ }
+ };
+
+ const auto CONST_127 = svdup_n_s32(127); // 127
+ const auto CONST_LN2 = svdup_n_f32(0.6931471805f); // ln(2)
+
+ // Extract exponent
+ auto m = svsub_s32_z(pg, svasr_n_s32_z(pg, svreinterpret_s32_f32(x), 23), CONST_127);
+ auto val = svreinterpret_f32_s32(svsub_s32_z(pg, svreinterpret_s32_f32(x), svlsl_n_s32_z(pg, m, 23)));
+
+ // Polynomial Approximation
+ auto poly = svtaylor_poly_f32_z(pg, val, log_tab);
+
+ // Reconstruct
+ poly = svmla_f32_z(pg, poly, svcvt_f32_s32_z(pg, m), CONST_LN2);
+
+ return poly;
+#endif /* defined(__ARM_FEATURE_SVE2) */
+}
+
+inline svfloat16_t svlog_f16_z(svbool_t pg, svfloat16_t x)
+{
+#if defined(__ARM_FEATURE_SVE2)
+ return svcvt_f16_s16_z(pg, svlogb_f16_z(pg, x));
+#else /* !defined(__ARM_FEATURE_SVE2) */
+
+ /** Logarithm polynomial coefficients */
+ const std::array<svfloat16_t, 8> log_tab
+ {
+ {
+ svdup_n_f16(-2.29561495781f),
+ svdup_n_f16(-2.47071170807f),
+ svdup_n_f16(-5.68692588806f),
+ svdup_n_f16(-0.165253549814f),
+ svdup_n_f16(5.17591238022f),
+ svdup_n_f16(0.844007015228f),
+ svdup_n_f16(4.58445882797f),
+ svdup_n_f16(0.0141278216615f),
+ }
+ };
+
+ const auto CONST_7 = svdup_n_s16(7); // 7
+ const auto CONST_LN2 = svdup_n_f16(0.6931471805f); // ln(2)
+
+ // Extract exponent
+ auto m = svsub_s16_z(pg, svasr_n_s16_z(pg, svreinterpret_s16_f16(x), 11), CONST_7);
+ auto val = svreinterpret_f16_s16(svsub_s16_z(pg, svreinterpret_s16_f16(x), svlsl_n_s16_z(pg, m, 11)));
+
+ // Polynomial Approximation
+ auto poly = svtaylor_poly_f16_z(pg, val, log_tab);
+
+ // Reconstruct
+ poly = svmla_f16_z(pg, poly, svcvt_f16_s16_z(pg, m), CONST_LN2);
+
+ return poly;
+#endif /* defined(__ARM_FEATURE_SVE2) */
+}
+} // namespace arm_compute
+#endif /* defined(__ARM_FEATURE_SVE) */