| // Auto-generated file. Do not edit! |
| // Template: src/f16-vsigmoid/neonfp16arith.c.in |
| // Generator: tools/xngen |
| // |
| // Copyright 2022 Google LLC |
| // |
| // This source code is licensed under the BSD-style license found in the |
| // LICENSE file in the root directory of this source tree. |
| |
| #include <assert.h> |
| |
| #include <arm_neon.h> |
| |
| #include <xnnpack/common.h> |
| #include <xnnpack/vunary.h> |
| |
| |
| void xnn_f16_vsigmoid_ukernel__neonfp16arith_rr1_p3_recpe_x32( |
| size_t batch, |
| const void* input, |
| void* output, |
| const union xnn_f16_sigmoid_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS |
| { |
| assert(batch % sizeof(__fp16) == 0); |
| |
| const float16x8_t vmagic_bias = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.magic_bias)); |
| const float16x8_t vminus_log2e = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.minus_log2e)); |
| const float16x8_t vln2 = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.ln2)); |
| const float16x8_t vc3 = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.c3)); |
| const float16x8_t vc2 = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.c2)); |
| const float16x8_t vone = vmovq_n_f16(1.0f); |
| const float16x8_t vdenorm_cutoff = vreinterpretq_f16_u16(vld1q_dup_u16(¶ms->neonfp16arith_rr1_p3.denorm_cutoff)); |
| |
| const __fp16* i = (const __fp16*) input; |
| __fp16* o = (__fp16*) output; |
| for (; batch >= 32 * sizeof(__fp16); batch -= 32 * sizeof(__fp16)) { |
| const float16x8_t vx0 = vld1q_f16(i); i += 8; |
| const float16x8_t vx1 = vld1q_f16(i); i += 8; |
| const float16x8_t vx2 = vld1q_f16(i); i += 8; |
| const float16x8_t vx3 = vld1q_f16(i); i += 8; |
| |
| const float16x8_t vz0 = vabsq_f16(vx0); |
| const float16x8_t vz1 = vabsq_f16(vx1); |
| const float16x8_t vz2 = vabsq_f16(vx2); |
| const float16x8_t vz3 = vabsq_f16(vx3); |
| |
| float16x8_t vn0 = vfmaq_f16(vmagic_bias, vz0, vminus_log2e); |
| float16x8_t vn1 = vfmaq_f16(vmagic_bias, vz1, vminus_log2e); |
| float16x8_t vn2 = vfmaq_f16(vmagic_bias, vz2, vminus_log2e); |
| float16x8_t vn3 = vfmaq_f16(vmagic_bias, vz3, vminus_log2e); |
| |
| const float16x8_t vs0 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn0), 10)); |
| const float16x8_t vs1 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn1), 10)); |
| const float16x8_t vs2 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn2), 10)); |
| const float16x8_t vs3 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn3), 10)); |
| |
| vn0 = vsubq_f16(vn0, vmagic_bias); |
| vn1 = vsubq_f16(vn1, vmagic_bias); |
| vn2 = vsubq_f16(vn2, vmagic_bias); |
| vn3 = vsubq_f16(vn3, vmagic_bias); |
| |
| float16x8_t vt0 = vfmaq_f16(vz0, vn0, vln2); |
| float16x8_t vt1 = vfmaq_f16(vz1, vn1, vln2); |
| float16x8_t vt2 = vfmaq_f16(vz2, vn2, vln2); |
| float16x8_t vt3 = vfmaq_f16(vz3, vn3, vln2); |
| |
| float16x8_t vp0 = vfmaq_f16(vc2, vc3, vt0); |
| float16x8_t vp1 = vfmaq_f16(vc2, vc3, vt1); |
| float16x8_t vp2 = vfmaq_f16(vc2, vc3, vt2); |
| float16x8_t vp3 = vfmaq_f16(vc2, vc3, vt3); |
| |
| vp0 = vfmsq_f16(vone, vp0, vt0); |
| vp1 = vfmsq_f16(vone, vp1, vt1); |
| vp2 = vfmsq_f16(vone, vp2, vt2); |
| vp3 = vfmsq_f16(vone, vp3, vt3); |
| |
| vt0 = vmulq_f16(vt0, vs0); |
| vt1 = vmulq_f16(vt1, vs1); |
| vt2 = vmulq_f16(vt2, vs2); |
| vt3 = vmulq_f16(vt3, vs3); |
| |
| const float16x8_t ve0 = vfmsq_f16(vs0, vp0, vt0); |
| const float16x8_t ve1 = vfmsq_f16(vs1, vp1, vt1); |
| const float16x8_t ve2 = vfmsq_f16(vs2, vp2, vt2); |
| const float16x8_t ve3 = vfmsq_f16(vs3, vp3, vt3); |
| |
| const float16x8_t vd0 = vaddq_f16(ve0, vone); |
| const float16x8_t vd1 = vaddq_f16(ve1, vone); |
| const float16x8_t vd2 = vaddq_f16(ve2, vone); |
| const float16x8_t vd3 = vaddq_f16(ve3, vone); |
| |
| float16x8_t vr0 = vrecpeq_f16(vd0); |
| float16x8_t vr1 = vrecpeq_f16(vd1); |
| float16x8_t vr2 = vrecpeq_f16(vd2); |
| float16x8_t vr3 = vrecpeq_f16(vd3); |
| |
| float16x8_t vf0 = vmulq_f16(ve0, vr0); |
| float16x8_t vf1 = vmulq_f16(ve1, vr1); |
| float16x8_t vf2 = vmulq_f16(ve2, vr2); |
| float16x8_t vf3 = vmulq_f16(ve3, vr3); |
| |
| vf0 = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf0), vcagtq_f16(vx0, vdenorm_cutoff))); |
| vf1 = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf1), vcagtq_f16(vx1, vdenorm_cutoff))); |
| vf2 = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf2), vcagtq_f16(vx2, vdenorm_cutoff))); |
| vf3 = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf3), vcagtq_f16(vx3, vdenorm_cutoff))); |
| |
| const uint16x8_t vm0 = vcltq_f16(vx0, vmovq_n_f16(0.0f)); |
| const uint16x8_t vm1 = vcltq_f16(vx1, vmovq_n_f16(0.0f)); |
| const uint16x8_t vm2 = vcltq_f16(vx2, vmovq_n_f16(0.0f)); |
| const uint16x8_t vm3 = vcltq_f16(vx3, vmovq_n_f16(0.0f)); |
| |
| vf0 = vbslq_f16(vm0, vf0, vsubq_f16(vone, vf0)); |
| vf1 = vbslq_f16(vm1, vf1, vsubq_f16(vone, vf1)); |
| vf2 = vbslq_f16(vm2, vf2, vsubq_f16(vone, vf2)); |
| vf3 = vbslq_f16(vm3, vf3, vsubq_f16(vone, vf3)); |
| |
| vst1q_f16(o, vf0); o += 8; |
| vst1q_f16(o, vf1); o += 8; |
| vst1q_f16(o, vf2); o += 8; |
| vst1q_f16(o, vf3); o += 8; |
| } |
| for (; batch >= 8 * sizeof(__fp16); batch -= 8 * sizeof(__fp16)) { |
| const float16x8_t vx = vld1q_f16(i); i += 8; |
| |
| const float16x8_t vz = vabsq_f16(vx); |
| |
| float16x8_t vn = vfmaq_f16(vmagic_bias, vz, vminus_log2e); |
| const float16x8_t vs = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn), 10)); |
| vn = vsubq_f16(vn, vmagic_bias); |
| float16x8_t vt = vfmaq_f16(vz, vn, vln2); |
| |
| float16x8_t vp = vfmaq_f16(vc2, vc3, vt); |
| vp = vfmsq_f16(vone, vp, vt); |
| |
| vt = vmulq_f16(vt, vs); |
| const float16x8_t ve = vfmsq_f16(vs, vp, vt); |
| const float16x8_t vd = vaddq_f16(ve, vone); |
| |
| const float16x8_t vr = vrecpeq_f16(vd); |
| float16x8_t vf = vmulq_f16(ve, vr); |
| vf = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf), vcagtq_f16(vx, vdenorm_cutoff))); |
| const uint16x8_t vm = vcltq_f16(vx, vmovq_n_f16(0.0f)); |
| vf = vbslq_f16(vm, vf, vsubq_f16(vone, vf)); |
| |
| vst1q_f16(o, vf); o += 8; |
| } |
| if XNN_UNLIKELY(batch != 0) { |
| const float16x8_t vx = vld1q_f16(i); |
| |
| const float16x8_t vz = vabsq_f16(vx); |
| |
| float16x8_t vn = vfmaq_f16(vmagic_bias, vz, vminus_log2e); |
| const float16x8_t vs = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn), 10)); |
| vn = vsubq_f16(vn, vmagic_bias); |
| float16x8_t vt = vfmaq_f16(vz, vn, vln2); |
| |
| float16x8_t vp = vfmaq_f16(vc2, vc3, vt); |
| vp = vfmsq_f16(vone, vp, vt); |
| |
| vt = vmulq_f16(vt, vs); |
| const float16x8_t ve = vfmsq_f16(vs, vp, vt); |
| const float16x8_t vd = vaddq_f16(ve, vone); |
| |
| const float16x8_t vr = vrecpeq_f16(vd); |
| float16x8_t vf = vmulq_f16(ve, vr); |
| vf = vreinterpretq_f16_u16(vbicq_u16(vreinterpretq_u16_f16(vf), vcagtq_f16(vx, vdenorm_cutoff))); |
| const uint16x8_t vm = vcltq_f16(vx, vmovq_n_f16(0.0f)); |
| vf = vbslq_f16(vm, vf, vsubq_f16(vone, vf)); |
| |
| float16x4_t vf_lo = vget_low_f16(vf); |
| if (batch & (4 * sizeof(__fp16))) { |
| vst1_f16(o, vf_lo); o += 4; |
| vf_lo = vget_high_f16(vf); |
| } |
| if (batch & (2 * sizeof(__fp16))) { |
| vst1_f16(o, vf_lo); o += 2; |
| vf_lo = vext_f16(vf_lo, vf_lo, 2); |
| } |
| if (batch & (1 * sizeof(__fp16))) { |
| vst1_lane_f16(o, vf_lo, 0); |
| } |
| } |
| } |