| // Auto-generated file. Do not edit! |
| // Template: src/f32-vsqrt/neonfma-nr2fma1adj.c.in |
| // Generator: tools/xngen |
| // |
| // Copyright 2020 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 <math.h> |
| |
| #include <arm_neon.h> |
| |
| #include <xnnpack/common.h> |
| #include <xnnpack/vunary.h> |
| |
| |
| void xnn_f32_vsqrt_ukernel__neonfma_nr2fma1adj_x24( |
| size_t n, |
| const float* x, |
| float* y, |
| const union xnn_f32_sqrt_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS |
| { |
| assert(n != 0); |
| assert(n % sizeof(float) == 0); |
| |
| const float32x4_t vhalf = vmovq_n_f32(0.5f); |
| for (; n >= 24 * sizeof(float); n -= 24 * sizeof(float)) { |
| const float32x4_t vx0123 = vld1q_f32(x); x += 4; |
| const float32x4_t vx4567 = vld1q_f32(x); x += 4; |
| const float32x4_t vx89AB = vld1q_f32(x); x += 4; |
| const float32x4_t vxCDEF = vld1q_f32(x); x += 4; |
| const float32x4_t vxGHIJ = vld1q_f32(x); x += 4; |
| const float32x4_t vxKLMN = vld1q_f32(x); x += 4; |
| |
| const float32x4_t vrsqrtx0123 = vrsqrteq_f32(vx0123); |
| const float32x4_t vrsqrtx4567 = vrsqrteq_f32(vx4567); |
| const float32x4_t vrsqrtx89AB = vrsqrteq_f32(vx89AB); |
| const float32x4_t vrsqrtxCDEF = vrsqrteq_f32(vxCDEF); |
| const float32x4_t vrsqrtxGHIJ = vrsqrteq_f32(vxGHIJ); |
| const float32x4_t vrsqrtxKLMN = vrsqrteq_f32(vxKLMN); |
| |
| float32x4_t vsqrtx0123 = vmulq_f32(vrsqrtx0123, vx0123); |
| float32x4_t vhalfrsqrtx0123 = vmulq_f32(vrsqrtx0123, vhalf); |
| float32x4_t vsqrtx4567 = vmulq_f32(vrsqrtx4567, vx4567); |
| float32x4_t vhalfrsqrtx4567 = vmulq_f32(vrsqrtx4567, vhalf); |
| float32x4_t vsqrtx89AB = vmulq_f32(vrsqrtx89AB, vx89AB); |
| float32x4_t vhalfrsqrtx89AB = vmulq_f32(vrsqrtx89AB, vhalf); |
| float32x4_t vsqrtxCDEF = vmulq_f32(vrsqrtxCDEF, vxCDEF); |
| float32x4_t vhalfrsqrtxCDEF = vmulq_f32(vrsqrtxCDEF, vhalf); |
| float32x4_t vsqrtxGHIJ = vmulq_f32(vrsqrtxGHIJ, vxGHIJ); |
| float32x4_t vhalfrsqrtxGHIJ = vmulq_f32(vrsqrtxGHIJ, vhalf); |
| float32x4_t vsqrtxKLMN = vmulq_f32(vrsqrtxKLMN, vxKLMN); |
| float32x4_t vhalfrsqrtxKLMN = vmulq_f32(vrsqrtxKLMN, vhalf); |
| |
| float32x4_t vresidual0123 = vfmsq_f32(vhalf, vsqrtx0123, vhalfrsqrtx0123); |
| float32x4_t vresidual4567 = vfmsq_f32(vhalf, vsqrtx4567, vhalfrsqrtx4567); |
| float32x4_t vresidual89AB = vfmsq_f32(vhalf, vsqrtx89AB, vhalfrsqrtx89AB); |
| float32x4_t vresidualCDEF = vfmsq_f32(vhalf, vsqrtxCDEF, vhalfrsqrtxCDEF); |
| float32x4_t vresidualGHIJ = vfmsq_f32(vhalf, vsqrtxGHIJ, vhalfrsqrtxGHIJ); |
| float32x4_t vresidualKLMN = vfmsq_f32(vhalf, vsqrtxKLMN, vhalfrsqrtxKLMN); |
| |
| vhalfrsqrtx0123 = vfmaq_f32(vhalfrsqrtx0123, vresidual0123, vhalfrsqrtx0123); |
| vsqrtx0123 = vfmaq_f32(vsqrtx0123, vresidual0123, vsqrtx0123); |
| vhalfrsqrtx4567 = vfmaq_f32(vhalfrsqrtx4567, vresidual4567, vhalfrsqrtx4567); |
| vsqrtx4567 = vfmaq_f32(vsqrtx4567, vresidual4567, vsqrtx4567); |
| vhalfrsqrtx89AB = vfmaq_f32(vhalfrsqrtx89AB, vresidual89AB, vhalfrsqrtx89AB); |
| vsqrtx89AB = vfmaq_f32(vsqrtx89AB, vresidual89AB, vsqrtx89AB); |
| vhalfrsqrtxCDEF = vfmaq_f32(vhalfrsqrtxCDEF, vresidualCDEF, vhalfrsqrtxCDEF); |
| vsqrtxCDEF = vfmaq_f32(vsqrtxCDEF, vresidualCDEF, vsqrtxCDEF); |
| vhalfrsqrtxGHIJ = vfmaq_f32(vhalfrsqrtxGHIJ, vresidualGHIJ, vhalfrsqrtxGHIJ); |
| vsqrtxGHIJ = vfmaq_f32(vsqrtxGHIJ, vresidualGHIJ, vsqrtxGHIJ); |
| vhalfrsqrtxKLMN = vfmaq_f32(vhalfrsqrtxKLMN, vresidualKLMN, vhalfrsqrtxKLMN); |
| vsqrtxKLMN = vfmaq_f32(vsqrtxKLMN, vresidualKLMN, vsqrtxKLMN); |
| |
| vresidual0123 = vfmsq_f32(vhalf, vsqrtx0123, vhalfrsqrtx0123); |
| vresidual4567 = vfmsq_f32(vhalf, vsqrtx4567, vhalfrsqrtx4567); |
| vresidual89AB = vfmsq_f32(vhalf, vsqrtx89AB, vhalfrsqrtx89AB); |
| vresidualCDEF = vfmsq_f32(vhalf, vsqrtxCDEF, vhalfrsqrtxCDEF); |
| vresidualGHIJ = vfmsq_f32(vhalf, vsqrtxGHIJ, vhalfrsqrtxGHIJ); |
| vresidualKLMN = vfmsq_f32(vhalf, vsqrtxKLMN, vhalfrsqrtxKLMN); |
| |
| vhalfrsqrtx0123 = vfmaq_f32(vhalfrsqrtx0123, vresidual0123, vhalfrsqrtx0123); |
| vsqrtx0123 = vfmaq_f32(vsqrtx0123, vresidual0123, vsqrtx0123); |
| vhalfrsqrtx4567 = vfmaq_f32(vhalfrsqrtx4567, vresidual4567, vhalfrsqrtx4567); |
| vsqrtx4567 = vfmaq_f32(vsqrtx4567, vresidual4567, vsqrtx4567); |
| vhalfrsqrtx89AB = vfmaq_f32(vhalfrsqrtx89AB, vresidual89AB, vhalfrsqrtx89AB); |
| vsqrtx89AB = vfmaq_f32(vsqrtx89AB, vresidual89AB, vsqrtx89AB); |
| vhalfrsqrtxCDEF = vfmaq_f32(vhalfrsqrtxCDEF, vresidualCDEF, vhalfrsqrtxCDEF); |
| vsqrtxCDEF = vfmaq_f32(vsqrtxCDEF, vresidualCDEF, vsqrtxCDEF); |
| vhalfrsqrtxGHIJ = vfmaq_f32(vhalfrsqrtxGHIJ, vresidualGHIJ, vhalfrsqrtxGHIJ); |
| vsqrtxGHIJ = vfmaq_f32(vsqrtxGHIJ, vresidualGHIJ, vsqrtxGHIJ); |
| vhalfrsqrtxKLMN = vfmaq_f32(vhalfrsqrtxKLMN, vresidualKLMN, vhalfrsqrtxKLMN); |
| vsqrtxKLMN = vfmaq_f32(vsqrtxKLMN, vresidualKLMN, vsqrtxKLMN); |
| |
| const float32x4_t vadjustment0123 = vfmsq_f32(vx0123, vsqrtx0123, vsqrtx0123); |
| const float32x4_t vadjustment4567 = vfmsq_f32(vx4567, vsqrtx4567, vsqrtx4567); |
| const float32x4_t vadjustment89AB = vfmsq_f32(vx89AB, vsqrtx89AB, vsqrtx89AB); |
| const float32x4_t vadjustmentCDEF = vfmsq_f32(vxCDEF, vsqrtxCDEF, vsqrtxCDEF); |
| const float32x4_t vadjustmentGHIJ = vfmsq_f32(vxGHIJ, vsqrtxGHIJ, vsqrtxGHIJ); |
| const float32x4_t vadjustmentKLMN = vfmsq_f32(vxKLMN, vsqrtxKLMN, vsqrtxKLMN); |
| |
| const float32x4_t vy0123 = vfmaq_f32(vsqrtx0123, vhalfrsqrtx0123, vadjustment0123); |
| const float32x4_t vy4567 = vfmaq_f32(vsqrtx4567, vhalfrsqrtx4567, vadjustment4567); |
| const float32x4_t vy89AB = vfmaq_f32(vsqrtx89AB, vhalfrsqrtx89AB, vadjustment89AB); |
| const float32x4_t vyCDEF = vfmaq_f32(vsqrtxCDEF, vhalfrsqrtxCDEF, vadjustmentCDEF); |
| const float32x4_t vyGHIJ = vfmaq_f32(vsqrtxGHIJ, vhalfrsqrtxGHIJ, vadjustmentGHIJ); |
| const float32x4_t vyKLMN = vfmaq_f32(vsqrtxKLMN, vhalfrsqrtxKLMN, vadjustmentKLMN); |
| |
| vst1q_f32(y, vy0123); y += 4; |
| vst1q_f32(y, vy4567); y += 4; |
| vst1q_f32(y, vy89AB); y += 4; |
| vst1q_f32(y, vyCDEF); y += 4; |
| vst1q_f32(y, vyGHIJ); y += 4; |
| vst1q_f32(y, vyKLMN); y += 4; |
| } |
| if XNN_UNLIKELY(n != 0) { |
| do { |
| const float vx = *x++; |
| const float vy = sqrtf(vx); |
| *y++ = vy; |
| n -= sizeof(float); |
| } while (n != 0); |
| } |
| } |