| // Auto-generated file. Do not edit! |
| // Template: src/f32-qs8-vcvt/scalar-lrintf.c.in |
| // Generator: tools/xngen |
| // |
| // Copyright 2021 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 <xnnpack/common.h> |
| #include <xnnpack/math.h> |
| #include <xnnpack/vcvt.h> |
| |
| |
| void xnn_f32_qu8_vcvt_ukernel__scalar_lrintf_x4( |
| size_t n, |
| const float* x, |
| uint8_t* y, |
| const union xnn_f32_qu8_cvt_params params[restrict XNN_MIN_ELEMENTS(1)]) |
| { |
| assert(n != 0); |
| assert(n % sizeof(float) == 0); |
| assert(x != NULL); |
| assert(y != NULL); |
| |
| const float vscale = params->scalar_lrintf.scale; |
| const float voutput_min_less_zero_point = params->scalar_lrintf.output_min_less_zero_point; |
| const float voutput_max_less_zero_point = params->scalar_lrintf.output_max_less_zero_point; |
| const int32_t voutput_zero_point = params->scalar_lrintf.output_zero_point; |
| |
| for (; n >= 4 * sizeof(float); n -= 4 * sizeof(float)) { |
| float vx0 = x[0]; |
| float vx1 = x[1]; |
| float vx2 = x[2]; |
| float vx3 = x[3]; |
| x += 4; |
| |
| vx0 *= vscale; |
| vx1 *= vscale; |
| vx2 *= vscale; |
| vx3 *= vscale; |
| |
| vx0 = math_max_f32(vx0, voutput_min_less_zero_point); |
| vx1 = math_max_f32(vx1, voutput_min_less_zero_point); |
| vx2 = math_max_f32(vx2, voutput_min_less_zero_point); |
| vx3 = math_max_f32(vx3, voutput_min_less_zero_point); |
| |
| vx0 = math_min_f32(vx0, voutput_max_less_zero_point); |
| vx1 = math_min_f32(vx1, voutput_max_less_zero_point); |
| vx2 = math_min_f32(vx2, voutput_max_less_zero_point); |
| vx3 = math_min_f32(vx3, voutput_max_less_zero_point); |
| |
| int32_t vy0 = (int32_t) lrintf(vx0); |
| int32_t vy1 = (int32_t) lrintf(vx1); |
| int32_t vy2 = (int32_t) lrintf(vx2); |
| int32_t vy3 = (int32_t) lrintf(vx3); |
| |
| vy0 += voutput_zero_point; |
| vy1 += voutput_zero_point; |
| vy2 += voutput_zero_point; |
| vy3 += voutput_zero_point; |
| |
| y[0] = (uint8_t) vy0; |
| y[1] = (uint8_t) vy1; |
| y[2] = (uint8_t) vy2; |
| y[3] = (uint8_t) vy3; |
| y += 4; |
| } |
| if XNN_UNLIKELY(n != 0) { |
| do { |
| float vx = *x++; |
| vx *= vscale; |
| vx = math_max_f32(vx, voutput_min_less_zero_point); |
| vx = math_min_f32(vx, voutput_max_less_zero_point); |
| |
| int32_t vy = (int32_t) lrintf(vx); |
| vy += voutput_zero_point; |
| |
| *y++ = (uint8_t) vy; |
| |
| n -= sizeof(float); |
| } while (n != 0); |
| } |
| } |