/external/XNNPACK/src/f32-vbinary/gen/ |
D | vminc-avx-x16.c | 37 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vminc_ukernel__avx_x16() local 41 __m256 vy89ABCDEF = _mm256_min_ps(va89ABCDEF, vb); in xnn_f32_vminc_ukernel__avx_x16()
|
D | vmaxc-avx-x16.c | 37 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vmaxc_ukernel__avx_x16() local 41 __m256 vy89ABCDEF = _mm256_max_ps(va89ABCDEF, vb); in xnn_f32_vmaxc_ukernel__avx_x16()
|
D | vmax-avx-x16.c | 36 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vmax_ukernel__avx_x16() local 44 __m256 vy89ABCDEF = _mm256_max_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vmax_ukernel__avx_x16()
|
D | vsqrdiffc-avx-x16.c | 37 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vsqrdiffc_ukernel__avx_x16() local 41 __m256 vy89ABCDEF = _mm256_sub_ps(va89ABCDEF, vb); in xnn_f32_vsqrdiffc_ukernel__avx_x16()
|
D | vmin-avx-x16.c | 36 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vmin_ukernel__avx_x16() local 44 __m256 vy89ABCDEF = _mm256_min_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vmin_ukernel__avx_x16()
|
D | vrdivc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vrdivc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_div_ps(vb, va89ABCDEF); in xnn_f32_vrdivc_minmax_ukernel__avx_x16()
|
D | vdivc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vdivc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_div_ps(va89ABCDEF, vb); in xnn_f32_vdivc_minmax_ukernel__avx_x16()
|
D | vsubc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vsubc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_sub_ps(va89ABCDEF, vb); in xnn_f32_vsubc_minmax_ukernel__avx_x16()
|
D | vrsubc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vrsubc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_sub_ps(vb, va89ABCDEF); in xnn_f32_vrsubc_minmax_ukernel__avx_x16()
|
D | vsqrdiff-avx-x16.c | 36 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vsqrdiff_ukernel__avx_x16() local 44 __m256 vy89ABCDEF = _mm256_sub_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vsqrdiff_ukernel__avx_x16()
|
D | vmulc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vmulc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_mul_ps(va89ABCDEF, vb); in xnn_f32_vmulc_minmax_ukernel__avx_x16()
|
D | vaddc-minmax-avx-x16.c | 39 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vaddc_minmax_ukernel__avx_x16() local 43 __m256 vy89ABCDEF = _mm256_add_ps(va89ABCDEF, vb); in xnn_f32_vaddc_minmax_ukernel__avx_x16()
|
D | vdiv-minmax-avx-x16.c | 38 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vdiv_minmax_ukernel__avx_x16() local 46 __m256 vy89ABCDEF = _mm256_div_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vdiv_minmax_ukernel__avx_x16()
|
D | vmul-minmax-avx-x16.c | 38 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vmul_minmax_ukernel__avx_x16() local 46 __m256 vy89ABCDEF = _mm256_mul_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vmul_minmax_ukernel__avx_x16()
|
D | vadd-minmax-avx-x16.c | 38 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vadd_minmax_ukernel__avx_x16() local 46 __m256 vy89ABCDEF = _mm256_add_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vadd_minmax_ukernel__avx_x16()
|
D | vsub-minmax-avx-x16.c | 38 const __m256 va89ABCDEF = _mm256_loadu_ps(a + 8); in xnn_f32_vsub_minmax_ukernel__avx_x16() local 46 __m256 vy89ABCDEF = _mm256_sub_ps(va89ABCDEF, vb89ABCDEF); in xnn_f32_vsub_minmax_ukernel__avx_x16()
|
/external/XNNPACK/src/f16-spmm/gen/ |
D | 32x1-minmax-neonfp16arith.c | 55 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith() local 61 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith() 103 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith() local 107 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith()
|
D | 24x1-minmax-neonfp16arith.c | 54 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith() local 59 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith() 96 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith() local 100 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith()
|
D | 24x1-minmax-neonfp16arith-x2.c | 84 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith_x2() local 89 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith_x2() 126 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith_x2() local 130 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_24x1__neonfp16arith_x2()
|
D | 32x1-minmax-neonfp16arith-x2.c | 92 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith_x2() local 98 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith_x2() 140 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith_x2() local 144 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_32x1__neonfp16arith_x2()
|
D | 16x1-minmax-neonfp16arith.c | 53 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_16x1__neonfp16arith() local 57 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_16x1__neonfp16arith()
|
D | 16x1-minmax-neonfp16arith-x2.c | 76 const float16x8_t va89ABCDEF = vld1q_f16(i + 8); in xnn_f16_spmm_minmax_ukernel_16x1__neonfp16arith_x2() local 80 vacc89ABCDEF = vfmaq_f16(vacc89ABCDEF, va89ABCDEF, vb); in xnn_f16_spmm_minmax_ukernel_16x1__neonfp16arith_x2()
|