Searched refs:vb01c0 (Results 1 – 4 of 4) sorted by relevance
/external/XNNPACK/src/f32-gemm/gen/ |
D | 4x2-minmax-neonfma-lane-ld64.c | 73 const float32x2_t vb01c0 = vld1_f32(w); w += 2; in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() local 76 vacc0x01 = vfma_lane_f32(vacc0x01, vb01c0, va0, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 77 vacc1x01 = vfma_lane_f32(vacc1x01, vb01c0, va1, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 78 vacc2x01 = vfma_lane_f32(vacc2x01, vb01c0, va2, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 79 vacc3x01 = vfma_lane_f32(vacc3x01, vb01c0, va3, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 85 vacc0x01 = vfma_f32(vacc0x01, va0c0, vb01c0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 86 vacc1x01 = vfma_f32(vacc1x01, va1c0, vb01c0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 87 vacc2x01 = vfma_f32(vacc2x01, va2c0, vb01c0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64() 88 vacc3x01 = vfma_f32(vacc3x01, va3c0, vb01c0); in xnn_f32_gemm_minmax_ukernel_4x2__neonfma_lane_ld64()
|
D | 4x2-minmax-neon-lane-ld64.c | 73 const float32x2_t vb01c0 = vld1_f32(w); w += 2; in xnn_f32_gemm_minmax_ukernel_4x2__neon_lane_ld64() local 75 vacc0x01 = vmla_lane_f32(vacc0x01, vb01c0, va0, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neon_lane_ld64() 76 vacc1x01 = vmla_lane_f32(vacc1x01, vb01c0, va1, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neon_lane_ld64() 77 vacc2x01 = vmla_lane_f32(vacc2x01, vb01c0, va2, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neon_lane_ld64() 78 vacc3x01 = vmla_lane_f32(vacc3x01, vb01c0, va3, 0); in xnn_f32_gemm_minmax_ukernel_4x2__neon_lane_ld64()
|
/external/XNNPACK/src/f32-igemm/gen/ |
D | 4x2-minmax-neonfma-lane-ld64.c | 95 const float32x2_t vb01c0 = vld1_f32(w); w += 2; in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() local 98 vacc0x01 = vfma_lane_f32(vacc0x01, vb01c0, va0, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 99 vacc1x01 = vfma_lane_f32(vacc1x01, vb01c0, va1, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 100 vacc2x01 = vfma_lane_f32(vacc2x01, vb01c0, va2, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 101 vacc3x01 = vfma_lane_f32(vacc3x01, vb01c0, va3, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 107 vacc0x01 = vfma_f32(vacc0x01, va0c0, vb01c0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 108 vacc1x01 = vfma_f32(vacc1x01, va1c0, vb01c0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 109 vacc2x01 = vfma_f32(vacc2x01, va2c0, vb01c0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64() 110 vacc3x01 = vfma_f32(vacc3x01, va3c0, vb01c0); in xnn_f32_igemm_minmax_ukernel_4x2__neonfma_lane_ld64()
|
D | 4x2-minmax-neon-lane-ld64.c | 95 const float32x2_t vb01c0 = vld1_f32(w); w += 2; in xnn_f32_igemm_minmax_ukernel_4x2__neon_lane_ld64() local 97 vacc0x01 = vmla_lane_f32(vacc0x01, vb01c0, va0, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neon_lane_ld64() 98 vacc1x01 = vmla_lane_f32(vacc1x01, vb01c0, va1, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neon_lane_ld64() 99 vacc2x01 = vmla_lane_f32(vacc2x01, vb01c0, va2, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neon_lane_ld64() 100 vacc3x01 = vmla_lane_f32(vacc3x01, vb01c0, va3, 0); in xnn_f32_igemm_minmax_ukernel_4x2__neon_lane_ld64()
|