| /external/XNNPACK/src/f32-dwconv2d-chw/gen/ |
| D | 5x5p2-minmax-wasmsimd-x86-splat-5x4.c | 167 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_5x4() local
421 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_5x4() local
668 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_5x4() local
|
| D | 5x5p2-minmax-wasmsimd-arm-splat-5x4.c | 167 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_splat_5x4() local
421 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_splat_5x4() local
668 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_splat_5x4() local
|
| D | 5x5p2-minmax-neonfma-5x4.c | 164 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neonfma_5x4() local
419 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neonfma_5x4() local
667 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neonfma_5x4() local
|
| D | 5x5p2-minmax-neon-5x4.c | 164 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neon_5x4() local
419 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neon_5x4() local
667 const float32x4_t vi8x3456 = vextq_f32(vi8x0123, vi8x4567, 3); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__neon_5x4() local
|
| D | 5x5p2-minmax-wasmsimd-x86-loadsplat-5x4.c | 193 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_loadsplat_5x4() local
447 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_loadsplat_5x4() local
694 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_loadsplat_5x4() local
|
| D | 5x5p2-minmax-wasmsimd-arm-loadsplat-5x4.c | 193 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_loadsplat_5x4() local
447 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_loadsplat_5x4() local
694 const v128_t vi8x3456 = wasm_v32x4_shuffle(vi8x0123, vi8x4567, 3, 4, 5, 6); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_arm_loadsplat_5x4() local
|
| D | 5x5p2-minmax-sse-5x4.c | 201 const __m128 vi8x3456 = _mm_move_ss(vi8x7456, vi8x3012); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__sse_5x4() local
454 const __m128 vi8x3456 = _mm_move_ss(vi8x7456, vi8x3012); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__sse_5x4() local
701 const __m128 vi8x3456 = _mm_move_ss(vi8x7456, vi8x3012); in xnn_f32_dwconv2d_chw_ukernel_5x5p2__sse_5x4() local
|