41   const v128_t vwCDEF = wasm_v128_load(weights + 12);  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()  local
117 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
118 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
119 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
145 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
146 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
147 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
208 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
209 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
210 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
243 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
244 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
245 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
297 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
298 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
299 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
325 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
326 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
327 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
388 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
389 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
390 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
423 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
424 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
425 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
472 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
473 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
474 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x4567, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 1, 1, 1,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
500 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
501 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
502 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x3456, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 0, 0, 0,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
556 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
557 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
558 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x5678, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 2, 2, 2,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
584 …vo0p0 = wasm_f32x4_add(vo0p0, wasm_f32x4_mul(vi2x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
585 …vo1p0 = wasm_f32x4_add(vo1p0, wasm_f32x4_mul(vi3x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()
586 …vo2p0 = wasm_f32x4_add(vo2p0, wasm_f32x4_mul(vi4x6789, wasm_v32x4_shuffle(vwCDEF, vwCDEF, 3, 3, 3,…  in xnn_f32_dwconv2d_chw_ukernel_5x5p2__wasmsimd_x86_splat_3x4()