145         const v128_t vi4x2 = wasm_v128_load(i4); i4 += 4;  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()  local
164 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk20c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 0, 0, 0, 0))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
185 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk20c2, wasm_v32x4_shuffle(vi4x2, vi4x2, 1, 1, 1, 1))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
206 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk21c0, wasm_v32x4_shuffle(vi4x2, vi4x2, 2, 2, 2, 2))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
227 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk21c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 3, 3, 3, 3))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
295 …vo1x0 = wasm_f32x4_add(vo1x0, wasm_f32x4_mul(vk22c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 0, 0, 0, 0))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
316 …vo1x0 = wasm_f32x4_add(vo1x0, wasm_f32x4_mul(vk22c2, wasm_v32x4_shuffle(vi4x2, vi4x2, 1, 1, 1, 1))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
398         v128_t vi4x2 = vzero;  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()  local
405           vi4x2 = wasm_v128_load(i4 + 4);  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
425 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk20c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 0, 0, 0, 0))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
446 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk20c2, wasm_v32x4_shuffle(vi4x2, vi4x2, 1, 1, 1, 1))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
472 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk21c0, wasm_v32x4_shuffle(vi4x2, vi4x2, 2, 2, 2, 2))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
499 …vo1x1 = wasm_f32x4_add(vo1x1, wasm_f32x4_mul(vk21c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 3, 3, 3, 3))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
566 …vo1x0 = wasm_f32x4_add(vo1x0, wasm_f32x4_mul(vk22c1, wasm_v32x4_shuffle(vi4x2, vi4x2, 0, 0, 0, 0))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()
581 …vo1x0 = wasm_f32x4_add(vo1x0, wasm_f32x4_mul(vk22c2, wasm_v32x4_shuffle(vi4x2, vi4x2, 1, 1, 1, 1))…  in xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2()