Add Sk4f::ToBytes(uint8_t[16], Sk4f, Sk4f, Sk4f, Sk4f)

src/opts/SkNx_neon.h

 /*
  * Copyright 2015 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #ifndef SkNx_neon_DEFINED
 #define SkNx_neon_DEFINED
 
@@ skipping 147 matching lines @@
     SkNx(float a, float b, float c, float d) { fVec = (float32x4_t) { a, b, c, d }; }
 
     void store(float vals[4]) const { vst1q_f32(vals, fVec); }
     void toBytes(uint8_t bytes[4]) const {
         uint32x4_t  fix8_32 = vcvtq_u32_f32(fVec);
         uint16x4_t  fix8_16 = vqmovn_u32(fix8_32);
         uint8x8_t   fix8    = vqmovn_u16(vcombine_u16(fix8_16, vdup_n_u16(0)));
         vst1_lane_u32((uint32_t*)bytes, (uint32x2_t)fix8, 0);
     }
 
+    static void ToBytes(uint8_t bytes[16],
+                        const SkNx& a, const SkNx& b, const SkNx& c, const SkNx& d) {
+        vst1q_u8(bytes, vuzpq_u8(vuzpq_u8((uint8x16_t)vcvtq_u32_f32(a.fVec),
+                                          (uint8x16_t)vcvtq_u32_f32(b.fVec)).val[0],
+                                 vuzpq_u8((uint8x16_t)vcvtq_u32_f32(c.fVec),
+                                          (uint8x16_t)vcvtq_u32_f32(d.fVec)).val[0]).val[0]);
+    }
+
     SkNx approxInvert() const {
         float32x4_t est0 = vrecpeq_f32(fVec),
                     est1 = vmulq_f32(vrecpsq_f32(est0, fVec), est0);
         return est1;
     }
     SkNx invert() const {
         float32x4_t est1 = this->approxInvert().fVec,
                     est2 = vmulq_f32(vrecpsq_f32(est1, fVec), est1);
         return est2;
     }
@@ skipping 142 matching lines @@
 #undef SHIFT8
 
 template<>
 inline SkNx<4, int> SkNx_cast<int, float, 4>(const SkNx<4, float>& src) {
     return vcvtq_s32_f32(src.fVec);
 }
 
 }  // namespace
 
 #endif//SkNx_neon_DEFINED