HalfFloat_SSE2 port from C algorithm to SSE2

source/planar_functions.cc

 /*
  *  Copyright 2011 The LibYuv Project Authors. All rights reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS. All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
@@ skipping 2452 matching lines @@
     src_y  = src_y  + (height - 1) * src_stride_y;
     src_stride_y = -src_stride_y;
   }
   // Coalesce rows.
   if (src_stride_y == width &&
       dst_stride_y == width) {
     width *= height;
     height = 1;
     src_stride_y = dst_stride_y = 0;
   }
+#if defined(HAS_HALFFLOATROW_SSE2)
+  if (TestCpuFlag(kCpuHasSSE2)) {
+    HalfFloatRow = HalfFloatRow_Any_SSE2;
+    if (IS_ALIGNED(width, 8)) {
+      HalfFloatRow = HalfFloatRow_SSE2;
+    }
+  }
+#endif
 #if defined(HAS_HALFFLOATROW_AVX2)
   if (TestCpuFlag(kCpuHasAVX2)) {
     HalfFloatRow = HalfFloatRow_Any_AVX2;
     if (IS_ALIGNED(width, 16)) {
       HalfFloatRow = HalfFloatRow_AVX2;
     }
   }
 #endif
+#if defined(HAS_HALFFLOATROW_AVX)
+  if (TestCpuFlag(kCpuHasAVX)) {
+//    HalfFloatRow = HalfFloatRow_Any_AVX2;
+    if (IS_ALIGNED(width, 16)) {
+      HalfFloatRow = HalfFloatRow_AVX;
+    }
+  }
+#endif
 
   for (y = 0; y < height; ++y) {
     HalfFloatRow(src_y, dst_y, scale, width);
     src_y += src_stride_y;
     dst_y += dst_stride_y;
   }
   return 0;
 }
 
 // Apply a lumacolortable to each ARGB pixel.
@@ skipping 373 matching lines @@
     }
     free_aligned_buffer_64(rows);
   }
   return 0;
 }
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif