Add MSA optimized TransposeWx8_MSA and TransposeUVWx8_MSA functions

source/rotate.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 44 matching lines @@
 #if defined(HAS_TRANSPOSEWX8_DSPR2)
   if (TestCpuFlag(kCpuHasDSPR2)) {
     if (IS_ALIGNED(width, 4) && IS_ALIGNED(src, 4) &&
         IS_ALIGNED(src_stride, 4)) {
       TransposeWx8 = TransposeWx8_Fast_DSPR2;
     } else {
       TransposeWx8 = TransposeWx8_DSPR2;
     }
   }
 #endif
+#if defined(HAS_TRANSPOSEWX8_MSA)
+  if (TestCpuFlag(kCpuHasMSA)) {
+    TransposeWx8 = TransposeWx8_Any_MSA;
+    if (IS_ALIGNED(width, 16)) {
+      TransposeWx8 = TransposeWx8_MSA;
+    }
+  }
+#endif
 
   // Work across the source in 8x8 tiles
   while (i >= 8) {
     TransposeWx8(src, src_stride, dst, dst_stride, width);
     src += 8 * src_stride;  // Go down 8 rows.
     dst += 8;               // Move over 8 columns.
     i -= 8;
   }
 
   if (i > 0) {
@@ skipping 150 matching lines @@
       TransposeUVWx8 = TransposeUVWx8_SSE2;
     }
   }
 #endif
 #if defined(HAS_TRANSPOSEUVWX8_DSPR2)
   if (TestCpuFlag(kCpuHasDSPR2) && IS_ALIGNED(width, 2) && IS_ALIGNED(src, 4) &&
       IS_ALIGNED(src_stride, 4)) {
     TransposeUVWx8 = TransposeUVWx8_DSPR2;
   }
 #endif
+#if defined(HAS_TRANSPOSEUVWX8_MSA)
+  if (TestCpuFlag(kCpuHasMSA)) {
+    TransposeUVWx8 = TransposeUVWx8_Any_MSA;
+    if (IS_ALIGNED(width, 8)) {
+      TransposeUVWx8 = TransposeUVWx8_MSA;
+    }
+  }
+#endif
 
   // Work through the source in 8x8 tiles.
   while (i >= 8) {
     TransposeUVWx8(src, src_stride, dst_a, dst_stride_a, dst_b, dst_stride_b,
                    width);
     src += 8 * src_stride;  // Go down 8 rows.
     dst_a += 8;             // Move over 8 columns.
     dst_b += 8;             // Move over 8 columns.
     i -= 8;
   }
@@ skipping 241 matching lines @@
     default:
       break;
   }
   return -1;
 }
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif