Add MSA optimized rotate functions (used 16x16 transpose)

include/libyuv/rotate_row.h

Index: include/libyuv/rotate_row.h
diff --git a/include/libyuv/rotate_row.h b/include/libyuv/rotate_row.h
index fda6b59f39340e8aa9b388329359310f5d39bb85..a3e69e01674cadd9c8da0bebc388e4c193798386 100644
--- a/include/libyuv/rotate_row.h
+++ b/include/libyuv/rotate_row.h
@@ -61,8 +61,8 @@ extern "C" {
 #endif  // defined(__mips__)
 
 #if !defined(LIBYUV_DISABLE_MSA) && defined(__mips_msa)
-#define HAS_TRANSPOSEWX8_MSA
-#define HAS_TRANSPOSEUVWX8_MSA
+#define HAS_TRANSPOSEWX16_MSA
+#define HAS_TRANSPOSEUVWX16_MSA
 #endif
 
 void TransposeWxH_C(const uint8* src,
@@ -77,6 +77,11 @@ void TransposeWx8_C(const uint8* src,
                     uint8* dst,
                     int dst_stride,
                     int width);
+void TransposeWx16_C(const uint8* src,
+                     int src_stride,
+                     uint8* dst,
+                     int dst_stride,
+                     int width);
 void TransposeWx8_NEON(const uint8* src,
                        int src_stride,
                        uint8* dst,
@@ -102,11 +107,11 @@ void TransposeWx8_Fast_DSPR2(const uint8* src,
                              uint8* dst,
                              int dst_stride,
                              int width);
-void TransposeWx8_MSA(const uint8* src,
-                      int src_stride,
-                      uint8* dst,
-                      int dst_stride,
-                      int width);
+void TransposeWx16_MSA(const uint8* src,
+                       int src_stride,
+                       uint8* dst,
+                       int dst_stride,
+                       int width);
 
 void TransposeWx8_Any_NEON(const uint8* src,
                            int src_stride,
@@ -128,11 +133,11 @@ void TransposeWx8_Any_DSPR2(const uint8* src,
                             uint8* dst,
                             int dst_stride,
                             int width);
-void TransposeWx8_Any_MSA(const uint8* src,
-                          int src_stride,
-                          uint8* dst,
-                          int dst_stride,
-                          int width);
+void TransposeWx16_Any_MSA(const uint8* src,
+                           int src_stride,
+                           uint8* dst,
+                           int dst_stride,
+                           int width);
 
 void TransposeUVWxH_C(const uint8* src,
                       int src_stride,
@@ -150,6 +155,13 @@ void TransposeUVWx8_C(const uint8* src,
                       uint8* dst_b,
                       int dst_stride_b,
                       int width);
+void TransposeUVWx16_C(const uint8* src,
+                       int src_stride,
+                       uint8* dst_a,
+                       int dst_stride_a,
+                       uint8* dst_b,
+                       int dst_stride_b,
+                       int width);
 void TransposeUVWx8_SSE2(const uint8* src,
                          int src_stride,
                          uint8* dst_a,
@@ -171,13 +183,13 @@ void TransposeUVWx8_DSPR2(const uint8* src,
                           uint8* dst_b,
                           int dst_stride_b,
                           int width);
-void TransposeUVWx8_MSA(const uint8* src,
-                        int src_stride,
-                        uint8* dst_a,
-                        int dst_stride_a,
-                        uint8* dst_b,
-                        int dst_stride_b,
-                        int width);
+void TransposeUVWx16_MSA(const uint8* src,
+                         int src_stride,
+                         uint8* dst_a,
+                         int dst_stride_a,
+                         uint8* dst_b,
+                         int dst_stride_b,
+                         int width);
 
 void TransposeUVWx8_Any_SSE2(const uint8* src,
                              int src_stride,
@@ -200,13 +212,13 @@ void TransposeUVWx8_Any_DSPR2(const uint8* src,
                               uint8* dst_b,
                               int dst_stride_b,
                               int width);
-void TransposeUVWx8_Any_MSA(const uint8* src,
-                            int src_stride,
-                            uint8* dst_a,
-                            int dst_stride_a,
-                            uint8* dst_b,
-                            int dst_stride_b,
-                            int width);
+void TransposeUVWx16_Any_MSA(const uint8* src,
+                             int src_stride,
+                             uint8* dst_a,
+                             int dst_stride_a,
+                             uint8* dst_b,
+                             int dst_stride_b,
+                             int width);
 
 #ifdef __cplusplus
 }  // extern "C"