Add MSA optimized RAW/RGB/ARGB to ARGB/Y/UV row functions

include/libyuv/row.h

 /*
  *  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 398 matching lines @@
 #define HAS_ARGBTOUV444ROW_MSA
 #define HAS_ARGBMULTIPLYROW_MSA
 #define HAS_ARGBADDROW_MSA
 #define HAS_ARGBSUBTRACTROW_MSA
 #define HAS_ARGBATTENUATEROW_MSA
 #define HAS_ARGBTORGB565DITHERROW_MSA
 #define HAS_ARGBSHUFFLEROW_MSA
 #define HAS_ARGBSHADEROW_MSA
 #define HAS_ARGBGRAYROW_MSA
 #define HAS_ARGBSEPIAROW_MSA
+#define HAS_ARGB1555TOARGBROW_MSA
+#define HAS_RGB565TOARGBROW_MSA
+#define HAS_RGB24TOARGBROW_MSA
+#define HAS_RAWTOARGBROW_MSA
+#define HAS_ARGB1555TOYROW_MSA
+#define HAS_RGB565TOYROW_MSA
+#define HAS_RGB24TOYROW_MSA
+#define HAS_RAWTOYROW_MSA
+#define HAS_ARGB1555TOUVROW_MSA
+#define HAS_RGB565TOUVROW_MSA
+#define HAS_RGB24TOUVROW_MSA
+#define HAS_RAWTOUVROW_MSA
 #endif
 
 #if defined(_MSC_VER) && !defined(__CLR_VER) && !defined(__clang__)
 #if defined(VISUALC_HAS_AVX2)
 #define SIMD_ALIGNED(var) __declspec(align(32)) var
 #else
 #define SIMD_ALIGNED(var) __declspec(align(16)) var
 #endif
 typedef __declspec(align(16)) int16 vec16[8];
 typedef __declspec(align(16)) int32 vec32[4];
@@ skipping 386 matching lines @@
 void ARGB1555ToUVRow_NEON(const uint8* src_argb1555,
                           int src_stride_argb1555,
                           uint8* dst_u,
                           uint8* dst_v,
                           int width);
 void ARGB4444ToUVRow_NEON(const uint8* src_argb4444,
                           int src_stride_argb4444,
                           uint8* dst_u,
                           uint8* dst_v,
                           int width);
+void RGB24ToUVRow_MSA(const uint8* src_rgb24,
+                      int src_stride_rgb24,
+                      uint8* dst_u,
+                      uint8* dst_v,
+                      int width);
+void RAWToUVRow_MSA(const uint8* src_raw,
+                    int src_stride_raw,
+                    uint8* dst_u,
+                    uint8* dst_v,
+                    int width);
+void RGB565ToUVRow_MSA(const uint8* src_rgb565,
+                       int src_stride_rgb565,
+                       uint8* dst_u,
+                       uint8* dst_v,
+                       int width);
+void ARGB1555ToUVRow_MSA(const uint8* src_argb1555,
+                         int src_stride_argb1555,
+                         uint8* dst_u,
+                         uint8* dst_v,
+                         int width);
 void BGRAToYRow_NEON(const uint8* src_bgra, uint8* dst_y, int width);
 void ABGRToYRow_NEON(const uint8* src_abgr, uint8* dst_y, int width);
 void RGBAToYRow_NEON(const uint8* src_rgba, uint8* dst_y, int width);
 void RGB24ToYRow_NEON(const uint8* src_rgb24, uint8* dst_y, int width);
 void RAWToYRow_NEON(const uint8* src_raw, uint8* dst_y, int width);
 void RGB565ToYRow_NEON(const uint8* src_rgb565, uint8* dst_y, int width);
 void ARGB1555ToYRow_NEON(const uint8* src_argb1555, uint8* dst_y, int width);
 void ARGB4444ToYRow_NEON(const uint8* src_argb4444, uint8* dst_y, int width);
+void RGB24ToYRow_MSA(const uint8* src_rgb24, uint8* dst_y, int width);
+void RAWToYRow_MSA(const uint8* src_raw, uint8* dst_y, int width);
+void RGB565ToYRow_MSA(const uint8* src_rgb565, uint8* dst_y, int width);
+void ARGB1555ToYRow_MSA(const uint8* src_argb1555, uint8* dst_y, int width);
 void BGRAToUVRow_DSPR2(const uint8* src_bgra,
                        int src_stride_bgra,
                        uint8* dst_u,
                        uint8* dst_v,
                        int width);
 void BGRAToYRow_DSPR2(const uint8* src_bgra, uint8* dst_y, int width);
 void ABGRToUVRow_DSPR2(const uint8* src_abgr,
                        int src_stride_abgr,
                        uint8* dst_u,
                        uint8* dst_v,
@@ skipping 40 matching lines @@
                              uint8* dst_y,
                              int width);
 void BGRAToYRow_Any_DSPR2(const uint8* src_bgra, uint8* dst_y, int width);
 void ARGBToYRow_Any_DSPR2(const uint8* src_argb, uint8* dst_y, int width);
 void ABGRToYRow_Any_DSPR2(const uint8* src_abgr, uint8* dst_y, int width);
 void RGBAToYRow_Any_DSPR2(const uint8* src_rgba, uint8* dst_y, int width);
 void ARGB4444ToYRow_Any_NEON(const uint8* src_argb4444,
                              uint8* dst_y,
                              int width);
 void ARGBToYRow_Any_MSA(const uint8* src_argb, uint8* dst_y, int width);
+void RGB24ToYRow_Any_MSA(const uint8* src_rgb24, uint8* dst_y, int width);
+void RAWToYRow_Any_MSA(const uint8* src_raw, uint8* dst_y, int width);
+void RGB565ToYRow_Any_MSA(const uint8* src_rgb565, uint8* dst_y, int width);
+void ARGB1555ToYRow_Any_MSA(const uint8* src_argb1555, uint8* dst_y, int width);
 
 void ARGBToUVRow_AVX2(const uint8* src_argb,
                       int src_stride_argb,
                       uint8* dst_u,
                       uint8* dst_v,
                       int width);
 void ARGBToUVJRow_AVX2(const uint8* src_argb,
                        int src_stride_argb,
                        uint8* dst_u,
                        uint8* dst_v,
@@ skipping 114 matching lines @@
 void ARGB1555ToUVRow_Any_NEON(const uint8* src_argb1555,
                               int src_stride_argb1555,
                               uint8* dst_u,
                               uint8* dst_v,
                               int width);
 void ARGB4444ToUVRow_Any_NEON(const uint8* src_argb4444,
                               int src_stride_argb4444,
                               uint8* dst_u,
                               uint8* dst_v,
                               int width);
+void RGB24ToUVRow_Any_MSA(const uint8* src_rgb24,
+                          int src_stride_rgb24,
+                          uint8* dst_u,
+                          uint8* dst_v,
+                          int width);
+void RAWToUVRow_Any_MSA(const uint8* src_raw,
+                        int src_stride_raw,
+                        uint8* dst_u,
+                        uint8* dst_v,
+                        int width);
+void RGB565ToUVRow_Any_MSA(const uint8* src_rgb565,
+                           int src_stride_rgb565,
+                           uint8* dst_u,
+                           uint8* dst_v,
+                           int width);
+void ARGB1555ToUVRow_Any_MSA(const uint8* src_argb1555,
+                             int src_stride_argb1555,
+                             uint8* dst_u,
+                             uint8* dst_v,
+                             int width);
 void BGRAToUVRow_Any_DSPR2(const uint8* src_bgra,
                            int src_stride_bgra,
                            uint8* dst_u,
                            uint8* dst_v,
                            int width);
 void ABGRToUVRow_Any_DSPR2(const uint8* src_abgr,
                            int src_stride_abgr,
                            uint8* dst_u,
                            uint8* dst_v,
                            int width);
@@ skipping 297 matching lines @@
                             int width);
 void RGB565ToARGBRow_AVX2(const uint8* src_rgb565, uint8* dst_argb, int width);
 void ARGB1555ToARGBRow_AVX2(const uint8* src_argb1555,
                             uint8* dst_argb,
                             int width);
 void ARGB4444ToARGBRow_AVX2(const uint8* src_argb4444,
                             uint8* dst_argb,
                             int width);
 
 void RGB24ToARGBRow_NEON(const uint8* src_rgb24, uint8* dst_argb, int width);
+void RGB24ToARGBRow_MSA(const uint8* src_rgb24, uint8* dst_argb, int width);
 void RAWToARGBRow_NEON(const uint8* src_raw, uint8* dst_argb, int width);
+void RAWToARGBRow_MSA(const uint8* src_raw, uint8* dst_argb, int width);
 void RAWToRGB24Row_NEON(const uint8* src_raw, uint8* dst_rgb24, int width);
 void RGB565ToARGBRow_NEON(const uint8* src_rgb565, uint8* dst_argb, int width);
+void RGB565ToARGBRow_MSA(const uint8* src_rgb565, uint8* dst_argb, int width);
 void ARGB1555ToARGBRow_NEON(const uint8* src_argb1555,
                             uint8* dst_argb,
                             int width);
+void ARGB1555ToARGBRow_MSA(const uint8* src_argb1555,
+                           uint8* dst_argb,
+                           int width);
 void ARGB4444ToARGBRow_NEON(const uint8* src_argb4444,
                             uint8* dst_argb,
                             int width);
 void RGB24ToARGBRow_DSPR2(const uint8* src_rgb24, uint8* dst_argb, int width);
 void RAWToARGBRow_DSPR2(const uint8* src_raw, uint8* dst_argb, int width);
 void RGB565ToARGBRow_DSPR2(const uint8* src_rgb565, uint8* dst_argb, int width);
 void ARGB1555ToARGBRow_DSPR2(const uint8* src_argb1555,
                              uint8* dst_argb,
                              int width);
 void ARGB4444ToARGBRow_DSPR2(const uint8* src_argb4444,
@@ skipping 29 matching lines @@
 void ARGB1555ToARGBRow_Any_AVX2(const uint8* src_argb1555,
                                 uint8* dst_argb,
                                 int width);
 void ARGB4444ToARGBRow_Any_AVX2(const uint8* src_argb4444,
                                 uint8* dst_argb,
                                 int width);
 
 void RGB24ToARGBRow_Any_NEON(const uint8* src_rgb24,
                              uint8* dst_argb,
                              int width);
+void RGB24ToARGBRow_Any_MSA(const uint8* src_rgb24, uint8* dst_argb, int width);
 void RAWToARGBRow_Any_NEON(const uint8* src_raw, uint8* dst_argb, int width);
+void RAWToARGBRow_Any_MSA(const uint8* src_raw, uint8* dst_argb, int width);
 void RAWToRGB24Row_Any_NEON(const uint8* src_raw, uint8* dst_rgb24, int width);
 void RGB565ToARGBRow_Any_NEON(const uint8* src_rgb565,
                               uint8* dst_argb,
                               int width);
+void RGB565ToARGBRow_Any_MSA(const uint8* src_rgb565,
+                             uint8* dst_argb,
+                             int width);
 void ARGB1555ToARGBRow_Any_NEON(const uint8* src_argb1555,
                                 uint8* dst_argb,
                                 int width);
+void ARGB1555ToARGBRow_Any_MSA(const uint8* src_argb1555,
+                               uint8* dst_argb,
+                               int width);
 void ARGB4444ToARGBRow_Any_NEON(const uint8* src_argb4444,
                                 uint8* dst_argb,
                                 int width);
 void RGB24ToARGBRow_Any_DSPR2(const uint8* src_rgb24,
                               uint8* dst_argb,
                               int width);
 void RAWToARGBRow_Any_DSPR2(const uint8* src_raw, uint8* dst_argb, int width);
 void RGB565ToARGBRow_Any_DSPR2(const uint8* src_rgb565,
                                uint8* dst_argb,
                                int width);
@@ skipping 1431 matching lines @@
                                  int width,
                                  const uint8* luma,
                                  uint32 lumacoeff);
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif
 
 #endif  // INCLUDE_LIBYUV_ROW_H_