Add MSA optimized ARGB/ABGR/BGRA/RGBA To 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 416 matching lines @@
 #define HAS_ARGB1555TOUVROW_MSA
 #define HAS_RGB565TOUVROW_MSA
 #define HAS_RGB24TOUVROW_MSA
 #define HAS_RAWTOUVROW_MSA
 #define HAS_NV12TOARGBROW_MSA
 #define HAS_NV12TORGB565ROW_MSA
 #define HAS_NV21TOARGBROW_MSA
 #define HAS_SOBELROW_MSA
 #define HAS_SOBELTOPLANEROW_MSA
 #define HAS_SOBELXYROW_MSA
+#define HAS_ARGBTOYJROW_MSA
+#define HAS_BGRATOYROW_MSA
+#define HAS_ABGRTOYROW_MSA
+#define HAS_RGBATOYROW_MSA
+#define HAS_ARGBTOUVJROW_MSA
+#define HAS_BGRATOUVROW_MSA
+#define HAS_ABGRTOUVROW_MSA
+#define HAS_RGBATOUVROW_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 192 matching lines @@
 #define IACA_SSC_MARK(MARK_ID)                        \
   __asm__ __volatile__("\n\t  movl $" #MARK_ID        \
                        ", %%ebx"                      \
                        "\n\t  .byte 0x64, 0x67, 0x90" \
                        :                              \
                        :                              \
                        : "memory");
 
 #define IACA_UD_BYTES __asm__ __volatile__("\n\t .byte 0x0F, 0x0B");
 
-#else  /* Visual C */
+#else /* Visual C */
 #define IACA_UD_BYTES \
   { __asm _emit 0x0F __asm _emit 0x0B }
 
 #define IACA_SSC_MARK(x) \
   { __asm mov ebx, x __asm _emit 0x64 __asm _emit 0x67 __asm _emit 0x90 }
 
 #define IACA_VC64_START __writegsbyte(111, 111);
 #define IACA_VC64_END __writegsbyte(222, 222);
 #endif
 
@@ skipping 176 matching lines @@
 void ARGBToYJRow_Any_AVX2(const uint8* src_argb, uint8* dst_y, int width);
 void ARGBToYJRow_SSSE3(const uint8* src_argb, uint8* dst_y, int width);
 void BGRAToYRow_SSSE3(const uint8* src_bgra, uint8* dst_y, int width);
 void ABGRToYRow_SSSE3(const uint8* src_abgr, uint8* dst_y, int width);
 void RGBAToYRow_SSSE3(const uint8* src_rgba, uint8* dst_y, int width);
 void RGB24ToYRow_SSSE3(const uint8* src_rgb24, uint8* dst_y, int width);
 void RAWToYRow_SSSE3(const uint8* src_raw, uint8* dst_y, int width);
 void ARGBToYRow_NEON(const uint8* src_argb, uint8* dst_y, int width);
 void ARGBToYJRow_NEON(const uint8* src_argb, uint8* dst_y, int width);
 void ARGBToYRow_MSA(const uint8* src_argb, uint8* dst_y, int width);
+void ARGBToYJRow_MSA(const uint8* src_argb, uint8* dst_y, int width);
 void ARGBToUV444Row_NEON(const uint8* src_argb,
                          uint8* dst_u,
                          uint8* dst_v,
                          int width);
 void ARGBToUVRow_NEON(const uint8* src_argb,
                       int src_stride_argb,
                       uint8* dst_u,
                       uint8* dst_v,
                       int width);
 void ARGBToUV444Row_MSA(const uint8* src_argb,
@@ skipping 43 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 ARGBToUVJRow_MSA(const uint8* src_argb,
+                      int src_stride_argb,
+                      uint8* dst_u,
+                      uint8* dst_v,
+                      int width);
+void BGRAToUVRow_MSA(const uint8* src_bgra,
+                     int src_stride_bgra,
+                     uint8* dst_u,
+                     uint8* dst_v,
+                     int width);
+void ABGRToUVRow_MSA(const uint8* src_abgr,
+                     int src_stride_abgr,
+                     uint8* dst_u,
+                     uint8* dst_v,
+                     int width);
+void RGBAToUVRow_MSA(const uint8* src_rgba,
+                     int src_stride_rgba,
+                     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 BGRAToYRow_MSA(const uint8* src_bgra, uint8* dst_y, int width);
+void ABGRToYRow_MSA(const uint8* src_abgr, uint8* dst_y, int width);
+void RGBAToYRow_MSA(const uint8* src_rgba, 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);
@@ skipping 43 matching lines @@
 void ARGB1555ToYRow_Any_NEON(const uint8* src_argb1555,
                              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 BGRAToYRow_Any_MSA(const uint8* src_bgra, uint8* dst_y, int width);
+void ABGRToYRow_Any_MSA(const uint8* src_abgr, uint8* dst_y, int width);
+void RGBAToYRow_Any_MSA(const uint8* src_rgba, uint8* dst_y, int width);
+void ARGBToYJRow_Any_MSA(const uint8* src_argb, 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,
@@ skipping 119 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 ARGBToUVJRow_Any_MSA(const uint8* src_argb,
+                          int src_stride_argb,
+                          uint8* dst_u,
+                          uint8* dst_v,
+                          int width);
+void BGRAToUVRow_Any_MSA(const uint8* src_bgra,
+                         int src_stride_bgra,
+                         uint8* dst_u,
+                         uint8* dst_v,
+                         int width);
+void ABGRToUVRow_Any_MSA(const uint8* src_abgr,
+                         int src_stride_abgr,
+                         uint8* dst_u,
+                         uint8* dst_v,
+                         int width);
+void RGBAToUVRow_Any_MSA(const uint8* src_rgba,
+                         int src_stride_rgba,
+                         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);
@@ skipping 1884 matching lines @@
                                  int width,
                                  const uint8* luma,
                                  uint32 lumacoeff);
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif
 
 #endif  // INCLUDE_LIBYUV_ROW_H_