swscale: add full bgra64 support

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

4 files changed, 77 insertions, 24 deletions

diff --git a/libswscale/input.c b/libswscale/input.c
index 919b232231..1539bd9307 100644
--- a/libswscale/input.c
+++ b/libswscale/input.c
@@ -37,8 +37,8 @@
 
 #define input_pixel(pos) (isBE(origin) ? AV_RB16(pos) : AV_RL16(pos))
 
-#define r ((origin == AV_PIX_FMT_BGR48BE || origin == AV_PIX_FMT_BGR48LE) ? b_r : r_b)
-#define b ((origin == AV_PIX_FMT_BGR48BE || origin == AV_PIX_FMT_BGR48LE) ? r_b : b_r)
+#define r ((origin == AV_PIX_FMT_BGR48BE || origin == AV_PIX_FMT_BGR48LE || origin == AV_PIX_FMT_BGRA64BE || origin == AV_PIX_FMT_BGRA64LE) ? b_r : r_b)
+#define b ((origin == AV_PIX_FMT_BGR48BE || origin == AV_PIX_FMT_BGR48LE || origin == AV_PIX_FMT_BGRA64BE || origin == AV_PIX_FMT_BGRA64LE) ? r_b : b_r)
 
 static av_always_inline void
 rgb64ToY_c_template(uint16_t *dst, const uint16_t *src, int width,
@@ -124,6 +124,8 @@ static void pattern ## 64 ## BE_LE ## ToUV_half_c(uint8_t *_dstU, uint8_t *_dstV
 
 rgb64funcs(rgb, LE, AV_PIX_FMT_RGBA64LE)
 rgb64funcs(rgb, BE, AV_PIX_FMT_RGBA64BE)
+rgb64funcs(bgr, LE, AV_PIX_FMT_BGRA64LE)
+rgb64funcs(bgr, BE, AV_PIX_FMT_BGRA64BE)
 
 static av_always_inline void rgb48ToY_c_template(uint16_t *dst,
                                                  const uint16_t *src, int width,
@@ -934,6 +936,12 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
         case AV_PIX_FMT_RGBA64LE:
             c->chrToYV12 = rgb64LEToUV_half_c;
             break;
+        case AV_PIX_FMT_BGRA64BE:
+            c->chrToYV12 = bgr64BEToUV_half_c;
+            break;
+        case AV_PIX_FMT_BGRA64LE:
+            c->chrToYV12 = bgr64LEToUV_half_c;
+            break;
         case AV_PIX_FMT_RGB48BE:
             c->chrToYV12 = rgb48BEToUV_half_c;
             break;
@@ -1013,6 +1021,12 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
         case AV_PIX_FMT_RGBA64LE:
             c->chrToYV12 = rgb64LEToUV_c;
             break;
+        case AV_PIX_FMT_BGRA64BE:
+            c->chrToYV12 = bgr64BEToUV_c;
+            break;
+        case AV_PIX_FMT_BGRA64LE:
+            c->chrToYV12 = bgr64LEToUV_c;
+            break;
         case AV_PIX_FMT_RGB48BE:
             c->chrToYV12 = rgb48BEToUV_c;
             break;
@@ -1279,6 +1293,11 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_RGBA64LE:
         c->lumToYV12 = rgb64LEToY_c;
         break;
+    case AV_PIX_FMT_BGRA64BE:
+        c->lumToYV12 = bgr64BEToY_c;
+        break;
+    case AV_PIX_FMT_BGRA64LE:
+        c->lumToYV12 = bgr64LEToY_c;
     }
     if (c->alpPixBuf) {
         if (is16BPS(srcFormat) || isNBPS(srcFormat)) {
@@ -1286,6 +1305,8 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
                 c->alpToYV12 = bswap16Y_c;
         }
         switch (srcFormat) {
+        case AV_PIX_FMT_BGRA64LE:
+        case AV_PIX_FMT_BGRA64BE:
         case AV_PIX_FMT_RGBA64LE:
         case AV_PIX_FMT_RGBA64BE:  c->alpToYV12 = rgba64ToA_c; break;
         case AV_PIX_FMT_BGRA:
diff --git a/libswscale/output.c b/libswscale/output.c
index f41b32ac63..4083c73116 100644
--- a/libswscale/output.c
+++ b/libswscale/output.c
@@ -663,8 +663,8 @@ yuv2422_1_c_template(SwsContext *c, const int16_t *buf0,
 YUV2PACKEDWRAPPER(yuv2, 422, yuyv422, AV_PIX_FMT_YUYV422)
 YUV2PACKEDWRAPPER(yuv2, 422, uyvy422, AV_PIX_FMT_UYVY422)
 
-#define R_B ((target == AV_PIX_FMT_RGB48LE || target == AV_PIX_FMT_RGB48BE) ? R : B)
-#define B_R ((target == AV_PIX_FMT_RGB48LE || target == AV_PIX_FMT_RGB48BE) ? B : R)
+#define R_B ((target == AV_PIX_FMT_RGB48LE || target == AV_PIX_FMT_RGB48BE || target == AV_PIX_FMT_RGBA64LE || target == AV_PIX_FMT_RGBA64BE) ? R : B)
+#define B_R ((target == AV_PIX_FMT_RGB48LE || target == AV_PIX_FMT_RGB48BE || target == AV_PIX_FMT_RGBA64LE || target == AV_PIX_FMT_RGBA64BE) ? B : R)
 #define output_pixel(pos, val) \
     if (isBE(target)) { \
         AV_WB16(pos, val); \
@@ -735,13 +735,13 @@ yuv2rgba64_X_c_template(SwsContext *c, const int16_t *lumFilter,
         B =                            U * c->yuv2rgb_u2b_coeff;
 
         // 8bit: 30 - 22 = 8bit, 16bit: 30bit - 14 = 16bit
-        output_pixel(&dest[0], av_clip_uintp2(B_R + Y1, 30) >> 14);
+        output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
         output_pixel(&dest[1], av_clip_uintp2(  G + Y1, 30) >> 14);
-        output_pixel(&dest[2], av_clip_uintp2(R_B + Y1, 30) >> 14);
+        output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
         output_pixel(&dest[3], av_clip_uintp2(A1      , 30) >> 14);
-        output_pixel(&dest[4], av_clip_uintp2(B_R + Y2, 30) >> 14);
+        output_pixel(&dest[4], av_clip_uintp2(R_B + Y2, 30) >> 14);
         output_pixel(&dest[5], av_clip_uintp2(  G + Y2, 30) >> 14);
-        output_pixel(&dest[6], av_clip_uintp2(R_B + Y2, 30) >> 14);
+        output_pixel(&dest[6], av_clip_uintp2(B_R + Y2, 30) >> 14);
         output_pixel(&dest[7], av_clip_uintp2(A2      , 30) >> 14);
         dest += 8;
     }
@@ -790,13 +790,13 @@ yuv2rgba64_2_c_template(SwsContext *c, const int32_t *buf[2],
             A2 += 1 << 13;
         }
 
-        output_pixel(&dest[0], av_clip_uintp2(B_R + Y1, 30) >> 14);
+        output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
         output_pixel(&dest[1], av_clip_uintp2(  G + Y1, 30) >> 14);
-        output_pixel(&dest[2], av_clip_uintp2(R_B + Y1, 30) >> 14);
+        output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
         output_pixel(&dest[3], av_clip_uintp2(A1      , 30) >> 14);
-        output_pixel(&dest[4], av_clip_uintp2(B_R + Y2, 30) >> 14);
+        output_pixel(&dest[4], av_clip_uintp2(R_B + Y2, 30) >> 14);
         output_pixel(&dest[5], av_clip_uintp2(  G + Y2, 30) >> 14);
-        output_pixel(&dest[6], av_clip_uintp2(R_B + Y2, 30) >> 14);
+        output_pixel(&dest[6], av_clip_uintp2(B_R + Y2, 30) >> 14);
         output_pixel(&dest[7], av_clip_uintp2(A2      , 30) >> 14);
         dest += 8;
     }
@@ -839,13 +839,13 @@ yuv2rgba64_1_c_template(SwsContext *c, const int32_t *buf0,
             G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
             B =                            U * c->yuv2rgb_u2b_coeff;
 
-            output_pixel(&dest[0], av_clip_uintp2(B_R + Y1, 30) >> 14);
+            output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
             output_pixel(&dest[1], av_clip_uintp2(  G + Y1, 30) >> 14);
-            output_pixel(&dest[2], av_clip_uintp2(R_B + Y1, 30) >> 14);
+            output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
             output_pixel(&dest[3], av_clip_uintp2(A1      , 30) >> 14);
-            output_pixel(&dest[4], av_clip_uintp2(B_R + Y2, 30) >> 14);
+            output_pixel(&dest[4], av_clip_uintp2(R_B + Y2, 30) >> 14);
             output_pixel(&dest[5], av_clip_uintp2(  G + Y2, 30) >> 14);
-            output_pixel(&dest[6], av_clip_uintp2(R_B + Y2, 30) >> 14);
+            output_pixel(&dest[6], av_clip_uintp2(B_R + Y2, 30) >> 14);
             output_pixel(&dest[7], av_clip_uintp2(A2      , 30) >> 14);
             dest += 8;
         }
@@ -878,13 +878,13 @@ yuv2rgba64_1_c_template(SwsContext *c, const int32_t *buf0,
             G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
             B =                            U * c->yuv2rgb_u2b_coeff;
 
-            output_pixel(&dest[0], av_clip_uintp2(B_R + Y1, 30) >> 14);
+            output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
             output_pixel(&dest[1], av_clip_uintp2(  G + Y1, 30) >> 14);
-            output_pixel(&dest[2], av_clip_uintp2(R_B + Y1, 30) >> 14);
+            output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
             output_pixel(&dest[3], av_clip_uintp2(A1      , 30) >> 14);
-            output_pixel(&dest[4], av_clip_uintp2(B_R + Y2, 30) >> 14);
+            output_pixel(&dest[4], av_clip_uintp2(R_B + Y2, 30) >> 14);
             output_pixel(&dest[5], av_clip_uintp2(  G + Y2, 30) >> 14);
-            output_pixel(&dest[6], av_clip_uintp2(R_B + Y2, 30) >> 14);
+            output_pixel(&dest[6], av_clip_uintp2(B_R + Y2, 30) >> 14);
             output_pixel(&dest[7], av_clip_uintp2(A2      , 30) >> 14);
             dest += 8;
         }
@@ -1117,6 +1117,10 @@ YUV2PACKED16WRAPPER(yuv2, rgba64, rgba64be, AV_PIX_FMT_RGBA64BE, 1)
 YUV2PACKED16WRAPPER(yuv2, rgba64, rgba64le, AV_PIX_FMT_RGBA64LE, 1)
 YUV2PACKED16WRAPPER(yuv2, rgba64, rgbx64be, AV_PIX_FMT_RGBA64BE, 0)
 YUV2PACKED16WRAPPER(yuv2, rgba64, rgbx64le, AV_PIX_FMT_RGBA64LE, 0)
+YUV2PACKED16WRAPPER(yuv2, rgba64, bgra64be, AV_PIX_FMT_BGRA64BE, 1)
+YUV2PACKED16WRAPPER(yuv2, rgba64, bgra64le, AV_PIX_FMT_BGRA64LE, 1)
+YUV2PACKED16WRAPPER(yuv2, rgba64, bgrx64be, AV_PIX_FMT_BGRA64BE, 0)
+YUV2PACKED16WRAPPER(yuv2, rgba64, bgrx64le, AV_PIX_FMT_BGRA64LE, 0)
 
 /*
  * Write out 2 RGB pixels in the target pixel format. This function takes a
@@ -2041,6 +2045,34 @@ av_cold void ff_sws_init_output_funcs(SwsContext *c,
                 *yuv2packedX = yuv2rgbx64be_X_c;
             }
             break;
+        case AV_PIX_FMT_BGRA64LE:
+#if CONFIG_SWSCALE_ALPHA
+            if (c->alpPixBuf) {
+                *yuv2packed1 = yuv2bgra64le_1_c;
+                *yuv2packed2 = yuv2bgra64le_2_c;
+                *yuv2packedX = yuv2bgra64le_X_c;
+            } else
+#endif /* CONFIG_SWSCALE_ALPHA */
+            {
+                *yuv2packed1 = yuv2bgrx64le_1_c;
+                *yuv2packed2 = yuv2bgrx64le_2_c;
+                *yuv2packedX = yuv2bgrx64le_X_c;
+            }
+            break;
+        case AV_PIX_FMT_BGRA64BE:
+#if CONFIG_SWSCALE_ALPHA
+            if (c->alpPixBuf) {
+                *yuv2packed1 = yuv2bgra64be_1_c;
+                *yuv2packed2 = yuv2bgra64be_2_c;
+                *yuv2packedX = yuv2bgra64be_X_c;
+            } else
+#endif /* CONFIG_SWSCALE_ALPHA */
+            {
+                *yuv2packed1 = yuv2bgrx64be_1_c;
+                *yuv2packed2 = yuv2bgrx64be_2_c;
+                *yuv2packedX = yuv2bgrx64be_X_c;
+            }
+            break;
         case AV_PIX_FMT_RGB48LE:
             *yuv2packed1 = yuv2rgb48le_1_c;
             *yuv2packed2 = yuv2rgb48le_2_c;
diff --git a/libswscale/utils.c b/libswscale/utils.c
index 3ced7ad3b9..983824b222 100644
--- a/libswscale/utils.c
+++ b/libswscale/utils.c
@@ -165,8 +165,8 @@ static const FormatEntry format_entries[AV_PIX_FMT_NB] = {
     [AV_PIX_FMT_Y400A]       = { 1, 0 },
     [AV_PIX_FMT_BGR48BE]     = { 1, 1 },
     [AV_PIX_FMT_BGR48LE]     = { 1, 1 },
-    [AV_PIX_FMT_BGRA64BE]    = { 0, 0, 1 },
-    [AV_PIX_FMT_BGRA64LE]    = { 0, 0, 1 },
+    [AV_PIX_FMT_BGRA64BE]    = { 1, 1, 1 },
+    [AV_PIX_FMT_BGRA64LE]    = { 1, 1, 1 },
     [AV_PIX_FMT_YUV420P9BE]  = { 1, 1 },
     [AV_PIX_FMT_YUV420P9LE]  = { 1, 1 },
     [AV_PIX_FMT_YUV420P10BE] = { 1, 1 },
diff --git a/libswscale/version.h b/libswscale/version.h
index 6f82d3d059..b0338636f5 100644
--- a/libswscale/version.h
+++ b/libswscale/version.h
@@ -27,8 +27,8 @@
 #include "libavutil/version.h"
 
 #define LIBSWSCALE_VERSION_MAJOR 2
-#define LIBSWSCALE_VERSION_MINOR 5
-#define LIBSWSCALE_VERSION_MICRO 102
+#define LIBSWSCALE_VERSION_MINOR 6
+#define LIBSWSCALE_VERSION_MICRO 100
 
 #define LIBSWSCALE_VERSION_INT  AV_VERSION_INT(LIBSWSCALE_VERSION_MAJOR, \
                                                LIBSWSCALE_VERSION_MINOR, \