Merge commit 'fa8fcab1e0d31074c0644c4ac5194474c6c26415'

* commit 'fa8fcab1e0d31074c0644c4ac5194474c6c26415':
  x86: h264_chromamc_10bit: drop pointless PAVG %define
  x86: mmx2 ---> mmxext in function names
  swscale: do not forget to swap data in formats with different endianness

Conflicts:
	libavcodec/x86/dsputil_mmx.c
	libavfilter/x86/gradfun.c
	libswscale/input.c
	libswscale/utils.c
	libswscale/x86/swscale.c
	tests/ref/lavfi/pixfmts_scale

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

5 files changed, 30 insertions, 21 deletions

diff --git a/libswscale/input.c b/libswscale/input.c
index 43a67aac90..d3aeb22151 100644
--- a/libswscale/input.c
+++ b/libswscale/input.c
@@ -935,6 +935,7 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUV420P14LE:
     case AV_PIX_FMT_YUV420P16LE:
     case AV_PIX_FMT_YUV422P16LE:
+    case AV_PIX_FMT_YUV444P16LE:
 
     case AV_PIX_FMT_YUVA444P9LE:
     case AV_PIX_FMT_YUVA422P9LE:
@@ -945,7 +946,6 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUVA420P16LE:
     case AV_PIX_FMT_YUVA422P16LE:
     case AV_PIX_FMT_YUVA444P16LE:
-    case AV_PIX_FMT_YUV444P16LE:
         c->chrToYV12 = bswap16UV_c;
         break;
 #else
@@ -1186,6 +1186,9 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUV422P16LE:
     case AV_PIX_FMT_YUV444P16LE:
 
+    case AV_PIX_FMT_GRAY16LE:
+        c->lumToYV12 = bswap16Y_c;
+        break;
     case AV_PIX_FMT_YUVA444P9LE:
     case AV_PIX_FMT_YUVA422P9LE:
     case AV_PIX_FMT_YUVA420P9LE:
@@ -1195,8 +1198,8 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUVA420P16LE:
     case AV_PIX_FMT_YUVA422P16LE:
     case AV_PIX_FMT_YUVA444P16LE:
-    case AV_PIX_FMT_GRAY16LE:
         c->lumToYV12 = bswap16Y_c;
+        c->alpToYV12 = bswap16Y_c;
         break;
 #else
     case AV_PIX_FMT_YUV444P9BE:
@@ -1215,6 +1218,9 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUV422P16BE:
     case AV_PIX_FMT_YUV444P16BE:
 
+    case AV_PIX_FMT_GRAY16BE:
+        c->lumToYV12 = bswap16Y_c;
+        break;
     case AV_PIX_FMT_YUVA444P9BE:
     case AV_PIX_FMT_YUVA422P9BE:
     case AV_PIX_FMT_YUVA420P9BE:
@@ -1224,8 +1230,8 @@ av_cold void ff_sws_init_input_funcs(SwsContext *c)
     case AV_PIX_FMT_YUVA420P16BE:
     case AV_PIX_FMT_YUVA422P16BE:
     case AV_PIX_FMT_YUVA444P16BE:
-    case AV_PIX_FMT_GRAY16BE:
         c->lumToYV12 = bswap16Y_c;
+        c->alpToYV12 = bswap16Y_c;
         break;
 #endif
     case AV_PIX_FMT_YUYV422:
diff --git a/libswscale/utils.c b/libswscale/utils.c
index f0f8a91b8f..7761441fcf 100644
--- a/libswscale/utils.c
+++ b/libswscale/utils.c
@@ -627,8 +627,9 @@ fail:
 }
 
 #if HAVE_MMXEXT_INLINE
-static int initMMX2HScaler(int dstW, int xInc, uint8_t *filterCode,
-                           int16_t *filter, int32_t *filterPos, int numSplits)
+static int init_hscaler_mmxext(int dstW, int xInc, uint8_t *filterCode,
+                               int16_t *filter, int32_t *filterPos,
+                               int numSplits)
 {
     uint8_t *fragmentA;
     x86_reg imm8OfPShufW1A;
@@ -1107,10 +1108,10 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
 #if HAVE_MMXEXT_INLINE
 // can't downscale !!!
         if (c->canMMXEXTBeUsed && (flags & SWS_FAST_BILINEAR)) {
-            c->lumMmxextFilterCodeSize = initMMX2HScaler(dstW, c->lumXInc, NULL,
-                                                         NULL, NULL, 8);
-            c->chrMmxextFilterCodeSize = initMMX2HScaler(c->chrDstW, c->chrXInc,
-                                                         NULL, NULL, NULL, 4);
+            c->lumMmxextFilterCodeSize = init_hscaler_mmxext(dstW, c->lumXInc, NULL,
+                                                             NULL, NULL, 8);
+            c->chrMmxextFilterCodeSize = init_hscaler_mmxext(c->chrDstW, c->chrXInc,
+                                                             NULL, NULL, NULL, 4);
 
 #if USE_MMAP
             c->lumMmxextFilterCode = mmap(NULL, c->lumMmxextFilterCodeSize,
@@ -1150,10 +1151,10 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
             FF_ALLOCZ_OR_GOTO(c, c->hLumFilterPos, (dstW       / 2 / 8 + 8) * sizeof(int32_t), fail);
             FF_ALLOCZ_OR_GOTO(c, c->hChrFilterPos, (c->chrDstW / 2 / 4 + 8) * sizeof(int32_t), fail);
 
-            initMMX2HScaler(      dstW, c->lumXInc, c->lumMmxextFilterCode,
-                            c->hLumFilter, (uint32_t*)c->hLumFilterPos, 8);
-            initMMX2HScaler(c->chrDstW, c->chrXInc, c->chrMmxextFilterCode,
-                            c->hChrFilter, (uint32_t*)c->hChrFilterPos, 4);
+            init_hscaler_mmxext(      dstW, c->lumXInc, c->lumMmxextFilterCode,
+                                c->hLumFilter, (uint32_t*)c->hLumFilterPos, 8);
+            init_hscaler_mmxext(c->chrDstW, c->chrXInc, c->chrMmxextFilterCode,
+                                c->hChrFilter, (uint32_t*)c->hChrFilterPos, 4);
 
 #if USE_MMAP
             mprotect(c->lumMmxextFilterCode, c->lumMmxextFilterCodeSize, PROT_EXEC | PROT_READ);
diff --git a/libswscale/x86/rgb2rgb.c b/libswscale/x86/rgb2rgb.c
index f7288d3e44..1e20176a03 100644
--- a/libswscale/x86/rgb2rgb.c
+++ b/libswscale/x86/rgb2rgb.c
@@ -102,7 +102,7 @@ DECLARE_ASM_CONST(8, uint64_t, mul16_mid)    = 0x2080208020802080ULL;
 #undef RENAME
 #undef COMPILE_TEMPLATE_MMXEXT
 #define COMPILE_TEMPLATE_MMXEXT 1
-#define RENAME(a) a ## _MMX2
+#define RENAME(a) a ## _MMXEXT
 #include "rgb2rgb_template.c"
 
 //SSE2 versions
@@ -142,7 +142,7 @@ av_cold void rgb2rgb_init_x86(void)
     if (INLINE_AMD3DNOW(cpu_flags))
         rgb2rgb_init_3DNOW();
     if (INLINE_MMXEXT(cpu_flags))
-        rgb2rgb_init_MMX2();
+        rgb2rgb_init_MMXEXT();
     if (INLINE_SSE2(cpu_flags))
         rgb2rgb_init_SSE2();
 #endif /* HAVE_INLINE_ASM */
diff --git a/libswscale/x86/swscale.c b/libswscale/x86/swscale.c
index e82666bec9..fd14021e1f 100644
--- a/libswscale/x86/swscale.c
+++ b/libswscale/x86/swscale.c
@@ -85,7 +85,7 @@ DECLARE_ALIGNED(8, const uint64_t, ff_w1111)        = 0x0001000100010001ULL;
 #undef RENAME
 #undef COMPILE_TEMPLATE_MMXEXT
 #define COMPILE_TEMPLATE_MMXEXT 1
-#define RENAME(a) a ## _MMX2
+#define RENAME(a) a ## _MMXEXT
 #include "swscale_template.c"
 #endif
 
@@ -211,7 +211,7 @@ static void yuv2yuvX_sse3(const int16_t *filter, int filterSize,
                            const uint8_t *dither, int offset)
 {
     if(((int)dest) & 15){
-        return yuv2yuvX_MMX2(filter, filterSize, src, dest, dstW, dither, offset);
+        return yuv2yuvX_MMXEXT(filter, filterSize, src, dest, dstW, dither, offset);
     }
     if (offset) {
         __asm__ volatile("movq       (%0), %%xmm3\n\t"
@@ -381,7 +381,7 @@ av_cold void ff_sws_init_swScale_mmx(SwsContext *c)
         sws_init_swScale_MMX(c);
 #if HAVE_MMXEXT_INLINE
     if (cpu_flags & AV_CPU_FLAG_MMXEXT)
-        sws_init_swScale_MMX2(c);
+        sws_init_swScale_MMXEXT(c);
     if (cpu_flags & AV_CPU_FLAG_SSE3){
         if(c->use_mmx_vfilter && !(c->flags & SWS_ACCURATE_RND))
             c->yuv2planeX = yuv2yuvX_sse3;
diff --git a/libswscale/x86/yuv2rgb.c b/libswscale/x86/yuv2rgb.c
index ea52e3c71b..3938e6bae3 100644
--- a/libswscale/x86/yuv2rgb.c
+++ b/libswscale/x86/yuv2rgb.c
@@ -63,7 +63,7 @@ DECLARE_ASM_CONST(8, uint64_t, pb_07) = 0x0707070707070707ULL;
 #undef RENAME
 #undef COMPILE_TEMPLATE_MMXEXT
 #define COMPILE_TEMPLATE_MMXEXT 1
-#define RENAME(a) a ## _MMX2
+#define RENAME(a) a ## _MMXEXT
 #include "yuv2rgb_template.c"
 #endif /* HAVE_MMXEXT_INLINE */
 
@@ -77,8 +77,10 @@ av_cold SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c)
 #if HAVE_MMXEXT_INLINE
     if (cpu_flags & AV_CPU_FLAG_MMXEXT) {
         switch (c->dstFormat) {
-        case AV_PIX_FMT_RGB24:  return yuv420_rgb24_MMX2;
-        case AV_PIX_FMT_BGR24:  return yuv420_bgr24_MMX2;
+        case AV_PIX_FMT_RGB24:
+            return yuv420_rgb24_MMXEXT;
+        case AV_PIX_FMT_BGR24:
+            return yuv420_bgr24_MMXEXT;
         }
     }
 #endif