Merge remote-tracking branch 'qatar/master'

* qatar/master:
  lavr: fix handling of custom mix matrices
  fate: force pix_fmt in lagarith-rgb32 test
  fate: add tests for lagarith lossless video codec.
  ARMv6: vp8: fix stack allocation with Apple's assembler
  ARM: vp56: allow inline asm to build with clang
  fft: 3dnow: fix register name typo in DECL_IMDCT macro
  x86: dct32: port to cpuflags
  x86: build: replace mmx2 by mmxext
  Revert "wmapro: prevent division by zero when sample rate is unspecified"
  wmapro: prevent division by zero when sample rate is unspecified
  lagarith: fix color plane inversion for YUY2 output.
  lagarith: pad RGB buffer by 1 byte.
  dsputil: make add_hfyu_left_prediction_sse4() support unaligned src.

Conflicts:
	doc/APIchanges
	libavcodec/lagarith.c
	libavfilter/x86/gradfun.c
	libavutil/cpu.h
	libavutil/version.h
	libswscale/utils.c
	libswscale/version.h
	libswscale/x86/yuv2rgb.c

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

6 files changed, 47 insertions, 47 deletions

diff --git a/libswscale/x86/rgb2rgb.c b/libswscale/x86/rgb2rgb.c
index cd46df719d..9e4938a81a 100644
--- a/libswscale/x86/rgb2rgb.c
+++ b/libswscale/x86/rgb2rgb.c
@@ -88,7 +88,7 @@ DECLARE_ASM_CONST(8, uint64_t, mul16_mid)    = 0x2080208020802080ULL;
 
 //Note: We have C, MMX, MMX2, 3DNOW versions, there is no 3DNOW + MMX2 one.
 
-#define COMPILE_TEMPLATE_MMX2 0
+#define COMPILE_TEMPLATE_MMXEXT 0
 #define COMPILE_TEMPLATE_AMD3DNOW 0
 #define COMPILE_TEMPLATE_SSE2 0
 
@@ -99,8 +99,8 @@ DECLARE_ASM_CONST(8, uint64_t, mul16_mid)    = 0x2080208020802080ULL;
 
 //MMX2 versions
 #undef RENAME
-#undef COMPILE_TEMPLATE_MMX2
-#define COMPILE_TEMPLATE_MMX2 1
+#undef COMPILE_TEMPLATE_MMXEXT
+#define COMPILE_TEMPLATE_MMXEXT 1
 #define RENAME(a) a ## _MMX2
 #include "rgb2rgb_template.c"
 
@@ -113,10 +113,10 @@ DECLARE_ASM_CONST(8, uint64_t, mul16_mid)    = 0x2080208020802080ULL;
 
 //3DNOW versions
 #undef RENAME
-#undef COMPILE_TEMPLATE_MMX2
+#undef COMPILE_TEMPLATE_MMXEXT
 #undef COMPILE_TEMPLATE_SSE2
 #undef COMPILE_TEMPLATE_AMD3DNOW
-#define COMPILE_TEMPLATE_MMX2 0
+#define COMPILE_TEMPLATE_MMXEXT 0
 #define COMPILE_TEMPLATE_SSE2 0
 #define COMPILE_TEMPLATE_AMD3DNOW 1
 #define RENAME(a) a ## _3DNOW
@@ -140,7 +140,7 @@ av_cold void rgb2rgb_init_x86(void)
         rgb2rgb_init_MMX();
     if (HAVE_AMD3DNOW && cpu_flags & AV_CPU_FLAG_3DNOW)
         rgb2rgb_init_3DNOW();
-    if (HAVE_MMX2     && cpu_flags & AV_CPU_FLAG_MMX2)
+    if (HAVE_MMXEXT   && cpu_flags & AV_CPU_FLAG_MMXEXT)
         rgb2rgb_init_MMX2();
     if (HAVE_SSE      && cpu_flags & AV_CPU_FLAG_SSE2)
         rgb2rgb_init_SSE2();
diff --git a/libswscale/x86/rgb2rgb_template.c b/libswscale/x86/rgb2rgb_template.c
index bb97a0b4a1..594524d9ed 100644
--- a/libswscale/x86/rgb2rgb_template.c
+++ b/libswscale/x86/rgb2rgb_template.c
@@ -35,7 +35,7 @@
 #if COMPILE_TEMPLATE_AMD3DNOW
 #define PREFETCH  "prefetch"
 #define PAVGB     "pavgusb"
-#elif COMPILE_TEMPLATE_MMX2
+#elif COMPILE_TEMPLATE_MMXEXT
 #define PREFETCH "prefetchnta"
 #define PAVGB     "pavgb"
 #else
@@ -49,7 +49,7 @@
 #define EMMS     "emms"
 #endif
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 #define MOVNTQ "movntq"
 #define SFENCE "sfence"
 #else
@@ -1136,7 +1136,7 @@ static inline void RENAME(shuffle_bytes_2103)(const uint8_t *src, uint8_t *dst,
         PREFETCH"     32(%1, %0)        \n\t"
         "movq           (%1, %0), %%mm0 \n\t"
         "movq          8(%1, %0), %%mm1 \n\t"
-# if COMPILE_TEMPLATE_MMX2
+# if COMPILE_TEMPLATE_MMXEXT
         "pshufw      $177, %%mm0, %%mm3 \n\t"
         "pshufw      $177, %%mm1, %%mm5 \n\t"
         "pand       %%mm7, %%mm0        \n\t"
@@ -1500,7 +1500,7 @@ static inline void RENAME(yuy2toyv12)(const uint8_t *src, uint8_t *ydst, uint8_t
 }
 #endif /* !COMPILE_TEMPLATE_AMD3DNOW */
 
-#if COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW
+#if COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW
 static inline void RENAME(planar2x)(const uint8_t *src, uint8_t *dst, int srcWidth, int srcHeight, int srcStride, int dstStride)
 {
     int x,y;
@@ -1590,7 +1590,7 @@ static inline void RENAME(planar2x)(const uint8_t *src, uint8_t *dst, int srcWid
                      SFENCE"     \n\t"
                      :::"memory");
 }
-#endif /* COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW */
+#endif /* COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW */
 
 #if !COMPILE_TEMPLATE_AMD3DNOW
 /**
@@ -1798,7 +1798,7 @@ static inline void RENAME(rgb24toyv12)(const uint8_t *src, uint8_t *ydst, uint8_
             "1:                                         \n\t"
             PREFETCH"    64(%0, %%"REG_d")              \n\t"
             PREFETCH"    64(%1, %%"REG_d")              \n\t"
-#if COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW
+#if COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW
             "movq          (%0, %%"REG_d"), %%mm0       \n\t"
             "movq          (%1, %%"REG_d"), %%mm1       \n\t"
             "movq         6(%0, %%"REG_d"), %%mm2       \n\t"
@@ -1859,7 +1859,7 @@ static inline void RENAME(rgb24toyv12)(const uint8_t *src, uint8_t *ydst, uint8_
             "packssdw                %%mm1, %%mm0       \n\t" // V1 V0 U1 U0
             "psraw                      $7, %%mm0       \n\t"
 
-#if COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW
+#if COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW
             "movq        12(%0, %%"REG_d"), %%mm4       \n\t"
             "movq        12(%1, %%"REG_d"), %%mm1       \n\t"
             "movq        18(%0, %%"REG_d"), %%mm2       \n\t"
@@ -2580,9 +2580,9 @@ static inline void RENAME(rgb2rgb_init)(void)
     yuyvtoyuv422       = RENAME(yuyvtoyuv422);
 #endif /* !COMPILE_TEMPLATE_SSE2 */
 
-#if COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW
+#if COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW
     planar2x           = RENAME(planar2x);
-#endif /* COMPILE_TEMPLATE_MMX2 || COMPILE_TEMPLATE_AMD3DNOW */
+#endif /* COMPILE_TEMPLATE_MMXEXT || COMPILE_TEMPLATE_AMD3DNOW */
     rgb24toyv12        = RENAME(rgb24toyv12);
 
     yuyvtoyuv420       = RENAME(yuyvtoyuv420);
diff --git a/libswscale/x86/swscale.c b/libswscale/x86/swscale.c
index eb33788168..132ec3e570 100644
--- a/libswscale/x86/swscale.c
+++ b/libswscale/x86/swscale.c
@@ -74,16 +74,16 @@ DECLARE_ALIGNED(8, const uint64_t, ff_w1111)        = 0x0001000100010001ULL;
 //MMX versions
 #if HAVE_MMX
 #undef RENAME
-#define COMPILE_TEMPLATE_MMX2 0
+#define COMPILE_TEMPLATE_MMXEXT 0
 #define RENAME(a) a ## _MMX
 #include "swscale_template.c"
 #endif
 
 //MMX2 versions
-#if HAVE_MMX2
+#if HAVE_MMXEXT
 #undef RENAME
-#undef COMPILE_TEMPLATE_MMX2
-#define COMPILE_TEMPLATE_MMX2 1
+#undef COMPILE_TEMPLATE_MMXEXT
+#define COMPILE_TEMPLATE_MMXEXT 1
 #define RENAME(a) a ## _MMX2
 #include "swscale_template.c"
 #endif
@@ -375,8 +375,8 @@ av_cold void ff_sws_init_swScale_mmx(SwsContext *c)
 #if HAVE_INLINE_ASM
     if (cpu_flags & AV_CPU_FLAG_MMX)
         sws_init_swScale_MMX(c);
-#if HAVE_MMX2
-    if (cpu_flags & AV_CPU_FLAG_MMX2)
+#if HAVE_MMXEXT
+    if (cpu_flags & AV_CPU_FLAG_MMXEXT)
         sws_init_swScale_MMX2(c);
     if (cpu_flags & AV_CPU_FLAG_SSE3){
         if(c->use_mmx_vfilter && !(c->flags & SWS_ACCURATE_RND))
@@ -439,7 +439,7 @@ switch(c->dstBpc){ \
     if (cpu_flags & AV_CPU_FLAG_MMX) {
         ASSIGN_MMX_SCALE_FUNC(c->hyScale, c->hLumFilterSize, mmx, mmx);
         ASSIGN_MMX_SCALE_FUNC(c->hcScale, c->hChrFilterSize, mmx, mmx);
-        ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmx2, cpu_flags & AV_CPU_FLAG_MMX2);
+        ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmx2, cpu_flags & AV_CPU_FLAG_MMXEXT);
 
         switch (c->srcFormat) {
         case PIX_FMT_Y400A:
@@ -471,7 +471,7 @@ switch(c->dstBpc){ \
             break;
         }
     }
-    if (cpu_flags & AV_CPU_FLAG_MMX2) {
+    if (cpu_flags & AV_CPU_FLAG_MMXEXT) {
         ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmx2, , 1);
     }
 #endif
diff --git a/libswscale/x86/swscale_template.c b/libswscale/x86/swscale_template.c
index 245bfdeadd..370a0ebe1b 100644
--- a/libswscale/x86/swscale_template.c
+++ b/libswscale/x86/swscale_template.c
@@ -23,13 +23,13 @@
 #undef MOVNTQ2
 #undef PREFETCH
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 #define PREFETCH "prefetchnta"
 #else
 #define PREFETCH  " # nop"
 #endif
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 #define REAL_MOVNTQ(a,b) "movntq " #a ", " #b " \n\t"
 #define MOVNTQ2 "movntq "
 #else
@@ -38,7 +38,7 @@
 #endif
 #define MOVNTQ(a,b)  REAL_MOVNTQ(a,b)
 
-#if !COMPILE_TEMPLATE_MMX2
+#if !COMPILE_TEMPLATE_MMXEXT
 static av_always_inline void
 dither_8to16(const uint8_t *srcDither, int rot)
 {
@@ -641,7 +641,7 @@ static void RENAME(yuv2rgb555_X)(SwsContext *c, const int16_t *lumFilter,
     "cmp  "#dstw", "#index"     \n\t"\
     " jb       1b               \n\t"
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 #undef WRITEBGR24
 #define WRITEBGR24(dst, dstw, index)  WRITEBGR24MMX2(dst, dstw, index)
 #else
@@ -1445,7 +1445,7 @@ static void RENAME(yuv2yuyv422_1)(SwsContext *c, const int16_t *buf0,
     }
 }
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 static void RENAME(hyscale_fast)(SwsContext *c, int16_t *dst,
                                  int dstWidth, const uint8_t *src,
                                  int srcW, int xInc)
@@ -1627,7 +1627,7 @@ static void RENAME(hcscale_fast)(SwsContext *c, int16_t *dst1, int16_t *dst2,
         dst2[i] = src2[srcW-1]*128;
     }
 }
-#endif /* COMPILE_TEMPLATE_MMX2 */
+#endif /* COMPILE_TEMPLATE_MMXEXT */
 
 static av_cold void RENAME(sws_init_swScale)(SwsContext *c)
 {
@@ -1691,17 +1691,17 @@ static av_cold void RENAME(sws_init_swScale)(SwsContext *c)
 
     if (c->srcBpc == 8 && c->dstBpc <= 14) {
     // Use the new MMX scaler if the MMX2 one can't be used (it is faster than the x86 ASM one).
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
     if (c->flags & SWS_FAST_BILINEAR && c->canMMX2BeUsed)
     {
         c->hyscale_fast = RENAME(hyscale_fast);
         c->hcscale_fast = RENAME(hcscale_fast);
     } else {
-#endif /* COMPILE_TEMPLATE_MMX2 */
+#endif /* COMPILE_TEMPLATE_MMXEXT */
         c->hyscale_fast = NULL;
         c->hcscale_fast = NULL;
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
     }
-#endif /* COMPILE_TEMPLATE_MMX2 */
+#endif /* COMPILE_TEMPLATE_MMXEXT */
     }
 }
diff --git a/libswscale/x86/yuv2rgb.c b/libswscale/x86/yuv2rgb.c
index 2a76716335..8cb7f1ae0d 100644
--- a/libswscale/x86/yuv2rgb.c
+++ b/libswscale/x86/yuv2rgb.c
@@ -52,20 +52,20 @@ DECLARE_ASM_CONST(8, uint64_t, pb_07) = 0x0707070707070707ULL;
 //MMX versions
 #if HAVE_MMX
 #undef RENAME
-#undef COMPILE_TEMPLATE_MMX2
-#define COMPILE_TEMPLATE_MMX2 0
+#undef COMPILE_TEMPLATE_MMXEXT
+#define COMPILE_TEMPLATE_MMXEXT 0
 #define RENAME(a) a ## _MMX
 #include "yuv2rgb_template.c"
 #endif /* HAVE_MMX */
 
 //MMX2 versions
-#if HAVE_MMX2
+#if HAVE_MMXEXT
 #undef RENAME
-#undef COMPILE_TEMPLATE_MMX2
-#define COMPILE_TEMPLATE_MMX2 1
+#undef COMPILE_TEMPLATE_MMXEXT
+#define COMPILE_TEMPLATE_MMXEXT 1
 #define RENAME(a) a ## _MMX2
 #include "yuv2rgb_template.c"
-#endif /* HAVE_MMX2 */
+#endif /* HAVE_MMXEXT */
 
 #endif /* HAVE_INLINE_ASM */
 
@@ -74,8 +74,8 @@ av_cold SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c)
 #if HAVE_INLINE_ASM
     int cpu_flags = av_get_cpu_flags();
 
-#if HAVE_MMX2
-    if (cpu_flags & AV_CPU_FLAG_MMX2) {
+#if HAVE_MMXEXT
+    if (cpu_flags & AV_CPU_FLAG_MMXEXT) {
         switch (c->dstFormat) {
         case PIX_FMT_RGB24:  return yuv420_rgb24_MMX2;
         case PIX_FMT_BGR24:  return yuv420_bgr24_MMX2;
diff --git a/libswscale/x86/yuv2rgb_template.c b/libswscale/x86/yuv2rgb_template.c
index 624de14252..79e48fd7f0 100644
--- a/libswscale/x86/yuv2rgb_template.c
+++ b/libswscale/x86/yuv2rgb_template.c
@@ -25,7 +25,7 @@
 #undef EMMS
 #undef SFENCE
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 #define MOVNTQ "movntq"
 #define SFENCE "sfence"
 #else
@@ -181,7 +181,7 @@
     "paddusb "GREEN_DITHER"(%4), %%mm2\n\t"      \
     "paddusb "RED_DITHER"(%4),   %%mm1\n\t"      \
 
-#if !COMPILE_TEMPLATE_MMX2
+#if !COMPILE_TEMPLATE_MMXEXT
 static inline int RENAME(yuv420_rgb15)(SwsContext *c, const uint8_t *src[],
                                        int srcStride[],
                                        int srcSliceY, int srcSliceH,
@@ -237,7 +237,7 @@ static inline int RENAME(yuv420_rgb16)(SwsContext *c, const uint8_t *src[],
     YUV2RGB_OPERANDS
     YUV2RGB_ENDFUNC
 }
-#endif /* !COMPILE_TEMPLATE_MMX2 */
+#endif /* !COMPILE_TEMPLATE_MMXEXT */
 
 #define RGB_PACK24(blue, red)\
     "packuswb  %%mm3,      %%mm0 \n" /* R0 R2 R4 R6 R1 R3 R5 R7 */\
@@ -254,7 +254,7 @@ static inline int RENAME(yuv420_rgb16)(SwsContext *c, const uint8_t *src[],
     "punpckhwd %%mm6,      %%mm5 \n" /* R4 G4 B4 R5 R6 G6 B6 R7 */\
     RGB_PACK24_B
 
-#if COMPILE_TEMPLATE_MMX2
+#if COMPILE_TEMPLATE_MMXEXT
 DECLARE_ASM_CONST(8, int16_t, mask1101[4]) = {-1,-1, 0,-1};
 DECLARE_ASM_CONST(8, int16_t, mask0010[4]) = { 0, 0,-1, 0};
 DECLARE_ASM_CONST(8, int16_t, mask0110[4]) = { 0,-1,-1, 0};
@@ -361,7 +361,7 @@ static inline int RENAME(yuv420_bgr24)(SwsContext *c, const uint8_t *src[],
     MOVNTQ "   %%mm5,       16(%1)\n\t"      \
     MOVNTQ "   %%mm"alpha", 24(%1)\n\t"      \
 
-#if !COMPILE_TEMPLATE_MMX2
+#if !COMPILE_TEMPLATE_MMXEXT
 static inline int RENAME(yuv420_rgb32)(SwsContext *c, const uint8_t *src[],
                                        int srcStride[],
                                        int srcSliceY, int srcSliceH,
@@ -448,4 +448,4 @@ static inline int RENAME(yuva420_bgr32)(SwsContext *c, const uint8_t *src[],
 }
 #endif
 
-#endif /* !COMPILE_TEMPLATE_MMX2 */
+#endif /* !COMPILE_TEMPLATE_MMXEXT */