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>

Doxyfile

@@ -1378,7 +1378,7 @@ PREDEFINED             = "__attribute__(x)=" \
                          "DEF(x)=x ## _TMPL" \
                          HAVE_AV_CONFIG_H \
                          HAVE_MMX \
-                         HAVE_MMX2 \
+                         HAVE_MMXEXT \
                          HAVE_AMD3DNOW \
                          "DECLARE_ALIGNED(a,t,n)=t n" \
                          "offsetof(x,y)=0x42"

configure

@@ -267,7 +267,7 @@ Optimization options (experts only):
   --disable-amd3dnow       disable 3DNow! optimizations
   --disable-amd3dnowext    disable 3DNow! extended optimizations
   --disable-mmx            disable MMX optimizations
-  --disable-mmx2           disable MMX2 optimizations
+  --disable-mmxext         disable MMXEXT optimizations
   --disable-sse            disable SSE optimizations
   --disable-ssse3          disable SSSE3 optimizations
   --disable-avx            disable AVX optimizations
@@ -1182,7 +1182,7 @@ ARCH_EXT_LIST='
     fma4
     mmi
     mmx
-    mmx2
+    mmxext
     neon
     ppc4xx
     sse
@@ -1459,7 +1459,7 @@ x86_64_suggest="cmov fast_cmov"
 amd3dnow_deps="mmx"
 amd3dnowext_deps="amd3dnow"
 mmx_deps="x86"
-mmx2_deps="mmx"
+mmxext_deps="mmx"
 sse_deps="mmx"
 ssse3_deps="sse"
 avx_deps="ssse3"
@@ -3194,9 +3194,9 @@ EOF
     # check whether xmm clobbers are supported
     check_asm xmm_clobbers '"":::"%xmm0"'
 
-    # check whether binutils is new enough to compile SSSE3/MMX2
+    # check whether binutils is new enough to compile SSSE3/MMXEXT
     enabled ssse3 && check_asm ssse3 '"pabsw %xmm0, %xmm0"'
-    enabled mmx2  && check_asm mmx2  '"pmaxub %mm0, %mm1"'
+    enabled mmxext && check_asm mmxext '"pmaxub %mm0, %mm1"'
 
     if ! disabled_any asm mmx yasm; then
         if check_cmd $yasmexe --version; then
@@ -3748,7 +3748,7 @@ echo "runtime cpu detection     ${runtime_cpudetect-no}"
 if enabled x86; then
     echo "${yasmexe}                      ${yasm-no}"
     echo "MMX enabled               ${mmx-no}"
-    echo "MMX2 enabled              ${mmx2-no}"
+    echo "MMXEXT enabled            ${mmxext-no}"
     echo "3DNow! enabled            ${amd3dnow-no}"
     echo "3DNow! extended enabled   ${amd3dnowext-no}"
     echo "SSE enabled               ${sse-no}"
@@ -4019,6 +4019,7 @@ cat > $TMPH <<EOF
 #define EXTERN_PREFIX "${extern_prefix}"
 #define EXTERN_ASM ${extern_prefix}
 #define SLIBSUF "$SLIBSUF"
+#define HAVE_MMX2 HAVE_MMXEXT
 EOF
 
 test -n "$assert_level" &&

doc/APIchanges

@@ -70,6 +70,11 @@ API changes, most recent first:
 2012-03-26 - a67d9cf - lavfi 2.66.100
   Add avfilter_fill_frame_from_{audio_,}buffer_ref() functions.
 
+2012-08-03 - xxxxxxx - lavu 51.37.1 - cpu.h
+                       lsws 2.1.1   - swscale.h
+  Rename AV_CPU_FLAG_MMX2  ---> AV_CPU_FLAG_MMXEXT.
+  Rename SWS_CPU_CAPS_MMX2 ---> SWS_CPU_CAPS_MMXEXT.
+
 2012-07-xx - xxxxxxx - lavf 54.13.0 - avformat.h
   Add AVFMT_FLAG_NOBUFFER for low latency use cases.
 

libavcodec/arm/vp56_arith.h

@@ -29,6 +29,14 @@
 #   define T(x)
 #endif
 
+#if CONFIG_THUMB || defined __clang__
+#   define L(x)
+#   define U(x) x
+#else
+#   define L(x) x
+#   define U(x)
+#endif
+
 #if HAVE_ARMV6 && HAVE_INLINE_ASM
 
 #define vp56_rac_get_prob vp56_rac_get_prob_armv6
@@ -42,8 +50,8 @@ static inline int vp56_rac_get_prob_armv6(VP56RangeCoder *c, int pr)
     __asm__ ("adds    %3,  %3,  %0           \n"
              "itt     cs                     \n"
              "cmpcs   %7,  %4                \n"
-           A("ldrcsh  %2,  [%4], #2          \n")
-           T("ldrhcs  %2,  [%4], #2          \n")
+           L("ldrcsh  %2,  [%4], #2          \n")
+           U("ldrhcs  %2,  [%4], #2          \n")
              "rsb     %0,  %6,  #256         \n"
              "smlabb  %0,  %5,  %6,  %0      \n"
            T("itttt   cs                     \n")
@@ -80,8 +88,8 @@ static inline int vp56_rac_get_prob_branchy_armv6(VP56RangeCoder *c, int pr)
     __asm__ ("adds    %3,  %3,  %0           \n"
              "itt     cs                     \n"
              "cmpcs   %7,  %4                \n"
-           A("ldrcsh  %2,  [%4], #2          \n")
-           T("ldrhcs  %2,  [%4], #2          \n")
+           L("ldrcsh  %2,  [%4], #2          \n")
+           U("ldrhcs  %2,  [%4], #2          \n")
              "rsb     %0,  %6,  #256         \n"
              "smlabb  %0,  %5,  %6,  %0      \n"
            T("itttt   cs                     \n")

libavcodec/arm/vp8dsp_armv6.S

@@ -1226,7 +1226,13 @@ vp8_mc_1                bilin,  8, v
 vp8_mc_1                bilin,  4, h
 vp8_mc_1                bilin,  4, v
 
-#define TMPSIZE \size * (8 - 8*(\size > 4) + \ytaps - 1)
+/* True relational expressions have the value -1 in the GNU assembler,
+   +1 in Apple's. */
+#ifdef __APPLE__
+#   define TMPSIZE \size * (8 + 8*(\size > 4) + \ytaps - 1)
+#else
+#   define TMPSIZE \size * (8 - 8*(\size > 4) + \ytaps - 1)
+#endif
 
 .macro  vp8_mc_hv       name, size, h, v, ytaps
 function ff_put_vp8_\name\size\()_\h\v\()_armv6, export=1

libavcodec/dct-test.c

@@ -87,7 +87,7 @@ static const struct algo fdct_tab[] = {
 
 #if HAVE_MMX && HAVE_INLINE_ASM
     { "MMX",            ff_fdct_mmx,           NO_PERM,   AV_CPU_FLAG_MMX     },
-    { "MMX2",           ff_fdct_mmx2,          NO_PERM,   AV_CPU_FLAG_MMX2    },
+    { "MMXEXT",         ff_fdct_mmx2,          NO_PERM,   AV_CPU_FLAG_MMXEXT  },
     { "SSE2",           ff_fdct_sse2,          NO_PERM,   AV_CPU_FLAG_SSE2    },
 #endif
 
@@ -132,7 +132,7 @@ static const struct algo idct_tab[] = {
 #endif
     { "SIMPLE-MMX",     ff_simple_idct_mmx,  MMX_SIMPLE_PERM, AV_CPU_FLAG_MMX },
     { "XVID-MMX",       ff_idct_xvid_mmx,      NO_PERM,   AV_CPU_FLAG_MMX,  1 },
-    { "XVID-MMX2",      ff_idct_xvid_mmx2,     NO_PERM,   AV_CPU_FLAG_MMX2, 1 },
+    { "XVID-MMXEXT",    ff_idct_xvid_mmx2,     NO_PERM,   AV_CPU_FLAG_MMXEXT, 1 },
     { "XVID-SSE2",      ff_idct_xvid_sse2,     SSE2_PERM, AV_CPU_FLAG_SSE2, 1 },
 #if ARCH_X86_64 && HAVE_YASM
     { "PR-SSE2",        ff_prores_idct_put_10_sse2_wrap,     TRANSPOSE_PERM, AV_CPU_FLAG_SSE2, 1 },

libavcodec/motion-test.c

@@ -116,8 +116,8 @@ int main(int argc, char **argv)
     AVCodecContext *ctx;
     int c;
     DSPContext cctx, mmxctx;
-    int flags[2] = { AV_CPU_FLAG_MMX, AV_CPU_FLAG_MMX2 };
-    int flags_size = HAVE_MMX2 ? 2 : 1;
+    int flags[2] = { AV_CPU_FLAG_MMX, AV_CPU_FLAG_MMXEXT };
+    int flags_size = HAVE_MMXEXT ? 2 : 1;
 
     if (argc > 1) {
         help();

libavcodec/x86/ac3dsp.asm

@@ -68,7 +68,7 @@ cglobal ac3_exponent_min_%1, 3,4,2, exp, reuse_blks, expn, offset
 %define LOOP_ALIGN
 INIT_MMX
 AC3_EXPONENT_MIN mmx
-%if HAVE_MMX2
+%if HAVE_MMXEXT
 %define PMINUB PMINUB_MMXEXT
 %define LOOP_ALIGN ALIGN 16
 AC3_EXPONENT_MIN mmxext

libavcodec/x86/ac3dsp_mmx.c

@@ -65,7 +65,7 @@ av_cold void ff_ac3dsp_init_x86(AC3DSPContext *c, int bit_exact)
             c->float_to_fixed24 = ff_float_to_fixed24_3dnow;
         }
     }
-    if (mm_flags & AV_CPU_FLAG_MMX2 && HAVE_MMX2) {
+    if (mm_flags & AV_CPU_FLAG_MMXEXT && HAVE_MMXEXT) {
         c->ac3_exponent_min = ff_ac3_exponent_min_mmxext;
         c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmx2;
     }

libavcodec/x86/cavsdsp_mmx.c

@@ -486,7 +486,7 @@ void ff_cavsdsp_init_mmx(CAVSDSPContext *c, AVCodecContext *avctx)
     int mm_flags = av_get_cpu_flags();
 
 #if HAVE_INLINE_ASM
-    if (mm_flags & AV_CPU_FLAG_MMX2)  ff_cavsdsp_init_mmx2 (c, avctx);
+    if (mm_flags & AV_CPU_FLAG_MMXEXT) ff_cavsdsp_init_mmx2(c, avctx);
     if (mm_flags & AV_CPU_FLAG_3DNOW) ff_cavsdsp_init_3dnow(c, avctx);
 #endif /* HAVE_INLINE_ASM */
 }