Merge remote-tracking branch 'qatar/master'

* qatar/master: (29 commits)
  amrwb: remove duplicate arguments from extrapolate_isf().
  amrwb: error out early if mode is invalid.
  h264: change underread for 10bit QPEL to overread.
  matroska: check buffer size for RM-style byte reordering.
  vp8: disable mmx functions with sse/sse2 counterparts on x86-64.
  vp8: change int stride to ptrdiff_t stride.
  wma: fix invalid buffer size assumptions causing random overreads.
  Windows Media Audio Lossless decoder
  rv10/20: Fix slice overflow with checked bitstream reader.
  h263dec: Disallow width/height changing with frame threads.
  rv10/20: Fix a buffer overread caused by losing track of the remaining buffer size.
  rmdec: Honor .RMF tag size rather than assuming 18.
  g722: Fix the QMF scaling
  r3d: don't set codec timebase.
  electronicarts: set timebase for tgv video.
  electronicarts: parse the framerate for cmv video.
  ogg: don't set codec timebase
  electronicarts: don't set codec timebase
  avs: don't set codec timebase
  wavpack: Fix an integer overflow
  ...

Conflicts:
	libavcodec/arm/vp8dsp_init_arm.c
	libavcodec/fraps.c
	libavcodec/h264.c
	libavcodec/mpeg4videodec.c
	libavcodec/mpegvideo.c
	libavcodec/msmpeg4.c
	libavcodec/pnmdec.c
	libavcodec/qpeg.c
	libavcodec/rawenc.c
	libavcodec/ulti.c
	libavcodec/vcr1.c
	libavcodec/version.h
	libavcodec/wmalosslessdec.c
	libavformat/electronicarts.c
	libswscale/ppc/yuv2rgb_altivec.c
	tests/ref/acodec/g722
	tests/ref/fate/ea-cmv

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

3 files changed, 142 insertions, 88 deletions

diff --git a/libavcodec/x86/h264_qpel_10bit.asm b/libavcodec/x86/h264_qpel_10bit.asm
index 51412e3977..bdacf9f472 100644
--- a/libavcodec/x86/h264_qpel_10bit.asm
+++ b/libavcodec/x86/h264_qpel_10bit.asm
@@ -619,7 +619,7 @@ MC MC33
 %define PAD 12
 %define COUNT 2
 %else
-%define PAD 0
+%define PAD 4
 %define COUNT 3
 %endif
 put_hv%2_10_%1:
diff --git a/libavcodec/x86/vp8dsp-init.c b/libavcodec/x86/vp8dsp-init.c
index a75fdf5bc5..e3b727d1b1 100644
--- a/libavcodec/x86/vp8dsp-init.c
+++ b/libavcodec/x86/vp8dsp-init.c
@@ -29,98 +29,98 @@
 /*
  * MC functions
  */
-extern void ff_put_vp8_epel4_h4_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h4_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_h6_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h6_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v4_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v4_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v6_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v6_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_epel8_h4_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h4_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h6_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h6_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v4_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v4_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v6_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v6_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_epel4_h4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_h6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_bilinear4_h_mmxext(uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_h_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_h_sse2  (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_h_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_h_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
 
-extern void ff_put_vp8_bilinear4_v_mmxext(uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_v_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_v_sse2  (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_v_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_v_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_v_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
 
 
-extern void ff_put_vp8_pixels8_mmx (uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels8_mmx (uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
-extern void ff_put_vp8_pixels16_mmx(uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels16_mmx(uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
-extern void ff_put_vp8_pixels16_sse(uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels16_sse(uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
 
 #define TAP_W16(OPT, FILTERTYPE, TAPTYPE) \
 static void ff_put_vp8_ ## FILTERTYPE ## 16_ ## TAPTYPE ## _ ## OPT( \
-    uint8_t *dst,  int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst,  ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
         dst,     dststride, src,     srcstride, height, mx, my); \
@@ -129,8 +129,8 @@ static void ff_put_vp8_ ## FILTERTYPE ## 16_ ## TAPTYPE ## _ ## OPT( \
 }
 #define TAP_W8(OPT, FILTERTYPE, TAPTYPE) \
 static void ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
-    uint8_t *dst,  int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst,  ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     ff_put_vp8_ ## FILTERTYPE ## 4_ ## TAPTYPE ## _ ## OPT( \
         dst,     dststride, src,     srcstride, height, mx, my); \
@@ -138,6 +138,7 @@ static void ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
         dst + 4, dststride, src + 4, srcstride, height, mx, my); \
 }
 
+#if ARCH_X86_32
 TAP_W8 (mmxext, epel, h4)
 TAP_W8 (mmxext, epel, h6)
 TAP_W16(mmxext, epel, h6)
@@ -148,6 +149,7 @@ TAP_W8 (mmxext, bilinear, h)
 TAP_W16(mmxext, bilinear, h)
 TAP_W8 (mmxext, bilinear, v)
 TAP_W16(mmxext, bilinear, v)
+#endif
 
 TAP_W16(sse2,   epel, h6)
 TAP_W16(sse2,   epel, v6)
@@ -161,8 +163,8 @@ TAP_W16(ssse3,  bilinear, v)
 
 #define HVTAP(OPT, ALIGN, TAPNUMX, TAPNUMY, SIZE, MAXHEIGHT) \
 static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v ## TAPNUMY ## _ ## OPT( \
-    uint8_t *dst, int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst, ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     DECLARE_ALIGNED(ALIGN, uint8_t, tmp)[SIZE * (MAXHEIGHT + TAPNUMY - 1)]; \
     uint8_t *tmpptr = tmp + SIZE * (TAPNUMY / 2 - 1); \
@@ -173,15 +175,21 @@ static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v ## TAPNUMY ## _ ## OPT
         dst, dststride, tmpptr, SIZE,      height,               mx, my); \
 }
 
+#if ARCH_X86_32
 #define HVTAPMMX(x, y) \
 HVTAP(mmxext, 8, x, y,  4,  8) \
 HVTAP(mmxext, 8, x, y,  8, 16)
 
+HVTAP(mmxext, 8, 6, 6, 16, 16)
+#else
+#define HVTAPMMX(x, y) \
+HVTAP(mmxext, 8, x, y,  4,  8)
+#endif
+
 HVTAPMMX(4, 4)
 HVTAPMMX(4, 6)
 HVTAPMMX(6, 4)
 HVTAPMMX(6, 6)
-HVTAP(mmxext, 8, 6, 6, 16, 16)
 
 #define HVTAPSSE2(x, y, w) \
 HVTAP(sse2,  16, x, y, w, 16) \
@@ -200,8 +208,8 @@ HVTAP(ssse3, 16, 6, 6, 4, 8)
 
 #define HVBILIN(OPT, ALIGN, SIZE, MAXHEIGHT) \
 static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
-    uint8_t *dst, int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst, ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     DECLARE_ALIGNED(ALIGN, uint8_t, tmp)[SIZE * (MAXHEIGHT + 2)]; \
     ff_put_vp8_bilinear ## SIZE ## _h_ ## OPT( \
@@ -211,43 +219,68 @@ static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
 }
 
 HVBILIN(mmxext, 8,  4,  8)
+#if ARCH_X86_32
 HVBILIN(mmxext, 8,  8, 16)
 HVBILIN(mmxext, 8, 16, 16)
+#endif
 HVBILIN(sse2,   8,  8, 16)
 HVBILIN(sse2,   8, 16, 16)
 HVBILIN(ssse3,  8,  4,  8)
 HVBILIN(ssse3,  8,  8, 16)
 HVBILIN(ssse3,  8, 16, 16)
 
-extern void ff_vp8_idct_dc_add_mmx(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, DCTELEM block[4][16], int stride);
-extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, DCTELEM block[4][16], int stride);
-extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, DCTELEM block[2][16], int stride);
+extern void ff_vp8_idct_dc_add_mmx(uint8_t *dst, DCTELEM block[16],
+                                   ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, DCTELEM block[16],
+                                    ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, DCTELEM block[4][16],
+                                      ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, DCTELEM block[4][16],
+                                      ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, DCTELEM block[2][16],
+                                      ptrdiff_t stride);
 extern void ff_vp8_luma_dc_wht_mmx(DCTELEM block[4][4][16], DCTELEM dc[16]);
 extern void ff_vp8_luma_dc_wht_sse(DCTELEM block[4][4][16], DCTELEM dc[16]);
-extern void ff_vp8_idct_add_mmx(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_add_sse(uint8_t *dst, DCTELEM block[16], int stride);
+extern void ff_vp8_idct_add_mmx(uint8_t *dst, DCTELEM block[16],
+                                ptrdiff_t stride);
+extern void ff_vp8_idct_add_sse(uint8_t *dst, DCTELEM block[16],
+                                ptrdiff_t stride);
 
 #define DECLARE_LOOP_FILTER(NAME)\
-extern void ff_vp8_v_loop_filter_simple_ ## NAME(uint8_t *dst, int stride, int flim);\
-extern void ff_vp8_h_loop_filter_simple_ ## NAME(uint8_t *dst, int stride, int flim);\
-extern void ff_vp8_v_loop_filter16y_inner_ ## NAME (uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter_simple_ ## NAME(uint8_t *dst, \
+                                                 ptrdiff_t stride, \
+                                                 int flim);\
+extern void ff_vp8_h_loop_filter_simple_ ## NAME(uint8_t *dst, \
+                                                 ptrdiff_t stride, \
+                                                 int flim);\
+extern void ff_vp8_v_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
+                                                     ptrdiff_t stride,\
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_h_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
+                                                    ptrdiff_t stride,\
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_v_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_h_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
                                                     int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter16y_inner_ ## NAME (uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
+                                                    ptrdiff_t stride,\
                                                     int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, int stride,\
+extern void ff_vp8_h_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
+                                                    ptrdiff_t stride,\
                                                     int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
                                                     int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);
+extern void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
+                                                    int e, int i, int hvt);
 
 DECLARE_LOOP_FILTER(mmx)
 DECLARE_LOOP_FILTER(mmxext)
@@ -288,15 +321,18 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
 
     if (mm_flags & AV_CPU_FLAG_MMX) {
         c->vp8_idct_dc_add    = ff_vp8_idct_dc_add_mmx;
-        c->vp8_idct_dc_add4y  = ff_vp8_idct_dc_add4y_mmx;
         c->vp8_idct_dc_add4uv = ff_vp8_idct_dc_add4uv_mmx;
+#if ARCH_X86_32
+        c->vp8_idct_dc_add4y  = ff_vp8_idct_dc_add4y_mmx;
         c->vp8_idct_add       = ff_vp8_idct_add_mmx;
         c->vp8_luma_dc_wht    = ff_vp8_luma_dc_wht_mmx;
         c->put_vp8_epel_pixels_tab[0][0][0]     =
         c->put_vp8_bilinear_pixels_tab[0][0][0] = ff_put_vp8_pixels16_mmx;
+#endif
         c->put_vp8_epel_pixels_tab[1][0][0]     =
         c->put_vp8_bilinear_pixels_tab[1][0][0] = ff_put_vp8_pixels8_mmx;
 
+#if ARCH_X86_32
         c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmx;
         c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmx;
 
@@ -309,17 +345,19 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
         c->vp8_h_loop_filter16y       = ff_vp8_h_loop_filter16y_mbedge_mmx;
         c->vp8_v_loop_filter8uv       = ff_vp8_v_loop_filter8uv_mbedge_mmx;
         c->vp8_h_loop_filter8uv       = ff_vp8_h_loop_filter8uv_mbedge_mmx;
+#endif
     }
 
     /* note that 4-tap width=16 functions are missing because w=16
      * is only used for luma, and luma is always a copy or sixtap. */
     if (mm_flags & AV_CPU_FLAG_MMX2) {
+        VP8_MC_FUNC(2, 4, mmxext);
+        VP8_BILINEAR_MC_FUNC(2, 4, mmxext);
+#if ARCH_X86_32
         VP8_LUMA_MC_FUNC(0, 16, mmxext);
         VP8_MC_FUNC(1, 8, mmxext);
-        VP8_MC_FUNC(2, 4, mmxext);
         VP8_BILINEAR_MC_FUNC(0, 16, mmxext);
         VP8_BILINEAR_MC_FUNC(1, 8, mmxext);
-        VP8_BILINEAR_MC_FUNC(2, 4, mmxext);
 
         c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmxext;
         c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmxext;
@@ -333,6 +371,7 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
         c->vp8_h_loop_filter16y       = ff_vp8_h_loop_filter16y_mbedge_mmxext;
         c->vp8_v_loop_filter8uv       = ff_vp8_v_loop_filter8uv_mbedge_mmxext;
         c->vp8_h_loop_filter8uv       = ff_vp8_h_loop_filter8uv_mbedge_mmxext;
+#endif
     }
 
     if (mm_flags & AV_CPU_FLAG_SSE) {
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index 833c88a1a0..140597031f 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
@@ -865,6 +865,7 @@ cglobal put_vp8_pixels8_mmx, 5,5
     jg .nextrow
     REP_RET
 
+%if ARCH_X86_32
 cglobal put_vp8_pixels16_mmx, 5,5
 .nextrow:
     movq  mm0, [r2+r3*0+0]
@@ -880,6 +881,7 @@ cglobal put_vp8_pixels16_mmx, 5,5
     sub   r4d, 2
     jg .nextrow
     REP_RET
+%endif
 
 cglobal put_vp8_pixels16_sse, 5,5,2
 .nextrow:
@@ -973,6 +975,7 @@ cglobal vp8_idct_dc_add_sse4, 3, 3, 6
 ; void vp8_idct_dc_add4y_<opt>(uint8_t *dst, DCTELEM block[4][16], int stride);
 ;-----------------------------------------------------------------------------
 
+%if ARCH_X86_32
 INIT_MMX
 cglobal vp8_idct_dc_add4y_mmx, 3, 3
     ; load data
@@ -1007,6 +1010,7 @@ cglobal vp8_idct_dc_add4y_mmx, 3, 3
     ADD_DC    m0, m6, 0, mova
     ADD_DC    m1, m7, 8, mova
     RET
+%endif
 
 INIT_XMM
 cglobal vp8_idct_dc_add4y_sse2, 3, 3, 6
@@ -1152,7 +1156,9 @@ cglobal vp8_idct_add_%1, 3, 3
     RET
 %endmacro
 
+%if ARCH_X86_32
 VP8_IDCT_ADD mmx
+%endif
 VP8_IDCT_ADD sse
 
 ;-----------------------------------------------------------------------------
@@ -1217,7 +1223,9 @@ cglobal vp8_luma_dc_wht_%1, 2,3
 %endmacro
 
 INIT_MMX
+%if ARCH_X86_32
 VP8_DC_WHT mmx
+%endif
 VP8_DC_WHT sse
 
 ;-----------------------------------------------------------------------------
@@ -1610,6 +1618,7 @@ cglobal vp8_%2_loop_filter_simple_%1, 3, %3, %4
 %endif
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 SIMPLE_LOOPFILTER mmx,    v, 4, 0
@@ -1617,6 +1626,8 @@ SIMPLE_LOOPFILTER mmx,    h, 5, 0
 %define SPLATB_REG SPLATB_REG_MMXEXT
 SIMPLE_LOOPFILTER mmxext, v, 4, 0
 SIMPLE_LOOPFILTER mmxext, h, 5, 0
+%endif
+
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2
 %define WRITE_8W   WRITE_8W_SSE2
@@ -2118,6 +2129,7 @@ cglobal vp8_%2_loop_filter16y_inner_%1, 5, %3, %5
     RET
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 INNER_LOOPFILTER mmx,    v, 6, 16, 0
@@ -2130,6 +2142,7 @@ INNER_LOOPFILTER mmxext, v, 6, 16, 0
 INNER_LOOPFILTER mmxext, h, 6, 16, 0
 INNER_LOOPFILTER mmxext, v, 6,  8, 0
 INNER_LOOPFILTER mmxext, h, 6,  8, 0
+%endif
 
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2
@@ -2814,6 +2827,7 @@ cglobal vp8_%2_loop_filter16y_mbedge_%1, 5, %3, %5
     RET
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 MBEDGE_LOOPFILTER mmx,    v, 6, 16, 0
@@ -2826,6 +2840,7 @@ MBEDGE_LOOPFILTER mmxext, v, 6, 16, 0
 MBEDGE_LOOPFILTER mmxext, h, 6, 16, 0
 MBEDGE_LOOPFILTER mmxext, v, 6,  8, 0
 MBEDGE_LOOPFILTER mmxext, h, 6,  8, 0
+%endif
 
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2